United Nations Conference on Trade and Development

WorldInvestmentReport

United NationsNew York and Geneva, 2005

2005 Transnational Corporations andthe Internationalization of R&D

ii

NOTE

As the focal point in the United Nations system for investment and technology, and building on30 years of experience in these areas, UNCTAD, through its Division on Investment, Technology andEnterprise Development (DITE), promotes the understanding, and helps build consensus, on matters relatedto foreign direct investment (FDI), transfer of technology and development. DITE also assists developingcountries to attract and benefit from FDI and in building their productive capacities and internationalcompetitiveness. The emphasis is on an integrated policy approach to investment, technological capacitybuilding and enterprise development.

The term “country” as used in this study also refers, as appropriate, to territories or areas; thedesignations employed and the presentation of the material do not imply the expression of any opinionwhatsoever on the part of the Secretariat of the United Nations concerning the legal status of any country,territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.In addition, the designations of country groups are intended solely for statistical or analytical convenienceand do not necessarily express a judgement about the stage of development reached by a particular countryor area in the development process. The reference to a company and its activities should not be construedas an endorsement by UNCTAD of the company or its activities.

The boundaries and names shown and designations used on the maps presented in this publicationdo not imply official endorsement or acceptance by the United Nations.

The following symbols have been used in the tables:

Two dots (..) indicate that data are not available or are not separately reported. Rows in tables have beenomitted in those cases where no data are available for any of the elements in the row;

A dash (-) indicates that the item is equal to zero or its value is negligible;

A blank in a table indicates that the item is not applicable, unless otherwise indicated;

A slash (/) between dates representing years, e.g., 1994/95, indicates a financial year;

Use of a hyphen (-) between dates representing years, e.g., 1994-1995, signifies the full period involved,including the beginning and end years;

Reference to “dollars” ($) means United States dollars, unless otherwise indicated;

Annual rates of growth or change, unless otherwise stated, refer to annual compound rates;

Details and percentages in tables do not necessarily add to totals because of rounding.

The material contained in this study may be freely quoted with appropriate acknowledgement.

UNITED NATIONS PUBLICATION

Sales No. E.05.II.D.10

ISBN 92-1-112667-3

Copyright © United Nations, 2005All rights reserved

Manufactured in Switzerland

iii

This Report is dedicated to thememory of Sanjaya Lall

PREFACE

Kofi A. AnnanNew York, July 2005 Secretary-General of the United Nations

The globalization of production is reshaping the international economic landscape. With that,the conventional wisdom of developed countries as capital and technology exporters and developingcountries as importers is gradually giving way to a more complex set of relationships. The geographyof international investment flows is changing. Developing countries are emerging as outward investors,and their importance as recipients of foreign direct investment in more knowledge-intensive activitiesis increasing. The World Investment Report 2005, focusing on the internationalization of researchand development by transnational corporations, illustrates some of these changes.

As global competition intensifies, transnational corporations are internationalizing even themost knowledge-intensive corporate functions, such as research and development. Until recently, thistrend was limited almost exclusively to developed countries. Today, corporations in industries suchas automobiles, electronics, biotechnology and pharmaceuticals are establishing research and developmentfacilities in selected developing countries. They do this to enhance their efficiency, to access expandingpools of scientists and engineers, and to meet the demands of increasingly sophisticated markets inthese countries.

These recent trends have important implications for the international division of labour. Thetraditional view, of more complex production activities being undertaken in the North and simplerones in the South, is less and less a true reflection of the reality. Firms now view parts of the developingworld as key sources not only of cheap labour, but also of growth, skills and even new technologies.As transnational corporations are the dominant players in the creation of new technologies, it matterswhere they undertake their research and development. Currently, only a few developing countriesattract such activities on a significant scale. Most low-income countries are not participating in globalresearch and development networks, and consequently do not reap the benefits that they can generate.

The internationalization of research and development by transnational corporations has importantimplications for policy-making. The World Investment Report 2005 stresses the need for coherentnational policies – particularly in the areas of science, technology and innovation, education andinvestment – to ensure greater benefits from this evolution. For many countries, however, this is adaunting task, which will necessitate the full support of the international community.

v

vi

ACKNOWLEDGEMENTS

The World Investment Report 2005 (WIR05) was prepared under the overall guidance of KarlP. Sauvant by a team led by Anne Miroux and comprising Diana Barrowclough, Harnik Deol, PersephoneEconomou, Torbjörn Fredriksson, Masataka Fujita, Masayo Ishikawa, Kálmán Kalotay, Dong Jae Lee,Guoyong Liang, Padma Mallampally, Nicole Moussa, Abraham Negash, Hilary Nwokeabia, Shin Ohinata,Jean-François Outreville and James Xiaoning Zhan. Specific inputs were prepared by Victoria Aranda,Americo Beviglia Zampetti, Kumi Endo, Hamed El-Kady, Anna Joubin-Bret, Victor Konde, MichaelLim, Helge Müller, Thomas Pollan, Prasada Reddy, Christoph Spennemann, Joerg Weber and KeeHwee Wee.

Principal research assistance was provided by Mohamed Chiraz Baly, Bradley Boicourt, JovanLicina, Lizanne Martinez and Tadelle Taye. Additional research assistance was provided by ClaudiaCardenas under the supervision of Henri Laurencin, and by Amare Bekele and Karen Lee. Katrin Arnold,Arnaud Guerreiro and Fennie Lansberger assisted as interns at various stages. The production of theWIR05 was carried out by Christopher Corbet and Esther Valdivia-Fyfe. WIR05 was desktop publishedby Teresita Sabico. It was edited by Michael Gordy and Praveen Bhalla.

Sanjaya Lall was the principal consultant. John H. Dunning was the senior economic adviser.

WIR05 benefited from inputs provided by participants in a Global Seminar in Geneva in May2005, and two regional seminars on FDI in research and development held in April 2005, one inMonterrey in cooperation with Texas A&M International University and Escuela de Graduados enAdministración y Dirección de Empresas (EGADE) at Tecnológico de Monterrey, and the other inBangkok in association with the ASEAN Secretariat. The former was organized by Tagi Sagafi-nejadand Alejandro Ibarra.

Inputs were also received from Rory Allan, Frank Barry, Nazha Benabbes-Taarji, John Daniels,Dieter Ernst, Vishwas Govitrikar, Robert Grosse, Yao-su Hu, Thomas Jost, Ruslan Lukach, MartinMolinuevo, Francisco Moris, Peter Muchlinski, Glenda Napier, Lisa Rydén and Martin Srholec.

Comments were received during various stages of preparation from Ismael Aguilar, HalehDaneshvar Alavi, Giovanni Balcet, Ricardo Bielshowsky, Peter Brimble, Mario Calderini, CristinaCasanueva-Reguart, Mario Cimoli, Csilla Endrödy, Elisa Cobas-Flores, Martha Corrales, RobertoEchandi, Fabienne Fortanier, Samuel Gayi, Andrea Goldstein, William C. Gruben, Miguel Giudicatti,Mongi Hamdi, Fabrice Hatem, Robert Hawkins, Gábor Hunya, Patarapong Intarakumnerd, JoachimKarl, Yves Kenfack, Tivadar Lippényi, Robert Lipsey, Henry Loewendahl, Jeffrey Lowe, GustavoLugones, Aimable Uwizeye Mapendano, Mina Mashayekhi, Riad Meddeb, Wolf R. Meier-Ewart, MichaelMortimore, Fiorina Mugione, Rajneesh Narula, Peter Nunnenkamp, Herbert Oberhänsli, Sheila Page,Gloria O. Pasadilla, Robert Pearce, Lucia Piscitello, Bruno von Pottelsberghe de la Potterie, AleksandraPrachowska, Sérgio Queiroz, Eric Ramstetter, Rajah Rasiah, Marie-Estelle Rey, Matfobhi Riba, PedroRoffe, Martin Roy, Reg Rumney, Pierre Sauvé, Carlos Scheel, Jon Sigurdson, Djisman S. Simandjuntak,Maurizio Sobrero, Shigeki Tejima, Jaroslav Tláskal, Douglas Thomas, Yasuyuki Todo, Mun Heng Toh,Elisabeth Tuerk, Sophia Twarog, Rob van Tulder, Christopher Wilkie, Maximilian von Zedwitz andZbigniew Zimny. Comments were also received from the United Nations Economic Commission forAfrica and the United Nations Economic and Social Commission for Western Asia.

Numerous officials of central banks, statistical offices, investment promotion and other governmentagencies, and officials of international organizations and non-governmental organizations, as wellas executives of a number of companies, also contributed to WIR05, especially through the provisionof data and other information. The Report also benefited from collaboration with Erasmus University,Rotterdam on data collection and analysis of the largest TNCs.

The financial support of the Governments of Norway and Sweden is gratefully acknowledged.

TABLE OF CONTENTSPage

PREFACE ..................................................................................................................................... v

ACKNOWLEDGEMENTS...................................................................................................... vi

OVERVIEW ..............................................................................................................................xix

��������������������� ��

CHAPTER I. GLOBAL TRENDS: FDI FLOWS RESUME GROWTH..................... 3

A. Signs of recovery .............................................................................................................................. 31. Overall analysis ........................................................................................................................................... 4

a. FDI inflows and outflows .................................................................................................................. 4b. Modes of FDI entry ............................................................................................................................ 9c. Components of FDI flows ............................................................................................................... 10d. Factors contributing to the recovery ............................................................................................. 12e. The importance of TNC activities in the world economy .......................................................... 13

2. The largest TNCs ...................................................................................................................................... 15a. The world’s top 100 TNCs ............................................................................................................... 15b. The top 50 TNCs from developing countries ................................................................................ 17c. Transnationality of the top TNCs .................................................................................................... 17d. The top 10 TNCs from South-East Europe and the CIS .............................................................. 19e. The world’s top 50 financial TNCs ................................................................................................. 19

3. FDI performance and potential .............................................................................................................. 20

B. Policy development ........................................................................................................................ 221. National policy changes .......................................................................................................................... 222. International investment agreements ................................................................................................... 23

a. Bilateral investment treaties ............................................................................................................. 24b. Double taxation treaties .................................................................................................................... 28c. Other international agreements ........................................................................................................ 28d. International investment disputes .................................................................................................... 30

C. Prospects: further FDI growth expected ................................................................................ 31

CHAPTER II. REGIONAL TRENDS: DEVELOPING REGIONSLEAD RISE IN FDI ................................................................................................................. 39

Introduction ............................................................................................................................................. 39

A. Developing countries ..................................................................................................................... 401. Africa: FDI inflows remain buoyant, sustained by investments in primary production ......... 40

a. Trends: FDI continues to flow, mostly to natural resources ....................................................... 40b. Policy developments: efforts to stabilize the environment for FDI inflows ............................ 43c. Prospects: cautiously positive .......................................................................................................... 48

2. Asia and Oceania: inflows at a record high ....................................................................................... 50a. Trends: strong growth in FDI flows ............................................................................................... 51b. Policy developments: favourable measures continue ................................................................... 58c. Prospects: increasingly bright .......................................................................................................... 60

3. Latin America and the Caribbean: FDI inflows rebound ................................................................ 62a. Trends: a resurgence of FDI inflows in many countries ............................................................. 62b. Policy developments: some changes in the area of natural resources ....................................... 69c. Prospects: growing opportunities .................................................................................................... 73

viii World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Page

B. South-East Europe and CIS: FDI rises for the fourth year in a row ........................... 741. Trends: FDI inflows sharply up ............................................................................................................ 742. Policy developments: diversity in policy approaches ...................................................................... 773. Prospects: continuing growth ................................................................................................................ 78

C. Developed countries: uneven performance ............................................................................ 801. Trends and developments: a turnaround in many countries ........................................................... 812. Policy developments: diverging tendencies ....................................................................................... 893. Prospects: positive overall ..................................................................................................................... 89

���������������������������������������������

INTRODUCTION..................................................................................................................... 99

CHAPTER III. INNOVATION, R&D AND DEVELOPMENT .................................101

A. Innovation matters for all countries ...................................................................................... 101

B. Global R&D trends ...................................................................................................................... 1051. R&D is geographically concentrated ................................................................................................. 1052. R&D by industry ..................................................................................................................................... 1073. Capability needs and benefits differ across activities ................................................................... 109

C. The innovation capability gap .................................................................................................. 1111. Measuring innovation capabilities ...................................................................................................... 1112. The UNCTAD Innovation Capability Index ..................................................................................... 113

D. Conclusion ....................................................................................................................................... 117

CHAPTER IV. R&D BY TNCS AND DEVELOPING COUNTRIES ...................... 119

A. TNCs are dominant R&D players ........................................................................................... 119

B. R&D by TNCs is internationalizing ....................................................................................... 1211. A growing share of TNCs’ R&D is performed abroad ................................................................... 1222. The growing role of foreign affiliates in host-country R&D ....................................................... 1253. Growing use of strategic alliances ..................................................................................................... 126

C. The emergence of developing economies as locations for TNCs’ R&D ...................... 1261. TNCs are expanding R&D to developing locations ....................................................................... 1272. Foreign affiliates in patenting by developing economies ............................................................. 134

D. Features of R&D undertaken in developing, South-EastEuropean and CIS markets ....................................................................................................... 1351. Industry composition of R&D by TNCs in developing countries ............................................... 1352. Types of R&D .......................................................................................................................................... 137

a. Asia and Oceania: dynamic trends ................................................................................................ 139b. Latin America and the Caribbean: limited R&D but with potential ........................................ 143c. Africa: generally marginal in R&D by TNCs ............................................................................. 147d. A comparison with economies in transition ................................................................................ 148

E. Developing-country TNCs are also expanding R&D abroad.......................................... 150

F. Prospects .......................................................................................................................................... 151

ixTABLE OF CONTENTS

Page

CHAPTER V. DRIVERS AND DETERMINANTS .......................................................157

A. What drives the internationalization of R&D? .................................................................. 157

B. Host-country determinants of R&D location ...................................................................... 160

C. How to internationalize R&D ................................................................................................... 1681. R&D outsourcing is growing ............................................................................................................... 1682. Greenfield versus acquisition .............................................................................................................. 170

Annex to Chapter V. The rise of chip design in Asia: a case study .................................... 1731. Pull factors ............................................................................................................................................... 1732. Policy factors ........................................................................................................................................... 1743. Push factors .............................................................................................................................................. 174

a. Changes in design methodology and organization ..................................................................... 174b. More outsourcing and multiple design interfaces ...................................................................... 175c. Changing skill requirements .......................................................................................................... 176

4. Enabling factors ...................................................................................................................................... 176

CHAPTER VI. DEVELOPMENT IMPLICATIONS ....................................................179

A. New development opportunities in the making .................................................................. 179

B. Implications for host countries ................................................................................................ 1811. Effects on the structure and performance of an NIS ...................................................................... 1812. Human resource implications ............................................................................................................... 1843. Knowledge spillovers from R&D by TNCs ...................................................................................... 1854. Contributions to industrial upgrading ............................................................................................... 1895. Potential concerns related to R&D internationalization ............................................................... 190

C. Implications for home countries .............................................................................................. 1931. Improved overall R&D efficiency ...................................................................................................... 1932. Reverse technology transfer implications ......................................................................................... 1943. Market expansion implications ............................................................................................................ 1944. Home country concerns ......................................................................................................................... 195

D. Conclusions ..................................................................................................................................... 197

CHAPTER VII. THE ROLE OF NATIONAL POLICIES .........................................201

A. Coherent policies and institutions make a difference ...................................................... 201

B. Strengthening the institutional framework for innovation ............................................ 2031. Fostering human resources ................................................................................................................... 203

a. Development of skilled human resources .................................................................................... 203b. Importing human resources ............................................................................................................ 204

2. The role of research capabilities in the public sector ................................................................... 2063. Policies related to intellectual property ............................................................................................ 2094. Competition policy and innovation .................................................................................................... 210

C. Promotion of R&D-related FDI ............................................................................................... 2121. The role of investment promotion agencies ..................................................................................... 2122. Performance requirements .................................................................................................................... 2143. The use of R&D incentives is expanding ......................................................................................... 2164. Using science parks as attractors ........................................................................................................ 218

D. Industry-specific policies to enhance the benefits of FDI in R&D .............................. 219

E. The role of home countries ........................................................................................................ 220

x World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Page

F. Concluding remarks .................................................................................................................... 222

CHAPTER VIII. THE INTERNATIONAL FRAMEWORK ....................................227

A. International investment agreements .................................................................................... 2271. Entry and establishment ........................................................................................................................ 2272. Performance requirements .................................................................................................................... 2293. Incentives .................................................................................................................................................. 2304. Key personnel .......................................................................................................................................... 2315. General protection of FDI in R&D..................................................................................................... 2316. Home-country measures and corporate social responsibility ....................................................... 232

B. International rules relating to IPRs ....................................................................................... 233

C. International cooperation in R&D.......................................................................................... 235

REFERENCES ........................................................................................................................239

SELECTED UNCTAD PUBLICATIONS ON TNCs AND FDI .................................329

QUESTIONNAIRE ................................................................................................................333

Boxes

I.1. Problems with FDI data ................................................................................................................................ 4I.2. Changes in geographical groupings used in WIR05 ................................................................................. 6I.3. FDI prospects: results of UNCTAD’s survey ......................................................................................... 33II.1. Africa: many producer-countries seek to improve policies as well as management of

petroleum industry ....................................................................................................................................... 44II.2. Kenya: UNCTAD’s Investment Policy Review recommends an alternative approach to

minimum capital requirements for FDI inflows ...................................................................................... 45II.3. AGOA Acceleration Act 2004: some new key provisions ..................................................................... 47II.4. Attracting FDI to South Africa through Government development assistance programmes ........... 48II.5. Report of the Commission for Africa: recommendations to help boost investment ......................... 49II.6. FDI inflows to West Asia increased, but remain concentrated ............................................................. 54II.7. Recent trends in FDI inflows in the Islamic Republic of Iran ............................................................. 56II.8. Some changes in national policies on inward FDI in Asia and Oceania in 2004-2005 ................... 58II.9. FTAs and economic partnership agreements between ASEAN or ASEAN member countries

and Japan: implications for FDI ................................................................................................................ 61II.10. MERCOSUR: FDI in the automobile industry is targeting broader export markets ........................ 68II.11. Can the apparel industry in Central America and the Caribbean compete with Asia for the

United States market? .................................................................................................................................. 70II.12. The need to weigh the costs and benefits of incentives to FDI: the experience of Argentina ........ 70II.13. China’s new investment interest in Latin America ................................................................................. 74II.14. What lies behind the negative FDI inflows to Germany in 2004? ...................................................... 85II.15. EU accession and its impact on FDI in the new member countries .................................................... 86III.1. Definition of R&D ..................................................................................................................................... 103III.2. Different ways of internationalizing innovation ................................................................................... 104III.3. Services-sector R&D in the United States ............................................................................................. 110III.4. Comparing the UNCTAD Innovation Capability Index with other indices ..................................... 112IV.1. Foreign R&D affiliates of German TNCs .............................................................................................. 124IV.2. Explanatory note on the UNCTAD survey on R&D internationalization ......................................... 124IV.3. Intel’s R&D network in developing countries ...................................................................................... 132IV.4. Taxonomy of R&D by foreign affiliates ................................................................................................ 138IV.5. The boom in R&D-related FDI inflows in China ................................................................................. 141IV.6. Motorola’s R&D network ......................................................................................................................... 143IV.7. Thailand in Toyota’s global R&D network ............................................................................................ 145IV.8. Innovative R&D by foreign affiliates in the Republic of Korea: Microsoft, Siemens and

Philips .......................................................................................................................................................... 146IV.9. General Motors in Brazil: from tropicalization to global innovative R&D ..................................... 146IV.10. STMicroelectronics’ design and software centre in Rabat .................................................................. 147

xiTABLE OF CONTENTS

Page

IV.11. R&D by TNCs in agriculture: Kenya ..................................................................................................... 148IV.12. R&D by foreign affiliates in the Czech Republic ................................................................................ 149IV.13. Alexandria Carbon Black: Indian FDI in R&D in Egypt .................................................................... 151V.1. The case for dispersing R&D from a centralized base ........................................................................ 158V.2. Tertiary enrolments by region and country ............................................................................................ 162V.3. IPR regimes and R&D location ............................................................................................................... 164V.4. Why are companies setting up R&D in China? .................................................................................... 166V.5. Why TNCs set up R&D in India .............................................................................................................. 167V.6. From closed to open innovation: the case of IBM ............................................................................... 169V.7. The determinants of the make/buy decision in R&D ........................................................................... 171V.8. Global design networks: the key players ............................................................................................... 175VI.1. Collaboration between foreign affiliates and local universities: selected examples ...................... 183VI.2. Reverse brain drain: the case of the Zhongguancun Science Park in Beijing ................................. 185VI.3. Protecting against the risks of technology leakage .............................................................................. 186VI.4. Asset-seeking foreign affiliates create more local R&D linkages: the case of Italy ...................... 188VI.5. R&D by TNCs and the biomedical science industry in Singapore .................................................... 190VI.6. Clinical trials in India ............................................................................................................................... 193VI.7. Nestlé’s R&D centre in Singapore .......................................................................................................... 195VI.8. Mobile telecommunications R&D by TNCs in China .......................................................................... 196VII.1. Engaging foreign affiliates in training: the Singapore case ............................................................... 205VII.2. Policies in the Republic of Korea to attract back scientists in the diaspora.................................... 206VII.3. Spurring innovation in Taiwan Province of China ............................................................................... 207VII.4. Government-supported research institutes in the Republic of Korea ................................................ 208VII.5. The IPA’s role in the Czech NIS .............................................................................................................. 213VII.6. Enhancing the benefits from R&D-related FDI: the case of Ireland ................................................. 214VII.7. Types of R&D incentives .......................................................................................................................... 217VII.8. The role of local governments in building domestic capabilities: the case of Shanghai ............... 221VIII.1. The OECD Guidelines for Multinational Enterprises .......................................................................... 232VIII.2. TRIPS minimum IPR standards of relevance to FDI in R&D and TRIPS flexibilities of

relevance to host-country R&D ............................................................................................................... 233VIII.3. TRIPS-plus provisions of potential relevance to FDI in R&D and local R&D............................... 235VIII.4. The promotion of R&D investment in regional agreements ............................................................... 236

Figures

I.1. FDI inflows, global and by group of economies, 1980-2004 ................................................................. 3I.2. Total resource flows to developing countries, by type of flow, 1990-2003 ......................................... 7I.3. FDI outflows from developing economies, and South-East Europe and CIS,

by group of economies, 1984-2004 ............................................................................................................. 8I.4. Share of different financing components in world FDI inflows, 1995-2004 ..................................... 10I.5. Growth rates of world FDI inflows and GDP, 1980-2004 .................................................................... 13I.6. Transnationality Index of host countries, by group of economies, 1998-2002 ................................. 15I.7. Transnationality Index of host economies, 2002 .................................................................................... 16I.8. Average TNI of the 100 largest TNCs in the world and the 50 largest TNCs

from developing countries, 1993-2003 ..................................................................................................... 19I.9. Distribution of foreign affiliates of the 50 largest financial TNCs, 2003 .......................................... 20I.10. Largest gains and losses in inward FDI performance, 2003-2004 ....................................................... 22I.11. Number of BITs and DTTs concluded, cumulative and annual, 1990-2004 ....................................... 27I.12. Total BITs concluded, by country group, end 2004 ............................................................................... 27I.13. Top 10 signatories of BITs, end 2004 ....................................................................................................... 28I.14. Top 10 signatories of DTTs, end 2004 ..................................................................................................... 28I.15. Total DTTs concluded, by country group, end 2004 .............................................................................. 29I.16. The growth of international investment agreements other than BITs and DTTs, 1957-2004 ......... 29II.1. FDI flows by region, 2003, 2004 .............................................................................................................. 39II.2. Africa: FDI inflows and their share in gross fixed capital formation, 1985-2004 ........................... 40II.3. Africa: FDI flows, top 10 economies, 2003, 2004 ................................................................................. 41II.4. Share of petroleum in FDI inflows to four major African countries, 2004 ........................................ 42II.5. Africa: BITs and DTTs concluded, cumulative and annual, 1990–2004 ............................................ 46II.6. Africa: prospects for FDI inflows, 2005-2006 ........................................................................................ 50II.7. Asia and Oceania: FDI inflows and their share in gross fixed

capital formation, 1985-2004 ..................................................................................................................... 51II.8. Asia and Oceania: FDI flows, top 10 economies, 2003, 2004 ............................................................. 52II.9. Top 10 economies in terms of cross-border M&A sales in Asia and Oceania, 2003, 2004 ............. 52

xii World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Page

II.10. Asia and Oceania: number of BITs and DTTs concluded, cumulative and annual, 1990-2004 ...... 60II.11. Asia and Oceania: prospects for FDI inflows, 2005-2006 .................................................................... 60II.12. Latin America and the Caribbean: FDI inflows and their share in gross fixed capital

formation, 1985-2004 .................................................................................................................................. 63II.13. Latin America and the Caribbean: FDI flows, top 10 economies, 2003, 2004 .................................. 64II.14. FDI inflows by sector in Brazil, 1996-2004 ............................................................................................ 65II.15. FDI inflows by sector in Argentina, 1996-2003 ..................................................................................... 66II.16. FDI inflows by sector in Mexico, 1996-2004 ......................................................................................... 66II.17. Automotive industry in Argentina and Brazil: production, domestic sales, exports and

imports, 1992-2004 ...................................................................................................................................... 67II.18. Maquila industry in Mexico, 1997-2004 ................................................................................................. 68II.19. United States imports of apparel and textile products from selected countries and regions,

1997-2004 ...................................................................................................................................................... 69II.20. Latin America and the Caribbean: number of BITs and DTTs concluded, cumulative and

annual, 1990-2004 ........................................................................................................................................ 72II.21. Latin America and the Caribbean: prospects for FDI inflows, 2005-2006 ........................................ 73II.22. South-East Europe and CIS: FDI inflows and their share in gross fixed capital formation,

1992-2004 ...................................................................................................................................................... 75II.23. South-East Europe and CIS: FDI flows, top 10 recipients, 2003, 2004 ............................................. 76II.24. South-East Europe and CIS: number of BITs and DTTs concluded, cumulative and annual,

1990-2004 ...................................................................................................................................................... 79II.25. The wage ladder: gross pay per annum in selected economies, 2004 ................................................. 80II.26. South-East Europe and CIS: prospects for FDI inflows, 2005-2006 .................................................. 81II.27. Developed countries: FDI inflows and their share in gross fixed capital formation,

1985-2004 ...................................................................................................................................................... 82II.28. Developed countries: FDI flows, top 10 economies, 2003, 2004 ........................................................ 83II.29. Current-account balance, net balance of FDI flows and net balance of portfolio flows in the

United States, 1990-2004 ............................................................................................................................ 84II.30. Developed countries: number of BITs and DTTs concluded, cumulative and annual,

1990-2004 ...................................................................................................................................................... 90II.31. Developed countries: prospects for FDI inflows, 2005-2006 ............................................................... 91III.1. Stages of technology development by innovation effort ..................................................................... 102III.2. Gross expenditure on R&D (GERD) and business enterprise R&D (BERD), by country

group, 1996 and 2002 ................................................................................................................................ 107III.3. Share of enterprise R&D in total R&D by country/region, 1996 and 2002 ..................................... 108III.4. Relationship between the UNCTAD Innovation Capability Index and log per capita

GDP, 2001 ................................................................................................................................................... 115IV.1. R&D expenditure by selected TNCs and economies, 2002 ................................................................ 120IV.2. R&D expenditure by Japanese foreign affiliates abroad and its share in the total R&D

spending of Japanese TNCs, 1986-2002 ................................................................................................ 123IV.3. Degree of R&D internationalization by home region or country in the UNCTAD survey,

2004-2005 .................................................................................................................................................... 125IV.4. Degree of R&D internationalization by industry, 2004-2005 ............................................................ 125IV.5. R&D expenditure by foreign affiliates, based on a sample of 30 economies, value and share

in business R&D, 1993-2002 ................................................................................................................... 126IV.6. Trends in R&D spending by foreign affiliates, selected economies, 1995-2003 ............................ 127IV.7. Worldwide locations of majority-owned foreign affiliates engaged in R&D, 2004 ....................... 128IV.8. Current foreign locations of R&D in the UNCTAD survey, 2004 ..................................................... 133IV.9. Industry composition of R&D by majority-owned foreign affiliates of United States TNCs,

2002 .............................................................................................................................................................. 137IV.10. Prospects of TNCs locating R&D abroad, 2005-2009 ......................................................................... 152IV.11. Most attractive prospective R&D locations in the UNCTAD survey, 2005-2009 ........................... 153VI.1. National innovation systems and FDI in R&D: a schematic diagram ............................................... 181VII.1. National innovation systems and FDI in R&D: the policy dimension .............................................. 202VIII.1. Schedules with commitments on commercial presence in R&D services ......................................... 228VIII.2. Level of commitments under commercial presence for R&D activities ........................................... 228VIII.3. Level of commitments of R&D services, by group of countries ........................................................ 229

Tables

I.1. Cross-border M&As with values of over $1 billion, 1987-2004 ........................................................... 9I.2. FDI inflows to the top 20 economies, ranked by size of different financing components,

2003 ................................................................................................................................................................ 12

xiiiTABLE OF CONTENTS

Page

I.3. Selected indicators of FDI and international production, 1982-2004 ................................................. 14I.4. Snapshot of the world’s 100 largest TNCs: assets, sales and employment, 2002, 2003 .................. 17I.5. Snapshot of the top 50 TNCs from developing countries: assets, sales and employment,

2002, 2003 ..................................................................................................................................................... 18I.6. Average TNI values for the world’s largest TNCs, 2002, 2003 ........................................................... 18I.7. Measures of efficiency and productivity of the world’s top 100 and

developing countries’ top50 TNCs, 2002, 2003 ...................................................................................... 19I.8. Snapshot of the top 10 TNCs from SEE and CIS: assets, sales and

employment, 2002, 2003 ............................................................................................................................. 20I.9. Inward FDI Performance Index, by region, 1990, 2003, 2004 ............................................................. 21I.10. Rankings by the Inward FDI Performance Index, 2004 ........................................................................ 23I.11. Top 25 economies by the Inward FDI Potential Index, 1990, 2002, 2003 ......................................... 24I.12. Matrix of inward FDI performance and potential, 2003 ....................................................................... 25I.13. Outward FDI Performance Index for the 20 leading investor economies,

1990, 2003, 2004 .......................................................................................................................................... 25I.14. National regulatory changes, 1991-2004 ................................................................................................. 26I.15. Changes in corporate income tax rates in selected economies, 2004 .................................................. 26II.1. Africa: country distribution of FDI inflows, by range, 2003, 2004 .................................................... 42II.2. Africa: distribution of cross-border M&A sales, by sector and industry, 2003, 2004 ..................... 43II.3. Asia and Oceania: country distribution of FDI inflows, by range, 2003, 2004 ................................ 53II.4. Asia and Oceania: distribution of cross-border M&A sales, by sector and industry,

2003, 2004 ..................................................................................................................................................... 53II.5. Latin America and the Caribbean: country distribution of FDI inflows, by range, 2003, 2004 ..... 64II.6. Latin America and the Caribbean: distribution of cross-border M&A sales, by sector and

industry, 2003, 2004 .................................................................................................................................... 65II.7. South-East Europe and CIS: country distribution of FDI inflows, by range, 2003, 2004 ............... 76II.8. South-East Europe and CIS: distribution of cross-border M&A sales, by sector and industry,

2003, 2004 ..................................................................................................................................................... 77II.9. Developed countries: country distribution of FDI inflows, by range, 2003, 2004 ........................... 82II.10. Developed countries: distribution of cross-border M&A sales, by sector and industry,

2003, 2004 ..................................................................................................................................................... 88II.11. Expected repatriation of profits from United States affiliates abroad to their parents,

selected TNCs, 2005 .................................................................................................................................... 90III.1. The 10 leading economies in R&D and business R&D spending, 1996 and 2002 ......................... 105III.2. Developing economies, South-East Europe and CIS: distribution of R&D, by region .................. 107III.3. Classification of manufacturing industries by R&D intensity ........................................................... 108III.4. Components of the UNCTAD Innovation Capability Index ............................................................... 113III.5. The UNCTAD Innovation Capability Index .......................................................................................... 114III.6. Regional unweighted averages for the UNCTAD Innovation Capability Index .............................. 115IV.1. The top 20 firms, by R&D expenditure in the world and in developing economies,

South-East Europe and CIS, 2003 ........................................................................................................... 120IV.2. Home economies of the 700 largest R&D spending firms of the world, 2003 ................................ 121IV.3. Industry breakdown of the 700 largest R&D performing firms, 2003 ............................................. 121IV.4. Global employment, R&D employment, and R&D expenditures of United States TNCs, by

domestic and overseas components, 1994, 1999, 2002 ....................................................................... 122IV.5. R&D expenditures of the 20 largest Swedish TNCs, 1995-2003 ....................................................... 123IV.6. R&D expenditure abroad by majority-owned foreign affiliates of United States parent

companies, by selected region/country, 1994-2002 ............................................................................. 129IV.7. R&D employment by majority-owned foreign affiliates of United States TNCs by

region/economy, 1999 ............................................................................................................................... 131IV.8. R&D bases by Japanese manufacturing companies, by host region, 2000-2004 ............................. 131IV.9. Greenfield FDI projects in R&D, 2002-2004 ........................................................................................ 132IV.10. United States Patent and Trademark Office patents granted to residents of selected

developing economies and countries in South-East Europe and CIS, 2001-2003 .......................... 135IV.11. United States Patent and Trademark Office patents assigned to institutions in selected

economies by the type of assignee, 2001-2003 ..................................................................................... 136V.1. Annual cost of employing a chip design engineer, 2002..................................................................... 174VI.1. Potential implications of R&D internationalization by TNCs ............................................................ 180VII.1. Do IPAs actively target FDI in R&D? .................................................................................................... 213VII.2. Policies and policy tools used by IPAs promoting FDI in R&D........................................................ 215

xiv World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Page

Box figures

I.3.1. Prospects for global FDI flows: responses of TNC, expert and IPAs, 2005-2006 ............................ 33I.3.2. Most attractive global business locations: responses of experts and TNCs ....................................... 34I.3.3. Major risks to global FDI flows, 2005-2006 ........................................................................................... 35I.3.4. Investment policy measures to attract FDI: responses by IPAs ........................................................... 35II.5.1. Africa: ODA inflows, 1980-2003 .............................................................................................................. 49II.6.1. Industry distribution of numbers of greenfield investment projects and cross-border M&A

deals in West Asia, 2002-2004 ................................................................................................................... 55II.7.1. FDI inflows to the Islamic Republic of Iran and its share in total inflows to Asia and

Oceania, 1993-2004 ..................................................................................................................................... 56II.7.2. Number and value of foreign investments approved under the foreign investment laws of

1955 and 2002 in the Islamic Republic of Iran, 1993-2004 ................................................................. 57

xvTABLE OF CONTENTS

Page

A.I.16. International investment agreements (other than BITs and DTTs) undernegotiation or consultation, as of end 2004 .......................................................................................... 278

A.II.1. Cross-border M&A deals with values of over $100 million concluded indeveloping and transition economies, 2004 .......................................................................................... 279

A.II.2. West Asia: selected FDI-related liberalization, 2004 ........................................................................... 282A.II.3. New projects announced by TNCs in the non-oil mining and oil and

gas industries in Latin America, January 2004-May 2005 .................................................................. 283A.II.4. New projects announced by TNCs in the automobile industry in Argentina,

Brazil and Mexico, January 2004-May 2005 ........................................................................................ 284A.II.5. Industry composition of FDI inflows in selected South-East European

countries and CIS, 2003-2004 ................................................................................................................. 285A.III.1. Illustrative matrix of technological and organizational capabilities within firms .......................... 286A.III.2. Gross domestic expenditure in R&D (GERD) and business

enterprise R&D (BERD), 1991-2003 ...................................................................................................... 287A.III.3. Patent applications from developing countries and South-East Europe

and CIS in the United States, by residence of investor, 1991-2003 .................................................. 289A.III.4. Technological Activity Index ................................................................................................................... 290A.III.5. The Human Capital Index ......................................................................................................................... 291A.IV.1. R&D expenditure by foreign affiliates in selected economies, 1993-2003 ...................................... 292A.IV.2. R&D performed by foreign affiliates of United States TNCs by country

and NAICS industry, 2002........................................................................................................................ 294A.IV.3. Selected cases of R&D by foreign TNCs in Singapore, March 2005 ............................................... 295A.V.1. Number of students enrolled at the tertiary level in all subjects and

technical subjects (science, engineering, mathematics, computing), 2000-2001 ............................ 296

Annex B: Statistical annex

DEFINITIONS AND SOURCES ........................................................................................297

A. General definitions ....................................................................................................................... 2971. Transnational corporations .................................................................................................. 2972. Foreign direct investment .................................................................................................... 2973. Non-equity forms of investment ........................................................................................ 298

B. Availability, limitations and estimates of FDI data presented in WIR ....................... 2981. FDI flows ................................................................................................................................. 2982. FDI stocks ............................................................................................................................... 2993. Special notes on recent changes in the methodology ................................................... 299

C. Data revisions and updates ....................................................................................................... 301

D. Data verification ........................................................................................................................... 301

E. Definitions and sources of the data in annex table B.3 ................................................... 301

F. Definitions and sources of the data on cross-border M&Asin annex tables B.4-B.5 ............................................................................................................... 301

Annex tables

B.1. FDI flows, by region and economy, 2002-2004 .................................................................................... 303B.2. FDI inward stock, by region and economy, 1990, 2000, 2004 ........................................................... 308B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI

stocks as a percentage of gross domestic product (GDP), 1990, 2000, 2004, by regionand economy ............................................................................................................................................... 313

B.4. Cross-border M&As, by region/economy of seller/purchaser, 2002-2004 ....................................... 325B.5. Cross-border M&As, by sector/industry, 2002-2004

AGOA African Growth and Opportunity Act (of the United States)ASEAN Association of South-East Asian NationsBERD business enterprise research and developmentBIT bilateral investment treatyCARICOM Caribbean CommunityCEMAC Communauté économique et monétaire d'Afrique centrale (Economic and

Monetary Community of Central Africa)CIS Commonwealth of Independent StatesCOMESA Common Market for Eastern and Southern AfricaDTT double taxation treatyECOWAS Economic Community of West African StatesEFTA European Free Trade AssociationEIU Economist Intelligence UnitEPA economic partnership agreementESCWA Economic and Social Commission for Western AsiaEU European UnionFDI foreign direct investmentFTA free trade agreementFTAA Free Trade Area of the AmericasGATS General Agreement on Trade in Services (WTO Agreement)GDP gross domestic productGERD gross expenditure on research and developmentICSID International Centre for Settlement of Investment DisputesICT information and communications technologyIIA international investment agreementIPA investment promotion agencyIPR intellectual property rightISDS investor-State dispute settlementLDC least developed countryM&A merger and acquisitionMERCOSUR Mercado Común del Sur (Southern Common Market)MFN most favoured nationMSTQ metrology, standards, testing and quality (system)NAFTA North American Free Trade AgreementNAICS North American Industry Classification SystemNEPAD New Partnership for Africa's DevelopmentNIE newly industrializing economyNIS national innovation systemNSB National Science Board (United States)OBM own brand manufactureODM own design manufactureOEM original equipment manufactureR&D research and developmentRTA regional trade agreementSACU Southern African Customs UnionSADC Southern African Development CommunitySAARC South Asian Association for Regional CooperationSCM subsidies and countervailing measures (also a Uruguay Round Agreement)SEE South-East EuropeSIC Standard Industrial ClassificationTNC transnational corporationTRIPS Trade-related Aspects of Intellectual Property Rights (WTO Agreement)UNDP United Nations Development ProgrammeUNICI UNCTAD Innovation Capability IndexUNIDO United Nations Industrial Development OrganizationUSPTO United States Patent and Trademark OfficeWTO World Trade Organization

�����������

Led by developing countries, globalFDI flows resumed growth in 2004…

On account of a strong increase in foreigndirect investment (FDI) flows to developingcountries, 2004 saw a slight rebound in globalFDI after three years of declining flows. At $648billion, world FDI inflows were 2% higher in2004 than in 2003. Inflows to developingcountries surged by 40%, to $233 billion, butdeveloped countries as a group experienced a14% drop in their inward FDI. As a result, theshare of developing countries in world FDIinflows was 36%, the highest level since 1997.The United States retained its position as thenumber one recipient of FDI, followed by theUnited Kingdom and China.

Many factors help to explain why thegrowth of FDI was particularly pronounced indeveloping countries in 2004. Intense competitivepressures in many industries are leading firmsto explore new ways of improving theircompetitiveness. Some of these ways are byexpanding operations in the fast-growing marketsof emerging economies to boost sales, and byrationalizing production activities with a viewto reaping economies of scale and loweringproduction costs. Higher prices for manycommodities have further stimulated FDI tocountries that are rich in natural resources suchas oil and minerals. In some developed as wellas developing countries, increased inflows in2004 were linked to an upturn in cross-bordermerger and acquisition (M&A) activity.Greenfield FDI continued to rise for the thirdconsecutive year in 2004. Provided economicgrowth is maintained, the prospects for a further

OVERVIEW

END OF THE DOWNTURN

increase in global FDI flows in 2005 arepromising.

FDI outflows increased in 2004 by 18%,to $730 billion, with firms based in developedcountries accounting for the bulk ($637 billion).In fact, almost half of all outward FDI originatedfrom three sources: the United States, the UnitedKingdom and Luxembourg in that order.Developed countries as a group remainedsignificant net capital exporters through FDI; netoutflows exceeded net inflows by $260 billion.While FDI outflows from the European Union(EU) declined by 25%, to $280 billion (a seven-year low), most other developed countriesincreased their investment abroad. In the case ofthe United States, outflows increased by over90%, to $229 billion, a record high.

The stock of FDI in 2004 is estimated at$9 tri l l ion. It is attributed to some 70,000transnational corporations (TNCs) and their690,000 affiliates abroad, with total sales byforeign affiliates amounting to almost $19 trillion.Ranked by foreign assets, General Electric(United States) remained the largest non-financialTNC worldwide, followed by Vodafone (UnitedKingdom) and Ford Motor (United States).Among the top 100 TNCs worldwide, fourcompanies, led by Hutchison Whampoa (HongKong, China), are based in developing economies.

The pace at which the top 100 TNCs areexpanding internationally appears to have sloweddown. Although their sales, employment andassets abroad all rose in absolute terms in 2003,their relative importance declined somewhat asactivities in the home countries expanded faster.Japanese and United States TNCs are generally

xx World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

less transnationalized than their Europeancounterparts. The top 50 TNCs based indeveloping economies, with a shorter history ofoutward expansion, are even lesstransnationalized, but the gap between TNCs fromdeveloped and developing countries is shrinkingin this respect.

International investment in services,particularly financial services, continued to growsteadily, accounting for the bulk of the world FDIstock. The services sector accounted for 63%of the total value of cross-border M&As in 2004,with financial services responsible for one-thirdof the value of cross-border M&As in this sector.For the first time, this year’s WIR ranks the top50 financial TNCs. Large TNCs dominate worldfinancial services, not only in terms of total assetsbut also in terms of the number of countries inwhich they operate. Citigroup (United States) topsthe list, followed by UBS (Switzerland) andAllianz (Germany). Financial TNCs from France,Germany, Japan, the United Kingdom and theUnited States accounted for 74% of the totalassets of the top 50 financial TNCs in 2003.

Low interest rates, higher profits and therecovery of asset prices, principally in developedcountries, contributed to an upturn in M&As,including cross-border M&As; their value shotup by 28% to $381 billion. These transactionsplayed an important part in the continuedrestructuring and consolidation process of manyindustries, especially in the developed world. Thelargest M&A deal in 2004 was the acquisitionof Abbey National (United Kingdom) bySantander Central Hispano (Spain), valued at $16billion. In developing countries, cross-borderM&As accounted for a more modest share ofoverall FDI activity, although firms from thesecountries were increasingly involved in M&As,including some high-profile cases. The upswingin FDI flows to developing countries was mainlyassociated with greenfield investments notablyin Asia. China and India together accounted forabout a half of all new registered greenfield (andexpansion) projects in developing countries in2004.

In terms of the three main forms of FDIfinancing, equity investment dominates at theglobal level. During the past decade, i t hasaccounted for about two-thirds of total FDI flows.The shares of the other two forms of FDI – intra-company loans and reinvested earnings – were

on average 23% and 12% respectively. These twoforms fluctuate widely, reflecting yearlyvariations in profit and dividend repatriations orthe need for loan repayment. There are notabledifferences in the pattern of FDI financingbetween developed and developing countries;reinvested earnings are consistently moreimportant in the latter.

FDI continues to surpass other privatecapital flows to developing countries as well asflows of official development assistance (ODA).In 2004, it accounted for more than half of allresource flows to developing countries and wasconsiderably larger than ODA. However, FDI isconcentrated in a handful of developing countries,while ODA remains the most important sourceof finance in a number of other developingcountries. This is particularly the case for mostleast developed countries (LDCs) even thoughFDI flows have surpassed ODA for individualcountries in that group.

Countries continue to adopt new laws andregulations with a view to making theirinvestment environments more investor friendly.Out of 271 such changes pertaining to FDIintroduced in 2004, 235 involved steps to openup new areas to FDI along with new promotionalmeasures. In addition, more than 20 countrieslowered their corporate income taxes in their bidto attract more FDI. In Latin America and Africa,however, a number of policy changes tended tomake regulations less favourable to foreigninvestment, especially in the area of naturalresources.

At the international level, the number ofbilateral investment treaties (BITs) and doubletaxation treaties (DTTs) reached 2,392 and 2,559respectively in 2004, with developing countriesconcluding more such treaties with otherdeveloping countries. More internationalinvestment agreements were also concluded atthe regional and global level, potentiallycontributing to greater openness towards FDI.The various international agreements aregenerally becoming more and more sophisticatedand complex in content, and investment-relatedprovisions are increasingly introduced intoagreements encompassing a broader range ofissues. There is also a rise in investor-Statedisputes, paralleling the proliferation ofinternational investment agreements.

xxiOVERVIEW

…with the Asia and Oceania regionthe largest recipient as well as sourceof FDI among developing countries.

The upturn in global FDI was marked bysignificant differences between countries andregions. Asia and Oceania was again the topdestination of FDI flows to developing regions.It attracted $148 billion of FDI, $46 billion morethan in 2003, marking the largest increase ever.East Asia saw a 46% increase in inflows, to reach$105 billion, driven largely by a significantincrease in flows to Hong Kong (China). InSouth-East Asia, FDI surged by 48% to $26billion, while South Asia, with India at theforefront, received $7 billion, corresponding toa 30% rise. FDI inflows to West Asia grew evenmore, rising from $6.5 billion to $9.8 billion, ofwhich more than half was concentrated in SaudiArabia, the Syrian Arab Republic and Turkey.China continued to be the largest developing-country recipient with $61 billion in FDI inflows.

The Asia and Oceania region is alsoemerging as an important source of FDI. In 2004the region's outward flows quadrupled to $69billion, due mainly to dramatic growth in FDIfrom Hong Kong (China) but also to increasedinvestments by TNCs from other parts of EastAsia and South-East Asia. Most of theseinvestments are intraregional, taking placeespecially among the economies of East andSouth-East Asia. However, interregionalinvestment from Asian economies also increased.For example, a key driver of Chinese outwardFDI was the growing demand for naturalresources. This has led to significant investmentprojects in Latin America. Indian TNCs alsoinvested large amounts in natural resources inother regions, primarily in African countries andthe Russian Federation. Asian investment indeveloped countries is on the rise as well: thepast year in particular has seen a few sizeableacquisitions of United States and EU firms byChinese and Indian TNCs – such as theacquisition by Lenovo (China) of the personalcomputers division of IBM (United States).

The growth of both inward and outwardFDI flows in Asia and Oceania is being facilitatedby various policy changes at the national andregional levels. For example, the Association ofSoutheast Asian Nations (ASEAN) and Chinasigned an agreement to establish a free trade areaby 2010, and several Asian countries signed freetrade agreements with the United States.

FDI rebounded in Latin Americafollowing four years of decline ...

Following four years of continuous decline,FDI flows to Latin America and the Caribbeanregistered a significant upsurge in 2004, reaching$68 billion – 44% above the level attained in2003. Economic recovery in the region, strongergrowth in the world economy and highercommodity prices were contributing factors.Brazil and Mexico were the largest recipients,with inflows of $18 bill ion and $17 bill ionrespectively. Together with Chile and Argentinathey accounted for two-thirds of all FDI flowsinto the region in 2004. However, FDI inflowsdid not increase in all the countries of LatinAmerica. There were notable declines in Boliviaand Venezuela, mainly linked to uncertaintyregarding legislation related to oil and gasproduction. In Ecuador the completion of thecrude oil pipeline construction explained thedecrease in FDI inflows. A number of countriesmodified their legislation and tax regimes toincrease the State's share in revenues from non-renewable natural resources. It is still too earlyto assess the impact of these changes on thevolume of FDI. Significant projects remain underdevelopment and additional ones were announcedduring 2004.

The sectoral composition of inward FDI toparts of Latin America and the Caribbean appearsto be changing. For several countries of theregion, natural resource and manufacturingindustries became more popular FDI destinationsthan services in 2004. In Argentina, Brazil andMexico, manufacturing attracted more FDI thanservices. FDI in Mexico's maquiladora industrysurged by 26% in response to growing demandin the United States after three consecutive yearsof decline. The completion of most privatizationprogrammes, coupled with financial difficultiesfacing foreign investors in the aftermath of therecent financial crisis and the ensuing economicstagnation in some countries, reduced theattractiveness of the services sector for FDI inLatin America. Firms in that sector suffered themost from the impact of the economic crisis,facing serious problems in reducing their largeforeign-currency liabilities while at the same timebeing unable (owing to the non-tradability of theiractivities) to shift towards export-orientedproduction. In Central America and theCaribbean, however, renewed privatizationactivity made services the largest FDI recipient

xxii World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

sector. In the Andean Community, high oil andmineral prices sustained the position of theprimary sector as the main recipient of FDI flows.

... remained stable in Africa …

FDI flows to Africa remained at almost thesame level – $18 billion – as in 2003. FDI innatural resources was particularly strong,reflecting the high prices of minerals and oil andthe increased profitability of investment in theprimary sector. High and rising prices ofpetroleum, metals and minerals induced TNCsto maintain relatively high levels of investmentin new exploration projects or to escalate existingproduction. Several large cross-border M&Aswere concluded in the mining industry last year.Despite these developments Africa's share in FDIflows worldwide remains low, at 3%.

Angola, Equatorial Guinea, Nigeria, Sudan(all rich in natural resources) and Egypt were thetop recipients, accounting for a little less thanhalf of all inflows to Africa. While FDI inflowsto the last three rose, those to South Africa,another important FDI recipient, fell. LDCs inAfrica received small amounts: around $9 billionin total in 2004. Most investment in Africaoriginated from Europe, led by investors fromFrance, the Netherlands and the United Kingdom,and from South Africa and the United States;together these countries accounted for more thanhalf of the region's inflows. FDI outflows fromAfrica more than doubled in 2004, to $2.8 billion.

A renewed wave of FDI-friendly measuresand initiatives at national and international levelshas sought to facilitate and attract more FDI tothe African continent. At the national level, manymeasures focused on liberalizing legalframeworks and improving the overallenvironment for FDI. However, failure to moverapidly on economic and social policies importantfor attracting and retaining FDI, and a weakemphasis on capacity building, have hamperedthe ability of many countries in the region toattract FDI, in particular in manufacturing. Thusfar, international market-access measures andinitiatives targeting African countries (such asthe United States’ African Growth andOpportunity Act) overall have not been verysuccessful in increasing FDI. In order to realizethe potential for increased FDI and to derivegreater benefits from it , African countriesgenerally need to develop stronger industrial andtechnological capabilities.

The need for international support toAfrica's development has been stressed in severalrecent initiatives. For example, the Commissionfor Africa (established by the United Kingdom)released a report in March 2005 recommendinga substantial increase in aid to Africa: anadditional $25 billion per year to be implementedby 2010. It also proposed several measures thatcould help the continent attract more FDI andenhance its benefits for development. Specificallythe report called for donors to double theirfunding for infrastructure, adopt a 100% externaldebt cancellation, support an Investment ClimateFacility for Africa under the New EconomicPartnership for Africa’s Development (NEPAD)initiative, and create a fund that would provideinsurance to foreign investors in post-conflictcountries in Africa.

… and increased in South-East Europeand the CIS for the fourth consecutiveyear.

FDI inflows to South-East Europe and theCIS, a new group of economies under the UnitedNations reclassification, recorded a fourth yearof growth in 2004, reaching an all-time high of$35 billion. This was the only region to escapethe three-year decline (2001-2003) in world FDIflows, and it maintained robust growth in inwardFDI in 2004 (more than 40%). Trends in inwardFDI to the two subregions have differedsomewhat, however, reflecting the influence ofvarious factors. In South-East Europe, FDIinflows started to grow only in 2003. Led by largeprivatization deals, these inflows nearly tripled,to $11 billion in 2004. In the CIS, inflows grewfrom $5 billion in 2000 to $24 billion in 2004,benefiting largely from the high prices ofpetroleum and natural gas. The RussianFederation is the largest recipient of FDI inflowsin the region.

By contrast, FDI inflows to developedcountries continued to decline.

FDI flows into developed countries, whichnow include the 10 new EU members, fell to $380billion in 2004. The decline was less sharp thanin 2003, possibly suggesting a bottoming out ofthe downward trend that started in 2001. Thedecline pertained to many major host countriesin the developed world. However, there were

xxiiiOVERVIEW

some significant exceptions; the United Statesand the United Kingdom recorded substantialincreases in inflows mainly as a result of cross-border M&As. Meanwhile, investment outflowsfrom developed countries turned upwards againin 2004 to reach $637 billion.

FDI flows into the EU as a whole fell to$216 bill ion – the lowest level since 1998.However, the performance of individual EUmembers varied, with Denmark, Germany, theNetherlands and Sweden registering the mostsignificant declines. To some extent thepersistence of the downward FDI trend in the EUreflected large repayments of intra-company loansand repatriation of earnings in a few members.At the same time, FDI inflows into all the 10 newEU countries increased, attracted by high ratesof economic growth, the availability of skilledhuman resources at competitive costs and reduceduncertainty with regard to the regulatoryframework for FDI following EU accession.Flows into Japan surged by 24% to $8 billion,while those to other developed countries (Israel,New Zealand, Norway and Switzerland) declined.

Further increases in FDI are expected.

Prospects for FDI worldwide appear to befavourable for 2005. For 2006, global FDI flowscan be expected to rise further if economic growthis consolidated and becomes more widespread,corporate restructuring takes hold, profit growthpersists and the pursuit of new markets continues.The continued need of firms to improve theircompetitiveness by expanding into new markets,reducing costs and accessing natural resourcesand strategic assets abroad provides strongincentives for further FDI in developing countriesin particular. Also, the improved profitability ofTNCs is likely to trigger greater M&A activity,which should also push up the levels of FDI indeveloped countries.

Surveys of TNCs, experts and investmentpromotion agencies (IPAs) undertaken byUNCTAD corroborate this relatively optimisticpicture, as do the findings of other recent surveys.In the UNCTAD surveys, more than half of theresponding TNCs as well as experts and four-fifths of the IPAs expected short-term (2005-2006) growth in FDI flows; very few predicteda decline of FDI in the near future. Thecompetitive pressure on firms, continuedoffshoring of services, ongoing liberalization and

the growth of TNCs from emerging markets wereidentified as factors that should lead to more FDI.

At the same time, there are grounds forcaution in forecasting FDI flows. The slowdownof growth in some developed countries, alongwith structural weaknesses and financial andcorporate vulnerabili t ies in some regions,continue to hinder a strong recovery of FDIgrowth. Continuing external imbalances in manycountries and sharp exchange-rate fluctuations,as well as high and volatile commodity prices,pose risks that may hinder global FDI flows.

There is some variation in the FDIprospects of individual regions. In view of theimproved economic situation in Asia and Oceania,its important role as a global production centre,its improved policy environment and significantregional integration efforts, the prospects for FDIflows to that region are strongly positive.According to the TNCs, experts and IPAssurveyed by UNCTAD, the region’s outlook forFDI is bright. FDI inflows to Latin America andthe Caribbean are expected to increase in 2005-2006 as most of the driving forces behind FDIgrowth in 2004 are set to continue. Prospects arealso positive for Africa, partly as a result ofhigher commodity prices and Africa’s naturalresource potential . One out of four TNCrespondents expected that inflows to Africa wouldincrease in 2005-2006, suggesting more cautiousoptimism vis-à-vis this region.

FDI inflows into South-East Europe andthe CIS are expected to grow further in the nearfuture, based on the expectation that theircompetitive wages, in particular in South-EastEurope, could attract an increasing number ofefficiency-seeking or export-oriented projects,while the natural-resource-rich CIS countriescould benefit from continued high oil and gasprices.

Despite the decline in 2004, prospects forrenewed growth in both inward and outward FDIflows for developed countries in 2005 remainpositive, underpinned by forecasts of moderateeconomic growth and a strong pick-up incorporate profits. Already, during the first sixmonths of 2005, cross border M&As in developedcountries increased significantly. For the largestrecipient country – the United States – prospectsfor FDI are good, although the inflows may notreach the high levels recorded in 2004.

xxiv World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

TNCs are internationalizing R&D,including in developing countries…

WIR05 focuses on the internationalizationof research and development (R&D) by TNCs. Thisis not a new phenomenon. When expandinginternationally, firms have always needed to adapttechnologies locally to sell successfully in hostcountries. In many cases, some internationalizationof R&D has been necessary to accomplish this.However, it was traditionally the case that R&Dwas reserved for the home countries of the TNCs.By contrast, now a number of new features areemerging in the internationalization process. Inparticular, for the first time, TNCs are setting upR&D facilities outside developed countries thatgo beyond adaptation for local markets;increasingly, in some developing and South-EastEuropean and CIS countries, TNCs’ R&D istargeting global markets and is integrated into thecore innovation efforts of TNCs.

Consider the following illustrations. Since1993 when Motorola established the first foreign-owned R&D lab in China, the number of foreignR&D units in that country has reached some 700.The Indian R&D activities of General Electric –the largest TNC in the world – employ 2,400people in areas as diverse as aircraft engines,consumer durables and medical equipment.Pharmaceutical companies such as Astra-Zeneca,Eli Lilly, GlaxoSmithKline, Novartis, Pfizer andSanofi-Aventis all run clinical research activitiesin India. From practically nothing in the mid-1990s, the contribution by South-East and EastAsia to global semiconductor design reachedalmost 30% in 2002. STMicroelectronics has someof its semiconductor design done in Rabat,Morocco. General Motors (GM) in Brazil competeswith other GM affiliates in the United States,Europe and Asia for the right to design and buildnew vehicles and carry out other core activitiesfor the global company. There are many suchexamples.

In theory, the internationalization of R&Dinto developing countries is both expected andunexpected. It is expected for two reasons. First,

as TNCs increase their production in developingcountries, some R&D (of the adaptive kind) canbe expected to follow. Second, R&D is a form ofservice activity and like other services (WIR04),it is “fragmenting”, with certain segments beinglocated where they can be performed mostefficiently. Indeed, according to a survey ofEurope’s largest firms conducted in 2004 byUNCTAD and Roland Berger, all service functions– including R&D – are now candidates foroffshoring. It is unexpected in that R&D is aservice activity with very demanding skill ,knowledge and support needs, traditionally metonly in developed countries with strong nationalinnovation systems. Moreover, R&D is taken tobe the least “fragmentable” of economic activitiesbecause it involves knowledge that is strategic tofirms, and because it often requires denseknowledge exchange (much of it tacit) betweenusers and producers within localized clusters.

It is clear that, to date, only a small numberof developing countries and economies in transitionare participating in the process of R&Dinternationalization. However, the fact that someare now perceived as attractive locations for highlycomplex R&D indicates that it is possible forcountries to develop the capabilities that are neededto connect with the global R&D systems of TNCs.From a host-country perspective, R&Dinternationalization opens the door not only forthe transfer of technology created elsewhere, butalso for the technology creation process itself. Thismay enable some host countries to strengthen theirtechnological and innovation capabilities. But itmay also widen the gap with those that fail toconnect with the global innovation network.

…with important implications forinnovation and development.

Innovative activity is essential for economicgrowth and development. Moreover, sustainableeconomic development requires more than simply“opening up” and waiting for new technologiesto flow in. It demands continuous technologicaleffort by domestic enterprises, along withsupportive government policies. With the

R&D INTERNATIONALIZATIONAND DEVELOPMENT

xxvOVERVIEW

increasing knowledge-intensity of production,the need to develop technological capabilitiesis growing. Greater openness to trade and capitalflows does not reduce the imperative of localtechnological effort. On the contrary,liberalization, and the open market environmentassociated with it, have made it necessary forfirms – be they large or small in developed ordeveloping countries – to acquire thetechnological and innovative capabilities neededto become or stay competitive.

R&D is only one source of innovation, butit is an important one. It takes various forms:basic research, applied research and product andprocess development. While basic research ismainly undertaken by the public sector, the othertwo forms are central to the competitiveness ofmany firms. In the early stages of technologicalactivity enterprises do not need formal R&Ddepartments. As they mature, however, they findit increasingly important to monitor, import andimplement new technologies. The role of formalR&D grows as a firm attempts significanttechnological improvements and tackles productor process innovation. For complex and fast-moving technologies it is an essential part of thetechnological learning process.

But the process of acquiring technologicalcapabilities is slow and costly. Technical changeand advanced science-based technologies in manyindustries call for more high-level skills andintense technical effort. These require betterinfrastructure, not least in information andcommunication technologies (ICTs). They alsorequire strong supporting institutions, as well asstable and efficient legal and governance systems.Finally, they require access to the internationalknowledge base, combined with a strategy toleverage this access for the benefit of localinnovation systems. The cumulative forces thatare increasing the gap between countries withrespect to innovation make the role of policyincreasingly important at both the national andinternational levels.

There are large differences in countries’capabilities to innovate and benefit from the R&Dinternationalization process. According to a newmeasure of national innovation capabilities – theUNCTAD Innovation Capability Index – thedifferences appear to be growing over time.Developed countries fall into the high capabilitygroup, as do Taiwan Province of China, the

Republic of Korea and Singapore, along withsome of the economies of South-East Europe andthe CIS. The medium capability group comprisesthe remaining economies in transition, most ofthe resource-rich and newly industrializingeconomies and two sub-Saharan Africaneconomies (Mauritius and South Africa). The lowcapability group contains most of the sub-SaharanAfrican countries as well as several countries inNorth Africa, West Asia and Latin America.Among developing countries, South-East and EastAsia are the leaders in innovation capability,while the position of Latin America and theCaribbean has deteriorated over time and has beenovertaken by North Africa and West Asia.

The innovative capabilities of a country aredirectly relevant to its attractiveness as a hostcountry for R&D by TNCs, as well as to its abilityto benefit from such R&D. The quality of R&Dperformed abroad depends on local capabilitiesof the host country. The same applies to theresulting externalities in terms of how much localfirms and institutions are able to absorb and learnfrom exposure to best practice R&D techniquesand skills. Whether or not R&D deepens overtime, and how far i t spreads over differentactivities, are the result of an interactive processbetween the TNCs and local actors in the hosteconomy, and this process is in turn affected bythe institutional framework and governmentpolicies of the host country.

TNCs are the drivers of global R&D.

Global R&D expenditure has grown rapidlyover the past decade to reach some $677 billionin 2002. It is highly concentrated. The top tencountries by such expenditure, led by the UnitedStates, account for more than four-fifths of theworld total. Only two developing countries (Chinaand the Republic of Korea) feature among the topten. However, the share of developed countriesfell from 97% in 1991 to 91% in 2002, while thatof developing Asia rose from 2% to 6%.Similarly, there has been a rise in innovationoutputs (as measured by the number of patentsissued). For example, between the two timeperiods of 1991-1993 and 2001-2003, the shareof foreign patent applications from developingcountries, South-East Europe and the CIS to theUnited States Patent and Trademark Office,jumped from 7% to 17%.

xxvi World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

TNCs are key players in this process. Aconservative estimate is that they account forclose to half of global R&D expenditures, andat least two-thirds of business R&D expenditures(estimated at $450 billion). These shares areconsiderably higher in a number of individualeconomies. In fact, the R&D spending of somelarge TNCs is higher than that of many countries.Six TNCs (Ford, Pfizer, DaimlerChrysler,Siemens, Toyota and General Motors) spent morethan $5 billion on R&D in 2003. In comparison,among the developing economies, total R&Dspending came close to, or exceeded, $5 billiononly in Brazil, China, the Republic of Korea andTaiwan Province of China. The world’s largestR&D spenders are concentrated in a fewindustries, notably IT hardware, the automotiveindustry, pharmaceuticals and biotechnology.

The R&D activities of TNCs are becomingincreasingly internationalized. This trend isapparent for all home countries, but starts fromdifferent levels. In the case of United StatesTNCs, the share of R&D of their majority-ownedforeign affiliates in their total R&D rose from11% in 1994 to 13% in 2002. German TNCs setup more foreign R&D units in the 1990s than theyhad done in the preceding 50 years. The shareof foreign to total R&D in Swedish TNCs shotup from 22% to 43% between 1995 and 2003.

Reflecting the increased internationa-lization of R&D, foreign affiliates are assumingmore important roles in many host countries’R&D activities. Between 1993 and 2002 the R&Dexpenditure of foreign affil iates worldwideclimbed from an estimated $30 billion to $67billion (or from 10% to 16% of global businessR&D). Whereas the rise was relatively modestin developed host countries, i t was quitesignificant in developing countries: the share offoreign affil iates in business R&D in thedeveloping world increased from 2% to 18%between 1996 and 2002. The share of R&D byforeign affiliates in different countries variesconsiderably. In 2003 foreign affiliates accountedfor more than half of all business R&D in Ireland,Hungary and Singapore and about 40% inAustralia, Brazil, the Czech Republic, Swedenand the United Kingdom. Conversely, it remainedunder 10% in Chile, Greece, India, Japan and theRepublic of Korea. Other indicators, such as therising number of R&D alliances and growingpatenting activity, similarly confirm the increased

internationalization of R&D activities indeveloping countries.

Their R&D is growing particularly fast,though unevenly, in developingcountries…

The share of host developing countries inthe global R&D systems of TNCs is rising, butunevenly. Only a few economies have attractedthe bulk of the R&D activity. Developing Asiais the most dynamic recipient. In the case of R&Dexpenditures by majority-owned foreign affiliatesof United States TNCs, for example, the shareof developing Asia soared from 3% in 1994 to10% in 2002. The increase was particularlynoticeable for China, Singapore, Hong Kong(China) and Malaysia. In the foreign R&Dactivities of Swedish TNCs the share of countriesoutside the Triad more than doubled, from 2.5%in 1995 to 7% in 2003. Survey findings and otherdata for Germany and Japan support the growingimportance of developing countries and someeconomies in transition as locations for TNCs’R&D.

Official statistics generally suffer from timelags, and may not fully capture the pace of R&Dinternationalization. More recent data on FDIprojects indicate that the expansion of R&D tonew locations is gaining momentum. Of 1,773FDI projects involving R&D worldwide duringthe period 2002-2004 for which information wasavailable, the majority (1,095) was in factundertaken in developing countries or in South-East Europe and the CIS. Developing Asia andOceania alone accounted for close to half of theworld total (861 projects). A survey of the world’slargest R&D spenders conducted by UNCTADduring 2004-2005 also shows the growingimportance of new R&D locations. More than halfof the TNCs surveyed already have an R&Dpresence in China, India or Singapore. In South-East Europe and the CIS, the Russian Federationwas the only significant target economymentioned by the responding firms as hostingR&D activities.

In the same survey, as many as 69% of thefirms stated that the share of foreign R&D wasset to increase; only 2% indicated the opposite,while the remaining 29% expected the level of

xxviiOVERVIEW

internationalization to remain unchanged. Themomentum appears to be particularly strongamong companies based in Japan and theRepublic of Korea, which until recently, have notbeen internationalizing their R&D to any largeextent. For example, nine out of ten Japanesecompanies in the sample planned to increase theirforeign R&D, while 61% of European firms statedsuch intentions. A further shift in terms of R&Dlocations towards some developing, South-EastEuropean and CIS markets is also envisaged.China is the destination mentioned by the largestnumber of respondents for future R&D expansion,followed by the United States. In third place isIndia, another significant newcomer location forR&D. Other developing economies mentioned ascandidates for further R&D by some respondentsinclude the Republic of Korea, Singapore, TaiwanProvince of China, Thailand and Viet Nam. Veryfew respondents indicated any plans to expandR&D to Latin America or Africa. The RussianFederation was also among the top 10 targetlocations.

Another new and notable trend in theinternationalization of R&D is the emergence andfast growth of foreign R&D activities ofdeveloping-country TNCs. This trend is drivenby the need to access advanced technologies andto adapt products to major export markets. Someof these TNCs are targeting the knowledge baseof developed countries, while others are settingup R&D units in other developing economies.

…and the type of R&D undertakenvaries by region.

The R&D conducted in different locationsvaries considerably by region and economy. Forexample, in 2002, three-quarters of the R&D ofUnited States majority-owned foreign affiliatesin developing Asia were related to computers andelectronic products, while in India over three-quarters of their R&D expenditure went intoservices (notably related to softwaredevelopment). In Brazil and Mexico, chemicalsand transport equipment together accounted forover half of all R&D by United States foreignaffiliates.

Moreover, TNCs carry out different typesof R&D abroad. Foreign affiliates of TNCs mayundertake adaptive R&D, which ranges from

basic production support to the modifying andupgrading of imported technologies. InnovativeR&D involves the development of new productsor processes for local, regional or (eventually)global markets. Technology monitoring units areestablished to keep abreast of technologicaldevelopment in foreign markets and to learn fromleading innovators and clients there.

While it is difficult to quantify R&D bytype, among developing host economies theevidence points to the predominance of Asia ininnovative R&D for international markets. R&Dactivities in selected Asian economies such asChina, India, the Republic of Korea and TaiwanProvince of China are becoming increasinglyimportant within the global R&D networks ofTNCs. Examples include the Toyota TechnicalCenter Asia Pacific in Thailand, Motorola’s R&Dnetwork in China and Microsoft’s sixth globalresearch centre in Bangalore, India. Some of theinnovative R&D conducted there is at the cuttingedge. The semiconductor industry is an example.One of the earliest to move production intodeveloping countries, it has also been among thefirst to move advanced design to selecteddeveloping economies in Asia. Some of the designis done by foreign affiliates and some by localfirms. A few firms from the Republic of Koreaand Taiwan Province of China, and to a lesserextent from China and India, for instance, are nowat the technology frontier of design work.

TNCs have so far located limited R&D inLatin America and the Caribbean. Relatively littleFDI in Latin America and the Caribbean is inR&D-intensive activities; when it is, the R&Dconducted is mostly confined to the adaptationof technology or products for local markets,called “tropicalization” in the Latin Americancontext. Some important exceptions exist inBrazil and Mexico in particular. In Africa, theR&D component of FDI is generally very low;with the exception of some countries such asMorocco and, especially, South Africa, R&D byTNCs is virtually non-existent. This is partlybecause of weak domestic R&D capabilities, andin many cases the absence of institutionalmechanisms that create sufficient incentives forinvestors to devote resources to R&D.

In some of the new EU members, foreignaffil iates have emerged as important R&Dplayers. In the Czech Republic, Hungary and

xxviii World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Poland, R&D by foreign affiliates is often linkedto manufacturing, mostly in the automotive andelectronics industries. Some foreign affiliates alsoconduct “innovative” R&D for regional or globalmarkets.

The process is driven by new pushand pull factors, and is facilitated byenabling technologies and policies...

The need to adapt products and processesto key host-country markets has always been animportant motive for TNCs to internationalizeR&D. The need to tap into knowledge centresabroad to source new technologies, recruit thebest skills and monitor the activities ofcompetitors is also well known in the literature.However, the recent surge of R&D by TNCs inselected developing host economies also reflectsthe quest for cost reduction and for accessingexpanding pools of talent in these locations. Itcan be seen as a logical next step in theglobalization of TNC production networks. It alsoresembles the international restructuring that hastaken place in export-oriented manufacturing andICT-based services through which TNCs seek toimprove their competitiveness by exploiting thestrengths of different locations.

R&D internationalization to new locationsoutside the Triad is driven by a complexinteraction of push and pull factors. On the pushside, intensifying competition, rising costs ofR&D in developed countries and the scarcity ofengineering and scientific manpower along withthe increasing complexity of R&D, reinforce theimperative to specialize as well as tointernationalize R&D work. On the pull side, thegrowing availability of scientific and engineeringskills and manpower at competitive costs, theongoing globalization of manufacturingprocesses, and substantial and fast-growingmarkets in some developing countries increasetheir attractiveness as new locations.

The expanding pool of talent in selecteddeveloping countries and economies in South-East Europe and the CIS is very important in thiscontext – notably in science-based activities –especially for companies that fail to find asufficient number of skilled people in their homecountries. In recent years, there has been adramatic increase in the number of peopleenrolled in higher education in developing

countries and economies in transition. In 2000-2001 China, India and the Russian Federationtogether accounted for almost a third of alltertiary technical students in the world. Inaddition, more scientists and engineers are stayingin, or returning to, China and India to performR&D work for foreign affiliates or local firmsor to start their own businesses. In Bangalore,for example, some 35,000 non-resident Indianshave lately returned with training and workexperience from the United States. Reflecting thegrowing importance of the human resource factor,both developed and developing countries are nowadopting new policy measures to attract skillsfrom abroad.

The internationalization of R&D is alsofacilitated by improvements in ICT and associatedcost decreases, new research techniques thatallow greater “fragmentation” of R&D and betterinformation on research capabilities that areavailable worldwide. At the same time, overallimprovements in host-country investmentclimates have all contributed to creating a moreenabling framework. Important policydevelopments relate, for example, to intellectualproperty rights (IPR) protection, reform of publicresearch activities, infrastructure development,and investment promotion efforts specificallytargeting R&D-related FDI and R&D incentives.

There are some fundamental reasons whythe current trend towards R&D internationalizationis set to continue. First, the competitive pressureon firms is likely to remain intense, forcing themto innovate more. Second, the need for greaterflexibili ty in R&D in response to rapidtechnological change requires sizeable numbersof research staff with a range of specializations,and it necessitates locating R&D activities wheresuch pools of researchers are available. Third,ageing populations in many developed countriesare likely to result in an insufficient supply ofspecialized, up-to-date skills, forcing TNCs tolook elsewhere for fresh talent. Fourth, throughcumulative learning processes involving localenterprises and institutions, the developingcountries that take part in the internationalizationof R&D will progressively enhance their ownability to conduct more R&D. At present however,it appears that only a few developing countriesled by China and India and some economies ofSouth-East Europe and the CIS, can effectivelymeet the conditions required to participate.

xxixOVERVIEW

…and has important implications forboth host and home countries.

The creation of knowledge is a driver ofeconomic growth, but no single country canproduce all the knowledge needed to staycompetitive and to grow in a sustained manner.Countries are therefore eager to connect withinternational networks of innovation. Outwardand inward FDI in R&D are two ways of doingso. R&D internationalization opens up newopportunities for developing countries to accesstechnology, build high-value-added products andservices, develop new skills and foster a cultureof innovation through spillovers to local firmsand institutions. FDI in R&D can help countriesstrengthen their innovation systems and upgradeindustrially and technologically, enabling themto perform more demanding functions, handlemore advanced equipment and make morecomplex products.

At the same time these benefits do notappear automatically, and unwanted effects canalso arise. The main concerns in economieshosting FDI in R&D relate to the potentialdownsizing of existing R&D when FDI involvestakeovers of domestic firms, unfair compensationto local firms and institutions collaborating withTNCs in the area of R&D, the crowding out oflocal firms from the market for researchers, a raceto the bottom in attracting R&D-related FDI andunethical behaviour by TNCs. There may also betensions between TNCs and host-countrygovernments, in that the former may seek to retainproprietary knowledge while the latter seek tosecure as many spillovers as possible.

A key determinant of the developmentimpact on a host economy is i ts absorptivecapacity. Indeed, technological capabilities in thedomestic enterprise sector and technologyinstitutions are necessary not only to attract R&Dbut also to benefit from its spillovers. Otherdeterminants are the type of R&D conducted, andwhether the R&D is linked to production. Themore a TNC interacts with a host developingcountry’s local firms and R&D institutions, andthe more advanced the country’s nationalinnovation system (NIS), the greater thelikelihood of positive effects on a host economy.

R&D internationalization also hasimplications for home countries – both developedand developing. It can help a country’s TNCs

improve their competitiveness by accessingstrategic assets and new technologies, acquiringunique knowledge at competitive prices,increasing specialization in their R&D, reducingcosts, increasing flexibility and expanding theirmarket shares. By extension, the improvedcompetitiveness of TNCs often has positiveimpacts on their home economies. Foreign R&Dcan generate opportunities and spillovers in thehome economy to the benefit of local firms andthe home economy as a whole.

At the same time, the transnationalexpansion of R&D may give rise to concerns inhome countries, especially with regard to the riskof hollowing out and the loss of jobs. Theseconcerns resemble those voiced in connectionwith the general debate on services offshoring.The trend is so new that any assessment must betentative. However, it does seem that protectionistmeasures to limit the expansion of R&D abroadwill not effectively address these concerns as theywould risk undermining the competitiveness ofthe country’s enterprises. Rather, to turn theinternationalization process into a win-winsituation for host and home countries alike,policies aimed at advancing the specificinnovation capabilities and the functioning of theNIS are key.

Appropriate policy responses areneeded at the national level…

Enterprises are the principal agents ofinnovation. However, they do not innovate andlearn in isolation, but in interaction withcompetitors, suppliers and clients, with publicresearch institutions, universities and otherknowledge-creating bodies like standards andmetrology institutes. The nature of theseinteractions, in turn, is shaped by the surroundinginstitutional framework. The complex web withinwhich innovation occurs is commonly referredto as the “national innovation system”. Itsstrength can be influenced by governmentintervention.

A number of policy and institutional areasneed to be addressed to attract FDI in R&D, tosecure the benefits that it can generate and toaddress potential costs. The starting point is tobuild an institutional framework that fostersinnovation. Particular policy attention is neededin four areas: human resources, public research

xxx World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

capabilities, IPR protection and competitionpolicy. Efforts to secure an adequate supply ofhuman resources with the right skills profileinvolve educational policies – not least at thetertiary level – and measures to attract expertisefrom abroad. For public R&D to contributeeffectively to the NIS, it is essential that it linkswith enterprise R&D and that public researchinstitutes promote the spin-off of new companies.The attractiveness of a location for conductingR&D may increase if the IPR regime is moreeffective, but a strong IPR regime is notnecessarily a prerequisite for TNCs to invest inR&D. The policy challenge is to implement asystem that encourages innovation and helps tosecure greater benefits from such activity, notablywhen it involves TNCs. At the same time, in orderto balance the interests of producers andconsumers, IPR protection needs to becomplemented by appropriate competitionpolicies.

Efforts in these areas need to reflect thecomparative advantage and technologicalspecialization of each country as well as thedevelopment trajectory along which a countryplans to move. FDI policy is also vital to promotethe desired forms and impacts of FDI. Selectivepolicies in this area can include targetedinvestment promotion, performance requirementsand incentives along with science and technologyparks.

IPAs can play an important role in acountry’s strategy to benefit from R&Dinternationalization by TNCs. It can potentiallyserve two prime functions. The first is tocommunicate and market existing investmentopportunities, for example, through targetedpromotion, based on a careful assessment of thelocations’ strengths and weaknesses and a goodunderstanding of the relevant locationaldeterminants. If a location is unlikely to be ableto offer the conditions needed to attract R&D byTNCs, an IPA may be better off focusing on itspolicy advocacy function. It may draw theattention of other relevant government bodies toareas that are important for making a locationbetter equipped to benefit from R&D by TNCs.

In a global survey of IPAs conducted byUNCTAD, a majority of the respondents werefound already to target FDI in R&D. A largemajority of IPAs in developed countries actively

promote FDI in R&D activities (79%), and 46%of those based in developing countries do so aswell. The highest percentage (94%) was notedfor IPAs in Asia and Oceania. Conversely, aminority of IPAs in Africa promote it actively,and only 11% of the IPAs in Latin American andthe Caribbean do so.

Finally governments need to pay attentionto more focused policies aimed at boosting thecapabilities of the domestic enterprise sector,notably through industry-specific and small andmedium-sized enterprise policies.

The various objectives of education,science and technology, competition andinvestment policies can be mutually reinforcing.Whether a country tries to connect with globalnetworks by promoting inward FDI, outward FDI,licensing technology, the inflow of skills orthrough any other mode, policies need to becoherent with broader efforts to strengthen theNIS. The stronger the NIS, the greater is thelikelihood of attracting R&D by TNCs and ofbenefiting from spillover benefits generated bysuch R&D. In essence the policies pursued needto be part of a broad strategy aimed at fosteringcompetitiveness and development.

Indeed, the emphasis on policy coherencemay be one of the most striking lessons learnedfrom those developing countries that are nowemerging as more important nodes in theknowledge networks of TNCs. In most of thesecountries, the starting point has been a long-termvision of how to move the economy towardshigher value-added and knowledge-basedactivities. The success of some Asian economiesis no coincidence; it is the outcome of coherentand targeted government policies aimed atstrengthening the overall framework forinnovation and knowledge inflows. In some form(and to varying degrees), they have activelysought to attract technology, know-how, peopleand capital from abroad. They have investedstrategically in human resources, typically witha strong focus on science and engineering;invested in infrastructure development for R&D(such as science parks, public R&D labs,incubators); used performance requirements andincentives as part of the overall strategy to attractFDI in targeted activities; and strategicallyimplemented IPR protection policies.

xxxiOVERVIEW

For many developing countries at the lowerend of the UNCTAD Innovation Capability Indexany expectation of a major influx of R&D byTNCs would be unrealistic in the short term.However, that is not an excuse for a lack ofaction. Rather, countries should consider how tobegin a process through which economic andtechnological upgrading could be fostered. Thecreation of innovative capabilities is a path-dependent and long-term task. For latecomers,ensuring that a process aimed at strengtheningtheir NIS gains momentum is an essential firststep.

For home countries, current trendsaccentuate the need to rely even more on thecreation, diffusion and exploitation of scientificand technological knowledge as a means ofpromoting growth and productivity. Rather thanregarding R&D internationalization as a threat,home countries should seize opportunities arisingfrom it. It is important to explore new ways ofcollaborating with the new R&D locations (e.g.through joint research programmes and carefulattention to the benefits and costs of outsourcingand R&D-related outward FDI). Countries shouldalso try to remove bottlenecks and “systemicinertia” in their NISs to be better positioned tobenefit from R&D internationalization. They mayalso see the need to specialize more in areaswhere they hold a competitive edge to strengthenexisting world-class centres of excellence andbuild new ones.

…taking developments at theinternational level into account.

Policy-making at the national level also hasto consider developments in internationalinvestment agreements at various levels. Manyinternational agreements give special attentionto investment in R&D activities. Key issues relateto the entry and establishment of R&D-relatedFDI, the treatment of R&D performancerequirements (whether by restricting or explicitlypermitting them), incentives encouraginginvestment in R&D activities, and the movementof key personnel.

In general, international investmentagreements confirm the importance of policiesthat seek to facilitate FDI in R&D. While most

countries welcome FDI in R&D, manygovernments do not allow foreign companies todraw on certain kinds of public R&D support.Many bilateral agreements also state explicitlythat governments are free to apply R&Drequirements as a condition for receivingpreferential treatment (e.g. an incentive). A smallnumber of agreements prohibit the use ofmandatory performance requirements in the areaof R&D.

Most international investment agreementsdo not have provisions that specifically protectR&D-related FDI; they protect FDI in general.Related provisions include the definition ofinvestment, the free transfer of returns arisingfrom R&D activities and the application of thenational treatment and most-favoured-nationstandards to foreign investors.

The protection of IPRs at the internationallevel and minimum standards set by internationaltreaties are of particular relevance for R&D-related FDI. The most important instrument inthis area is the WTO Agreement on Trade-relatedAspects of Intellectual Property Rights (TRIPS).Some recent agreements at the bilateral andregional levels have extended the minimumstandards set in the TRIPS Agreement. Theprotection of IPRs enshrined in these agreementsis intended to encourage the development ofproprietary knowledge; but at the same time, itlimits the policy space of States in an area thatis directly relevant to R&D activities. Fordeveloping countries it is therefore important tounderstand and make use of the flexibilitiescontained in the TRIPS Agreement. There is alsoa clear need for additional technical assistanceto facilitate the implementation of IPRs in adevelopment-friendly manner.

Some international investment agreementsalso encourage home countries to support thestrengthening of NISs in developing countries,by promoting outward R&D-related investmentin developing countries. In addition, internationalcooperation agreements in the areas of science,technology and innovation help create an enablingframework for R&D internationalization byfacilitating the flow of information, the formationof alliances, the pooling of financial resources,the improvement of access to technological

xxxii World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

expertise, matchmaking and the establishmentof private-public sector partnerships.

But there is scope for more cooperation tofoster policy formulation and stronger innovationsystems in developing countries. One key areais human resource development. The internationalcommunity could play a more active role in thisarea, for example, by supporting the strengtheningof the local educational infrastructure and bymaking educational opportunities to developingcountries available in developed countries. Homecountries could contribute to the improvementof the institutional framework for innovation indeveloping countries by assisting in the

establishment of technical standards andcertification systems through access to andprovision of testing equipment for standardsetting and quality assessment. Similar stepscould be taken with regard to the implementationof IPR systems and through R&D collaborationbetween institutions in developed and developingcountries.

Policies at the international level havedirect implications for the ability of developingcountries to formulate their R&D policies andto create the conditions that will enable them tobenefit from the internationalization of R&D byTNCs.

Supachai PanitchpakdiGeneva, September 2005 Secretary-General of UNCTAD

PART ONE

END OF THE DOWNTURN

A. Signs of recovery

Global FDI inflows rose modestly in 2004following large declines in their value in 2001(41%), 2002 (13%) and 2003 (12%). At $648billion in 2004, they were 2% higher than in2003. This growth reflected increased flows todeveloping countries as well as to South-EastEurope and the Commonwealth of IndependentStates (CIS) (figure I.1), which more than offsetthe decline (for the fourth year in a row) in flowsto developed countries. The difference betweeninflows to developed countries and developingcountries shrank to $147 billion – a significantnarrowing of the gap compared with previousyears.1 The United States was the largestrecipient in 2004, ahead of the United Kingdomand China as well as Luxembourg,2 the top FDIrecipients in 2003.

��������

GLOBAL TRENDS: FDI FLOWSRESUME GROWTH

Cross-border mergers and acquisitions(M&As) – key modes of global FDI since the late1980s – started to pick up in 2004 following threeyears of decline. Greenfield FDI continued to risefor a third consecutive year, strengthening thelikelihood of a reversal of the global downwardtrend in flows. Data on the financing componentsof FDI show that the overall magnitude andtrends of FDI in both developed and developingcountries are determined to a significant extentby equity investment. However, fluctuations inother components can occasionally influenceannual FDI flows to individual countries as inthe case of Germany in 2004. The degree oftransnationality – a measure of the relativeeconomic importance of foreign affiliates in totaleconomic activity – continued to rise for hosteconomies as international production maintainedgrowth.

Figure I.1. FDI inflows, global and by groups of economies, 1980-2004(Billions of dollars)

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics).

Developing economies

Developed economies

0

200

400

600

800

1 000

1 200

1 400

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

World total

South-East Europeand CIS

4 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

1. Overall analysis

a. FDI inflows and outflows

Global inflows of FDI rose in 2004 for thefirst t ime in four years. Notwithstandingstatistical problems in FDI data collection and

reporting that make comparisons of FDI betweencountries and regions difficult (box I.1), a numberof observations can be made regarding FDI flowsby region and sector.

Developed countries – a category nowdefined to include also the 10 new EuropeanUnion (EU) countries (box I.2) – saw FDI inflows

Box I.1. Problems with FDI data

The analysis of FDI trends in Part One of WIRis largely based on FDI flow data collected fromnational balance-of-payments statistics. Values ofFDI flows in national currencies are converted toUnited States dollars to calculate global FDI flowsand compare FDI inflows to and outflows fromdifferent countries and country groups. Balance-of-payments data on FDI flowsa are available formost countries for many years with a short timelag.b But there are some problems with these datathat have to be kept in mind when interpreting them.Many countries still deviate one way or anotherfrom the recommendations of the InternationalMonetary Fund (IMF) and the Organisation forEconomic Co-operation and Development (OECD)in their collection, definition and reporting of FDIdata (IMF/OECD 2004).

FDI is an investment involving a lastinginterest by a home-economy entity in an enterprisein a host economy. For data collection purposes,FDI has been defined as involving an equity stakeof 10% or more in a foreign enterprise. FDI hasthree components: equity capital, intra-companyloans and reinvested earnings. Different countrieshave different recording practices relating to thesethree components. Some countries deviate from thesuggested 10% threshold value for foreign equityownership. Most countries report long-term intra-

company loans, but not all countries record short-term loans and trade credits (annex B, Definitionsand sources). Some countries are still not able toreport reinvested earnings, as the data are not easilyavailable from company reports or balance-of-payments surveys; those that report often do sowith a considerable time lag. Out of 34 developedeconomies, only Greece did not report reinvestedearnings at all in 2003, and 78% of developingcountries reported such data that year.

Differences in how countries measure andreport FDI complicate the interpretation of FDItrends for the following reasons:

• Bilateral discrepancies between FDI flows asreported by home and host countries can bequite large. The following table on FDI flowsto China as reported by China (the host) andby a number of the investing (home) countrieshighlights this problem (box table I.1.1). Thusglobal FDI inflows and outflows differ. In 2004for example, global FDI outflows were 13%higher than global FDI inflows. This imbalanceis due to various factors such as: differentmethods of data collection by host and homecountries, different data coverage of FDI (i.e.all three components of FDI may not beincluded), different time periods used forrecording FDI transactions, and different

Box table I.1.1. FDI flows to China as reported by China and by the investing economy(Millions of dollars)

2000 2001 2002As reported As reported As reported

As reported by investing As reported by investing As reported by investingEconomy by China economy by China country by China economy

France 853 324 533 166 576 563Germany 1 041 819 1 213 976 928 887Hong Kong, China 15 500 46 361 16 717 8 496 17 861 15 938Japan 2 916 937 4 348 2 161 4 190 2 608Malaysia 203 40 263 82 368 81Netherlands 790 56 776 388 572 156Thailand 204 9 194 11 188 16United Kingdom 1 164 620 1 052 953 896 1 135United States 4 384 1 817 4 433 1 912 5 424 924

Source: UNCTAD FDI/TNC database (www.unctad.org/fdistatistics)./...

5CHAPTER I

Box I.1. Problems with FDI data (concluded)

treatment of round-trip investments and of FDIin special-purpose entities.

• As recording practices change over time, timeseries data on FDI flows have structural breaks.For example, Japanese data on FDI flows startedto include reinvested earnings (in addition to theother components) only in 1996, the same yearGerman FDI flows began to cover short-term,intra-company loans.

Furthermore, to facilitate a comparativeanalysis of worldwide FDI, data on flows in variouscurrencies are converted into a single currency, the

United States dollar, and growth rates of dollar-denominated FDI flows may diverge from growthrates of FDI flows in national currencies.c In 2004for instance, the United States dollar depreciatedagainst most currencies of the developed countries.Therefore the 9% decline in the dollar value of FDIinflows into developed countries using constantexchange rates was smaller than the decline in FDIinflows calculated with current exchange rates.Similarly, as FDI flows are expressed in nominalor current prices of a country, the conversion of theseflows into constant prices yields different results (boxtable I.1.2).

Source: UNCTAD.a The IMF’s Balance of Payments Manual (fifth edition, 1993) and the OECD Benchmark Definition of Foreign

Direct Investment (third edition, 1995) provide agreed guidelines for compiling FDI flows. Both of them are nowbeing revised. New methodologies and definitions of FDI are scheduled to be released in 2008.

b In the case of FDI stock, reliable data are available for considerably fewer countries because they are normally basedon company surveys.

c For example, if the currency of country A devalues by 10% against the dollar while FDI inflows in national currencyare constant, then FDI inflows into country A expressed in dollar terms would drop by 10%.

Box table I.1.2. FDI inflows to developed countries in various prices, 1980-2004(Billions of dollars and per cent)

In current exchange Percentage In constant Percentage In real PercentageYear rates and pricesa change exchange ratesb change prices c change

1980 46.6 39.0 55.8 81.1 46.4 13.31981 45.9 -1.6 49.9 -10.7 45.3 -2.31982 31.8 -30.6 30.9 -38.0 32.6 -28.11983 32.9 3.6 30.6 -1.1 35.1 7.81984 40.6 23.2 35.5 16.1 44.0 25.11985 42.5 4.6 35.9 1.1 46.7 6.31986 70.1 65.0 70.5 96.4 75.6 61.91987 115.6 64.9 129.1 83.1 113.8 50.61988 133.6 15.6 158.5 22.7 125.7 10.41989 163.3 22.2 187.5 18.3 151.4 20.51990 172.1 5.4 206.4 10.1 146.8 -3.11991 117.1 -32.0 141.2 -31.6 101.6 -30.81992 112.6 -3.9 138.9 -1.6 101.6 0.01993 144.0 27.9 171.8 23.7 138.6 36.41994 151.8 5.4 183.5 6.8 142.3 2.71995 218.7 44.1 273.5 49.1 186.3 30.91996 234.9 7.4 281.7 3.0 203.2 9.01997 284.0 20.9 317.3 12.6 261.8 28.81998 503.9 77.4 525.6 65.7 491.6 87.81999 849.1 68.5 891.1 69.5 844.8 71.92000 1 134.3 33.6 1 134.3 27.3 1 134.3 34.32001 596.3 -47.4 555.1 -51.1 618.6 -45.52002 547.8 -8.1 512.0 -7.8 568.2 -8.12003 442.2 -19.3 451.1 -11.9 416.0 -26.82004 380.0 -14.1 410.3 -9.0 331.4 -20.3

Source: UNCTAD.a FDI inflows to developed countries calculated by converting FDI inflows in national currencies and in current

prices into dollar values on the basis of the annual average exchange rate of the respective currencies againstthe dollar.

b Calculated by using the real effective exchange rate of the United States dollar (base year 2000).c FDI inflows to developed countries calculated by using the import price indices of industrialized countries

with 2000 as the base year (as reported by the IMF), as a proxy for constant prices.

6 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

fall by another 14% (to $380 billion) in 2004,despite economic recovery in many countries andsubregions, returning investor confidence andimproved corporate earnings (chapter II). Afterthe significant fall of 2001-2003, the furtherdecline brought FDI inflows to developedcountries to just 30% of their peak level of $1.1trillion in 2000. The decline was particularly

marked in the EU, where FDI fell by 36% toreach its lowest level since 1996. This declinewas concentrated in a few members. Denmark,Germany, the Netherlands and Sweden aloneaccounted for 86% of the total decline that wasspread over 10 countries. Other developedcountries in Western Europe (particularlyNorway, Switzerland) also experienced a fall (of

Box I.2. Changes in geographical groupings used in WIR05

Source: UNCTAD.

Major changes in the classification of groupsof economies have been introduced in the WorldInvestment Report beginning this year followingthe reclassification of some countries by the UnitedNations Statistical Office (UNSO). The EU nowhas 25 members, including the 10 countries thatbecame new members on 1 May 2004. Eightcountries (the Czech Republic, Estonia,Hungary, Latvia, Lithuania, Poland,Slovakia and Slovenia) have beenreclassified from Central and EasternEurope (CEE) to EU, and Cyprus fromWest Asia to EU. Malta has now beenreclassified from “other developedcountries” to EU. These ten countries arenow included among the “developedcountries”. All references to the EU inWIR05 refer to the new classification (i.e.the EU following the accession of the newmembers); growth rates have beencalculated on the basis of adjusted seriesunless stated otherwise. For the purposeof analysis in WIR05, EU-15 refers to thegroup of countries that were members ofthe EU before 2004 and EU-10 to the 10new EU members.

After the reclassification of the eightEU-accession countries as developedcountries instead of CEE, the rest of theCEE countries, along with countriesformerly in the group Central Asia (underdeveloping countries) are now classifiedby UNSO under South-East Europe in anew grouping comprising South-EastEurope and the Commonwealth ofIndependent States (CIS) (box table I.2.1).The CIS was created in December 1991and includes all of the republics that werepart of the former USSR, except the BalticStates.

In addition to the reclassificationsmentioned above, the nomenclature usedfor the developing Pacific Island countries

classified in previous WIRs under the Pacificsubregion of the Asia-Pacific region is changed to“Oceania” in order to bring WIR usage in line withthat of other UNCTAD publications. The countrycomposition of the subregion and region remainsthe same as in previous WIRs.

Box table I.2.1. Reclassification of countrygroupings in WIR05

New classification

New EU South-East Europe (SEE)countries and Commonwealth of

(classified Independent States (CIS)under

“developedOld classification countries”) SEE CIS

Former Central and Eastern EuropeAlbania AlbaniaBelarus BelarusBosnia and Herzegovina Bosnia and HerzegovinaBulgaria BulgariaCroatia CroatiaCzech Republic Czech RepublicEstonia EstoniaHungary HungaryLatvia LatviaLithuania LithuaniaRepublic of Moldova Republic of MoldovaPoland PolandRomania RomaniaRussian Federation Russian FederationSerbia and Montenegro Serbia and MontenegroSlovakia SlovakiaSlovenia SloveniaTFYR Macedonia TFYR MacedoniaUkraine UkraineCentral Asia (Developing countries)

Armenia ArmeniaAzerbaijan AzerbaijanGeorgia GeorgiaKazakhstan KazakhstanKyrgyzstan KyrgyzstanTajikistan TajikistanTurkmenistan Turkmenistan

7CHAPTER I

66%) in their combined inflows. Conversely, FDIflows to the United States rose for the first timesince 2000, to more than three times their 2003level; however, they too were at about one-thirdof their peak level of 2000. The United Kingdomwas another developed country that received largeFDI inflows in 2004 – nearly four times their2003 level. Flows to Australia, Japan and NewZealand also rose.

In contrast to developed-country inflows,flows to developing countries rose by 40% (to$233 billion) in 2004. As a result, their share inworld FDI inflows reached 36% – the highestsince 1997. While flows to Africa remainedvirtually unchanged, all other regions andsubregions experienced a significant increase:

• Africa attracted constant but relatively highlevels of FDI inflows at $18 bill ion,following an increase of 39% in 2003.

• Inbound FDI to the Asia-Oceania regionreached $148 billion, up from $101 billion.3

• FDI flows to Latin America and theCaribbean rose by 44% (to $68 billion) afterfour years of consecutive decline.

FDI flows to developing countries remainconcentrated: the top five recipients, China, HongKong (China), Brazil, Mexico and Singapore, inthat order, accounted for over 60% of total flows.

FDI inflows to the least developedcountries (LDCs)4 also rose, by 3% in 2004, to

reach $11 billion, the highest level ever for thesecountries. Thirty-five of the 50 LDCs receivedhigher inflows. FDI growth in this group in 2004was largely due to an increase in flows to suchcountries as the Democratic Republic of theCongo, Myanmar and Equatorial Guinea; theyexperienced growth rates of 470%, 91% and 16%respectively (annex table B.1). (Flows to themajor oil-producing countries in this group hadrisen considerably in 2003; for example, flowsto Angola and Sudan doubled.) However, FDIflows to LDCs still remain low; in spite of therise registered in 2004, their share in world anddeveloping-country FDI inflows was no morethan 2% and 5% respectively. Nonetheless, theshares of FDI inflows in gross fixed capitalformation are more significant for the LDCs asa group than for other developing countries: 20%vs. 10% in 2002-2004 (annex table B.3).

In the new regional category of South-EastEurope and the CIS, FDI flows amounted to $35billion in 2004 compared with $24 billion in 2003(chapter II). In the Russian Federation alone FDIgrew from $8 billion to $12 billion.

Of all capital flows to developingcountries, FDI continued to be the largestcomponent and is increasing (figure I.2): i taccounted for 51% of all resource flows todeveloping countries and has been several timeslarger than official flows in recent years.

Figure I.2. Total resource flowsa to developing countriesb, by type of flow, 1990–2003(Billions of dollars)

Source: UNCTAD, based on World Bank 2005a.a Defined as net liability transactions of original maturity of greater than one year.b The World Bank classification is used here. It differs from UNCTAD’s classification in that it includes CEE countries

under developing countries.

-50

0

50

100

150

200

250

300

350

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Total resource flows

FDI inflows

Portfolio flows

Commercial bank loans

Official flows

8 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Unsurprisingly, there was no markedchange in the sectoral distribution of FDI in2003-2004. FDI in the services sector continuedto grow, particularly in financial services (annextables A.I.4-A.I.7). Services accounted for 63%of the total value of cross-border M&As in 2004compared to 54% in 2003 (annex table B.5) andone-third of M&As in services were in financialservices. In the primary sector, FDI, driven byrising demand for various commodities,particularly oil, started to grow significantly insome regions in 2004, especially in mining andoil-related industries in Africa and Latin America(chapter II).

In terms of corporate functions there wasa large increase in FDI as seen in the number ofnewly established regional headquarters: in 2004more than 350, of which nearly 60% wereestablished in developing countries. A noteworthydevelopment is the continued growth of FDI inresearch and development (R&D), a phenomenonthat is extending increasingly to developingcountries (chapter IV). For instance, the numberof foreign greenfield investment projects in R&Drose from 516 in 2003 to 642 in 2004 (annextable A.I.3).5 The increase was higher in the caseof host developing economies, which received429 new R&D projects in 2004 compared with316 in 2003. The increasing internationalizationof TNCs’ R&D activities and the implications

of this, particularly for developing countries, arethe special focus of Part Two of this WIR.

FDI outflows increased in 2004 by 18% to$730 billion, of which $637 billion were fromdeveloped countries. These countries remainsignificant net capital exporters through FDI:outflows exceeded inflows of developed countriesby nearly $260 billion. While FDI outflows fromthe EU declined by 25% to $280 billion (a seven-year low), those from most other developedcountries increased in 2004. FDI outflows fromthe United States increased by 90%, to $229billion, its highest amount ever, and from Canadaand Switzerland by 121% and 67% respectively(to $47 billion and $25 billion).

While developed countries remain themajor source of FDI, outflows from developingcountries have also risen, from a negligibleamount in the early 1980s to $83 billion in 2004(figure I.3).6 The outward FDI stock fromdeveloping countries reached more than $1trillion in 2004, with a share in world stock of11% (annex table B.2). A number of notableM&As were undertaken recently by firms fromdeveloping countries (especially Asian firms),including in developed countries (chapter II).Developing countries are beginning to recognizethe importance of such investment for their firms’competitiveness and their economies’performance. A few of them even invest relatively

Figure I.3. FDI outflows from developing economies, and South-East Europe and CIS,by group of economies, 1984-2004

(Billions of dollars)

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics).

-10

10

30

50

70

90

110

130

150South-East Europe and CIS

Africa

Latin America and the Caribbean

Asia and Oceania

Developing economies and South-East Europe and CIS

19

84

19

85

19

86

19

87

19

88

19

89

19

90

19

91

19

92

19

93

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

20

03

20

04

9CHAPTER I

more abroad than some developed countries(WIR04). For example, the ratio of FDI outflowsto gross fixed capital formation was 25% forSingapore in 2002-2004 compared to 8% for theUnited States (annex table B.3). This rise of FDIfrom developing economies’ TNCs has takenplace largely in the context of governmentpolicies that have paid little attention to outwardinvestment, have been restrictive or have not beenactively supportive.7

b. Modes of FDI entry

Firms may enter host economies throughgreenfield investments or M&As.8 The choice ofmode is influenced by industry-specific factors.For example, greenfield investment is more likelyto be used as a mode of entry in industries inwhich technological skills and productiontechnology are key. The choice may also beinfluenced by institutional, cultural andtransaction cost factors (WIR00), in particular,the attitude towards takeovers, conditions incapital markets, l iberalization policies,privatization, regional integration, currency risksand the role played by intermediaries (e.g.investment bankers) actively seeking acquisitionopportunities and taking initiatives in making deals.

In 2004, cross-border M&As rose by 28%,to $381 billion (annex tables B.4-B.5), amidstan overall expansion of total (cross-border plusdomestic) M&As by nearly 50%, to over $2trillion. The number of cross-border deals reachedsome 5,100 – 12% higher than the previous year.An increase in the number of mega cross-borderdeals (with transaction values exceeding $1billion) contributed to the growth in the valueof cross-border M&As (table I.1). The largestdeal in 2004 was the acquisition of AbbeyNational (United Kingdom) by Santander CentralHispano (Spain) for $15.8 billion (annex tableA.I.1), almost the same value as that of the largestdeal in 2003 but only one-thirteenth of the largestdeal ever (the Vodafone-Mannesmann deal in2000).

Cross-border M&As rose more markedlyat the domestic and regional levels than at theglobal level. For instance, between companiesof the EU-15 such deals increased in value by57% to $99 billion, accounting for 57% of thevalue of all cross-border deals in that region in2004 (as compared with 52% in 2003).

In addition to low interest rates in majoreconomies and rising corporate profits, therecovery of asset prices since 2003 (as reflectedin the rise in stock exchange indices) contributedto the rise in M&As. Indeed, partly as a resultof increased stock prices, the number of cross-border deals using stock swaps rose from 123to 161 in 2004 (close to the number of such dealsin 1999), accounting for 16% of the total valueof cross-border M&As.9

The growth in the value and number ofcross-border M&As in 2004 was largely due totransactions taking place among developed-country firms: their value rose by 29%. Indeveloping countries – where such transactionsare normally less common, as fewer companiesattract foreign investors and restrictions continueto be imposed on M&As – cross-border M&Asalso rose in 2004 by 36% in value, to reach $55billion, two-thirds of the peak reached in 2001(annex table B.4). There was a significant risein cross-border M&A purchases in China andIndia, with a doubling of value in both countries,to record highs of $6.8 billion and $1.8 billionrespectively. For the first time, China became thelargest target country for cross-border M&As indeveloping countries.

Greenfield FDI, for its part, expanded froman estimated 9,300 projects in 2003 to 9,800projects in 2004.10 As in 2003, developing and

Table I.1. Cross-border M&As withvalues of over $1 billion, 1987-2004

Number of Percentage Value PercentageYear deals of total ($ bil l ion) of total

1987 14 1.6 30.0 40.31988 22 1.5 49.6 42.91989 26 1.2 59.5 42.41990 33 1.3 60.9 40.41991 7 0.2 20.4 25.21992 10 0.4 21.3 26.81993 14 0.5 23.5 28.31994 24 0.7 50.9 40.11995 36 0.8 80.4 43.11996 43 0.9 94.0 41.41997 64 1.3 129.2 42.41998 86 1.5 329.7 62.01999 114 1.6 522.0 68.12000 175 2.2 866.2 75.72001 113 1.9 378.1 63.72002 81 1.8 213.9 57.82003 56 1.2 141.1 47.52004 75 1.5 199.8 52.5

Source: UNCTAD, cross-border M&A database.

10 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

transition (South-East Europe and the CIS)economies attracted a larger number of greenfieldinvestments than developed countries. Thisillustrates the tendency for developing countriesto receive more FDI through greenfield projectsthan through M&As; greenfield investment is thekey driver behind the recent recovery of FDI.However, in developing countries such investmentis somewhat concentrated geographically: basedon some 4,800 projects for which informationwas collected in 2004, for instance, only 11economies11 received more than 100 greenfieldinvestments each in 2004 (annex table A.I.2).This concentration is in line with that of FDI asa whole in developing countries (chapter II). Asin the case of M&As, China and India attractedsignificant numbers of such FDI projects,together accounting for nearly half of the totalnumber in developing countries. Recentliberalization measures in India and strongeconomic growth in China, combined withincreased liberalization after its accession toWTO (chapter II), contributed to this trend.Three-fifths of all greenfield projects in the worldwere in the services sector (annex table A.I.3).

c. Components of FDI flows

FDI is financed by TNCs through equitycapital, intra-company loans and/or reinvestedearnings.12 The availability of data on eachcomponent of FDI flows varies by country:between 66 and 110 of the 212 economies forwhich FDI flows are reported provided data on

all of these three components for the period 1995-2004.13 Equity capital is the largest componentof FDI financing. Worldwide, its share in totalinflows fluctuated between 58% and 71% duringthe period 1995-2004; the higher shares wereregistered during the recent decline in world FDIflows (figure I.4). During the same period, intra-company loans, on average, accounted for 23%,and reinvested earnings for 12%, of world FDIinflows. The latter two components are much lessstable. The share of reinvested earnings in FDIfinancing reached a low of 2% of worldwide FDIinflows in 2001, but i t has been risingsubstantially since then. The share of intra-company loans, on the other hand, has fallencontinuously and significantly (figure I.4).

The lion’s share of FDI flows to developedcountries comprises equity capital (around 67%of total FDI flows over the period 1995-2004)(figure I.4). Its importance varies by country andover time. For instance, the average share ofequity capital in annual FDI flows was 85% inthe United States, 78% in Germany and rangedbetween 50% and 70% in Finland, Norway,Switzerland and the United Kingdom. In contrast,in Ireland and the Netherlands the shares wereonly 23% and 35%, respectively, during thatperiod. Equity capital was also the most importantcomponent of FDI flows to developing countriesin 1995-2003, but to a lesser extent than fordeveloped countries: its share in total FDI flowsfluctuated between 49% and 67%. In 2004 it fellto only 29%.14 Here again there are substantialdifferences between countries. In the case of

Figure I.4. Share of different financing components in world FDI inflows, 1995-2004(Per cent)

Source: UNCTAD, based on national sources and IMF Balance of Payments Statistics, CD-ROM, June 2005.

Note: Based on data only for countries for which all three components of FDI inflows were available. This number rangesfrom 66 to 110 economies and it accounts for an average of 87% of total FDI inflows.

-10

0

10

20

30

40

50

60

70

80

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Equity capital Reinvested earnings Intra-company loans

11CHAPTER I

some host economies such as Brazil, inward FDIrelied heavily on equity capital, while in someothers like Hong Kong (China), the share ofequity was only 28% during the period 1998-2004, with reinvested earnings and intra-companyloans assuming greater importance.

In a number of countries the share of equitycapital in FDI financing has also variedsubstantially over time. This reflects more thevolatility of the two other components of FDI– reinvested earnings and, especially, intra-company loans – than that of equity capital. Inthe United States, for instance, the contributionof equity capital to FDI inflows varied from alow of 58% in 1997 to a high of 153%15 in 2003(72% in 2004), in Germany, from 27% in 1998to 168%16 in 2003 (70% in 2004) and inArgentina, from 72% in 1996 to 282%17 in 2002(53% in 2004).

As noted above, the share of intra-companyloans in worldwide FDI inflows has fallensharply since 2001 (figure I.4). This is mainlydue to developments in a few large developedeconomies, such as the repatriation by TNCs oflarge amounts of credit from their affiliates inGermany ($10.1 billion in 2003 and $57.4 billionin 2004) and the United States ($31.7 billion in2003 and $17.8 billion in 2004) (chapter II),resulting in negative flows of intra-companyloans to the two countries in those years.Australia, Japan, the Netherlands and Portugalalso experienced negative inflows of intra-company loans due to large-scale repatriationsof such loans, but to a smaller extent thanGermany and the United States. Similar trendshave occurred in some developing economies.In Hong Kong (China), for instance, foreignTNCs withdrew credits of nearly $10 billion in2002 and $3 billion in 2003, but resumed lendingto their Hong Kong affiliates in 2004.

The share of intra-company loans differsbetween host countries. During the period 1995-2004 they contributed 40-50% of inward FDIflows in Germany18 and France but less than 10%in Argentina, Australia and Switzerland. Thisvariation can be explained partly by differencesin the structural features of the host and homeeconomies. Cross-border, intra-company loansoften depend on the financial management ofTNCs, which is in turn influenced by taxes andinterest-rate differentials as well as by thecharacteristics of home- and host-country capitalmarkets. For instance if the interest on a loan isreceived in a low-tax home country but the

interest payment is deductible (as cost) in a high-tax host country, TNCs can save on their globaltaxes by using intra-firm lending.19

Empirical studies on FDI in the UnitedStates (Desai, Foley and Hines 2004, Altshulerand Grubert 2003) and Germany (Ramb andWeichenrieder 2004) have highlighted the roleof tax differentials in intra-company lendingacross borders: low taxes in the United Statescompared to those in the home countries offoreign TNCs investing in the United States werefound to reduce the incentive to finance FDI inthe United States through intra-company loans.On the other hand, foreign TNCs were found toreact to the high German tax rate by preferringintra-company loans to equity financing for theirinvestments in Germany (chapter II).

As far as reinvested earnings are concerned(i.e. foreign affiliates’ earnings not distributedas dividends to the parent company) their sharein FDI flows has grown recently in all groupsof economies. In developed countries as a group,it rose to 15% of FDI inflows in 2003 – more thandouble the average of the previous ten years. In2004, the corresponding share was 33%, mainlydue to negative flows of intra-company loans.As with other components of FDI inflows, theimportance of reinvested earnings differs fromcountry to country (table I .2). While mostdeveloped countries received positive FDIinflows in the form of reinvested earnings in2003, France and Germany recorded negativereinvested earnings.20 In the case of France, thisseems to be a temporary phenomenon. InGermany, however, negative reinvested earningsof foreign affiliates have been registered for manyyears. This does not necessarily mean thataffiliates of foreign TNCs located in Germanyhave been enduring sustained losses; data showthat over a period of 30 years, aggregateddividends have been higher than the aggregatedprofits of all reporting foreign affiliates.21 Inprinciple, the distribution of large dividendpayments by foreign affil iates in Germanyreduces their retained profits, which can helpreduce the taxes they pay in Germany (chapter II).

In developing countries the picture isslightly different, with reinvested earnings beingmore prominent: these earnings accounted forabout 30% of FDI flows, on average, during theperiod 1995-2004, reaching 36% in 2003. Suchearnings are therefore becoming crucial tosustained flows of FDI to developing countries,which is why a number of countries have

12 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Table I.2. FDI inflows to the top 20 economies, ranked by size of differentfinancing components, 2003

Equity capital Reinvested earnings Intra-company loans

Bill ions of Bill ions of Bill ions ofRank Economy dollars Economy dollars Economy dollars

1 United States 87.0 Ireland 19.4 France 27.72 Luxembourg 80.9 Hong Kong, China 16.0 Spain 14.23 Germany 45.7 United Kingdom 12.2 Italy 8.84 China 37.4 China 7.2 Luxembourg 6.45 Belgium 26.2 Russian Federation 7.1 Belgium 5.96 France 17.0 Canada 6.7 Mexico 5.87 Netherlands 14.6 Australia 5.7 Switzerland 5.38 Spain 13.0 Netherlands 5.2 Sweden 3.29 Brazil 9.3 Italy 4.8 Angola 2.8

10 Switzerland 8.3 Luxembourg 3.7 Russian Federation 2.811 Portugal 7.7 Switzerland 2.9 United Kingdom 2.812 Japan 7.6 Malaysia 2.8 China 2.513 Ireland 6.0 Mexico 2.3 New Zealand 2.314 United Kingdom 5.4 Finland 2.3 Ireland 1.515 Poland 4.6 Czech Republic 2.2 Norway 1.416 Austria 4.4 Hungary 2.1 Austria 1.317 Thailand 4.1 Chile 1.9 Ecuador 1.318 Azerbaijan 3.3 Nigeria 1.9 Venezuela 1.219 Argentina 3.0 Spain 1.9 Chad 1.020 Israel 2.9 India 1.8 Kazakhstan 0.9

Source: UNCTAD (www.unctad.org/fdistatistics) and UNCTAD’s own estimates.

introduced fiscal incentives to encouragereinvestment of earnings by foreign affiliates.

d. Factors contributing to therecovery

The recovery of FDI flows in 2004 is theresult of favourable developments with respectto the macro, micro and institutional factorsdetermining these flows.

Macroeconomic factors. After the sharpslowdown in 2001, global economic growthrecovered gradually in 2002 and 2003. In 2004,world economic growth reached 5.1%, thestrongest growth rate since the mid-1980s (figureI.5). As in the past, improved economic growthhelped many countries attract more FDI (WIR03).

Most of the countries and regions with higheconomic growth rates recorded a sharp increasein FDI inflows in 2004. A number of developingcountries in Asia, Africa and Latin Americaexperienced a generally strong economic growthand, partly as a result, received significantlyhigher FDI inflows. This was also the case in theUnited Kingdom, the United States and the newEU member countries, which registered growth

rates in 2004 of 3.1% (2.2% in 2003), 4.4% (3.0%in 2003) and 4.9% (3.7% in 2003) respectively(chapter II).22 In contrast, several EU countriesthat grew at slower rates than the developedcountries mentioned above, saw declining orstagnating FDI inflows.

The sharp increase in FDI inflows into theUnited States and some other countries (e.g.China) may also have been driven by theweakening dollar, which made investment in theUnited States – and in other countries withexchange rates pegged to the dollar – less costlyfor foreign investors. This is similar to the waveof FDI inflows into the United States in the 1980sin reaction to the dollar’s weakness (Froot andStein 1991). The declining dollar also improvedthe price competitiveness of companies locatedin these countries, therefore attracting efficiency-seeking FDI. The dollar’s depreciation boostedtheir exports, which further stimulated FDIflows.23 Rising exports are often accompaniedby increasing FDI for improving distribution andmarketing facilities for exports and for meetingthe specific needs of exporters (Blomström,Lipsey and Kulchycky 1998, Pfaffermayer 1996,Egger 2001).

13CHAPTER I

Source: UNCTAD, based on UNCTAD FDI/TNC database (www.unctad.org/fdistatistics) for FDI and International MonetaryFund, World Economic Outlook Database, April 2005 for GDP.

Figure I.5. Growth rates of world FDI inflows and GDP, 1980-2004(Per cent)

Country risks, overall, declined worldwidein 200424 and business and consumer confidenceincreased.25 The gradual decline of risk may havecontributed to the recovery of FDI flows,although the empirical evidence for this is mixed(Moosa 2003, chapter 5).26

Microeconomic factors. Strong economicgrowth as well as large-scale restructuring andconsolidation of business brought manycompanies back firmly to profit-making in 2004.Corporate profitability in the large economiesimproved even more.27 Increased profits andfavourable financing conditions have helpedexpand investments abroad. In addition, as manyas 48 out of 49 major stock exchanges recordedrising share prices in 2004, which eased thefinancing of investments.28 Increasing stockmarket values produce positive wealth effects andfacilitate takeovers, especially through stockswapping. Higher stock market valuations alsoboost the value of cross-border M&As.

The recovery of FDI flows in many regionsof the world was also influenced by fast risingcommodity prices, at a rate of 11% for four yearsin a row.29 Consequently, by 2004 such pricesreached a record high. The higher prices andsupply shortages induced TNCs to invest in newexploration and production facilities, especiallyin Africa and Latin America. Rising incomes ofproducers of oil, gas and other raw materialscontributed to increasing FDI by TNCs in thoseindustries.

Institutional factors. The process ofprivatization has come to an end in manydeveloping and transition economies, and hencedid not contribute much to FDI in 2004. But twoother relatively new developments did. Privateindividual and institutional equity investors (asdistinct from TNCs) gained significantimportance in FDI. The value of cross-borderM&As by private equity companies30 rose froman estimated $69 billion in 2003 to $107 billionin 2004, accounting for 28% of all cross-borderM&As, up from 23% in 2003.31 Anotherdevelopment was the liberalization of FDI in realestate, traditionally closed to foreign investmentin many countries (chapter II). In Germany32 andPoland, for instance, l iberalization andprivatizations played a major role in attractingFDI into real estate. FDI in real estate grewrapidly worldwide in 2004, helped also by therise in real estate prices: for example, the valueof cross-border M&As in real estate tripled to$30 billion.33

e. The importance of TNC activitiesin the world economy

The universe of TNCs is large, diverse andexpanding. By the early 1990s, there were anestimated 37,000 TNCs in the world, with at least170,000 foreign affiliates. Of these, 33,500 wereparent corporations based in developed countries.By 2004 the number of TNCs had risen to some70,000 with at least 690,000 foreign affiliates,

FD

Iin

flow

s

-60.0

-40.0

-20.0

0.0

20.0

40.0

60.0

80.01

98

0

19

81

19

82

19

83

19

84

19

85

19

86

19

87

19

88

19

89

19

90

19

91

19

92

19

93

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

20

03

20

04

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

RealG

DP

FDI inflows Real GDP

14 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

almost half of which are now located in developingcountries (annex table A.I.8).

The role of TNCs in the world economy hasthus continued to grow, as reflected in theexpansion of FDI stock and in the operations offoreign affiliates (table I.3). Sales, value added(gross product), assets, employment and exportsof foreign affiliates have all resumed an upwardtrend since 2002.

The degree of transnationality of hostcountries stagnated during 2000-2002 in bothdeveloped and developing countries according to

the transnationality indices for host economies(figure I.6). This reflects the decline of FDI flowsin these regions during that period. There are alsosignificant differences in the degree oftransnationality of different countries. The mosttransnationalized economies in 2002 were Belgiumand Luxembourg, among developed countries, andHong Kong (China), among developing economies(figure I.7) – positions held by those economiessince this index was developed in 1996 (WIR99).While India has been catching up in inward FDI,it sti l l ranks near the bottom in 2002. Thetransnationality of host countries depends on the

Table I.3. Selected indicators of FDI and international production, 1982-2004(Billions of dollars and per cent)

Value at current prices Annual growth rate

(Bil l ions of dollars) (Per cent)

1986- 1991- 1996-Item 1982 1990 2003 2004 1990 1995 2000 2001 2002 2003 2004

FDI inflows 59 208 633 648 22.8 21.2 39.7 -40.9 -13.3 -11.7 2.5FDI outflows 27 239 617 730 25.4 16.4 36.3 -40.0 -12.3 -5.4 18.4FDI inward stock 628 1 769 7 987 8 902 16.9 9.5 17.3 7.1 8.2 19.1 11.5FDI outward stock 601 1 785 8 731 9 732 18.0 9.1 17.4 6.8 11.0 19.8 11.5Cross-border M&As a .. 151 297 381 25.9 b 24.0 51.5 -48.1 -37.8 -19.6 28.2Sales of foreign affi l iates 2 765 5 727 16 963 c 18 677 c 15.9 10.6 8.7 -3.0 14.6 18.8 c 10.1 c

Gross product of foreign affi l iates 647 1 476 3 573 d 3 911 d 17.4 5.3 7.7 -7.1 5.7 d 28.4 d 9.5Total assets of foreign affi l iates 2 113 5 937 32 186 e 36 008 e 18.1 12.2 19.4 -5.7 41.1 e 3.0 e 11.9 e

Exports of foreign affi l iates 730 1 498 3 073 f 3 690 f 22.1 7.1 4.8 -3.3 f 4.9 f 16.1 f 20.1 f

Employment of foreign affi l iates (thousands) 19 579 24 471 53 196 g 57 394 g 5.4 2.3 9.4 -3.1 10.8 g 11.1 g 7.9 g

GDP (in current prices) h 11 758 22 610 36 327 40 671 10.1 5.2 1.3 -0.8 3.9 12.1 12.0Gross fixed capital formation 2 398 4 905 7 853 8 869 12.6 5.6 1.6 -3.0 0.5 12.9 12.9Royalties and licence fee receipts 9 30 93 98 21.2 14.3 8.0 -2.9 7.5 12.4 5.0Exports of goods and non-factor services h 2 247 4 261 9 216 11 069 12.7 8.7 3.6 -3.3 4.9 16.1 20.1

Source: UNCTAD, based on its FDI/TNC database (www.unctad.org/fdi statistics), and UNCTAD estimates.a Data are available only from 1987 onward.b 1987-1990 only.c Based on the following regression result of sales against FDI inward stock (in millions of dollars) for the period 1980-

2002: Sales = 2 003.858+1.87288*FDI inward stock.d Based on the following regression result of gross product against FDI inward stock (in millions of dollars) for the period

1982-2002: Gross product = 622.0177+0.369482*FDI inward stock.e Based on the following regression result of assets against FDI inward stock (in millions of dollars) for the period 1980-

2002: Assets = -1 179.838+4.177434*FDI inward stock.f For 1995-1998, based on the regression result of exports of foreign affiliates against FDI inward stock (in millions of

dollars) for the period 1982-1994: Exports = 357.6124+0.558331*FDI inward stock. For 1999-2004, the share of exportsof foreign affiliates in world exports in 1998 (33.3 per cent) was applied to obtain the values.

g Based on the following regression result of employment (in thousands) against FDI inward stock (in millions of dollars)for the period 1980-2002: Employment = 16 552.15+4.587846*FDI inward stock.

h Based on data from the International Monetary Fund, World Economic Outlook, April 2005.

Note: Not included in this table are the values of worldwide sales by foreign affiliates associated with their parent firmsthrough non-equity relationships and the sales of the parent firms themselves. Worldwide sales, gross product,total assets, exports and employment of foreign affiliates are estimated by extrapolating the worldwide data offoreign affiliates of TNCs from Austria, Finland, France, Germany, Italy, Japan, Portugal, Sweden, Switzerlandand the United States for employment; those from Austria, Finland, France, Germany, Italy, Japan, Portugal andthe United States for sales; those from Japan and the United States for exports; those from the United Statesfor gross product; and those from Austria, Germany and the United States for assets, on the basis of the sharesof those countries in worldwide outward FDI stock.

15CHAPTER I

extent to which TNCs are expanding their foreignactivities in various locations. The next sectionlooks at the universe of the largest TNCs, whichplay an important role in that process.

2. The largest TNCs

TNCs are mainly based in developedcountries, and are increasingly being establishedin developing countries as well. This sectionlooks at developments among the largest TNCs:the 100 largest non-financial TNCs worldwideand the 50 largest ones from developingeconomies ranked by foreign assets. It alsoincludes an analysis of the ten largest TNCs fromSouth-East Europe and the CIS (also ranked byforeign assets), and, for the first time in the WIR,an analysis of the transnationalization of the 50largest financial TNCs worldwide ranked by totalassets.

a. The world’s top 100 TNCs

The 100 largest TNCs play a major role ininternational production; they account for 12%,18% and 14%, respectively, of the estimated

foreign assets, sales and employment of all TNCsin the world. Following a slowdown in theirexpansion in 2000, they resumed growth in 2002.In 2003, their assets and sales, both foreign andtotal, grew significantly (table I.4). Overall, therankings in the top 100 list in 2003 (the latestyear for which data on the top TNCs wereavailable) were fairly similar to those in 2002(annex table A.I.9). The top 10 companiesmaintained almost the same order as in 2002,General Electric and Vodafone heading the listeach with foreign assets of about $250 billion.Despite the overall stability at the top of the list,there were 15 newcomers, including somemanufacturing firms such as BAE Systems,Robert Bosch and United Technologies, as wellas some petroleum and mining companies, likePetronas, Statoil and Rio Tinto.

Over the past decade or so, a number of newcompanies from the services sector have joinedtop rankings on the list, yet some companies intraditional industries have remained in the highestrankings. In the petroleum industry, for instance,Shell and ExxonMobil, which were numbers oneand two, respectively, in 1992, are still amongthe top 10 TNCs. Motor vehicle companies like

Ford, General Motors and Toyota arealso still among the top 10. Globally,10 of the top 20 companies in 2003were already in the top 20 in 1992.

The three industries dominatingthe list are motor vehicles, petroleumand electrical/electronic equipment with11, 10 and 9 entries each. Together,more than half of the 30 leadingcompanies listed among the top 100were in these industries. A large groupof new TNCs has emerged in recentyears in service industries that arerelatively new to FDI – notably,telecommunications, electricity, waterand postal services – many of whichwere former State-owned monopolies.In 2003, TNCs in these industriesaccounted for almost 20% of the top100 firms. The two companies thatclimbed the most in the rankings in2003, Suez (11th) and DeutscheTelekom (14th), operate in serviceindustries.

The largest TNCs remaingeographically concentrated in a fewhome countries. The United Statesdominated the list with 25 entries. Five

Figure I.6. Transnationality Index of host countries,aby group of economies, 1998-2002

Source: UNCTAD.a Average of four shares: three-year average of FDI inflows as a

percentage of gross fixed capital formation; FDI inward stock as apercentage of GDP; value added of foreign affiliates as a percentageof GDP; and employment of foreign affiliates as a percentage of totalemployment. Data cover 73 economies: 22 developed countries, 32developing countries and 19 countries which are classified under Centraland Eastern Europe.

Note: For each group of economies, the weighted average is used. Fordetails, see the note in figure I.7. For the country compositionof each group of economies, see also figure I.7.

0

3

6

9

12

15

18

1998 1999 2000 2001 2002

Developed economies Developing economies

South-East Europe and CIS

16 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Figure I.7. Transnationality Index of host economies, 2002(Per cent)

Source: UNCTAD estimates.a Average of four shares: FDI inflows as a percentage of gross fixed capital formation for the past three years, 2000-

2002; FDI inward stock as a percentage of GDP in 2002; value added of foreign affiliates as a percentage of GDPin 2002; and employment of foreign affiliates as a percentage of total employment in 2002.

b Only the economies for which data for all of these four shares are available were selected. Data on value added areavailable only for Belarus, Czech Republic, Finland (2001), France (2001), Hungary (2000), Ireland (2000), Italy (1997),Japan (1999), the Netherlands (1996), Norway (1998), Poland, Portugal, Sweden (2000), the United Kingdom (1997),the United States, China, India (1995), Malaysia (1995), Singapore (2000), Taiwan Province of China (1994) and theRepublic of Moldova. For Albania, the value added of foreign owned firms was estimated on the basis of the per capitainward FDI stocks and the corresponding ratio refers to 1999. For the other economies, data were estimated by applyingthe ratio of value added of United States affiliates to United States outward FDI stock to total inward FDI stock of thecountry. Data on employment are available only for Austria (2001), the Czech Republic, Denmark (1996), Finland (2001),France (2001), Germany, Hungary (2000), Ireland, Italy (1999), Japan (2001), the Netherlands. For Albania, employmentof foreign-owned affiliates was estimated on the basis of their per capita inward FDI stock, and the corresponding ratiorefers to 1999. For the remaining countries, data were estimated by applying the ratio of employment of Finnish, German,Japanese, Swedish, Swiss and United States affiliates to Finnish, German, Japanese, Swedish, Swiss and United Statesoutward FDI stock to total inward FDI stock of the economy. Data for France, the Netherlands, Norway, Sweden andthe United Kingdom refer to majority-owned foreign affiliates only.

Note: The simple average refers to the simple mean of the indices of the individual countries within each group, whilethe weighted average takes into account the weight that each country has in each the four shares (as explainedin footnote a above).

(a) Developed economies (b) Developing economies (c) South East Europe and CIS

Weighted average of group

Simple average of group

Japan

Italy

Greece

United States

Norway

France

Germany

Austria

Portugal

Poland

Israel

United Kingdom

Latvia

Switzerland

Australia

Finland

Spain

Canada

Slovenia

Lithuania

New Zealand

Slovakia

Sweden

Hungary

Czech Republic

Denmark

Netherlands

Estonia

Ireland

Belgium and Luxembourg

Thailand

Weighted average of group

Simple average of group

Indonesia

Saudi Arabia

India

United Arab Emirates

Republic of Korea

Turkey

Barbados

Philippines

Taiwan Province of China

Egypt

China

Costa Rica

Peru

Brazil

Mexico

Colombia

Bahamas

Guatemala

Dominican Republic

Venezuela

South Africa

Argentina

Malaysia

Ecuador

Jamaica

Panama

Honduras

Chile

Trinidad and Tobago

Singapore

Hong Kong, China77.1

69.3

39.0

38.4

35.3

30.9

30.1

28.5

27.5

27.4

81.6

60.3

51.2

27.0

26.8

TFYR Macedonia

Republic ofMoldova

RussianFederation

Bulgaria

Croatia

Romania

Albania

Ukraine

Bosnia andHerzegovina

Serbia andMontenegro

Belarus

Weighted average of group

Simple average of group

0 5 10 15 20 25 00 5 510 1015 15 2020 25 25-5

41.4

28.9

17CHAPTER I

countries (France, Germany, Japan, the UnitedKingdom and the United States) accounted for71 out of the 100, while the EU alone accountedfor 50. Four companies are from developingeconomies, Hutchison-Whampoa of Hong Kong(China) being the largest among them (16th).

b. The top 50 TNCs from developingeconomies

Since UNCTAD began publishing the listof the top 50 TNCs from developing economiesin 1995, these companies have expanded theiractivities abroad. In 2003 their foreign assetsclimbed to $249 billion from $195 billion in 2002(table I.5). As in 2002, the five largest TNCsaccounted for almost half of the total foreignassets of the top 50. With foreign assets of $59billion, Hutchison Whampoa (Hong Kong, China)continues to hold the leading position, with 25%of the total foreign assets of the top 50. Singtel(Singapore), Petronas (Malaysia), SamsungElectronics (Republic of Korea) and Cemex(Mexico) remained, in that order, in the next fourpositions. Although the top TNCs remained thesame, 14 newcomers also entered the top 50 listin 2003 mainly from Asia (annex table A.I.10).

Asia has reinforced its dominance in thetop 50 with 39 enterprises on the list. The other11 enterprises came from South Africa (4),Mexico (4) and Brazil (3). Hong Kong (China)

Table I.4 Snapshot of the world’s 100largest TNCs: assets, sales and

employment, 2002, 2003(Billions of dollars, thousand of

employees, per cent)

Variable 2002 2003 % change

AssetsForeign 3 317 3 993 20.4Total 6 891 8023 16.4Foreign as % of total 48.1 49.8 1.7a

SalesForeign 2 446 3 003 22.8Total 4 749 5 551 16.9Foreign as % of total 51.5 54.1 2.6a

EmploymentForeign 7 036 7 242 2.9Total 14 332 14 626 2.1Foreign as % of total 49.1 49.5 0.4a

Source: UNCTAD/Erasmus University database.a In percentage points.

and Singapore remained the most important homeeconomies, with ten and nine entries in the listrespectively. Taiwan Province of China, witheight companies in the top 50, became the homeeconomy with the third largest contingent ofTNCs on the list largely owing to its electronicscompanies. The growing significance of thiseconomy was mainly at the expense of SouthAfrica, which had four companies listed in thetop 50 in 2003 compared to seven in 2002.

The top 50 TNCs operate in a wide rangeof industries, the most important being electrical/electronic equipment and computers (mainlycompanies from Asia), followed by food andbeverages. Other relatively significant industriesfor the top 50 include petroleum (6 TNCs),telecommunications (3), transportation (3),utilities (3) and hotels (3).

Four companies in the top 50 list(Hutchison Whampoa, Singtel, Petronas andSamsung) are also among the world’s top 100TNCs discussed above. It is likely that in thefuture more TNCs from developing economieswill enter the list of the top 100, since outwardFDI from these countries is expanding.Meanwhile, though, there remains a large gap insize between TNCs from the developed anddeveloping groups. For instance, the total foreignassets of all the top 50 TNCs from developingeconomies in 2003 was barely equal to those ofGeneral Electric, the world’s largest TNC.

In 2003, the assets, sales and employment,both foreign and total, of the largest TNCs fromdeveloping economies registered a large increaseover previous years. However, the share of theforeign component of the three indicatorsdeclined. Moreover, when comparing the threeratios for the TNCs from developing economieswith those from developed countries it is clearthat the degree of internationalization of theformer is lower (table I.5), as discussed in thefollowing section.

c. Transnationality of the top TNCs

The degree of transnationality (or theimportance of foreign as compared with the totalactivity of TNCs) stagnated during 2001-2003,for both the world’s top 100 TNCs and the top50 TNCs from developing countries, accordingto UNCTAD’s Transnationality Indices (TNIs)34

(figure I.8). An analysis of the TNI of the 100largest TNCs suggests that the TNI, measured

18 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

as the simple average value of the TNIs of allthe TNCs on the top 100 list, decreased againin 2003, from 57 to 55.8 (table I.6). However,if the value of the TNI is based on global figuresfor the assets, sales and employment of the top100 (a weighted average), its value rose slightlyin 2003, by 1.5 percentage points, suggesting thatthe degree of transnationality of the top quartileof the largest TNCs has recovered faster than thatof the bottom quartile. This reflects the fact thatTNCs are focusing more on their domesticmarkets at a t ime of worldwide economicslowdown of their activities, and that the largestTNCs are able to recover faster than the average-sized TNCs.

Of the top 100, firms from Japan and theUnited States are, on average, lesstransnationalized than their Europeancounterparts (table I .6). Firms from smallEuropean economies have the highest averageTNI, partly reflecting the need to go abroad tocompensate for smaller home markets. Exceptin 2003, the TNI of the top 50 TNCs fromdeveloping countries has increased substantiallyover the past decade, and has been catching upwith that of the world’s largest TNCs (figure I.8).

The sales-to-assets ratio is an indicator ofcapital efficiency. The ratio of sales-to-employment shows the value of sales peremployee, and provides an indication of labourproductivity, which may in turn indicate

Table I.5. Snapshot of the top 50 TNCsfrom developing countries: assets,sales and employment, 2002, 2003

(Billions of dollars, thousands ofemployees, per cent)

Variable 2002 2003 % change

AssetsForeign 195.2 248.6 27.4Total 464.3 710.9 53.1Foreign as % of total 42.0 35.0 - 7.0a

SalesForeign 140.0 202.2 45.9Total 308.4 512.5 66.1Foreign as % of total 45.4 39.9 - 5.5a

EmploymentForeign 713.6 1 077.2 50.9Total 1 503.3 3 096.6 106.0Foreign as % of total 47.5 34.8 - 12.7a

Source: UNCTAD/Erasmus University database.a In percentage points.

Table I.6. Average TNI values for theworld’s largest TNCs, 2002, 2003

(Per cent)

Variable 2002 2003

Top 100 TNCs 57.0 55.8United States 43.8 45.8United Kingdom 70.4 69.2Japan 43.6 42.8France 69.0 59.5Germany 46.9 49.0Small European countries 88.5 72.2

Top 50 TNCs 49.2 47.8

Source: UNCTAD/Erasmus University database.

Note: A simple average value is used. It is thesum of the TNI values of all the companies,divided by the total number of companies.

differences in the types of activities andtechnologies involved. A comparison of the sales-to-assets ratio for the top 100 TNCs worldwideand for the top 50 from developing economiesshows a marginal difference. On the other hand,the indicator of labour productivity shows a muchhigher value for the world’s 100 largest TNCscompared with the 50 largest TNCs fromdeveloping countries (table I.7). It should benoted that these ratios are highly dependent onthe industry composition of the top 100 and top50, and that the indicators differ across sectorsof activity much more than between firms withinthe same sector.

The geographic spread of a company’soperations and interests is captured by theInternationalization Index, the ratio of the numberof foreign affil iates to the total number ofaffiliates: it shows that, on average, 66% of theaffiliates of the top 100 TNCs are located abroad(annex table A.I.9). Like the TNI, theInternationalization Index is highest for top TNCsfrom small economies (such as Finland, Spainand Switzerland) and for the pharmaceuticalindustry. On average, the top TNCs have affiliatesin 39 foreign economies. Ranking TNCs by thenumber of host countries shows that firms fromEuropean countries rank high, with affiliates inan average of 71 host economies.35 The hostcountry most favoured by these 100 largest TNCsis the Netherlands, where 91 of the 100 have atleast one affil iate, followed by the UnitedKingdom and Canada. Among developingcountries, Brazil hosts the largest number ofaffiliates of the top 100 TNCs (75), followed byChina, with 60.

19CHAPTER I

Figure I.8. Average TNIa of the 100 largestTNCs in the world and of the 50 largest TNCs

from developing countries, 1993-2003

Source: UNCTAD/Erasmus University database.a A simple average (for definition, see table I.6).

The Internationalization Index also showsthat, on average, 49% of the affiliates of the top50 TNCs are located abroad (annex table A.I.10).This index is highest for TNCs from Hong Kong(China), the Republic of Korea and Singapore,and for those in the electrical/electronics industry.On average, the top 50 TNCs have affiliates in13 host economies, which is much less than thoseof the top 100 TNCs, though the East Asian firmsat the top of the 50 list come close (with anaverage of 36 host economies) to theircounterparts from developed countries.

d. The top 10 TNCs from South-EastEurope and the CIS

During 2002-2003 the 10 largest non-financial TNCs from South-East Europe and theCIS continued to expand both at home and abroadin terms of assets, sales and employment (tableI.8). Firms in natural resources and transportationdominate the list. The largest TNC, Lukoil, ranks

within the top 10 of the largest TNCs fromdeveloping countries (annex table A.I.11).

Russian TNCs dominate the list, but onaverage they are less transnationalized than thetop 50 TNCs from developing economies. Thesimple average TNI for the top 10 (36.6) is alsomuch lower than that for the top 50. Althoughthe sales-to-assets ratio is high, the ratio of salesto employment is much lower than for TNCs fromdeveloping economies.

e. The world’s top 50 financial TNCs

During the past decade or so, deregulationof financial services in Europe and NorthAmerica, technological change and competitivepressures have contributed to the creation offinancial conglomerates that provide bankingservices, mortgages, all lines of insurance, assetmanagement, and treasury and securities services.According to Fortune , the largest financialservices companies by revenues did not rankamong the top 50 of the world’s biggestcorporations in 1989. In 2003, the largestfinancial services company from Germany(Allianz) ranked 11th, and 13 financial groupsfrom the Triad (EU, Japan and the United States)were listed among the top 50 corporations in theworld in terms of revenues.36

The rise in the value of the assets offinancial TNCs in the 1990s is mainly attributedto growth through M&As. The growth oftransnational financial conglomerates is notconfined to developed economies: foreignparticipation in the financial sectors of emergingmarkets also increased rapidly during the 1990sparticularly in Latin America, the new EUmember countries and South-East Europe.Mexico alone accounted for about 50% of thecumulative FDI flows in financial services inLatin America and the Caribbean region from1990 to 2003. The new EU members andcountries in South-East Europe became majorrecipients of FDI flows in the financial industrywhen privatizations and preparations for EUmembership took place in the second half of the1990s. The proportion of cross-border M&As inthe financial sectors of Asia has been smallcompared to other regions (BIS 2004).

Large groups dominate world financialservices, not only in terms of total assets but alsoin terms of the number of countries in which theyoperate.37 This year, for the first time, WIR

0

10

20

30

40

50

60

70

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Top 100 Top 50

Table I.7 Measures of efficiency andproductivity of the world’s top 100

and developing countries’ top 50 TNCs,2002, 2003

Top 100 Top 50

Measure 2002 2003 2002 2003

Sales/assets 68.9 69.3 66.4 72.0Sales/employmenta 0.33 0.38 0.21 0.16

Source: UNCTAD/Erasmus University database.a In millions of dollars per employee.

20 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Table I.8. Snapshot of the top 10 TNCsfrom SEE and CIS: assets, sales and

employment, 2002, 2003(Billions of dollars, thousands of

employees, per cent)

Variable 2002 2003 % change

AssetsForeign 8.4 12.0 43.6Total 42.7 48.9 14.6Foreign as % of total 19.7 24.6 4.9a

SalesForeign 14.5 24.9 72.0Total 23.7 44.1 86.3Foreign as % of total 61.2 56.5 -4.7a

EmploymentForeign 19.1 39.9 108.4Total 382.3 469.0 22.7Foreign as % of total 5.0 8.5 3.5a

Source: UNCTAD/Erasmus University database.a In percentage points.

introduces a list of the top 50 largest financialTNCs. These are ranked by total assets since dataon foreign assets, foreign sales or foreignemployment are not available.

TNCs from five countries (France,Germany, Japan, the United Kingdom and theUnited States) dominate the list, accounting for70% of all companies in the top 50 and 74% oftheir total assets. However, there are companiesfrom seven different countries in the top 10,accounting for 34% of total assets. In addition,

the top 10 companies account for only 26% oftotal employment (annex table A.I.12).

The degree of transnationality of financialTNCs can only be measured by the physicalspread and location of their operations. TheInternationalization Index shows that, on average,46% of the affiliates of the top 50 financial TNCsare located abroad. The index is highest forfinancial groups from Switzerland that facedomestic growth constraints due to the small sizeof the domestic market, amd have built up strongcompetitive advantages over a long period oftime. The top 50 financial TNCs have, onaverage, affiliates in 25 countries. The largestshare of affiliates is in Europe (figure I.9). Thereis a strong correlation between the size of acompany and its transnationalization: the top 10companies on the list have, on average, 58% oftheir affiliates located abroad in 44 countries,while the average for the whole group of affiliatesis 43% in 25 host countries.

3. FDI performance and potential

The UNCTAD Inward FDI Performance38

and Potential39 Indices, as well as the OutwardFDI Performance Index,40 showed somenoticeable changes for individual countries in2004, reflecting uneven developments of FDIinflows and improvements in general economicperformance (annex tables A.I.13-A.I.14).

The Inward FDI Performance Index fordeveloping countries as well as the transitioneconomies of South-East Europe and the CIS

Figure I.9. Distribution of foreign affiliates of the 50 largest financial TNCs, 2003

Source: UNCTAD, based on Who Owns Whom database (London: Dun & Bradstreet).

15 690

3 371

57984

68837

63772

0

2 000

4 000

6 000

8 000

10 000

12 000

14 000

16 000

Europe North America Other

developed

countries

Africa Asia and

Oceania

West Asia Latin America

and the

Caribbean

South-East

Europe and

CIS

21CHAPTER I

improved in 2004,41 notably in South, East andSouth-East Asia, South-East Europe and the CIS(table I.9). However, it worsened in developedcountries compared to 2003, although as a groupthey were well ahead of developing countries(table I.9). The United States, where FDI inflowsrose by 69% in 2004, had a lower PerformanceIndex and ranked at 114th out of 140 countriesin the world, due to its lower FDI flows in 2002-2003; these are taken into account in the 2004index (see annex table A.I.13 for rankings of all140 countries). Denmark, the Netherlands,Portugal and Sweden fell by more than30 positions in the country rankings (figure I.10).With large negative FDI inflows in 2004,Denmark fell by nearly 100 positions and wasranked second from the bottom. The top positionin 2004 was held by Azerbaijan due to large oil-related FDI flows relative to the small size ofits economy. In 2004, Tajikistan rose the mostin the country rankings to 19th in the world (tableI.10), reflecting a significant increase of FDI

inflows in mining in 2002-2004 (annex tableB.1).

In contrast to the changes in rankings bythe Inward FDI Performance Index (see annextable A.I.13 for rankings of all 140 countries),there were almost no changes in the Inward FDIPotential Index rankings of the top rankedcountries between 2002 and 200342 (table I.11).This reflects the stabili ty of the structuralvariables comprising the Index. In other words,this index shows how the structural variablesmove in relation to each other. Comparing therankings by the Potential Index with those of thePerformance Index gives an indication of howeach country performs against its potential.Countries in the world can be divided into thefollowing four categories: front-runners(countries with high FDI potential andperformance); above potential (countries with lowFDI potential but strong FDI performance); belowpotential (countries with high FDI potential butlow FDI performance); and under-performers

(countries with both low FDIpotential and performance (tableI.12). The data for thiscategorization are limited to 2003(due to unavailability of the 2004data for the Potential Index), thelast year of the global FDIdownturn period. As in past years,there are no significant changesin the first and last groups, withmany developed and newlyindustrializing economies in theformer and many LDCs or poordeveloping countries in the latter.The second and third groups alsoinclude mostly the same countriesas in the previous year. Thequestion remains for the above-potential countries as to how theycan continue to sustain their FDIperformance at levels comparablewith those of the past whileaddressing structural problems(i.e. FDI potential). The concernfor the below-potential countries,on the other hand, is how theycould raise their FDI performanceto match their potential.

Performance in FDI outflowsrelative to the size of economiesas measured by the Outward FDIPerformance Index (annex table

Table I.9. Inward FDI Performance Index, by region,1990, 2003, 2004a

Region 1990 2003 2004

World 1.000 1.000 1.000Developed countries 1.022 0.947 0.891

Western Europe 1.310 1.837 1.625European Union 1.310 1.866 1.647Other Western Europe 1.307 1.261 1.175

North America 1.129 0.474 0.402Other developed countries 0.290 0.202 0.372

Developing countries 0.977 1.187 1.353Africa 0.731 1.253 1.226

North Africa 0.847 0.925 1.031Other Africa 0.650 1.508 1.360

Latin America and the Caribbean 0.898 1.394 1.523South America 0.741 1.399 1.648Other Latin America and the Caribbean 1.302 1.386 1.359

Asia and Oceania 1.075 1.092 1.306Asia 1.063 1.092 1.306

West Asia 0.141 0.415 0.478South, East and South-East Asia 1.312 1.230 1.482

South Asia 0.115 0.320 0.418East and South-East Asia 1.735 1.444 1.729East Asia 1.193 1.523 1.821

South-East Asia 3.104 1.180 1.423Oceania 7.358 0.936 0.795

South-East Europe and CIS 0.955 b 1.254 1.787South-East Europe 0.835 b 2.273 3.064CIS 0.981 b 1.044 1.533

Source: UNCTAD.a Three-year moving average, using data for the three years ending with the

year in question.b As most of the countries in this region did not exist in their present form before

1992, the period for the index is 1992-1994.

22 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

A.I.14) shows some changes in country positionsin 2004 as compared with those in 2003. Thereare three newcomers to the list of the top 20outward investment economies: Australia, Austriaand Estonia (table I.13). However, Denmark,Finland and Ireland are no longer in the list,unlike other small economies that rank relativelyhigh. Denmark and Finland also fell in rankingon the Inward FDI Performance Index in 2004.

B. Policy developments

1. National policy changes

With a view to upgrading or enhancingtheir ability to attract and benefit from FDI,countries are continuing to adopt measuresintended to improve their investment climates.

In 2004, both the number of national policymeasures affecting FDI and TNCs that wereintroduced and the number of economies involvedin the process increased. A total of 271 newmeasures were adopted by 102 economies (tableI.14).

The vast majority (87%) of regulatorychanges tended to make conditions morefavourable for foreign companies to enter andoperate. Most of these measures implied furtherliberalization of investment regimes; 95 involvednew promotional efforts (including various typesof incentives) and 37 greater investor protection.In terms of regional distribution, Asia andOceania accounted for 30% of the new measures,followed by the transition economies (22%),Africa (21%), developed countries (14%) andLatin America and the Caribbean (13%).

While the trend towards morewelcoming policies for FDI continued, 36were less favourable in 2004 – an unusuallyhigh share. This is the highest numberreported since UNCTAD started monitoringchanges in national laws in 1991. In LatinAmerican and the Caribbean countries, asmany as 24% of all changes wereunfavourable, and the share was alsorelatively high in Africa (19%). In terms oftheir nature, 11 involved less promotionalefforts (e.g. making incentives less generous),9 involved new restrictions to FDI entry andestablishment, while 5 affected the operationsof foreign investors. The relatively highincidence of such measures may reflect thegrowing disappointment of many developingcountries in the ability of liberalization,generous incentives and promotion to attractthe level of FDI inflows that is commensuratewith their potential.

An area in which many changes wereundertaken in 2004 was corporate taxation.Reflecting the growing competition for FDI(as well as the need to stimulate investmentgenerally), significant reductions in corporateincome tax rates were noted in manycountries.43 According to UNCTAD’sfindings, about 20 economies reduced theircorporate income tax rates during 2004 (tableI.15) – nine were developed economies, fivetransition economies and six developingeconomies. From a regional perspective,developed countries as a group showed themost significant reduction in their average

Figure I.10. Largest gains and losses in inwardFDI performance, 2003-2004a

(Changes in country ranking)

Source: UNCTAD, based on annex table A.I.13.a Three-year moving average, using data for the three years ending

with the year in question.

- 100 - 50 0 50 100

Tajikistan

Congo, Democratic Republic of

Gabon

Australia

Jordan

Bahrain

Kyrgyzstan

Zambia

Latvia

Malta

Romania

TFYR Macedonia

Bolivia

France

Canada

Germany

Morocco

Portugal

Sweden

South Africa

Netherlands

Denmark

23CHAPTER I

corporate tax rate from 29.7% to 26.5% (KPMG2005). Among individual economies, Romaniamade the largest tax cut, from 25% to 16%,followed by Uruguay and Bulgaria. Only threecountries reported increased rates (Germany,India and Viet Nam).

Corporate taxes may affect a country’sinternational attractiveness in the eyes of foreigninvestors (OECD 2002a).44 Studies show thatlocation of FDI is becoming more sensitive totaxation, and that corporate income tax rates caninfluence a TNC’s decision to undertake FDI,especially if competing jurisdictions have similar“enabling conditions”. For instance, EU investorswere found to increase their FDI positions inother EU member States by approximately 4%if the latter reduced their effective corporate

income tax rates by one percentage point relativeto the European mean (Gorter and Parikh 2003).

While policy changes overall are in thedirection of more liberalization and deregulation,there are some differences between regions. FDIpolicy changes at the regional level are describedin the analysis of regional trends in chapter II.

2. International investmentagreements

The past year saw a further proliferationof international investment agreements (IIAs)45

at the bilateral, regional and interregional levels.Several developments are worth noting in thiscontext. First, the universe of bilateral investmenttreaties (BITs) and bilateral double taxationtreaties (DTTs) continued to expand, albeit at a

Table I.10. Rankings by the Inward FDI Performance Index, 2004 a

1 Azerbaijan 36 Tanzania, United Republic of 71 Ukraine 106 Thailand2 Belgium and Luxembourg 37 Mali 72 Macedonia, TFYR 107 Paraguay3 Brunei Darussalam 38 Zambia 73 El Salvador 108 Egypt4 Angola 39 Syrian Arab Republic 74 New Zealand 109 Korea, Republic of5 Ireland 40 Australia 75 Poland 110 Oman6 Gambia 41 Botswana 76 Iceland 111 Turkey7 Hong Kong, China 42 Albania 77 Kyrgyzstan 112 India8 Singapore 43 Bolivia 78 United Kingdom 113 Zimbabwe9 Mongolia 44 Nigeria 79 Mexico 114 United States

10 Congo 45 China 80 France 115 Burkina Faso11 Kazakhstan 46 Hungary 81 Portugal 116 Libyan Arab Jamahiriya12 Bulgaria 47 Latvia 82 Argentina 117 Myanmar13 Georgia 48 Jordan 83 Israel 118 Germany14 Cyprus 49 Spain 84 Malta 119 Malawi15 Trinidad and Tobago 50 Viet Nam 85 Guinea 120 Guatemala16 Estonia 51 Costa Rica 86 Venezuela 121 Saudi Arabia17 Jamaica 52 Bahamas 87 Côte d'Ivoire 122 Bangladesh18 Sudan 53 Honduras 88 Russian Federation 123 Madagascar19 Tajikistan 54 Uganda 89 Austria 124 Rwanda20 Congo, Democratic Republic of 55 Finland 90 Lebanon 125 Taiwan Province of China21 Chile 56 Malaysia 91 Ghana 126 South Africa22 Armenia 57 Gabon 92 Papua New Guinea 127 Kenya23 Mozambique 58 Dominican Republic 93 Sweden 128 Niger24 Ethiopia 59 Lithuania 94 Canada 129 Greece25 Slovakia 60 Slovenia 95 Algeria 130 Iran, Islamic Republic of26 Moldova, Republic of 61 Switzerland 96 Sri Lanka 131 Sierra Leone27 Bahrain 62 Brazil 97 Benin 132 Yemen28 Czech Republic 63 Qatar 98 Italy 133 Haiti29 Panama 64 Peru 99 Belarus 134 Japan30 Nicaragua 65 Morocco 100 Philippines 135 Nepal31 Guyana 66 Togo 101 Senegal 136 Indonesia32 Namibia 67 Tunisia 102 Pakistan 137 Cameroon33 Croatia 68 Netherlands 103 Norway 138 Kuwait34 Ecuador 69 Colombia 104 United Arab Emirates 139 Denmark35 Romania 70 Uruguay 105 Uzbekistan 140 Suriname

Source: UNCTAD, based on annex table A.I.13.a Three-year moving average, using data for the three years ending with the year in question.

24 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

slower pace than in previous years. Second,international investment rules are becomingincreasingly sophisticated and complex incontent, and are also being formulated as partof agreements that encompass a broader rangeof issues (including trade in goods and servicesas well as the movement of other factors ofproduction). Third, among the new BITs, someare re-negotiated treaties that replace earlier BITsbetween the same partners, either because theoriginal treaty has reached its expiry date orbecause of changed circumstances. Fourth, South-South cooperation in the area of internationalinvestment policy is intensifying. And fifth, thereis a marked rise in investor-State disputes. Asa result of these developments, countries andfirms have to operate within an increasinglycomplicated framework of investment rules thatis both multilayered and multifaceted, withoverlapping obligations and commitments as wellas gaps in its coverage.

Table I.11. Top 25 economies by theInward FDI Potential Index,

1990, 2002, 2003 a

Economy 1990 2002 2003

United States 1 1 1Norway 5 2 2United Kingdom 3 3 3Canada 2 5 4Singapore 15 4 5Sweden 6 7 6Qatar 19 6 7Germany 4 10 8Belgium and Luxembourg 10 8 9Ireland 27 9 10Netherlands 8 11 11France 7 15 12Finland 9 12 13Iceland 14 14 14Hong Kong, China 20 13 15Japan 13 16 16Switzerland 11 18 17Denmark 16 17 18Australia 12 21 19Korea, Republic of 21 19 20Taiwan Province of China 22 20 21United Arab Emirates 26 22 22Israel 31 23 23Austria 18 24 24Spain 24 25 25

Source: UNCTAD, based on annex table A.I.13.a Three-year moving average, using data for the three

years ending with the year in question.

a. Bilateral investment treaties

The number of BITs worldwide hascontinued to expand over the past year, but ata slower pace than before. During 2004, 73 newBITs were concluded, 10 of which replacedearlier BITs, bringing the total number to 2,392(figure I.11). However, this represents aslowdown in the conclusion of BITs since 2001.The largest number of the new BITs signed during2004 was between developing countries, with 28BITs or 38% of the total, followed closely byBITs between developed and developingcountries with 27 of all BITs signed.

As of the end of 2004, the share of BITssigned between developed and developingcountries in total BITs worldwide was 40%. BITsconcluded among developing economiesaccounted for 25%, while those betweendeveloping and transition economies (South-EastEurope and CIS) rose to 10% of the total (figureI.12). BITs typically are not concluded betweendeveloped economies because, with a fewexceptions, investment relations between thesecountries are traditionally governed by otherinternational instruments.46 Developed countriesdominate the list of economies with the highestnumber of BITs. Only two countries within thetop ten are developing economies (figure I.13).

Within the South-South BITs universe,China, Egypt, the Republic of Korea andMalaysia have each signed more than 40 treatieswith other developing countries. Each of thesefour countries has signed more agreements withother developing countries than with developedcountries. The recent increase in developing-country BITs reflects a greater emphasis onSouth-South cooperation on investment, as wellas the rise of outward FDI from developingcountries (UNCTAD forthcoming a).

Not all BITs signed are in force (i .e.ratified and/or enacted). In fact, only about 70%of the 2,392 BITs signed by the end of 2004 werein force. For 46% of the BITs that had not enteredinto force, the time period since signatureexceeded five years (i.e. longer than the averageperiod of two to three years that it takes to ratifya BIT and for i t to enter into force). Thisproportion is higher for BITs concluded bydeveloping economies: 51% of them exceed thefive-year span. The same ratio for BITs concludedby LDCs is 33% (UNCTAD forthcoming b). This

25CHAPTER I

Table I.12. Matrix of inward FDI performance and potential, 2003a

High FDI performance Low FDI performance

Front-runners Below potential

Source: UNCTAD.a Three-year moving average, using data for the three years ending with the year in question.

Albania, Angola, Armenia, Azerbai jan, Bol iv ia,Colombia, Congo, Ecuador, Ethiopia, Gambia,Georgia, Guyana, Honduras, Jamaica, Mal i ,Mongol ia, Morocco, Mozambique, Namibia,Nicaragua, Nigeria, Peru, Republic of Moldova,Romania, Sudan, Syr ian Arab Republ ic, TFYRMacedonia, Togo, Uganda, the United Republic ofTanzania and Zambia.

Bahamas, Bahrain, Belg ium and Luxembourg,Botswana, Brazil, Brunei Darussalam, Bulgaria,Chi le, China, Costa Rica, Croatia, Cyprus, theCzech Republic, Denmark, Dominican Republic,Estonia, Fin land, France, Hong Kong (China),Hungary, I re land, Israel , Kazakhstan, Latv ia,L i thuania, Mexico, the Nether lands, Panama,Portugal, Qatar, Singapore, Slovakia, Slovenia,Spain, Sweden, Switzerland, Trinidad and Tobago,Tunisia and Viet Nam.

High FDI potential

Low FDI potential

Argent ina, Austral ia, Austr ia, Belarus, Canada,Germany, Greece, Iceland, Islamic Rep.of Iran, Italy,Japan, Jordan, Kuwai t , Lebanon, L ibyan ArabJamahiriya, Malaysia, Malta, New Zealand, Norway,the Oman, the Philippines, Poland, the Republic ofKorea, the Russian Federat ion, Saudi Arabia,Taiwan Province of China, Thailand, Ukraine, UnitedArab Emirates, the United Kingdom and the UnitedStates.

Above potential Under-performers

Alger ia, Bangladesh, Benin, Burk ina Faso,Cameroon, the Democratic Republic of the Congo,Côte d’Ivoire, Egypt, El Salvador, Gabon, Ghana,Guatemala, Guinea, Haiti, India, Indonesia, Kenya,Kyrgyzstan, Madagascar, Malawi, Myanmar, Nepal,Niger, Pakistan, Papua New Guinea, Paraguay,Rwanda, Senegal, Sierra Leone, South Africa, SriLanka, Sur iname, Taj ik istan, Turkey, Uruguay,Uzbekistan, Venezuela, Yemen and Zimbabwe.

Table I.13. Outward FDI PerformanceIndex for the 20 leading investor

economies, 1990, 2003, 2004a

Rank Economy 1990 2003 2004

1 Belgium and Luxembourg 2.740 22.331 20.0702 Panama 7.800 9.479 9.7913 Hong Kong, China 3.451 3.526 7.0024 Azerbaijan .. 3.313 6.535

5 Iceland 0.067 1.937 5.6046 Bahrain 0.588 2.244 3.7747 Singapore 2.961 5.792 3.5268 Sweden 4.649 2.499 2.8709 Switzerland 3.525 2.485 2.786

10 Spain 0.439 2.390 2.649 11 Netherlands 3.965 4.623 2.62712 Cyprus 0.037 1.915 2.28213 Canada 0.926 1.835 2.01414 United Kingdom 3.034 1.822 1.79915 Portugal 0.165 1.800 1.69716 France 1.890 2.097 1.57417 Austria 0.609 1.205 1.43118 Australia 0.970 1.347 1.38019 Botswana 0.069 1.824 1.33220 Estonia .. 1.172 1.123

Source: UNCTAD.a Three-year moving average, using data for the three

years ending with the year in question.

Notes: Economies are ranked in descending order of theirperformance index in 2002-2004.

reflects, among other things, the fact that theformal requirement for the ratification andenactment of BITs varies from country to countryaccording to their constitutions and legislativeprocedures. In some countries, for example, theratification of a treaty may require the enactmentof an implementing legislation, which in turn mayrequire major adaptations of relevant legislation.In other countries, ratification and entry intoforce of international treaties takes place onlyafter a certain number of treaties ready to beratified have been accumulated. Non-ratificationmay also be due to lack of coordination andcommunication within the government, changesin government and/or changes in governmentpolicy, political upheaval, civil unrest or war,or a deliberate policy choice of the government.

It is important to note in this context thatthe signature of a treaty itself has legalimplications for its parties. According to Article18 of the Vienna Convention on the Law ofTreaties, “A State is obliged to refrain from actswhich would defeat the object and purpose of atreaty when: (a) it has signed the treaty or hasexchanged instruments constituting the treatysubject to ratification, acceptance or approval,until it shall have made its intention clear not

26 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

to become a party to the treaty; or (b) it hasexpressed its consent to be bound by the treaty,pending the entry into force of the treaty andprovided that such entry into force is not undulydelayed”.

Two issues arise. The first concerns theapplicability of the substantive provisions of atreaty even though not ratified. The second issueconcerns the availabili ty of recourse for aninvestor or a government to international

Table I.15. Changes in corporateincome tax rates in selected

economies, 2004(Per cent)

Economy 1 January 2004 1 January 2005

DecreaseAlbania 25.00 23.00Austria 34.00 25.00Barbados 33.00 30.00Bulgaria 19.50 15.00Czech Republic 28.00 26.00Denmark 30.00 28.00Finland 29.00 26.00France 34.33 33.83Greece 35.00 32.00Israel 36.00 34.00Japan 42.05 40.69Korea, Republic of 29.70 27.50Latvia 19.00 15.00Mexico 33.00 30.00Netherlands 34.50 31.50Romania 25.00 16.00Singapore 22.00 20.00Switzerland 24.10 21.30Turkey 33.00 30.00Uruguay 35.00 30.00

IncreaseGermany 38.29 38.31India 35.875 36.5925Viet Nam 26.00 28.00

Source: UNCTAD, based on national sources and KPMG,2005.

arbitration. While the case law on this matteris limited,47 it appears that it could be difficultfor an investor or a government to invoke consentto arbitration under a treaty that has not yetentered into force.

It is also worth noting that countries areincreasingly renegotiating their existing BITs.While BITs generally provide for tacit renewalafter their expiration, in some cases countriesundertake re-negotiation of these agreements,either to obtain stronger commitments or becauseof the need to make existing BITs comply withthe parties’ commitments made under otherinvestment agreements.48 In such cases, the newBIT supersedes the earlier one. The trend towardsrenegotiation accelerated in the late 1990s andcontinued at an increasing pace thereafter,reaching 34 renegotiated BITs by the year 2000,and over 85 renegotiations by 2004.

Some of the BITs concluded most recentlymay have been influenced in some respect by theexperience in the application and implementationof the investment chapter of the North AmericanFree Trade Agreement (NAFTA) and of a fewother IIAs. The United States-Uruguay BIT(2004) and – to a lesser degree – the BIT betweenJapan and the Republic of Korea (2002) reflectthis phenomenon. In particular, some recent BITs(and BIT models) deviate from the traditionalopen-ended asset-based definition of investment,with a view to striking a balance betweenmaintaining a comprehensive investmentdefinition, on the one hand, and excluding fromcoverage those assets that are not intended bythe parties to fall under an agreement’s protectivewings, on the other.49

Furthermore, some recent BITs includesignificant revisions to the wording of varioussubstantive treaty obligations. For instance,drawing on the implementation legacy of the

Table I.14. National regulatory changes, 1991-2004

Item 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Number of countries that introduced changes in their investment regimes 35 43 57 49 64 65 76 60 63 69 71 70 82 102Number of regulatory changes 82 79 102 110 112 114 151 145 140 150 208 248 244 271 of which: More favourable to FDI a 80 79 101 108 106 98 135 136 131 147 194 236 220 235 Less favourable to FDI b 2 - 1 2 6 16 16 9 9 3 14 12 24 36

Source: UNCTAD database on national laws and regulations.a Includes liberalizing changes or changes aimed at strengthening market functioning, as well as increased incentives.b Includes changes aimed at increasing control, as well as reducing incentives.

27CHAPTER I

Figure I.11. Number of BITs and DTTs concluded, cumulative and annual,1990-2004

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).

investment chapter of NAFTA, the new modelBITs of Canada and the United States elaboratethe language and clarify the meaning ofprovisions dealing with absolute standards ofprotection. This is notably the case with themeaning of the minimum standard of treatment

concept in accordance with international law andthe concept of indirect expropriation.50

Some new BITs also address a broader setof issues, including not only specific economicaspects such as investment in financial services,but also other issues where greater policy spacefor host-country regulation may be sought. In thisregard, language is sometimes included to clarifythat the investment protection and liberalizationprovisions cannot be pursued at the expense ofthe protection of key public policy objectivessuch as health, safety, the environment and thepromotion of internationally recognized labourrights.

Finally, some recent BITs have madesignificant innovations regarding investor-Statedispute settlement procedures, in an effort tosecure greater transparency in arbitralproceedings, including open hearings, publicationof related legal documents and the possibility forrepresentatives of civil society to submit “amicuscuriae” (i.e. “friends of the court”) briefs toarbitral tribunals. In addition, other very detailedprovisions on investor-state dispute settlementare included in order to provide for more legallyoriented, predictable and orderly conduct at thedifferent stages of the ISDS process. Thus, forexample, the Canadian BIT model includesspecific standard waiver forms to facilitate thefiling of waivers as required by Article 26 of theAgreement for purposes of filing an ISDS claim.The United States-Uruguay BIT, on the other

0

50

100

150

200

250

19

90

19

91

19

92

19

93

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

20

03

20

04

0

500

1 000

1 500

2 000

2 500

3 000

BITs (total per year: left scale) DTTs (total per year: left scale)

Total BITs (cumulative: right scale) Total DTTs (cumulative: right scale)

Figure I.12. Total BITs concluded,by country group,a end 2004

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).a Due to the accession of ten countries to the EU on 1 May

2004, the BITs previously signed by those countries havebeen added to the BITs involving developed countries.

Note: SEE: South-East Europe.

13% 4%

25%

40%

10%

8%

Between developing countries

Between developed and developing countries

Between developing countries and countries of SEE and CIS

Between developed countries

Between developed countries and countries of SEE and CIS

Between countries of SEE and CIS

28 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

hand, not only provides for a special procedureavailable at the early stages of the ISDS processaimed at discarding frivolous claims or to seekinterim injunctive relief, but also envisages thepossibility to set up a mechanism for appellatereview, in order to foster a more consistent andrigorous application of international law inarbitral awards. A number of these proceduralissues have also been taken up in the debate aboutchanges to ICSID’s rules and regulations.51

b. Double taxation treaties

In 2004, 84 new DTTs were concludedbetween 79 countries. This represents a continuedgrowth of DTTs, albeit at a slightly slower pacecompared to 2003. The total number of DTTs roseto 2,559 by the end of 2004 (figure I.11). Austriaset the pace by concluding ten new DTTs,Azerbaijan concluded six, while South Africa andLithuania each concluded five. Unlike in the caseof BITs, the top ten economies in terms of numberof DTTs signed are all developed economies(figure I.14).

As of the end of 2004 about 39% of allDTTs were concluded between developed anddeveloping countries. DTTs among developedcountries accounted for 29%, another 19%involved countries in South-East Europe and theCIS and the remaining 13% were concludedamong developing economies (figure I.15).

As far as developing-country DTTs areconcerned, a trend can be observed that is similar,but less pronounced, than that of BITs regarding

Figure I.13. Top 10 signatories of BITs,end 2004

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).

increasing South-South investment cooperation.Although the first South-South DTT wasconcluded as early as 1948 (by Argentina andPeru), such DTTs proliferated only during thesecond half of the 1990s. During the 1990s, 156new DTTs were signed between 69 developingcountries, bringing the total number of South-South treaties to 256 by the end of 1999. Growthpersisted until 2004, with the number of South-South DTTs reaching 345 between 90 countries.

c. Other international investmentagreements

Besides BITs and DTTs, internationalinvestment rules are increasingly being adoptedas part of bilateral, regional and interregionalagreements that address trade and investmenttransactions. These agreements contain, inaddition to a range of trade liberalization andpromotion provisions, commitments to liberalize,protect and/or promote investment flows betweenthe parties. They respond to the increasing globalcompetition facing national economies forresources and markets. The number of suchagreements has been growing steadily, and byApril 2005 exceeded 212 (209 at the end of2004). The large majority of these agreements(about 87%) were concluded since 1990 (figureI.16). In 2004 and early 2005 at least 32 newagreements were concluded and about 66 otherswere under negotiation or consultation (annextables A.I.15 and A.I.16). Until the late 1980s,investment facilitation through these agreementsremained confined mainly to intraregional

Figure I.14. Top 10 signatories of DTTs,end 2004

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).

0 20 40 60 80 100 120 140

Czech Republic

Romania

Italy

Netherlands

Egypt

France

United Kingdom

Switzerland

China

Germany

Number of BITs

0 20 40 60 80 100 120 140 160 180

Canada

Denmark

Germany

Norway

Switzerland

Sweden

Netherlands

France

United Kingdom

United States

Number of DTTs

29CHAPTER I

processes, with some exceptions (e.g. earlyagreements between the European Communityand developing countries). Since 1990, countriesand groups located in different regions havebegun to conclude trade and investmentagreements with one another, with the result thatinterregional agreements now account for morethan half of the total, and for about 49% of the182 concluded since 1990.

The growth of IIAs (other than BITs andDTTs) is partly the result of two importantqualitative changes that took place during the1990s. First, these agreements, which previouslyhad been used mainly by countries at similarlevels of development, started to be concludedbetween developed and developing countries: byApril 2005, 81 had been signed (77 since 1990)and 39 were under negotiation (annex tableA.I.16). Second, there has also been a dramaticincrease in such agreements between developingcountries since the 1990s. By April 2005 at least70 of them had been signed (59 since 1990) andanother 24 were under negotiation, suggestingthat developing countries are increasinglypursuing development strategies based oncooperation among themselves.

Compared to BITs, these other IIAs showfar more variation in their scope, approach andcontent. Moreover, they increasingly encompassa broader range of economic transactions,including, notably, trade in goods and services,investment and capital flows, as well asmovement of labour. The more issues that areaddressed, the more complex the agreement, andthe greater the likelihood of overlaps andinconsistencies between provisions. At the sametime, their greater variation presents anopportunity for experimenting with differentapproaches to promoting international investmentflows that better reflect the special circumstancesof countries at different levels of economicdevelopment and in different regions. A numberof patterns have emerged concerning investmentprovisions in recent IIAs, though with manysignificant variations.

With respect to investment liberalization,IIAs other than BITs and DTTs have typicallyfollowed two main approaches. One is to providefor actual l iberalization subject to a l ist ofcountry exceptions (negative list approach). Thisapproach is typical of most agreements signedbetween countries of the Western Hemispherefollowing the NAFTA model. The secondapproach is to provide for the progressiveabolition of restrictions to the entry,establishment and operation of investment. Thispattern has been followed notably in theagreements between the European Communityand third countries, as well as by the membersof the Association of South-East Asian Nations(ASEAN) in the Framework Agreement on the

Figure I.16. The growth of internationalinvestment agreements other than BITs and

DTTs, 1957-2004(Number)

Source: UNCTAD (www.unctad.org/iia).

Figure I.15. Total DTTs concluded,by country group,a end 2004

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).a Due to the accession of ten countries to the EU on 1 May

2004, the DTTs previously signed by those countries havebeen added to the DTTs involving developed countries.

Note: SEE: South-East Europe.

1957-1979 1980-1989 1990-1999 2000-2004

0

50

100

150

200

250

Annual Cumulative

Between developing countries

Between developed and developing countries

Between developing countries and countries of SEE and CIS

Between developed countries

Between developed countries and countries of SEE and CIS

Between countries of SEE and CIS

11%

29%

6%

39%

13%

2%

30 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

ASEAN Investment Area and several agreementssigned by ASEAN members with third countries.Under the latter approach, the level ofliberalization sought varies considerably. Whilesome agreements commit to achieving fullliberalization of investment by a particular date(e.g. the ASEAN Investment Area), others aimat completing the process of investmentliberalization in several stages (e.g. the EuropeAssociation Agreements signed by the EuropeanCommunity with Central European countries).Still others establish a framework for futurenegotiations to liberalize investment (e.g. theEuro-Mediterranean Agreements signed betweenthe European Community with countries inNorthern Africa and the Middle East; the AfricanEconomic Community; the ASEAN Agreementwith China).

The more recent agreements that providefor investment protection in addition toliberalization, concluded by countries such asChile, Japan, Singapore, Morocco and the UnitedStates, are more comprehensive, detailed, and,for the most part , more rigorous than priorNAFTA-style agreements. While theseagreements address many of the same topics, theyalso deal with additional issues, or modify theNAFTA approach to these issues on the basis ofaccumulated experience. They typically dealextensively with trade in services, while separatechapters or provisions are devoted to topics suchas competition policy, government procurement,intellectual property rights, labour, environment,trade and investment in particular industries,temporary entry for business persons, andtransparency.

On the other hand, other recent agreementshave remained narrow in their coverage ofinvestment issues, l imiting themselves toestablishing a framework for cooperation oninvestment promotion. Recent examples includethe free trade agreements signed between themembers of the European Free Trade Association(EFTA) and Central European countries, bilateralagreements between Canada and countries invarious regions, as well as a number offramework agreements on trade and investmentrelations between the United States and countriesin Africa and the Middle East. The cooperationprovided for under the latter type of agreementsis typically aimed at creating favourableconditions for encouraging investment, notablythrough the exchange of information. It is also

common for such agreements to set upconsultative committees, or a similar institutionalarrangement involving the parties, to follow upon the implementation of negotiated commitmentsand to discuss and study possible obstacles tomarket access for trade and investment.

d. International investment disputes

A new and significant development is therise of investor-State disputes. These involve thewhole range of investment activities and all kindsof investments, including privatization contractsand State concessions.52

Numerous IIAs allow investors to choosebetween the arbitral proceedings of the WorldBank Group’s International Centre for Settlementof Investment Disputes (ICSID) (includingICSID’s Additional Facili ty) and ad hocarbitration procedures, using arbitration rules ofthe United Nations Commission on InternationalTrade Law (UNCITRAL) for example. Otherinstitutional facilities available for use are theInternational Chamber of Commerce (ICC) Courtof Arbitration in Paris, the Stockholm Chamberof Commerce Arbitration, the London Court ofInternational Arbitration and various regionalarbitration centres, particularly in Singapore andCairo. However, only ICSID provides a list ofcases. And even under ICSID, decisions of thetribunals have not all been made public. Whilethis situation may gradually be changing, it isnot possible to know the actual number of casesto date, nor is it possible to learn about the legalissues or factual circumstances theyencompassed.

The cumulative number of treaty-basedcases brought before ICSID and other arbitrationfora has been rising dramatically over the pastfive years, reaching 171 known claims byDecember 2004 and at least 183 by June 2005.53

At least 57 governments – 36 of them ofdeveloping countries, 12 of developed countriesand 9 of South-East Europe and the CIS – areinvolved in investment treaty arbitration.Argentina leads them all with 40 claims, 37 ofwhich relate at least in part to that country’sfinancial crisis. Mexico has the second highestnumber of known claims (15), most of themfalling under NAFTA and a handful under variousBITs. The United States has also faced a sizeablenumber (10), all of them pursuant to NAFTA.Poland (7 claims), Egypt (6) and the Russian

31CHAPTER I

Federation (6) also figure prominently, along withnine countries that have each faced four claims:Canada, Chile, the Czech Republic, theDemocratic Republic of Congo, Ecuador, India,Kazakhstan, Ukraine and Venezuela.

This rise in investment disputes poses aparticular challenge for developing countries. Thefinancial implications of the investor-Statedispute-settlement process can be substantial,both from the point of view of the costs of thearbitration proceedings and the awards rendered.Information about the level of damages beingsought by investors tends to be patchy andunreliable. Even ascertaining the amounts soughtby foreign investors can be difficult, as most ofthe cases are still at a preliminary stage and,under the ICSID system, claimants are notobliged to quantify their claims until after thejurisdictional stage has been completed. Claimsproceeding under other rules of arbitration arealso difficult to quantify. It is, nonetheless, clearthat some claims involve large sums.54

Furthermore, even defending against claims thatmay not ultimately be successful costs money.A cursory review of cost decisions in recentawards suggests that the average legal costsincurred by governments are between $1 millionand $2 million including lawyers’ fees, the costsfor the tribunal of about $400,000 or more, andthe costs for the claimant, which are about thesame as for the defendant.55

The surge in investment disputes arisingfrom IIAs and the costs incurred from thesedisputes signify that governments that decide toenter into IIAs need to be judicious in negotiatingsuch agreements. They also need to follow thedevelopments of disputes in order to be sensitiveto actions that could trigger l i t igation.Furthermore, it is important to review experiencesin implementing international commitments inIIAs and to draw lessons from them.

C. Prospects: further FDIgrowth expected

Economic growth, continuing liberalizationof investment policies and trade regimes, andincreased competition among firms are likely todrive the global expansion of TNC activity.Following slow growth or recession during 2002-2003, the world economy has entered a periodof recovery. Projections indicate that world realGDP, which grew by 5.1% in 2004, will increasemore moderately, by 4.3% in 2005 and 4.4% in

2006 (IMF 2005). The rate of growth is likelyto slow down in developed countries from 3.4%to 2.6% in 2005 and 3.0% in 2006, while stillregistering a high level in developing countriesof above 6% during 2005-2006. Estimates by theUnited Nations and the World Bank corroboratethese projections (UNDESA-UNCTAD 2005,World Bank 2005a). With the substantial increaseregistered in the rate of world economic growthsince 2003, and moderate downward adjustmentsin projected growth, FDI flows should continueto rise, at least over the next couple of years.

Meanwhile, the slowdown of growth insome developed countries and structuralweaknesses, along with financial and corporatevulnerabilities in some regions, continue tohinder a strong recovery in FDI. Continuingexternal imbalances in some countries and sharpexchange rate fluctuations, as well as high andvolatile commodity prices, pose additional risksthat may also limit global FDI flows.

Looking at prospects by sector, FDI isexpected to pick up in natural resources,reflecting high demand for such resources partlystemming from China’s growing economy andthe opening up of new and potentially profitableopportunities, for instance in the oil and gasindustries. Announcements abound, for example,two Japanese general trading companies, Ito Chuand Mitsui, plan to invest jointly a total of $3billion in iron ore in Australia with BHP Billiton(Australia), while Rio Doce (Brazil) and RioTinto (Australia) plan to expand their productioncapacities in Brazil.56 The anticipated increasein the offshoring of services also augurs well forFDI in that sector. One exception istelecommunications: in the United States alone,a reduction of more than $2 billion in investmentin that industry is expected in 2006, in order torationalize investment after the merger boom.57

For developing countries overall, FDI inflowsin telecommunications are now well below theirhistorical highs in the 1990s (World Bank 2005b).Prospects for FDI in manufacturing are positiveoverall , especially as regards investment inspecial economic zones, encouraged by a varietyof incentives offered by most developingcountries.

The need for private financing ofinfrastructure in developing countries remainsstronger than ever, with new modalities ofinvestment (e.g. public-private partnerships thatare gaining in popularity). A recent study by theWorld Bank, the Japan Bank for International

32 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Cooperation (JBIC) and the Asian DevelopmentBank, for example, estimated that theinfrastructure financing needs of developingcountries in Asia will exceed $1 trillion over thenext five years.58 It is likely that countries willseek to attract FDI to meet at least part of theseneeds.

Trends in cross-border M&As also pointto increased investment activity. M&As, whichaccount for the largest proportion of FDI flowsto developed countries, rose in 2004 and areexpected to do so again in 2005. Almost 40% ofthe United States tax and finance executives andsenior professionals participating in a surveyundertaken by KPMG in 2004 predicted that thenumber of worldwide M&A transactions wouldexceed 30,000 in 2005.59 Nearly 90% ofrespondents indicated that their company expectsto complete at least one merger or acquisitionin 2005, compared with roughly 70% who saidso in 2004. In developing countries, greenfieldFDI is expected to increase as a proportion ofall FDI, as investment channelled viaprivatization is declining, and because severalcountries (e.g. India) are actively seeking thisform of investment via regulatory reforms andincentives.

Outward investment by TNCs based in anumber of developing countries is likely to growfurther. Like their counterparts in developedcountries, these TNCs are in search of resources,markets and technology, driven by the samefactors that determine FDI in countries with along history of outward investment (UNCTAD2005a). In some countries, government policiesseek to encourage this trend.

On the policy front, l iberalization iscontinuing, and has intensified in key developingeconomies such as China and India. China, whosetransition period in the context of the WTO iscoming to an end, has introduced legislationopening up several new industries to FDI (chapterII). India has also been opening up importantindustries, such as telecommunications,construction and real estate, to FDI (chapter II).At the same time privatization continues to winddown in many countries, especially in LatinAmerica and the transition economies of South-East Europe and the CIS; moreover, recentprivatization deals have also been smaller in size.While this reduces FDI potential via this channel,i t may lead to expansion and sequentialinvestment.

At the international level, the continuedtrend towards greater liberalization, in particular,the pursuit of negotiations on a number ofbilateral, regional and international agreements(chapters I.B and II), may facilitate increasedflows in years to come. On the trade front,eligibili ty under the African Growth andOpportunity Act (AGOA) has been extended to37 countries in Africa, while the Central AmericaFree Trade Agreement (CAFTA) is awaitingratification and the free trade agreement (FTA)between the Southern Common Market(MERCOSUR) and the Andean Pact was signedin 2004.

A number of specific policy developmentsin 2005 are also likely to have an impact on thesize and direction of FDI flows. First, a one-off tax amnesty on foreign earnings awarded bythe United States has already led toannouncements of the repatriation of sizeablefunds by several United States TNCs (chapterII). Had these earnings been reinvested, theywould have been counted as part of FDI outflowsfor 2005. This repatriation of earnings by firmsfrom the United States, the largest outwardinvestor in 2004, is likely to lead to a substantialdecline in United States FDI outflows. While theexact magnitude of the repatriation is difficultto predict, it will be a force holding back globalFDI flows.

Second, the value of the dollar will havean effect on all cross-border financial flows byTNCs, be they in the form of equity, earnings orloans. It is not certain at the time of writing howthe dollar exchange rate will develop. Forforeign-based TNCs, a dollar depreciation meansthat United States assets become cheaper. Forforeign affiliates of United States-based TNCs,this means that it is a good time to repay intra-firm dollar-denominated debt or repatriate foreignearnings. The appreciation of the United Statesdollar that started in 2005, if continued, will meanthe opposite. In any event, the net impact willdepend on the relative magnitudes of the currencyfluctuations.

Third, a likely outcome of the tsunamidisaster is increased investment, both domesticand foreign, in infrastructure in the affectedcountries over the next few years. During thereconstruction phase, foreign and domesticinvestors are expected to be called upon toparticipate in tenders for the rebuilding of largeinfrastructure projects such as seaports and power

33CHAPTER I

utilities. In both Indonesia and Sri Lanka, forexample, public-private partnerships, includingsome with foreign investors, are expected to playan important role in the rebuilding ofinfrastructure and in the revival of the tourismindustry.60 Complemented by foreign aid andgrants from multilateral and regional developmentbanks, these partnerships will boost foreigninvestor involvement in post-tsunamireconstruction.

A number of surveys confirm promisingprospects for FDI flows in 2005, and evenbeyond, although respondents do not seem to beas optimistic as they were last year. This is thecase, for instance, with the McKinsey GlobalSurvey of Business Executives Confidence Index(McKinsey 2005). This report revealed optimismamong the more than 9,300 business executivesfrom 130 countries surveyed; however their viewswere less positive than a year ago. The CEOBriefing 2005 compiled by the EconomistIntelligence Unit found that competition for

global offshoring is intensifying, with 57% ofexecutives viewing offshoring as a critical forcereshaping the global marketplace in 2005, upfrom 51% in 2004 (EIU 2005a). As regardsJapanese TNCs, the annual survey undertakenby JBIC found that about half of themanufacturing firms surveyed in 2004 wouldstrengthen and expand foreign operations in thefollowing three years and that 5% would reducethem (compared to 42% and 7%, respectively,in the 2003 survey) (JBIC 2005).

A survey undertaken by UNCTAD (box I.3)also points to increased world FDI flows in thenear future.61 Expectations, however, vary byregion, being more positive for developingregions such as Asia and Oceania than for otherregions (chapter II examines regional prospectsseparately). In the longer term, FDI is poised tocontinue its upward trend, although it may besome time before FDI flows reach levelscomparable to those of the late 1990s.

Box I.3. FDI prospects: results of UNCTAD’s survey

The overall findings of the 2005 UNCTADsurveya on FDI prospects is that prospects for FDIin 2005-2006 are promising, although forecastsare not as optimistic as in the 2004 survey (WIR04,p. 32). More than half of the responding TNCsand experts as well as four-fifths of the IPAsexpected short-term (2005-2006) growth in FDIflows, while almost all the remaining respondentsexpected FDI levels to be stable (box figure I.3.1).Only a small fraction expected that FDI woulddecrease in the immediate future.

Prospects for FDI vary significantly byindustry:b

• In the primary sector, FDI in mining andpetroleum is expected to increase: over two-thirds of the IPA respondents, and a slightlylower percentage of the experts, expectedimproved FDI prospects. This is not surprising,since demand for natural resources is forecastto remain strong (chapter II). Expectationsregarding FDI in agriculture were less upbeat,with less than half of the IPAs and only aquarter of the experts forecasting improvedprospects. This might be due to ongoing tradedisputes in agriculture, lack of furtherliberalization in this area, and the fact that thesector as a whole has traditionally attracted lessFDI. Source: UNCTAD (www.unctad.org/fdiprospects).

Box figure I.3.1. Prospects for globalFDI flows: responses of TNCs, experts

and IPAs, 2005-2006(Per cent)

/...

• In manufacturing, expectations are high forincreased flows in electrical and electronicproducts, machinery and equipment, and metalsand metal products. A majority of respondents(IPAs as well as experts) expected a growth ofFDI in these industries. On the other hand, thereis less optimism regarding prospects for FDI

0

10

20

30

40

50

60

70

80

90

100

TNCs Experts IPAs

Increase Remain the same Decrease

34 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Box I.3. FDI prospects: results of UNCTAD’s survey

flows in textiles and clothing, rubber andplastic products, non-metallic minerals ormedia and publishing.

• The FDI outlook for the services sectorcontinues to be more positive than that for themanufacturing and primary sectors. A majorityof the respondents – experts as well as IPAs– expected improved prospects in most serviceindustries. The industries expected to be atthe forefront of FDI growth in services includecomputing/ICT, public utilities (such as thegeneration and distribution of electricity, waterand gas), transportation and tourism-relatedservices.

In terms of the investment locationsselected as the most attractive, four of the topfive countries ranked by the percentage ofresponses from experts and TNCs combined, arein the developing world. China is considered themost attractive location by 85% of TNCs andexperts (box figure I.3.2). India’s high ranking,albeit with 30% fewer responses than China’s,is even more remarkable, given that FDI flowsto the country have been modest until recently.The United States, Germany, the United Kingdomand Canada (in the ranking by TNC responses)only made it to the lower half of the top tenrankings.

Views on the risks for global FDI differamong the three groups of respondents to the2005 survey (box figure I.3.3). Judging from the

rates of response, TNCs and FDI experts considerprotectionism and slow growth in developedcountries to be the major threats. Indeed, everyTNC respondent felt that potential trade frictioncould undermine FDI growth in 2005-2006. Thefact that TNCs and experts regarded protectionismas a major risk for global FDI growth is alsoevident from other parts of the survey. Forexample, the lowest number of respondentsexpected an “increase” in FDI in industriesrecently affected by trade disputes, such as textilesand agriculture.

In contrast, IPAs were more concerned aboutthe financial instability of major economies andthe volatility of raw material prices than aboutany other factors listed. This difference could wellbe due to the fact that a larger proportion of IPArespondents are from developing countries. It alsoexplains why “political instability and civil war”is the third greatest concern of IPAs accordingto the percentage of respondents, while the othertwo groups of respondents rank it last.

Countries employed a variety of measuresto attract FDI in 2004 (box figure I.3.4). Theoverwhelming majority of them plan to adoptfurther FDI policy measures in 2005-2006. Over95% of responding IPAs expect to employ newand different policy measures to compete for FDI,

including additional incentives,further liberalization and otherpromotion measures. Thissuggests that global and regionalcompetition for FDI is increasingand will continue to do so in thefuture. Furthermore, given thelimited resources at theirdisposal, most countries intendto use much more targetedapproaches to investmentpromotion.

The positive outlook forglobal FDI in the short term isdriven largely by the potentialof specific regions, primarilydeveloping regions along withSouth-East Europe and the CIS.UNCTAD surveys at the regional

level find that FDI growth is being led bydeveloping economies rather than by developedcountries. FDI prospects in each of the individualregions are discussed in chapter II.

/...

Source: UNCTAD (www.unctad.org/fdiprospects).a Countries are ranked according to the number of responses that rated

each as the most attractive location.

Box figure I.3.2. Most attractive global businesslocations: responses of experts and TNCsa

Responses fromexperts1. China (85%)

Responses fromTNCS

Responses from experts

1. China (85%)2. United States (55%)3. India (42%)4. Brazil (24%)5. Russian Federation (21%)6. United Kingdom (21%)7. Germany (12%)8. Poland (9%)9. Singapore (9%)10. Ukraine (9%)

1. China (87%)2. India (51%)3. United States (51%)4. ( %)Russian Federation 335. (20%)Brazil6. Mexico (16%)7. Germany (13%)8. United Kingdom (13%)9. Thailand (11%)10. Canada (7%)

Responses from TNCs

35CHAPTER I

Box I.3. FDI prospects: results of UNCTAD’s survey (concluded)

Box figure I.3.3. Major risks to global FDI flows,a 2005-2006(TNC, expert and IPA respondents)

Source: UNCTAD (www.unctad.org/fdiprospects).

a Percentage of respondents that considered each factor as important or veryimportant.

Box figure I.3.4. Investment policy measures to attract FDI: responses by IPAs

Source: UNCTAD (www.unctad.org/fdiprospects).

Source: UNCTAD (www.unctad.org/fdiprospects).

a UNCTAD’s survey on FDI prospects analyses expected future patterns of FDI flows at the global, regional, nationaland industry levels based on the perspectives of global investors, host countries and international FDI experts. The2005 Survey of FDI Prospects for 2005-2008 involved IPAs of 109 countries, 81 of the largest TNCs (ranked bythe size of their foreign assets) from developed, developing and transition economies as well as 74 internationalinvestment experts. Their replies are based on their perceptions.

b Only IPAs and FDI experts were questioned about the prospects for FDI by industry, since TNCs are generally notwell placed to provide forecasts for industries other than their own.

0 10 20 30 40 50 60 70 80 90 100

Price volatility of petroleum and other raw materials

Protectionism

Exchange rate volatility

Global terrorism threat

Slow growth in industrialized countries

Political instability and civil war

Financial instability of major economies

IPAs Experts TNCs

0 10 20 30 40 50 60 70 80 90

No new measures

Further liberalization

Additional incentives

Other promotion measures

Greater targeting

2005-2006 2004

36 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Notes1 In 2000 for instance, the gap between developed and

developing country FDI flows was $881 billion.2 Luxembourg was the largest recipient of FDI inflows

in the world in both 2002 and 2003 due to massiveFDI in special purpose entities (holding companies)that was transhipped to other countries (for details onthis kind of FDI, see WIR03, p. 69).

3 The fact that Central Asia is now excluded from theregion (box I.2) had a small effect (-$10 billion).

4 Countries are designated by the United Nations as “leastdeveloped” on the basis of national income per capita,human assets and economic vulnerability. This categoryincluded 50 countries as of May 2005. For more detailssee UNCTAD 2004a.

5 The figures refer to the number of primary activitiesof the projects.

6 The data must be interpreted with caution. They areover-stated for some economies, as they include round-tripping (which may, for example, be around 25% inthe case of Hong Kong, China); investment by foreignaffiliates of (typically) developed-country TNCsestablished in developing economies (investment thatis particularly large in economies such as Cyprus, HongKong (China), Mauritius, Singapore and a number oftax havens); and capital flight. On the other hand, otherfactors may lead to under-reporting of outflows.Moreover, firms from some developing economies arenot allowed to transfer funds from their home countries,but rather need to raise them locally or in internationalmarkets; in that case, the extent of their internationalproduction activities is not reflected in FDI statistics.

7 Some countries, however, are relaxing their policieson outward investment and are encouraging their firmsto go abroad as international players. The 9th sessionof the Commission on Investment, Technology andRelated Financial Issues of UNCTAD, 7-11 March2005, noted important aspects of the links betweenoutward FDI and the competitiveness of firms indeveloping countries as well as the role host- and home-country governments can play. See UNCTAD,“Emerging FDI from developing countries”, noteprepared by the UNCTAD secretariat for theCommission on Investment, Technology and RelatedFinancial Issues, TD/B/COM.2/64, 4 February 2005.

8 Greenfield investment refers to investment in newfacilities and the establishment of new entities throughentry as well as expansion, while M&As refer toacquisitions of, or mergers with, existing local firms.For both, data used in WIR are original data collectedby private firms (OCO Consulting for greenfieldinvestments and Thomson Financial for cross-borderM&As). Data on greenfield FDI from OCO Consulting’sLOCOmonitor database (www.locomonitor.com) includenew and expanding FDI projects worldwide, bothannounced and realized. The data are available from2002 onwards. For an explanation of the data on cross-border M&As used in WIR, see annex B, “Definitionsand sources”.

9 Data from UNCTAD’s cross-border M&A database.10 Information from OCO Consulting, LOCOmonitor

website (www.locomonitor.com).

11 Brazil, China, Hong Kong (China), India, Malaysia,Mexico, the Republic of Korea, Singapore, Thailand,the United Arab Emirates and Viet Nam. Bulgaria alsoreceived more than 100.

12 For definitions of each of these components of FDI,see “Definitions and sources” in Annex B of WIR05.

13 For developed countries, almost all of the FDI inflowsover the period 1995-2004 can be broken down intothe three components of FDI financing, whereas only54% of total FDI inflows into developing countriescan be classified under these three categories.

14 Based on data for 31 countries that account for about38% of the total FDI flows to developing countries.

15 More than 100% due to negative figures for the othercomponents.

16 More than 100% due to negative figures for the othercomponents.

17 More than 100% due to negative figures for the othercomponents.

18 The sum of the shares of equity capital and intra-company loans is more than 100% because of negativereinvested earnings.

19 Thus, if a parent company in the United States givesa loan to a foreign affiliate located in Germany theinterest income of the parent firm (received from theaffiliate located in Germany) is taxed in the UnitedStates at a low tax rate, whereas the interest paymentof the German affiliate can be deducted from itsrevenue, lowering its taxed profits in Germany.

20 Reinvested earnings represent additions to a directinvestor’s stake in its foreign affiliates. In the balanceof payments they are recorded, therefore, as FDIinflows into the host county of the foreign affiliates(with a positive sign). If foreign affiliates’ activitiesresult in losses, the direct investor’s equity claims onthe foreign affiliates decrease. The losses are recordedunder reinvested earnings in the balance of payments,but with a negative sign as it indicates a reduction ordisinvestment of accumulated FDI.

21 Data from Deutsche Bundesbank, Balance of PaymentsStatistics.

22 IMF 2005. The data on growth rates of the new EUmembers are obtained from Eurostat(www.eurostat.cec.eu.int).

23 The volume of world trade in goods and services in2004 grew by nearly 20%, much faster than in 2002and 2003 (5% and 16%, respectively) (table I.3; IMF2005), and well above the long-term trend.

24 According to PRS Group/International Country RiskGuide, the average of the composite risk ratings (basedon three factors – political, financial and economicrisks) of some 150 countries improved from 69 in 2003to 71 in 2004, and is expected to be 73 in 2005 and78 in 2009.

25 Many indicators in 2004 show more favourablebusiness and consumer sentiments than in 2003: in theUnited States, for example, the Personal ConsumptionExpenditure Price Index of the Department ofCommerce and the Consumer Sentiment Index of theUniversity of Michigan were up by 6% and 8.6%respectively; for the EU, the Economic SentimentIndicator was up by 9.1%, the Industrial ConfidenceIndicator by 64% and the Consumer ConfidenceIndicator by 25%, all of the European Commission;

37CHAPTER I

and in Japan, the Business Conditions Diffusion Indexwas up by 97% and the Consumer Confidence Indexby 17%.

26 The country risk is also one of the 12 variables usedby UNCTAD for constructing the FDI Potential Index.

27 For example, net profits of Japanese firms reached arecord high in the year ending March 2005 (31% largerthan in fiscal year 2003 for all firms listed in the stockmarkets – Nihon Keizai Shimbun, 1 June 2005) whilethose of the 500 largest firms in terms of sales of theUnited States and Europe improved by 12% and 71%respectively in 2004 (source: UNCTAD, based on datafrom Thomson One Banker).

28 Data from the World Federation of Exchanges(www.fibv.com).

29 Based on the Reuters-CRB-Index of 17 raw materials.30 Investment, commodity and exchange firms and dealers.31 Cross-border investments of private equity funds that

lead to an ownership of 10% or more are in most casesrecorded as FDI even if private equity funds do notalways have the motivation for a lasting interest ora long-term relationship with the acquired enterprise.The figures in the text refer to these investments.

32 In Germany, for instance, public communities andpublic entities also sold houses and apartments becauseof budgetary problems.

33 Data from UNCTAD cross-border M&A database.34 The Transnationality Index is calculated as the average

of the following three ratios: foreign assets to totalassets, foreign sales to total sales and foreignemployment to total employment.

35 UNCTAD’s calculations, based on data from Dun &Bradstreet, Who Owns Whom database.

36 Fortune, 26 July 2004, pp. F1-F10.37 According to the Wall Street Journal Market Data

Group, the top 30 companies represented 60% of totalassets of the top 100 largest public financial companiesin 2003, and the top 50 almost 77%.

38 The UNCTAD Inward FDI Performance Index is ameasure of the extent to which a host country receivesinward FDI relative to its economic size. It iscalculated as the ratio of a country’s share in globalFDI inflows to its share in global GDP.

39 The UNCTAD Inward FDI Potential Index is basedon 12 economic and structural variables measured bytheir respective scores on a range of 0-1 (raw dataavailable on www.unctad.org/wir). It is the unweightedaverage of scores on the following: GDP per capita,the rate of growth of GDP, the share of exports in GDP,telecoms infrastructure (the average of telephone linesper 1,000 inhabitants, and mobile phones per 1,000inhabitants), commercial energy use per capita, theshare of R&D expenditures in gross national income,the share of tertiary students in the population, countryrisk, exports of natural resources as a percentage ofthe world total, imports of parts and components ofelectronics and automobiles as a percentage of theworld total, exports in services as a percentage of theworld total, and inward FDI stock as a percentage ofthe world total. For the methodology for building theindex, see WIR02, pp. 34-36.

40 The UNCTAD Outward FDI Performance index iscalculated in the same way as the Inward FDI

Performance Index: the ratio of a country’s share inglobal FDI outflows to its share in world GDP.

41 A three-year moving average is used. Thus the dataused for calculating the 2004 index are for those of2002, 2003 and 2004.

42 Because of late availability of the data used for thePotential Index, the most recent available year is alwaysone year behind that for the Performance Index.

43 It should be noted that a reduction of the tax rate doesnot necessarily signify a lowering of the overall taxburden. For example, a widening of the tax base orless generous rules on depreciation may counteract alower rate.

44 Corporate tax incentives may be provided in a numberof ways, including tax holidays, statutory corporateincome tax reductions, enriched capital cost allowances,investment tax credits, reductions of withholding taxon dividends and the extension of imputation reliefto non-resident shareholders (OECD 2000).

45 IIAs include bilateral treaties for the promotion andprotection of investment (or bilateral investmenttreaties), treaties for the avoidance of double taxation(or double taxation treaties), other bilateral and regionaltrade and investment agreements as well as variousmultilateral agreements that contain a commitment toliberalize, protect and/or promote investment.

46 The number of BITs involving developed countries alsoincreased due to the accession of ten countries to theEU on 1 May 2004, whereupon the earlier BITs signedby these countries began to be counted as developed-country BITs. For the same reason, the total numberof BITs signed between transition economies andbetween these and developed and developing countriesshows a corresponding reduction.

47 See the case of Ceskoslovenska Obchodni Banka, A.S.v. the Slovak Republic, Decision on jurisdiction, 24May 1999, available at (www.worldbank.org/ICSID/cases).

48 BITs signed by Central European countries prior totheir accession to the EU in 2004 have been affectedby these countries’ EU membership. In thesecircumstances, the United States and the EuropeanCommission signed a Memorandum of Understanding(MoU) in September 2003 concerning the applicabilityand the preservation of BITs concluded between theUnited States and the new EU members or countriescandidates for accession (see WIR04, box II.20). Asimilar exercise is currently taking place with Canada.In addition, Finland renegotiated its BITs with China,Egypt and Ukraine.

49 For example, in the new Canada model BIT (2004),the open asset-based definition of investment wasreplaced by a comprehensive, but finite, definition ofinvestment. The recently negotiated BIT between theUnited States and Uruguay, on the other hand, optedto define the term “investment” in economic terms.Such a definition covers, in principle, every asset thatan investor owns and controls, but with the qualificationthat such assets must have the “characteristics of aninvestment” such as “the commitment of capital or otherresources, the expectation of gain or profit, or theassumption of risk”. This approach is complementedby the explicit exclusion of several kinds of assets from

38 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

the category of covered investment under the agreement(e.g. certain debt instruments).

50 For instance, the new treaty models make clear thatan adverse effect on the economic value of aninvestment does not per se establish that an indirectexpropriation has occurred. It is further stated that,except in rare circumstances, non-discriminatoryregulatory actions by a Party aimed at protectinglegitimate public welfare objectives, such as publichealth, safety, and the environment, do not constituteindirect expropriations.

51 See the ICSID website, www.worldbank.org/icsid.52 For an analysis in the rise of treaty-based investment

disputes, see UNCTAD forthcoming c.53 UNCTAD database on investor-State dispute-settlement

cases.54 For instance, the Czech Republic’s payout of some $270

million plus substantial interest in the Lauder case;the recent award in CSOB v Slovakia (29 December2004) of $824 million plus an additional $10 millionas partial contribution to CSOB’s costs; or Occidental’s2002 award against Ecuador of $71 million plusinterest.

55 Preliminary results of a CEPMLP/Dundee researchproject on economic analysis of transnational disputemanagement.

56 Nihon Keizai Shimbun, 21 March 2005.57 Nihon Keizai Shimbun, 17 February 2005.58 “East Asia needs $1 trillion for infrastructure over next

five years” (www.worldbank.org).59 “Economic confidence will drive M&A activity through

2005, according to KPMG survey”,www.biz.yahoo.com.

60 See interview with Sri Lanka’s tourism minister in“Plans to bring back the tourists”, FDI Magazine, 7February 2005 (www.fdimagazine.com).

61 As far as developing and transition economies(according to the IMF’s classification) are concerned,the International Monetary Fund’s World EconomicOutlook (April 2005) estimates FDI flows will increaseto $217.4 billion in 2005 and to $222.3 billion in 2006(www.imf.org). The Institute of International Finance(March 2005) forecast an increase in FDI in 29emerging markets in 2005, to $148.2 billion from$138.3 billion in 2004 (www.iif.com). The WorldBank’s Global Development Finance 2005 (April 2005)projected an annual growth rate of 9% for FDI flowsto developing countries (or low-income and middle-income countries according to the World Bank’sclassification) (nominal value) over the next two years(www.siteresources.worldbank.org).

Introduction

As chapter I shows, FDI inflows todeveloped countries dropped again in 2004, adecline that was offset by rising flows todeveloping countries and South-East Europe andthe Commonwealth of Independent States (CIS)(figure II.1). Not only did this put an end to thedownturn that had begun in 2001, i t alsorepresented the highest ever level of investmentflows to these countries. Increases were notedfor all developing regions except Africa whereFDI inflows remained stable at a high level. Asin 2003, the continued decline of inflows todeveloped countries was due primarily to largerepayments of intra-company loans by foreignaffiliates in some host countries, particularlyGermany and the Netherlands. France and

��������

REGIONAL TRENDS: DEVELOPING REGIONSLEAD RISE IN FDI

Figure II.1. FDI flows by region, 2003, 2004(Bill ions of dollars)

Source: UNCTAD (www.unctad.org/fdistatistics) and annex table B.1.

Luxembourg, both major recipients of FDI in2003, received less of it in 2004, while inflowsto the United Kingdom and the United Statesrecovered. The Russian Federation accounted forthe bulk of the higher flows to South-East Europeand the CIS, a new country grouping (box I.2).

Developed countries remain the mainsources of FDI globally (figure II.1). As in thecase of inflows, the United States and the UnitedKingdom, in that order, accounted for the largestshares of FDI outflows in 2004. France andGermany also ranked among the top four homeeconomies. Developing economies, particularlythose from Asia, are emerging sources of FDI;in 2004 Asia and Oceania contributed more thanfour-fifths of outward FDI from developingcountries.

0

100

200

300

400

500

600

700

Developed

countries

Africa Asia and

Oceania

Latin

America

and the

Caribbean

South-East

Europe

and CIS

2003

2004

Developingcountries

Developingcountries

Developed

countries

Africa Asia and

Oceania

Latin

America

and the

Caribbean

South-East

Europe

and CIS

2003

2004

0

100

200

300

400

500

600

700

(a) FDI inflows (b) FDI outflows

40 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

A. Developing countries

1. Africa: FDI inflows remain buoyant,sustained by investments inprimary production

In 2004, Africa’s FDI inflows remained atthe relatively high level reached in 2003 ($18billion) (figure II.2), following a 39% increasein 2003.1 High prices for minerals such ascopper, diamonds, gold and platinum, andparticularly for oil, along with the consequentimproved profitability of investment in naturalresources encouraged TNC investment in theregion. Cross-border M&As in the miningindustry increased to more than three times their2003 value. Inflows rose in 40 out of the 53countries in Africa and fell in 13, including insome of the region’s top FDI recipients such asAngola, Morocco and Nigeria. The five top homecountries of FDI for Africa in 2004 were France,the Netherlands, South Africa, the UnitedKingdom and the United States, togetheraccounting for well over half of the flows to theregion. Although inflows in 2004 were relativelyhigh, Africa’s share in world FDI inflowsremained small at 3%. Continued high demandfor commodities, a more stable policyenvironment and increasing participation ininfrastructure networks by African TNCs areexpected to boost FDI in Africa in 2005. At the

same time, FDI outflows from African countriesmore than doubled in 2004.

a. Trends: FDI continues to flow,mostly to natural resources

The level of FDI flows to Africa remainedvirtually unchanged in 2004, at $18 billion. Mostof the inflows were in natural-resourceexploitation, spurred by rising commodityprices.2 The profitability of natural-resourceexploitation in the region increased,3 which alsoinduced TNCs to engage in cross-border M&Asin the primary sector. This further pushed up FDIinflows (see annex table A.II.1 for major cross-border M&A deals).

Still, Africa’s share of world FDI flows wasonly 3% in 2004. Over the past ten years thisshare has risen by less than one percentage point.On a per capita basis, FDI inflows to Africa rosefrom $8 in 1995 to $20 in 2004, but thisrepresented only about half of the per capita FDIinflows to China, for example, which stood at$46 in 2004. FDI inflows accounted for 5.5%of Africa’s gross fixed capital formation in 2004(figure II.2).

Among the different subregions, NorthAfrica4 attracted the highest inflows in 2004, withall the countries in the subregion, except theLibyan Arab Jamahiriya, on the list of the top10 host countries for FDI in Africa (figure II.3).

Figure II.2. Africa: FDI inflows and their share in gross fixed capital formation,1985-2004

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex tables B.1 and B.3.

-2

0

2

4

6

8

10

12

14

16

18

20

0

1

2

3

4

5

6

7

8

9

%

North AfricaWest AfricaCentral Africa

East AfricaSouthern AfricaFDI inflows as a percentage of gross fixed capital formation

$b

illio

n

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

1989

1988

1987

1986

1985

41CHAPTER II

The subregion attracted 29% of Africa’s totalinflows, particularly in oil and gas. Sudan toppedthe list, mainly as a result of FDI in petroleumfrom China, India and Malaysia. Investment linkshave also been established with several membersof the CIS (e.g. the Russian Federation) and withsome Gulf countries. Oil and natural gasexploitation also contributed to inflows to Algeriaand Egypt. Inflows to Morocco declined by morethan half to $0.9 billion in 2004 because of aslowdown in the privatization of the country’spublic enterprises. In Tunisia inflows were stable.

East Africa5 and West Africa6 also receivedhigher inflows in 2004, but they declined inCentral Africa7 and Southern Africa.8 While FDIflows to South Africa fell, most of the small host

economies received higher inflows. However, asin previous years, such flows remained belowthe $0.1 bill ion level in 2004 (table II .1),especially in the natural-resource-poor and leastdeveloped countries (LDCs). In countries longaffected by political conflict such as Burundi andSomalia, there were virtually no inflows until2003, with a few exceptions. In many of theseLDCs, the size of the domestic market is smalland some of the market-access initiatives put inplace to encourage investment in export-orientedindustries have been constrained by the lack ofappropriate human and other resources. Markinga change in this regard, Coca-Cola opened a newbottling plant worth $8.3 million in Mogadishu,Somalia in 2004, the largest single investmentin that country since 1991.9

Rising oil prices contributed torelatively high levels of FDI inflows to themajor oil-producing African countries,especially Sudan and Equatorial Guinea(figure II.3). Although FDI inflowsdecreased in Angola and Nigeria, the levels,nevertheless, remained high in those twocountries.10 These four countries, togetherwith Egypt, were the top recipients of FDIto Africa in 2004. With over $1 billion eachin inflows, their combined total amountedto $8.6 billion (or a little under 50% ofAfrica’s total inflows), while the top tenhost countries accounted for 69% in 2004.

As a result, the composition of FDIinflows to Africa in 2004 (as well as in2003) was significantly ti l ted towardsnatural resources, particularly in thepetroleum industry. The share of thisindustry exceeded 60% of total inflows inAngola, Egypt, Equatorial Guinea andNigeria, four of the five largest hostcountries in Africa (figure II.4). It has alsoaccounted for the largest share of FDI inAlgeria, the Libyan Arab Jamahiriya andSudan in recent years. In South Africa aswell, a major transaction in the oil industrydominated FDI inflows in 2004: Tullow OilPlc of the United Kingdom merged withEnergy Africa Ltd of South Africa, resultingin a $0.5 billion investment.

In some countries efforts to diversifythe economy, and in some cases to reducedependence on the hydrocarbons industryby opening up new industries to foreignparticipation, are beginning to pay off. In2004, for example, there were sizeable

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex table B.1.

a Ranked on the basis of the magnitude of FDI flows in 2004.

Figure II.3. Africa: FDI flows, top 10 economies,a

2003, 2004(Bill ions of dollars)

0.0 0.5 1.0 1.5 2.0 2.5

Tunisia

Congo

Morocco

Algeria

Congo, Democratic Republic of

Egypt

Sudan

Equatorial Guinea

Angola

Nigeria

2003

2004

0.0 0.5 1.0 1.5

Morocco

Mauritius

Kenya

Liberia

Libyan Arab Jamahiriya

Egypt

Algeria

Nigeria

Botswana

South Africa

2003

2004

(a) FDI inflows

(b) FDI outflows

3.5

1.6

42 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Table II.1. Africa: country distribution of FDI inflows, by range, 2003, 2004

2003 2004

Range Economy a Economy a

More than $2.0 bil l ion Angola, Morocco and Nigeria Nigeria and Angola

$1.0-1.9 bil l ion Equatorial Guinea and Sudan Equatorial Guinea, Sudan and Egypt

$0.5-0.9 bil l ion South Africa, Chad, Algeria, Tunisia and Democratic Republic of the Congo, Algeria, Morocco,United Republic of Tanzania Congo, Tunisia, South Africa and Ethiopia

$0.1-0.4 bil l ion Ethiopia, Botswana, Mozambique, Congo, Chad, United Republic of Tanzania, Côte d’Egypt, Mauritania, Uganda, Gabon, Zambia, Ivoire, Zambia, Gabon, Mauritania, Namibia,Côte d’Ivoire, Democratic Republic of the Uganda, Mali, Ghana, Mozambique, Libyan ArabCongo, Namibia, Libyan Arab Jamahiriya, Jamahiriya and GuineaGhana and Mali

Less than $0.1 bil l ion Kenya, Guinea, Mauritius, Seychelles, Senegal, Swaziland, Mauritius, Benin, Gambia,Senegal, Benin, Lesotho, Togo, Zimbabwe, Togo, Seychelles, Zimbabwe, Sao Tome andBurkina Faso, Gambia, Eritrea, Cape Verde, Principe, Lesotho, Botswana, Kenya,Madagascar, Niger, Djibouti, Malawi, Sao Madagascar, Burkina Faso, Djibouti, Eritrea,Tome and Principe, Rwanda, Guinea-Bissau, Cape Verde, Liberia, Niger, Malawi, Rwanda,Central African Republic, Sierra Leone, Somalia, Guinea-Bissau, Sierra Leone, Burundi,Liberia, Comoros, Cameroon, Somalia, Comoros, Cameroon and Central AfricanBurundi and Swaziland Republic

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex table B.1.a Listed in order of the magnitude of FDI inflows for each respective year.

investments in the telecommunications industryin Algeria.11 In Morocco a 16% stake of MarocTelecom (MT) was sold to Vivendi, which wasdue to be paid in early 2005.12 In Egypt,liberalization and privatization have promptedFDI in a range of industries such as cement,telecoms and tourism. In Sudan, inflows of FDIfrom China are expected for the building of a new

Figure II.4. Share of petroleum in FDI inflows tofour major African countries, 2004

Source: UNCTAD, based on nat ional sources and off ic ia lcommunications.

power plant and a refinery north of Khartoumand for the refurbishing of a long-neglectedrailway system. In Tunisia, FDI inflows in themanufacturing industry constituted 39% of totalflows to the country, and in recent years, theyhave also gone to major infrastructure projectsin energy and telecommunications.

About 63% of the cross-borderM&As in Africa in 2004 were related tomining activities, up from 13% in 2003(table II.2). Greenfield FDI inflows tonatural resources also increasedmarginally (annex table A.I.3). Forinstance, Gold Fields (South Africa),Junior Orezone Resources (Canada) andRiverstone Resources (Canada) increasedtheir investment in the Essakan gold jointventure in Burkina Faso. Reefton Miningof Australia enlarged its diamondactivities in Namibia. In addition, WestAfrica Gold Inc. (now Great West GoldInc.) of the United States expanded itsinvestment in gold, platinum andpalladium extraction in Mali. About athird of all registered greenfield FDIprojects were in manufacturing and nearlyhalf were in the services sector (annextable A.I.3).

Notwithstanding growing interestamong Asian investors, most of Africa’sFDI inflows originate mainly from

Angola

Petroleum93%

Others7%

Egypt

Others36%

Petroleum64%

Equatorial Guinea

Petroleum94%

Others6%

Nigeria

Others10%Petroleum90%

43CHAPTER II

Table II.2. Africa: distribution of cross-border M&A sales,by sector and industry, 2003, 2004

(Millions of dollars and per cent)

2003 2004 Growth rate Sector/industry Value % Value % in 2004 (%)

Primary 828 12.9 2 918 63.5 252Mining 828 12.9 2 918 63.5 252

Manufacturing 5 066 78.8 1 144 24.9 -77Food, beverages and tobacco 1 657 25.8 46 1.0 -97Wood and wood products 3 - - - -Printing, publishing and allied services - - 10 0.2 -Oil and gas; petroleum refining 3 130 48.7 1 076 23.4 -66Chemicals and chemical products 110 1.7 - - -Stone, clay, glass and concrete products - - - - -Metals and metal products 166 - - - -Machinery - - 4 0.1 -Miscellaneous manufacturing - - 9 0.2 -

Services 532 8.3 533 11.6 -Electricity, gas and water distribution 329 5.1 19 0.4 -94Hotels and restaurants - - 33 0.7 -Trade 2 - 44 1.0 2 059Transport, storage and communications 2 - 331 7.2 16 472Finance 89 1.4 65 1.4 -27Business activities 107 1.7 25 4.9 -76Community, social and personal service activities 3 - 15 0.3 497.5

All industries 6 427 100.0 4 595 100.0 -28

Source: UNCTAD, cross-border M&A database (www.unctad.org/fdistatistics).

developed countries (Western Europe, the UnitedStates) and South Africa. The top five homecountries for FDI flows to Africa are France, theNetherlands, South Africa, the United Kingdomand the United States, which together accountedfor more than half of total inflows to Africa in2003.

FDI outflows from Africa more thandoubled, to $2.8 billion in 2004. Most of theseoutflows, about 57%, were the result of cross-border acquisitions by TNCs from South Africa,following an increasingly liberalized outwardinvestment policy in that country. For instance,AngloGold (South Africa) purchased AshantiGoldfields (Ghana) which has major FDI projectsin Guinea, the United Republic of Tanzania andZimbabwe, and Gold Fields (South Africa)acquired IAMGOLD (Canada). In another deal,Allied Technologies (South Africa) acquired theEconet Wireless Group of Botswana. TNCs fromsome other African countries are also investingwithin and outside the region. Examples includethe expansion of the operations of OrascomTelecom Holding (Egypt) into Iraq and otherAsian countries, and the expansion of productionby Oriental Resources of Nigeria in Chad.Algeria, Egypt, Nigeria and South Africa togetheraccounted for 81% of the FDI outflows fromAfrica in 2004 (annex table B.1).

b. Policy developments: efforts tostabilize the environment for FDIinflows

In terms of policy changes, there was afurther wave of FDI-friendly measures andinitiatives at the national, regional and globallevels to attract more FDI into African countriesin 2004. Most of these measures focused onliberalizing legal frameworks and improving theinvestment climate.

Egypt, the Libyan Arab Jamahiriya andMauritius introduced at least four policy changeseach. Among the countries implementing policyreform, Algeria, the Democratic Republic of theCongo, Egypt, Ghana, Madagascar, Mauritania,Mauritius, Senegal, the United Republic ofTanzania and Uganda generally simplified aspectsof their FDI regulations, including through theestablishment of more transparent FDI regimes.Nigeria implemented reforms allowing foreignbanks to merge with local commercial banks. TheDemocratic Republic of the Congo and the UnitedRepublic of Tanzania reduced the levels of taxand royalty payments. Other specific changesincluded the adoption in Egypt of an antitrust lawas part of a concerted drive to improve thecountry’s business environment, and the

44 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

announcement by the Central Bank of Zimbabwe ofa new guarantee to pay back the entire capitalwithin three months if investors decided to leave.13

Some noticeable national policy andinstitutional changes are also taking place in thepetroleum industry, the main attraction in severalAfrican countries for FDI inflows in 2004 (boxII.1), in an attempt to enhance the favourableimpact of oil revenues on national development.

In Kenya, the Government completed abidding process to privatize Kenyan Telkom.However, FDI policy in Kenya appears to havebecome stricter in some areas (box II.2).

Many African countries also stepped up theirinvestment promotion efforts in 2004. Forexample, Egypt initiated a number of measuresincluding the simplification of investmentprocedures; it is also reviewing the fiscal regime.In addition, i t is restructuring the GeneralAuthority for Investment and Free Zones (GAFI).Similar efforts are under way in Moroccoregarding the Investment Directorate. A numberof countries, including Egypt, Morocco andTunisia, are trying to promote their countries asinvestment destinations through the organizationof investors’ meetings and annual conferences.

Source: UNCTAD, based on national sources.

Box II.1. Africa: several producer-countries seek to improve policies and management ofthe petroleum industry

Several African petroleum-producer countriesadopted or proposed new policies and institutionalchanges with respect to petroleum exploration andexploitation in 2004. Some of these changes aimat improving the management of the oil industry inorder to enhance the benefits to the local economy.Others aim at creating a better environment forproduction activities in the oil industry. Major newpolicies and institutional changes have included thefollowing:

• The Government of Angola proposed a newlegislation requiring oil companies to route alltheir payments through the domestic bankingsystem. This measure is expected to lead to a largeinflux of FDI-related foreign exchange intoAngola, sharply boosting transactions and revenuefor domestic banks and increasing the bankingsector’s ability to offer credit to domesticenterprises.The legislation also sets out requirements on theprocurement of goods and hiring of services byoil companies operating in Angola. Oil companiesare expected to:- hold competitive tenders to contract the supply

of goods and the provision of support servicesfor their operations;

- ensure that Angolan companies benefit frompreferential treatment in competitive tendersfor services and goods. Domestic firms shouldbe awarded the relevant contract when theirbid is no more than 10% higher than the bidssubmitted by foreign competitors. If theAngolan authorities enforce the order strictly,it will have a significant impact on the scope

of services that may be directly provided byforeign contractors to oil operators. As a result,foreign service companies wishing to dobusiness in Angola are likely to optincreasingly for structuring their businessesthrough joint ventures with local partners.

• The Democratic Republic of the Congo isreorganizing the corporate structure of its nationaloil company, Société Nationale des Pétroles duCongo (SNPC), into a holding company withseven affiliates. Of particular interest to investorsis SNPC Refining, which is to be privatized.

• The Government of the Libyan Arab Jamahiriyaadopted a new exploration and production-sharingagreement called EPSA-IV. The Government isintended to offer fresh incentives to foreigncompanies to invest in oil and gas explorationand development, and it will make the contractingprocess more efficient and transparent.

• In Mali, a new oil code was adopted in June 2004.The initial time span allowed for oil prospectingis four years, renewable for two further periodsof four years each. The attribution of prospectingand exploration permits as well as their renewalis subject to the payment of fixed taxes. Permitholders are liable for the payment of charges onthe production of oil and a tax of 35% on profits,but they benefit from tax exemption on petroleumproducts.

• In Mauritania, a bill proposing a simplified taxsystem for oil producers was adopted. The newtext complements an act dating back to 1988 anddefines the framework for the execution ofcontracts and the rights and obligations of allparties.

45CHAPTER II

Box II.2. Kenya: UNCTAD’s Investment Policy Review recommends an alternativeapproach to minimum capital requirements for FDI inflows

Source: UNCTAD forthcoming d.

In the 1970s, Kenya was a prime locationfor FDI inflows in East Africa. However,deteriorating infrastructure and a poor track recordof policies in the 1980s and 1990s discouragedinflows of FDI for about two decades. Inflowsdeclined to one-fifth of those of neighbouringUganda in 2004, and stood at $46 million. On aper capita basis, this represented $1.4 comparedwith Uganda’s $8.5. As a result, Kenya is nowamong the developing countries that have attractedthe least FDI relative to their size over the pastdecade. FDI inflows have nevertheless had a crucialimpact on the development of the country’s export-oriented horticulture industry, contributed to therevival of Kenya Airways and accelerated thedevelopment of the mobile telecommunicationsnetwork in the country.

In 2002 the new Government indicated itsinterest in improving the investment framework soas to support private sector development and wealthcreation. In 2004, the Parliament adopted anInvestment Promotion Bill to promote and facilitateinvestment by assisting investors to obtain licencesand providing other incentives for related purposes.Its two core incentives are entitlements to businesslicences for an initial period along with theallotment of six residence and work permits forforeign staff in FDI projects.

However, the new Act requires all foreigninvestors to have their projects screened andapproved, and it imposes a minimum investmentrequirement of $500,000 on prospective foreigninvestors. This requirement was introduced to avoidthe crowding out of small national investors, andto encourage only “serious” foreign investors intoKenya. However, this approach is unlikely torespond adequately to the country’s legitimateconcerns; it could even create a barrier to beneficialFDI inflows: almost 75% of foreign investmentprojects registered in 2000-2004 were worth less

than $500,000. The minimum investment is likelyto deter FDI in low-capital but knowledge-intensiveservice industries that could bring benefits toKenya in some areas in which it has a comparativeadvantage. As a concrete example, Homegrown,which has evolved into Kenya’s largest horticultureand floriculture company and a major source ofemployment and spillovers, started with an initialinvestment well below the current requirement of$500,000.

The Investment Policy Review of Kenyacompleted by UNCTAD in early 2005recommends the adoption of an alternativeapproach to regulating FDI entry which wouldeffectively lift the screening and minimumcapital requirements and make investmentcertificates optional. Targeted protection tosensitive industries, in turn, could be considered,if deemed necessary. The Government of Kenyahas recognized that the general restrictionsimposed on FDI entry are likely to be counter-productive and has introduced a few keyamendments to the Investment Promotion Act.If adopted by the Parliament, these amendmentswill remove the compulsory screening of FDIand the minimum capital requirement. In turn,optional investment certificates would remaina condition for specific incentives and be subjectto a lower capital requirement of $100,000.

Like many other African countries, Kenyahas not attracted significant FDI inflows intomanufacturing and R&D activities. In this context,it might be useful to target FDI promotion effortsto attract FDI in projects in areas such astechnological inputs, R&D activities, andprocessing and manufacturing activities. Thatwould imply that projects that may initially havelow initial financial capital values but bring, forexample, valuable manufacturing and R&D inputswould be allowed to operate.

Various bilateral, regional and multilateraltreaties were also concluded, whichcomplemented national regulations for promotingFDI. African countries concluded 33 new bilateralinvestment treaties (BITs) and 15 new doubletaxation treaties (DTTs) in 2004 (figure II.5).These brought the cumulative numbers of BITsand DTTs for the region to 615 and 404respectively. In addition, the Libyan Arab

Jamahiriya and India agreed on liberalizing visaregimes for business people from the twocountries, and signed a bilateral investmentpromotion agreement in 2004. Tunisia concludeda free trade agreement (FTA) with members ofthe European Free Trade Area (EFTA), andMorocco concluded one with the United States.Egypt concluded a framework agreement withthe Southern Common Market (MERCOSUR),

46 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Figure II.5. Africa: BITs and DTTs concluded, cumulative and annual, 1990-2004(Number)

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).

and ratified the EU-Egypt Association Agreement(signed in 2001), which is expected to promotetrade and exports, improve bilateral relations withthe EU and encourage European investment inEgypt. Five economic and partnership agreementsbetween the EU and regional groupings ofAfrican countries were being negotiated in 2004(but have yet not been concluded).

The Government of the United Statesamended key provisions of the African Growthand Opportunity Act (AGOA) in 2004 (box II.3)that allow more flexible rules of origin. From2005, however, with the ending of the quotaslimiting some countries’ exports under the WorldTrade Organization’s (WTO) Agreement onTextiles and Clothing (ACT), the preferentialadvantage provided by the AGOA may not sufficeto attract FDI into textiles and clothing. Therewill be increased competition, especially fromAsian countries, the exports of which werepreviously restricted by the quotas.

In 2004, the Multilateral InvestmentGuarantee Agency (MIGA) of the World Bank,through its guarantee programme, supported fournew FDI projects in power generation, businessservices, banking and IT services, and undertook28 technical assistance activities in the region.14

At the same time, the African Trade InsuranceAgency (ATI) – the region’s only pan-Africanmultilateral import and export credit and politicalrisk guaranty agency15 – adopted measures toprotect foreign investors in Africa against traderisks. The region now has better market access

(as a result of the Everything-but-Arms (EBA)initiative of the EU, Japan’s 99% rule16 forLDCs, AGOA and the Generalized System ofPreferences (GSP)), and national policies aremore stable. Despite these measures and efforts,African countries’ capacity to target FDIstrategically in manufacturing and services hasbeen constrained by economic and social factors.Impediments range from small market size andpoor regulation to meagre financial resources andlow skills. The annual gross national income percapita, for instance, is around $500 in sub-Saharan Africa, and investment in sectors suchas education remains insufficient.

The continued low levels of FDI inmanufacturing in many African countries areexplained by two main factors: a failure to moverapidly on developing economic and socialpolicies that are important for FDI inflows (aswell as on development in general); and yearsof reforms in the 1980s that placed insufficientemphasis on capacity building. As a result, theinternational market-access measures andinitiatives provided for African countries havenot been very successful in attracting FDI,particularly in manufacturing, given the lack ofcapacity to exploit FDI in a number of countries.The future of FDI in Africa’s development liesin an integrated and genuine partnership betweenthe private sector and governments to strengthenhuman resource capabilities, for example throughtraining of the labour force (WIR03). Initiativessuch as AGOA can only have a stronger impact

0

10

20

30

40

50

60

70

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 20040

100

200

300

400

500

600

700

BITs (total per year: left scale) DTTs (total per year: left scale)

Total BITs (cumulative: right scale) Total DTTs (cumulative: right scale)

47CHAPTER II

Box II.3. AGOA Acceleration Act 2004: some new key provisions

Source: “AGOA Acceleration Act for 2004 (AGOA III) summary”, AGOA website (www.agoa.gov).

a The 37 African countries are: Angola, Benin, Botswana, Burkina Faso, Cameroon, Cape Verde, Chad, Congo,the Democratic Republic of the Congo, Djibouti, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau,Kenya, Lesotho, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Namibia, Niger, Nigeria,Rwanda, Sao Tome and Principe, Senegal, Seychelles, Sierra Leone, South Africa, Swaziland, the United Republicof Tanzania, Uganda and Zambia.

b For a description of progress with respect to exports and FDI in export-oriented production in some AGOAbeneficiary countries, including Lesotho, Mauritius, Mozambique, South Africa, Swaziland, and Uganda, seeWIR04, p.91, ff4. In Mali, a $12.5-million cotton-thread factory opened in February 2004. This facility is oneof the sub-Saharan African plants outside South Africa capable of producing quality thread for use inmanufacturing apparel for export under AGOA. Mauritians were among the investors. The factory created 200new jobs (www.agoa.gov).

The United States has made AGOA acornerstone of its policy of promoting trade andinvestment in Africa. In 2004, the United StatesGovernment enacted a law – the AGOAAcceleration Act of 2004 – that amended theoriginal initiative. The law now has the followingkey features:

The Act extends the expiration of theprogramme from 2008 until 2015, and the third-country fabric provision is extended for three years,from September 2004 until September 2007,including a phase-down in year three. The cap ofthe third-country provision will remain at the fullcurrent level available in years one and two. In thethird year, the cap will be phased down by 50%.

The law includes a statement ofCongressional policy that textile and apparelprovisions under the programme should beinterpreted in a broad and trade-expanding mannerto maximize opportunities for imports from Africa.This is accompanied by minor technical correctionsto prevent restrictive interpretations by customsofficials. The Act includes a modification of therules of origin to allow use of non-AGOA productsfor all import categories and continued use offabrics from AGOA countries – such as South

Africa – which also become free trade partnerswith the United States.

The Act increases the de minimis rule fromits current 7% to 10%. It states that apparelproducts assembled in sub-Saharan Africa, whichwould otherwise be considered eligible for AGOAbenefits except for the presence of some fibresor yarns not wholly formed in the United Statesor the beneficiary sub-Saharan African country,will still be eligible for benefits as long as thetotal weight of all such fibres and yarns is not morethan a certain percentage (currently 7%) of theweight of the article.

The Act also expands the current “folklore”AGOA coverage to include ethnic fabric made onmachines, and supports many of the aims of theNew Partnership for Africa’s Development(NEPAD) initiative, including regional integrationamong African countries.

AGOA was intended to apply to 48 Africancountries, but by the end of 2004 only 37 hadqualified.a To date, only 18 of these countries metthe rules-of-origin requirements, creating the legalconditions required for taking advantage of thescheme. However, only seven countries attractedany FDI inflows.b

on FDI inflows if African countries implementdevelopment-oriented economic and socialpolicies.

Africa’s ability to industrialize successfullycould weaken unless supported by strongdomestic investment capacity, which isparticularly important given the region’sdeclining share of global FDI inflows inmanufacturing. The scope for industrializationlies not just in improving its market access andthe investment climate but, more significantly,in strengthening its domestic industrial

capabilities. For the latter, governments maychoose to use public policies and finance toattract the type of FDI they need in themanufacturing industries, as illustrated by somepolicies in South Africa (box II.4).

However, attracting FDI into themanufacturing sector in Africa is becomingdifficult as competition grows from the otherdeveloping countries, particularly in Asia.Factors such as good physical infrastructure andappropriate human skill levels have becomeincreasingly important in attracting FDI projects,

48 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Box II.4. Attracting FDI to South Africa through Government development assistanceprogrammes

South Africa’s FDI flows over the past fiveyears have fluctuated between $6.8 billion in 2001and $600 million in 2004. Two of its currentdevelopment assistance programmes, the NationalIndustrial Participation Programme and theForeign Investment Grant (FIG), were designedto use the government’s financial capacity toattract FDI inflows to manufacturing projects, withsome success.

The National Industrial ParticipationProgramme is an offset scheme that requires acommitment by suppliers doing more than $10million worth of business with the Governmentor the companies it owns to facilitate industrialdevelopment in the country.a Under the scheme,when the Government purchases goods or servicesin which the import content exceeds $10 million,the foreign suppliers incur an obligation toreinvest a portion of their profits from sales insidethe country. Procurement programmes tied tothis arrangement include the Government’sstrategic defence procurement package andpurchases made by State-owned enterprises suchas Telkom, South African Airways, Eskom,Transnet and Petro S.A. The programme isobligatory and is focused on the transport, energy,and information and telecommunications

Source: Department of Trade and Industry website (www.dti.gov.za).

a “Jet-propelled investment”, FDI Magazine, April/May 2005 (www.fdimagazine.com).b Data from the Department of Trade and Industry. Even though South Africa has had successes with the offset

programme, some of the past commitments did not materialize.c “Jet-propelled investment”, op. cit.

industries. About 125 FDI projects have so farbeen facilitated by this programme resulting ininvestments of $750 million and exports of $1.5billion by the end of 2004.b The value of purchaseobligations currently being monitored by theDepartment of Trade and Industry isapproximately $14 billion, the bulk of whichcomes from the Government’s strategic defencepackage. In 2003, the programme yielded a bigoffset package: an $8.7 billion commitment fromaircraft supplier BAE Systems of the UnitedKingdom and Saab of Sweden.c The full offsetobligations are due to be discharged over a periodof seven years (by April 2011).

The FIG was created as a cash incentivescheme for foreign investors who invest in newmanufacturing enterprises in South Africa. In theFIG programme, a foreign entrepreneur can becompensated for up to 15% of the costs of movingnew machinery and equipment to South Africa,up to a maximum amount of 3 million rand ($0.5million) per entity. The scheme aims at promotingFDI as well as enhancing the level of technologyand overall economic growth in South Africa. Itis open to foreign investors who hold at least 50%of the shares in the relevant company.

especially as a number of international tradeadvantages such as those provided by the Multi-Fibre Arrangement (MFA), AGOA and othershave already, or will eventually, come to an end.This scenario may, however, change with newinitiatives for Africa such as those proposed bythe renewed emphasis on the MillenniumDevelopment Goals by the United Nations andby the Commission for Africa that was set up in2004 by the Government of the United Kingdom(box II.5).

c. Prospects: cautiously positive

The significant rise in commodity pricesthat started in 2004, and the resulting highprofitability of investments, are expected to lead

to further increases in FDI in Africa in 2005.Furthermore, the United States is expected toincrease its share of oil imports from Africa fromthe current level of 18% to 25% by 2015.17

Pressure on TNCs to access more petroleumresources, slash costs and take advantage of highprices is expected to set off a new wave of cross-border M&As in the region. United States andEuropean TNCs (such as Chevron Corp. (UnitedStates) in Angola and Total (France) in Nigeria)are already expanding or planning to expand theirinvestments. In the mining industry, significantprojects are planned as well, for instance indiamond, copper and cobalt in the DemocraticRepublic of the Congo.18

In infrastructure projects, TNCs are alsolikely to invest in some African countries. Eskomof South Africa, for instance, is already involved

49CHAPTER II

in the first phase of an infrastructure project torehabilitate the Inga hydroelectric power stationin the Democratic Republic of the Congo as partof the “Unified African Grid”. In 2004, Germaninvestors had announced plans to build acomputerized railway line from Rongai to Jubain Southern Sudan. Morocco might also receiveincreased FDI inflows in 2005 as a result offurther privatization of public enterprises and theconclusion of an FTA with the United States.

Improving economic conditions in SouthAfrica are encouraging FDI in the country’sbanking industry. The acquisition of 60% ofABSA (South Africa) by Barclays of the UnitedKingdom in 2005 may herald a wave of M&Asand greenfield FDI in South Africa and in othercountries in the region. Opportunities exist forFDI in key service industries in Africa,

particularly telecommunications, electricity andtransport. FDI inflows to processing and otherindustries in the manufacturing sector areexpected to be small, going mainly to the LibyanArab Jamahiriya, Nigeria, South Africa andUganda.

A 2005 survey of international FDI experts,TNCs and investment promotion agencies (IPAs)undertaken by UNCTAD (box I.3) revealedcautious optimism concerning the prospects forFDI in Africa. Among the TNCs, one out of fourrespondents expected FDI inflows to Africa toincrease in 2005-2006 (figure II.6). An equalnumber of TNCs believed that inflows woulddecrease. FDI experts and IPAs were moreoptimistic: one out of three FDI experts and nineout of 10 African IPAs expected FDI inflows togrow in 2005-2006. Experts and TNCs judge FDI

Box II.5. The Report of the Commission for Africa: recommendations to help boostinvestment

Africa is a major recipient of officialdevelopment assistance (ODA) as a source offinancing for development. After declining for muchof the 1990s, ODA to the region has risensubstantially in recent years, from $16 billion in2000 to $26 billion in 2003 (box figure II.5.1).Most of the region’s ODA comes fromdeveloped countries, with the United Kingdombeing one of the major donor countries(box table II.5.1).

In 2004, the Prime Minister of the UnitedKingdom established a Commission for Africa“to define the challenges facing Africa, andprovide clear recommendations on how tosupport the changes needed to reduce poverty”(Commission for Africa 2005, p. 1). Its Report,released in March 2005, recommends asubstantial increase in aid to Africa – anadditional $25 billion per year to beimplemented by 2010 – emphasizing the needfor innovative financial methods to securefunding.a It calls for changes by the recipientsas well as donors in an integrated packagefocusing on governance and capacity building,peace and security, investment in people, growthand poverty reduction, and trade to ensure that aidis well spent. It proposes a “Marshall Plan” to pullAfrica out of poverty, just as the Marshall Planinvolving large amounts of aid from the UnitedStates enabled Europe to rebuild its industrialinfrastructure after the Second World War.

Several of the report’s recommendations aredirectly relevant to boosting both local and foreigninvestment in African economies. The Reportnotes that infrastructure and policy measures inAfrica have not been adequate, nor have they beenimproved or expanded. It points out that private

investment cannot be expected to flow withoutdecent transportation systems, a stable policyclimate, human capital and reliable utilities.

The report underlines concrete prioritiesfor the use of additional aid in areas that couldencourage investment in the region. It calls for

Box figure II.5.1. Africa: ODA inflows, 1980-2003(Billions of dollars)

Source: UNCTAD, based on OECD ODA/OA database.

/...

0

5

10

15

20

25

30

G7

All other donors

Total ODA to Africa

19

80

19

81

19

82

19

83

19

84

19

86

19

87

19

88

19

89

19

90

19

91

19

92

19

93

19

94

19

95

19

96

19

98

19

99

20

01

20

02

20

03

19

85

19

97

2000

50 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Source: UNCTAD, based on the United Kingdom, Commission for Africa 2005.

a At the end of the summit of the G-8 countries in Gleneagles, United Kingdom, in July 2005, the countries and otherdonors made substantial commitments to increase aid by a variety of means, including through traditional developmentassistance, debt relief and innovative financing mechanisms, which would lead to an increase in ODA to Africa of $25billion a year by 2010.

Box table II.5.1. Top 10 ODA donors toAfrica, 2000-2003a

(Millions of dollars)

Donor country 2000 2001 2002 2003

United States 2 107 1 975 3 189 5 063France 1 812 1 531 2 603 3 587Germany 871 830 1 009 2 061United Kingdom 1 151 1 204 1 048 1 508Belgium 219 245 363 1 053Netherlands 601 853 956 1 026Italy 252 196 811 744Japan 1 226 1 091 700 704Sweden 399 352 409 683Norway 339 325 452 581

G7b to Africa 7 638 7 044 9 748 14 184All donors to Africa 15 732 16 691 21 261 26 318

MemorandumG7b to all recipients 167 773 153 514 184 551 223 633All donors to all recipients 314 378 320 487 368 712 426 330

Source: UNCTAD, based on OECD, ODA/OA database.a Ranked according to 2003 figures.b Canada, France, Germany, Italy, Japan, the United Kingdom and

the United States.

Box II.5. The Report of the Commission for Africa: recommendations to help boostinvestment (concluded)

donors to double their spending on infrastructure– from rural roads to regional highways, powerprojects and information and communicationstechnologies (ICT) – and proposes a 100%external debt cancellation for African countries.The report recognizes the need to reverse yearsof chronic underinvestment in education (partlyas a result of budget cuts made in order to complywith the IMF’s structural adjustment programmes).It also calls on developed countries to supportan Investment Climate Facility for Africa underthe NEPAD initiative, and to insure foreigninvestors in post-conflict countries in Africathrough a risk-bearing fund of the MultilateralInvestment Guarantee Agency.

New ODA inflows into Africa, if allocatedaccording to the priorities outlined in the report,could help improve the investment climate byproviding opportunities for foreign firms to investproductively, creating jobs, and contributing tosustainable progress in reducing poverty whileimproving living standards in the region.

prospects for North African countries to be morepositive than those for sub-Saharan Africancountries.

FDI outflows from Africa are also poisedfor a rapid expansion in 2005. The major homesources of this expansion are likely to be South

Africa, Egypt and Nigeria. For instance, severalSouth African TNCs are committed to largeprojects inside and outside Africa, particularlyin the Democratic Republic of the Congo andWestern Asian countries. Orascom TelecomHolding of Egypt has offered to buy the WindSpA phone company of Italy in 2005.19 OrientalEnergy Resources of Nigeria is seeking to acquirepetroleum exploration rights in Angola.

2. Asia and Oceania: inflows at arecord high

FDI inflows to Asia and Oceania reacheda new high at $148 billion in 2004, registeringthe largest increase ever. The region’s share ofFDI inflows worldwide also increased from 16%in 2003 to 23% in 2004. Almost all parts of Asiaand Oceania received higher flows than in 2003.FDI inflows also rose as a percentage of grossfixed capital formation (figure II.7). Outwardflows from the region quadrupled to $69 billion,

Figure II.6. Africa: prospects for FDI inflows,2005-2006

(Per cent of responses from TNCs, experts and IPAs)

Source: UNCTAD (www.unctad.org/fdiprospects).

0

20

40

60

80

100

TNCs Experts IPAs

Increase Remain the same Decrease

51CHAPTER II

the second highest level ever, driven by FDI frommost major economies, and particularly fromHong Kong (China). The policy environment forFDI continued to improve, and the prospects forFDI in and from the region remain promising.

a. Trends: strong growth in FDIflows

FDI flows to Asia and Oceania20 increasedby 46% in 2004; 34 out of 54 economies receivedhigher flows than in 2003. However, they remainconcentrated: the top 10 host economies (figureII.8) accounted for 92% of FDI inflows to theregion.

The distribution of inflows by size changedsignificantly compared with 2003: a few largeFDI-recipient economies saw an increase in thelevel of FDI flows, and the number of economiesthat received less than $100 million decreased(table II .3). Bangladesh, China, India, theRepublic of Korea, Macao (China), Mongolia,Pakistan, Qatar, Singapore, the Syrian ArabRepublic and Viet Nam received record levelsof flows (annex table B.1).

While greenfield investment remains themost important mode of FDI in the region, cross-border M&As increased from $22 billion in 2003to $25 billion in 2004 largely due to transactionsin East Asia (annex table B.4). The top threetargets in terms of the value of cross-border M&Asales in 2004 were China, the Republic of Korea

and Hong Kong (China) (figure II.9). The mostsignificant increase took place in China, makingthe value of its cross-border M&A sales thelargest in the region in 2004. The surge of M&Asin China was driven largely by policy changesin that country.21

Cross-border M&As in Asia and Oceaniaprimarily targeted service industries (and inparticular financial services), which accountedfor two-thirds of total cross-border M&A salesin 2004 (table II.4). Cross-border M&A salesalmost doubled in the chemical industry, makingit the largest recipient industry of cross-borderM&As in manufacturing in the region.

In contrast to cross-border M&As,greenfield investment by TNCs concentrated onmanufacturing followed by sales and marketing,retail and business services (annex table A.I.3).FDI in R&D, a relatively new area for TNCexpansion in developing countries, has gainedimportance in recent years, accounting for 11%of all greenfield projects in Asia and in Oceaniain 2004 (annex table A.I.3).

With a 46% increase in FDI inflows, EastAsia remains the most important subregion forFDI inflows. However in terms of increase ininflows, the performance of West Asia (with a51% increase) and South-East Asia (48%) wasmore impressive. FDI inflows to South Asia alsoincreased, by 31%, to reach a record high. Incontrast, Oceania witnessed a 54% decrease inflows.

Figure II.7. Asia and Oceania: FDI inflows and their share in gross fixed capital formation,1985-2004

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex tables B.1 and B.3.

0

30

60

90

120

150

19

85

19

86

19

87

19

88

19

89

19

90

19

91

19

92

19

93

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

20

03

20

04

0

2

4

6

8

10

12

14

%

Oceania West AsiaEast Asia South AsiaSouth-East Asia FDI inflows as a percentage of gross fixed capital formation

$b

illio

n

52 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

• East Asia22 accounted for the lion’sshare (71%) of FDI flows to Asia andOceania. These rose from $72 billionin 2003 to $105 billion in 2004, mainlyon account of higher FDI flows toHong Kong (China), China and theRepublic of Korea. FDI flows to HongKong (China) increased by 150%, to$34 bill ion, led by flows to theservices sector. An increase in cross-border M&A transactions in theRepublic of Korea, especially large-value ones, helped push that country’sinflows to $8 billion.

China was again the largest recipientof FDI, not only in the region but alsoamong all developing countriesworldwide, with flows reaching thehighest level ($61 billion).23 Strongeconomic growth, an improved policyenvironment and further opening upto FDI in certain industries – such asbanking and other financial services– contributed to the increase. In 2004,five Chinese banks attracted $2.7billion in FDI24 and total FDI flowsto the banking sector reached $3.8billion. Investments by private equityand venture capital funds, especiallyfrom the United States, have becomeimportant sources of foreigninvestment in China.25 Theimplementation of large-scale FDIprojects also led to a significantincrease in FDI in the automotiveindustry26 and the semiconductorindustry.

• South-East Asia27 witnessed a furtherrise in flows from $17 billion in 2003to $26 billion in 2004. The decline inrepayments of intra-company loans byforeign affiliates in the subregion toparent firms helped, as did theincrease in the level of cross-borderM&As in the region (annex table B.4).Higher flows to Singapore, Malaysia,Indonesia, Myanmar, Viet Nam, thePhilippines and Cambodia contributedto the subregion’s increased FDIreceipts. In Indonesia, the successfulprivatization of State assets andforeign acquisitions of private firmshelped putting an end to thecontinuous period of negative FDIinflows that began in 1998.Acquisition by an investor group (ledby Standard Chartered of the UnitedSource: UNCTAD, cross-border M&As database (www.unctad.org/

fdistatistics) and annex table B.4.

5

0

1

2

3

4

6

7

2003

2004

$b

illio

n

Ch

ina

Ko

rea

,R

ep

ub

lico

f

Ho

ng

Ko

ng

,C

hin

a

Ind

ia

Ind

on

esia

Th

aila

nd

Sin

ga

po

re

Ph

ilip

pin

es

Ma

laysia

Pa

kis

tan

Figure II.9. Top 10 economies in terms of cross-border M&A sales in Asia and Oceania: 2003, 2004

(Bill ions of dollars)

Figure II.8. Asia and Oceania: FDI flows, top 10economies,a 2003, 2004

(Bill ions of dollars)

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics)and annex table B.1.

a Ranked on the basis of the magnitude of FDI flows in 2004.

(a) FDI inflows

(b) FDI outflows

0 2 4 6 8 10

2003

2004

-2 0 2 4 6 8 10

Viet Nam

Saudi Arabia

Taiwan Province of China

Turkey

Malaysia

India

Korea, Republic of

Singapore

Hong Kong, China

China

Philippines

Turkey

Bahrain

China

Malaysia

India

Korea, Republic of

Taiwan Province of China

Singapore

Hong Kong, China

54

14

2003

2004

61

34

16

40

11

53CHAPTER II

Kingdom) of a controlling interest inPT Bank Permata Tbk for $305million is an example of suchprivatization (annex table A.II.1). The

value of cross-border M&As inMalaysia, the Philippines andThailand also rose significantly.

The rapid rise of FDI inflows to thesubregion and the narrowing gapbetween flows to ASEAN membersand China assuaged those concernedthat China is crowding out FDI fromits neighbouring countries. A recentstudy suggests that FDI in China didnot crowd out FDI inflows to South-East Asian countries during 1992-2001 (Zhou and Lall 2005).28 Thiswas based on the fact that there islittle competition between countriesin market- and resource-seeking FDIand that efficiency-seeking, export-oriented FDI in China may have beenso far complementary to that in South-East Asian countries.

• FDI inflows to South Asia29 alsoclimbed in 2004 for the fourthconsecutive year. Inflows to India –at a record level of $5 billion – wereencouraged by an improvingeconomic situation and a more openFDI climate. Cross-border M&As inIndia rose in 2004 as thetelecommunications, business processoutsourcing and pharmaceutical

Table II.3. Asia and Oceania: country distribution of FDI inflows,by range, 2003, 2004

2003 2004Range Economy a Economy a

More than $5 bil l ion China, Hong Kong (China) and Singapore China, Hong Kong (China), Singapore, Republicof Korea and India

$2.0-4.9 bil l ion India, Republic of Korea, Malaysia, and Malaysia and TurkeyBrunei Darussalam

$1.0-1.9 bil l ion Thailand, Turkey, Viet Nam, and Syrian Taiwan Province of China, Saudi Arabia, Viet Nam,Arab Republic Syrian Arab Republic, Thailand and Indonesia

$0.1-0.9 bil l ion Saudi Arabia, Qatar, Pakistan, Oman, Bahrain, Pakistan, Bahrain, United Arab Emirates, Qatar,Islamic Republic of Iran, Taiwan Province of Jordan, Macao (China), Myanmar, IslamicChina, Jordan, Macao (China), Lebanon, Republic of Iran, Phil ippines, Bangladesh, Iraq,Phil ippines, Myanmar, Bangladesh, Sri Lanka, Lebanon, Sri Lanka, Mongolia, Cambodia andDemocratic People’s Republic of Korea, Brunei DarussalamMongolia and Papua New Guinea

Less than $0.1 bil l ion Cambodia, United Arab Emirates, Fij i, Democratic People’s Republic of Korea, PapuaLao People’s Democratic Republic, Vanuatu, New Guinea, Vanuatu, Lao People’s DemocraticNepal, Maldives, Tonga, Yemen, Iraq, Republic, Maldives, Marshall Islands, Nepal,Timor-Leste, Marshall Islands, Palau, Tuvalu, New Caledonia, Palau, Tonga, Timor-Afghanistan, Nauru, Bhutan, Samoa, Tokelau, Leste, Afghanistan, Bhutan, Samoa, SolomonTuvalu, New Caledonia, Solomon Islands, Islands, Fij i, Oman, Kuwait and YemenFrench Polynesia, Kuwait and Indonesia

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex table B.1.a Listed in order of the magnitude of FDI inflows for each respective year.

Table II.4. Asia and Oceania: distribution of cross-border M&A sales, by sector

and industry, 2003, 2004(Millions of dollars and per cent)

2003 2004 Growthrate in

Sector/industry Value % Value % 2004 (%)

Primary 42 0.2 215 0.9 419Manufacturing 7 401 34.2 8 125 32.7 10

Chemicals and chemicalproducts 1 248 5.8 2 392 9.6 92

Electrical and electronicequipment 943 4.4 1 691 6.8 79

Food, beverages and tobacco 1 276 5.9 1 652 6.7 30Oil and gas; petroleum refining 1 757 8.1 614 2.5 -65Motor vehicles and other

transport equipment 1 312 6.1 516 2.1 -61Other manufacturing 866 4.0 1 260 5.1 45

Services 14 212 65.6 16 480 66.4 16Finance 6 052 27.9 10 947 44.1 81Business activit ies 2 388 11.0 2 825 11.4 18Electricity, gas, and water

distribution 885 4.1 891 3.6 1Transport, storage and

communications 3 787 17.5 846 3.4 -78Trade 481 2.2 426 1.7 -11Other services 618 0.2 545 2.2 -12

All industries 21 654 100.0 24 820 100.0 15

Source: UNCTAD, cross-border M&As database (www.unctad.org/fdistatistics).

54 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

industries saw an increase in large deals.Improved investment environments and theprivatization of assets in Pakistan andBangladesh contributed to higher FDI flowsto those countries. Improvements in theregional political situation also played arole. In Afghanistan, investors from 25countries have set up operations (Eedes2005).30

• FDI inflows to West Asia31 increased from$6.5 bill ion in 2003 to $9.8 bill ion in2004.32 Countries such as Bahrain, Jordan,Saudi Arabia, Turkey and the United ArabEmirates saw a sharp rise in inflows (boxII.6). While high oil prices might have

influenced oil-related FDI, it is difficult toassess precisely their impact on FDI in theregion. Efforts by a number of countries topromote non-oil investment in theireconomies contributed, to some extent, tothe subregion’s improved FDI flows (boxII.6), as illustrated by developments in theIslamic Republic of Iran (box II.7).

• Oceania33 witnessed a sharp fall in FDIinflows, from $146 million in 2003 to $67million in 2004. This was mainly caused bythe significant decline of flows to PapuaNew Guinea (from $101 million to $25million) and Fiji (from $23 million to -$9million). Flows to Vanuatu and Tuvalu roseto $22 million and $9 million respectively.

In 2004, FDI flows to West Asia rose by 51%.This increase was spread unevenly among theeconomies of the subregion, and FDI inflows wereconcentrated in particular in Turkey, Saudi Arabiaand the Syrian Arab Republic in that order; the threecountries together accounting for 59% of totalinflows. The Triad was the main source of FDI flowsto West Asian countries. South Africa was anotherrelatively significant source of investment, whileintraregional investment from within Asia alsocontributed to the upward trend. The growth in FDIinflows in 2004 largely reflected an increase in somelarge-scale greenfield investments by internationaloil and gas firms, as well as cross-border M&Asin business and financial services, mining (includingoil and gas) and manufacturing.

The relatively low importance of FDI in WestAsian economies is reflected in the ratio of FDIflows to gross fixed capital formation: at 4.9%, itis below the developing-country average not tomention that of South, East and South-East Asia.This is partly due to the economic structure of theWest Asian economies, the size of their markets,the importance of oil revenues to some of them andthe overall level of political uncertainty affectingthe subregion. Indeed, a difficult geopoliticalsituation in parts of the subregion heightens the riskperceptions of investors, while sanctions imposedon several countries in West Asia have impeded theirintegration into the world economy (Yousef 2005).

The primary sector remains dominant in termsof inward FDI stock, but FDI in manufacturing andservices is rising in some countries such as Bahrain,the Islamic Republic of Iran, Saudi Arabia, Turkeyand the United Arab Emirates. For instance, thenumber of cross-border M&As and greenfield FDI

Box II.6. FDI flows to West Asia increased but remain concentrated

projects in the subregion between 2002 and 2004were larger in business services and inmanufacturing, including the oil refining industry,than in natural resource extraction (box figureII.6.1). Greenfield FDI projects in manufacturingwere mainly in the chemical (28% of totalmanufacturing), automotive (28%) and food anddrink (19%) industries. Large oil firms such asChevron, ExxonMobil and Royal Dutch ShellGroup announced large investments in the chemicaland energy industries, especially in liquefiednatural gas-related projects. Finally, spurred bythe liberalization of regulatory restrictions on realestate investment, FDI in real estate andconstruction also increased, particularly in Bahrain,Jordan, Lebanon and the Syrian Arab Republic(UNDESA and UNCTAD 2005). This has beenbolstered by the robust oil prices of the last fewyears and significant developments in the tourismsector. Bahrain, Dubai (part of the United ArabEmirates), and Qatar are the leading markets forintraregional FDI in real estate and tourism-relatedconstruction.a

The ICT industries have also attracted FDIfollowing, in particular, efforts by some countries,in the context of their “e-Government Strategy”,to attract FDI flows to such industries. For exampleDubai Internet City, a free trade zone, has attracteda large number of companies such as Canon, CiscoSystems, Compaq, Dell, IBM, Microsoft, Oracle,Siemens and Sony Ericsson. In 2004, the DubaiInternational Financial Centre, a financial free zoneallowing full foreign ownership, a zero tax rateand freedom to repatriate capital and profits withoutrestrictions, was established as an onshore capitalmarket.

/...

55CHAPTER II

Box II.6. FDI flows to West Asia increased but remain concentrated (concluded)

Box figure II.6.1. Industry distribution of numbers of greenfield investment projects andcross-border M&A deals in West Asia, 2002-2004

(Per cent)

Source: UNCTAD, cross-border M&A database (www.unctad.org/fdistatistics) as well as data from OCOConsulting, LOCOmonitor website (www.locomonitor.com).

Note: With regard to greenfield investments the industry refers to the key business function or the primaryactivity of each project. Figures in parentheses show the number of projects/deals.

Source: UNCTAD.

a “How long can the Middle East real estate boom last?”, AME Info, 4 December 2004, www.ameinfo.com, “Desirefor diversity drives building boom”, FDI Financial Times Business, 10 December 2004 (www.fdimagazine.com).

b For instance, international institutions like OECD, the United Nations Development Programme (UNDP) andthe World Bank are already involved in assisting the reform process in the West Asia’s and North Africa’s 19economies. This includes an initiative developed by the governments of these countries on “Governance andInvestment for Development”, which was approved by the OECD Council on 10 November 2004 (www.oecd.org).

Countries in West Asia continue to pursueeconomic and regulatory reforms to improve theirinvestment environment. However, despite a seriesof liberalization efforts, the past decade has not seenlarge increases in the activities of the private sectorin West Asia. The subregion is partly affected bya low “level of freedom” (UNDP 2002, p. 27) andby weaknesses in competitiveness, in particular asregards the countries’ ability to absorb new

technologies (Lopez-Claros 2004, Blanke andLopez-Claros 2005). Significant efforts toimplement financial, administrative and judicialreforms would be necessary for the subregion toenhance its attractiveness to investors and increaseFDI inflows, in keeping with its size and economicsignificance. In this process, regional initiativesand international cooperation and assistance couldplay an important role.b

Greenfield FDI

0 10 20 30 40 50 60 70 80 90 100

2002-2004

2004

2003

2002

ManufacturingTradeConstructionICT infrastructure

ExtractionBusiness servicesSales, marketing and supportLogistics and distributionElectricity Others

(302)

(176)

(798)

(320)

Manufacturing Services

Primary

Cross-border M&As

0 10 20 30 40 50 60 70 80 90 100

2002-2004

2004

2003

2002

ManufacturingTransport, storage and communicationsTelecommunicationsConstruction

ExtractionBusiness and finance servicesTradeElectricity, gas and waterOthers

(135)

(55)

(40)

(40)

Manufacturing

Primary

Services

56 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Intraregional FDI flows in Asia andOceania have grown over the years, encouragedby regional integration efforts, the expansion ofproduction networks and the relocation ofproduction to lower cost areas within the region.Intraregional FDI accounted for an estimated 46%of total flows to the region in 2002.34 Significantintraregional FDI flows took place between Eastand South-East Asia, in particular from Hong

Kong (China) to the more developed South-EastAsian countries such as Singapore and Malaysia,from Taiwan Province of China and the Republicof Korea to less developed countries such as thePhilippines and Viet Nam, and from Singaporeto China and Hong Kong (China). These flowsare also important within East Asia – originatinglargely from Hong Kong (China), TaiwanProvince of China and the Republic of Korea and

Box II.7. Recent trends in FDI inflows in the Islamic Republic of Iran

Although there were large increases in FDIflows to the Islamic Republic of Iran followingthe adoption of its new FDI law of 2002, suchflows remain modest, amounting to $0.5 billionon average over the period 2002-2004 (box figureII.7.1). Although the presence of foreign investorsin the country is indeed on the rise, it is not fullycaptured by data on FDI inflows. This is becausea large number of projects with foreignparticipation are not covered by FDI statisticscompiled on a balance-of-payments basis as theyinvolve low levels of equity or non-equityarrangements.a

In the past few years, the Islamic Republicof Iran has enjoyed strong GDP growth due inpart to high oil prices and to the implementationof regulatory reforms under the country’s thirdFive-year Development Plan, 2000-2005 (IMF2004). The main goal of the reforms is to diversifythe country’s economic structure. Efforts havebeen directed towards fostering private sectordevelopment and growth, including through

financial sector reform, privatization, further tradeliberalization and improvements in the businessclimate (box II.8). In 2002, the country enacteda foreign investment law, the Foreign InvestmentPromotion and Protection Act, which is moreliberal than the former law of 1955 (Law on theAttraction and Protection of Foreign Investment).

In the non-oil and gas sector, FDI inflowswent into a wider range of industries (includingservice industries, chemicals and machinery) in2002-2004 than in previous years. For example,no FDI was recorded in the tourism,telecommunications and electricity generation anddistribution industries in 1999-2001, while theseindustries accounted for over 60% of flows in non-oil and gas industries in 2002-2004.b

Approved data, however, show a differentpicture of foreign presence in the country fromthat based on actual data (box figures II.7.1 andII.7.2). The value of foreign investment approvedby the Organization for Investment, Economic and

Technical Assistance of Iran(OIETAI)c increased significantlyafter 2002 (box figure II.7.2). Datafrom OIETAI include FDI as wellas various types of non-equityarrangements, referred to as“indirect” investments.d Foreignparticipation in projects in the oiland gas upstream activities and innational projects that are normallyclosed to FDI can be implementedonly through contractual schemes,including buy-back arrangements(Islamic Republic of Iran, OIETAI2004). Under the buy-backarrangements, as applied especiallyto the oil and gas industries,investors receive payments over afixed period of time, rather than

/...

Box figure II.7.1. FDI inflows to the Islamic Republic ofIran and its share in total inflows to Asia and Oceania,

1993-2004

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) andannex table B.1.

0

100

200

300

400

500

600

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 20040.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

%

FDI inflows Share in total inflows to Asia and Oceania region

$m

illio

n

57CHAPTER II

targeting particularly China. FDI flows withinSouth-East Asia are also significant, withSingapore and Malaysia as the main sources ofintraregional investment in that subregion.Although intra- and inter-regional FDI flows aremuch smaller in other subregions including SouthAsia, India is emerging as a key investor fromthat subregion.

Outward FDI flows from Asia and Oceaniagrew to $69 billion (annex table B.1), driven bystronger outflows from most major economiesin the region (figure II.8). Supportive government

policies have played a role.35 Outward FDI fromHong Kong (China) witnessed the mostsignificant increase, jumping from $5 billion in2003 to $40 billion in 2004. FDI from Singaporeand the Republic of Korea also rose sharply, asdid flows from China and India. For mostdeveloping Asian economies, FDI outflows aredirected primarily at locations within the region.However, FDI outflows from Asia to otherdeveloping regions are increasing. For instance,in 2004, Latin America became the largestdestination for Chinese investment, accountingfor half of the total outflows from China due to

Source: UNCTAD.

a For example, FDI is not allowed in upstream activities in the oil and gas industries.b Based on information provided by OIETAI.c OIETAI was established in 1975 as an affiliate of the Ministry of Economic Affairs and Finance, and is legally

empowered to serve as an IPA of the country under the 2002 FDI law.d The investment law of 2002 defines two types of foreign investments, FDI and foreign “indirect” investment.e See www.petroleumiran.com.

Box figure II.7.2. Number and value of foreign investmentsa approved under the foreigninvestment laws of 1955 and 2002 in the Islamic Republic of Iran, 1993-2004

Source: UNCTAD, based on Islamic Republic of Iran, OIETAI 2004.

a Includes, under the FDI law of 2002, FDI and foreign indirect (non-equity) investments(such as buy-back financing arrangements and build-operate-transfer schemes).

equity shares, in return for their outlay on thegoods and services required for the execution ofthe projects.e As the Iranian Constitution currentlyprohibits the granting of petroleum rights on aconcessionary or equity ownership basis, theGovernment supports buy-back arrangements asa way of attracting foreign capital and servicesin oil and gas industries (Islamic Republic of Iran,Chamber of Commerce, Industries and Mines,undated).

Box II.7. Recent trends in FDI inflows in the Islamic Republic of Iran (concluded)

Political uncertainty in the region,however, is casting a shadow over the country’sforeign investment climate and future growth.The escalation of international political tensionsis an additional obstacle to attracting foreigninvestments to the Islamic Republic of Iran. Thismay affect FDI flows to the country for the nextfew years.

0

5

10

15

20

25

30

35

40

45

50

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 20040

500

1 000

1 500

2 000

2 500

3 000

3 500

4 000

4 500

5 000

Number Value on an approval basis

Nu

mb

er

$m

illio

n

58 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

massive investments in natural resources. Thelargest FDI transactions by Indian companieswere also in the natural resource sector in otherregions: in 2004, the Oil and Natural GasCorporation decided to invest $1.1 billion in theRussian Federation and $660 million in Angola.Asian investments in developed countries are alsoon the rise as illustrated by the acquisition ofIBM’s personal computers division by Lenovo(China), and by investment in FLAG Telecom(United States) and Tyco Global Network (UnitedStates) by India’s Reliance and VSNL industrialgroups respectively, in 2004.

b. Policy developments: favourablemeasures continue

The policy environment for FDI in theregion improved further over the past year (boxII.8) as more countries introduced favourablepolicy measures with a view to increasing theireconomies’ attractiveness for FDI. Countries also

cooperated in promoting investment: the ASEANFinance Ministers conducted investment roadshows in the United States in September 2004and the First Asia Summit is scheduled to takeplace in December 2005 in Malaysia to strengtheneconomic cooperation and encourage intra-regional trade and FDI flows.

At the international level, countries of Asiaand Oceania signed 33 new BITs in 2004 (figureII.10), accounting for 45% of the world total andbringing that region’s total to 956. Afghanistanconcluded its first BIT in that year (with Turkey),while China and the Republic of Korea added sixand four new treaties, respectively, to theiralready long BIT lists. In West Asia, Lebanonconcluded eight BITs, of which six were withAfrican countries. Asian countries also signed26 DTTs in 2004, bringing the total number ofDTTs involving countries of this region to 870.The Islamic Republic of Iran was the most activein that respect, concluding four new DTTs.

In China in 2004, several important policychanges took place. The Catalogue for theIndustrial Guidance of FDI was revised inNovember to take into account commitments madeby China in the context of its accession to theWTO. A number of industries have been addedto the “encouraged” category, while some havebeen re-categorized from “encouraged” to“permitted” in order to control overheatinginvestment of the domestic economy. China isfurther opening its services sector to foreigninvestment, for example by liberalizing rules onFDI in financial services, distribution services,media and education. In particular, stringentqualifications, ownership restrictions andgeographical limitations previously imposed onFDI in distribution services (such as wholesale,retail and franchising) have been removed.Meanwhile, the National Economy and SocialDevelopment Plan 2005 emphasized the need toimprove the quality of FDI by encouraging it inhigh-technology industries, advancedmanufacturing, modern services and agriculture,and environmental protection. The planencourages the establishment of R&D centres,regional headquarters and bases of advanced

manufacturing. It also welcomes the participationof foreign investors in the reform of State-ownedenterprises.

In India, the Indian Investment Commissionwas charged with the responsibility of wooingprivate investors, both domestic and foreign. TheForeign Investment Promotion Board will becomea one-stop service centre and facilitator for FDI.In 2004, foreign-equity ceilings in aviationservices, private banks, non-news printpublications and the petroleum industry wereadjusted upwards.

In early 2005, the Government of Indonesiaadopted the Jakarta Declaration outlining theGovernment’s vision for infrastructuredevelopment, and underscoring its commitmentto removing bureaucratic impediments to privateinvestment. It also introduced a one-stopinvestment service.a A number of other measuresare contemplated such as abolishing therequirement for foreign affiliates to sell part oftheir shares to local investors after a certainnumber of years of operation and removal of the30-year limit on the validity of business licencesfor foreign investors.

Box II.8. Some changes in national policies on inward FDI in Asia and Oceaniain 2004-2005

/...

59CHAPTER II

An increasing number of countries in 2003-2004 also signed or negotiated bilateral andregional FTAs that include investment provisions.ASEAN and China signed an agreement pavingthe way for establishing the world’s largest freetrade zone by 2010. ASEAN also concluded aFramework Agreement with India in October2003 and a similar process is underway withJapan (box II.9). Members of the South AsianAssociation for Regional Cooperation (SAARC)are considering signing a regional agreement forthe promotion and protection of FDI within theSAARC region.

In West Asia, a number of FTAs with FDIprovisions at both bilateral and regional levelswere signed or are under negotiation. Bahrain

and Jordan each signed an FTA with Singaporein 2004; Bahrain (2004) signed an FTA with theUnited States with a view to preparing for theUnited States-Middle East Free Trade Area by2013. At the regional level, the Gulf CooperationCouncil (GCC) signed a Framework Agreementon Economic Cooperation with India in August2004 to pave the way for a future FTA with India.The GCC is also in negotiations with China fora similar agreement. Lebanon signed an FTA withEFTA in 2004 and a draft agreement to establisha free trade area with the GCC. The GCC mayalso sign an FTA with the EU before the end of2005. Finally, the Aghadir Agreement signed inFebruary 2004 by Egypt, Jordan, Morocco andTunisia is a crucial step towards the creation ofa subregional free trade zone.

Box II.8. Some changes in national policies on inward FDI in Asia and Oceaniain 2004-2005 (concluded)

Source: UNCTAD.

a It takes 151 days in Indonesia to start a business due to the long process of obtaining a licence, compared with 33days in Thailand, 30 days in Malaysia, 56 days in Viet Nam, 50 days in the Philippines and 41 days in China (WorldBank 2005d).

In the Republic of Korea, the Korea Trade-Investment Promotion Agency and its investmentarm, Invest Korea, began to construct the InvestKorea Plaza in 2004, which will provideincubating facilities during initial investmentstages and offer easy settlement services forforeign investors, in addition to existing one-stopservices. Newly initiated corporate town projectsas well as more free trade zones were launchedin 2005. There has also been growing attentionin recent years to attracting FDI in R&D (seeChapter VII).

In December 2004, the Philippines adopteda measure allowing the establishment of wholly-owned foreign affiliates in natural-resource-relatedactivities.

In Thailand in 2004, the Board ofInvestment launched new investment packagesfor specific industries including the agro-industry,the high-end clothing (fashion) industry, theautomotive industry, the ICT industry (inparticular the hard disk drive industry) and highvalue-added services. The Skills, Technology, andInnovation tax privilege scheme was introducedto raise the technology levels and innovativecapabilities of firms, while introducing specialprivileges to promote investment in the fournortheastern provinces.

In West Asia, most of the economies aremaking efforts to liberalize their FDI regimes andimprove their investment climate (annex tableA.II.2). All countries in the region (except forQatar) have already established IPAs. In SaudiArabia’s negotiations for membership in the WTOhave accelerated the country’s liberalization ofits FDI regulatory framework. Since 2003, Turkeyhas been implementing a series of investment-related reforms as well as a privatizationprogramme in line with its planned negotiationson accession to the EU. In Bahrain and the UnitedArab Emirates, a noteworthy development is theliberalization of the real estate sector, a sectorthat is driving an intraregional investment boomboth in construction and tourism developmentprojects. Further liberalization in the financialsector in Lebanon may encourage large capitalinflows, including from the Lebanese diaspora.

In Oceania, the amendment to the ForeignInvestment Act in Fiji in 2004 applied theprinciples of the Convention on the Settlementof Investment Disputes, to which Fiji is a party.This amendment also provides for non-discrimination on grounds of nationality amongforeign investors.

60 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Figure II.10. Asia and Oceania: number of BITs and DTTs concluded, cumulative andannual, 1990-2004

(Number)

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).

c. Prospects: increasingly bright

In view of the improved economic situationin the region, a better policy environment, andsignificant regional integration efforts, theprospects for FDI flows to Asia and Oceania in2005 are highly positive: 85% of internationalexperts, 90% of TNCs and 96% of IPAsresponding to UNCTAD’s 2005 survey (box I.3)anticipated increased FDI flows to Asia (figureII.11). This is even more optimistic than in thepast, and is corroborated by a number of othersurveys and reports (A.T. Kearney 2004, IIF2005, PriceWaterhouseCoopers 2005, JBIC2005). The recent increase in cross-border M&Asin countries such as China, India and the Republicof Korea supports this optimistic assessment ofFDI prospects in the region. However, flows arelikely to remain concentrated in a few economies.

In 2003-2004 the increase in globaldemand for electronics and textiles augurs wellfor FDI in the region. FDI in ICT, as well asoffshoring and outsourcing activities willcontinue to rise as services TNCs are driven bypressures to keep costs down. Many countriesin the region will benefit because of their skills,cost and infrastructure advantages for suchactivities. Services FDI, encouraged byliberalization policies in industries such asfinance, will continue to rise, thereby increasingthe share of this sector in FDI flows to the region.

• East Asia is expected to receive the largestshare of inflows, led by a further increasein flows to China. In this country, forinstance, FDI will continue to rise inservices, in particular in the bankingindustry. Large-scale foreign investmentsare expected in China’s four largest State-owned banks before their initial publicofferings.36 Cross-border M&As areexpected to rise in service industries in othercountries. For example in finance in theRepublic of Korea, Standard Chartered(United Kingdom) acquired Korea FirstBank in 2005.

Figure II.11. Asia and Oceania: prospects forFDI inflows, 2005-2006

(Per cent of responses from TNCs, experts and IPAs)

Source: UNCTAD (www.unctad.org/fdiprospects).

0

10

20

30

40

50

60

70

80

90

100

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 20040

100

200

300

400

500

600

700

800

900

1 000

BITs (total per year: left scale) DTTs (total per year: left scale)

Total BITs (cumulative: right scale) Total DTTs (cumulative: right scale)

0

20

40

60

80

100

TNCs Experts IPAs

Increase Remain the same Decrease

61CHAPTER II

• FDI flows to South-East Asia shouldincrease in 2005 for the third consecutiveyear. Japanese companies foresee thatdemand in their host country markets inASEAN will expand, leading to higherprofits in 2005.37 Japanese manufacturersview Viet Nam in particular as a promisinglocation for production. Agreements betweenJapan and ASEAN as a group, or its membercountries individually, are expected tostrengthen FDI relationships between Japanand countries in the subregion (box II.9).Intra-regional investment will also continueto rise as the region integrates further. FDIin natural resource-related activities isexpected to rise significantly in thePhilippines.

• In South Asia, flows to India shouldcontinue to increase, especially in steel,telecommunications, infrastructure andfinance. In India, the Government aims toattract $150 billion in the next decade bysetting up special economic zones, scienceparks and free trade and warehousingzones.38 Bangladesh will receive increasedinflows as compared to 2004 primarilybecause of an increase in FDI from India.Flows to Pakistan are expected to increase

partly as a result of privatization, especiallyin the telecommunications industry. Finally,the end of the textiles and clothing quotasshould benefit countries such as Bangladesh,India and Pakistan in attracting moretextiles-related FDI (UNCTAD 2005b).

• The global oil markets will largely determinethe West Asia’s economic outlook in 2005.Although oil production and prices may notremain at their present high levels(UNDESA and UNCTAD 2005), FDI in thesubregion should rise in 2005, notably inthe production and distribution of petroleumand liquefied natural gas. While FDI growthper se will be modest, foreign presencecould rise as a result of non-equitycontractual arrangements. Significant effortsby Turkey in the investment area willcontinue, including privatization in oilrefining and telecommunications in the nextfew years.

• In the Oceania subregion 2005 is likely tobe a year of recovery in FDI flows.Countries such as Samoa will experiencehigher FDI flows as a result of relativelylarge M&A deals including the acquisitionby Virgin Blue (Australia) of a stake in thecountry’s State airline in 2005.

Box II.9. FTAs and economic partnership agreements between ASEAN or ASEAN membercountries and Japan: implications for FDI

Source: UNCTAD.

a Information from JETRO, “Japanese business sentiment in Asia improved in April”, press release of 21 April 2005(www.jetro.go.jp).

Following the 2002 Agreement betweenJapan and Singapore for a New Age EconomicPartnership, recent negotiations between otherASEAN member countries (in particular,Malaysia, the Philippines and Thailand) and Japanalso cover a broad range of provisions oninvestment, movement of personnel, intellectualproperty rights (IPRs) and competition policies.According to the Japan External TradeOrganization (JETRO) survey released in April2005, on Japanese-affiliated manufacturersoperating in six ASEAN countries (Indonesia,Malaysia, the Philippines, Singapore, Thailandand Viet Nam) and India, some 60% of thecompanies surveyed expect that FTAs or economicpartnership agreements (EPAs) between Japan andthe region where they operate will benefit their

business activities.a On a country basis, more firmsoperating in Indonesia and Thailand than in othercountries expect that such agreements will havefavourable effects. Few respondents, however,expect improvements in their business activitiesas a result of FTAs or EPAs between China andJapan or between China and ASEAN: only 22%,for instance, foresee favourable effects from theEPA between China and ASEAN.

In another survey – the 2004 survey onoverseas business operations of Japanesemanufacturing companies by the Japan Bank forInternational Cooperation (JBIC) – 72% of allrespondents expect to benefit from the conclusionof FTAs with Japan (JBIC 2005).

62 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Prospects for FDI outflows from Asia andOceania are also promising and should lead toincreased intraregional FDI. An increasingproportion of the growth in outward FDI will befrom Chinese, Indian and Korean firms, includingthrough large-scale overseas M&As. Theinternationalization of Chinese enterprises willcontinue, including through investments outsideAsia. In particular, significant Chineseinvestments are planned in natural resources(mainly in Latin America), steel (in Brazil inparticular)39 and real estate (for example, in theRussian Federation).40 China is set to becomea major foreign investor in Latin America (boxII.13). Chinese investments in developedcountries will also increase, as suggested by therecent bid made by CNOOC to acquire the UnitedStates oil firm, Unocal Corp.41 Recentappreciation of the Chinese currency maycontribute further to the increase in Chineseoutward FDI.

3. Latin America and the Caribbean:FDI inflows rebound

Following four years of continuous decline,FDI flows to Latin America and the Caribbeanregistered a significant upsurge in 2004.Economic recovery in Latin America – after halfa decade of economic stagnation – and strongergrowth of the world economy were the mainreasons for the rebound. High prices of primarycommodities also played a role. At the same timethe sectoral composition of inward FDI isshowing signs of change in some parts of theregion. In the MERCOSUR subregion, themanufacturing sector has re-emerged as theleading recipient of FDI inflows. Policy changes,particularly those related to extraction activities,could also affect FDI in some countries. Overall,FDI inflows in Latin America are projected tostrengthen further in 2005.

a. Trends: a resurgence of FDIinflows in many countries

In 2004, FDI inflows into Latin Americaand the Caribbean rose for the first time in fiveyears (figure II.12). They reached $68 billion,44% more than in 2003. However, they were stillfar below their average of the second half of the1990s when large-scale privatizations and cross-border acquisitions of private firms triggered an

FDI boom. FDI as a percentage of gross fixedcapital formation increased from 13% in 2003to 15.5% in 2004 (figure II.12). Brazil andMexico consolidated their positions as the largestrecipients of FDI in the region (figure II.13 andtable II.5). The steepest rises were seen inArgentina (125%), Brazil (79%) and Chile (73%).In Central America and the Caribbean, FDIinflows rose by 32%, to $30 bill ion, owingmainly to a sharp increase in flows to Mexico.The situation was different in the AndeanCommunity where total inflows remainedunchanged from 2003, although the trend variedfor different countries: FDI inflows rose inColombia and Peru by 53% and 37%,respectively, while they fell in Venezuela,Ecuador and Bolivia.

A combination of internal and externalfactors contributed to the strong increase in FDIinflows to Latin America and the Caribbean in2004:

• Strong economic growth in most of thecountries in the region resulted in asignificant increase in domestic demand,which attracted market-seeking FDI.

• Exchange rates remained at levels thatfavour competitiveness, although somecurrencies appreciated during 2004.42 Thisstimulated FDI in export activities and inmarket-seeking activities in manufacturing.

• The boom in demand for commodities,especially in China, helped fuel FDI inminerals in Argentina, Brazil, Chile andPeru, as well as in oil and gas in Colombia,Peru and Trinidad & Tobago. It also had anindirect impact on FDI in other relatedactivities such as the manufacture of trucks,farm machinery and extraction andexploration machinery, mainly located inMERCOSUR and dominated by TNCs.

• Windfall profits from higher commodityprices have increased reinvested earningsof resource-seeking TNCs in countries likeChile where undistributed corporate profitsare subject to a lower tax rate thandistributed dividends (17% instead of 35-42%). In Chile, reinvested earnings offoreign affiliates amounted to $6.2 billionin 2004, corresponding to 82% of totalinward FDI. These earnings were mainlygenerated by foreign affiliates in the miningsector, a sector that benefited from highermineral prices.

63CHAPTER II

• The continued recovery of the United Stateseconomy had positive effects on export-oriented FDI in the manufacturing sector inMexico and Central America.

• Cross-border M&As made a strongcomeback in the region with an increase of109% in total value, their first upturn since2000 (table II.6).

The decline in FDI inflows to Bolivia,Ecuador and Venezuela, most of which targethydrocarbon activities, is due to changes in oiland gas contracts in Venezuela, delays inadopting a new hydrocarbon law in Bolivia, andto the completion of the Crude Oil Pipeline(OCP) construction in Ecuador in 2003 that hadpreviously been associated with significantamounts of FDI.

FDI outflows from Latin America grew ata modest 3.6% in 2004, their first increase since2000, reaching $11 billion, most of which camefrom Brazil ($9.5 bill ion). The $4 bill ionacquisition of the controlling shares of thebrewer, Ambev (Brazil) , by Interbrew(Belgium),43 as well as unusual amounts of intra-company loans by Brazilian companies explainsthis high level of FDI from Brazil. Among theother 10 largest outward-investor countries inthe region, only Mexico and Costa Rica increasedtheir FDI outflows in 2004 (figure II.13).

The sectoral distribution of FDI in LatinAmerica varies by subregion and country, and

is changing. The services sector has lostimportance as a recipient of FDI in Argentina andBrazil since 2001. In Brazil, it was overtaken bythe manufacturing sector in 2004, for the firsttime since 1996 (figure II.14). In Argentina, FDIinflows to services reached negative values in2002 (figure II.15). In Mexico, FDI flows to themanufacturing sector recovered in 2004 andsurpassed those in services for the first time since2000. Conversely, in Central America and theCaribbean, the recent privatizations of publicutil i ty services in a number of countriescontributed to the growing importance of servicesas recipients of FDI. In the Andean Community,high oil and mineral prices sustained the positionof the primary sector as the main recipient of FDIinflows.

Several factors are behind the decliningflows of FDI into services in Argentina andBrazil:

• the completion of most of the privatizationprogrammes;

• strategic changes of some parent companiesfacing financial difficulties; and

• economic stagnation (1999-2003),devaluations and the rise of regulatoryconflicts, which have made this sector lessattractive to FDI since the early 2000s.

These factors provoked a number ofdivestments by foreign companies in the servicessector, particularly in the telecoms, electricity,

Figure II.12. Latin America and the Caribbean: FDI inflows and their share in gross fixedcapital formation, 1985-2004

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex tables B.1 and B.3.

0

20

40

60

80

100

120

0

3

6

9

12

15

18

21

24

27

%

Caribbean and other America

Central America

South America

FDI inflows as a percentage of gross fixed capital formation

$b

illio

n

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

64 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

banking and retailing industries (ECLAC2003, 2004b). The service firms sufferedmost from the impact of the economiccrisis. They faced serious difficulties inreducing their large foreign-currencyliabilities incurred during their expansionphase. Because of the non-tradability oftheir activities they were often unable torefocus their strategy towards export-oriented production to take advantage ofdevalued currencies as some TNCs inmanufacturing did.

In the case of Mexico, manufacturingbegan losing importance as a recipient ofFDI in the early 2000s (figure II.16) for twomain reasons: first, the emergence of thefinancial sector as an increasingly attractivearea for FDI owing to the removal of allremaining market-share limitations onforeign ownership of national banks inDecember 1998; and second, the significantdrop in FDI flows to the maquila industryduring 2001-2003 due to a downturn indemand from the United States and risingcompetition from China. The strongrecovery of FDI in the manufacturing sectorin 2004 (by 64%), exceeding that inservices, reflected new investments in themaquiladora industry, some large-scaleM&A transactions44 and improved domesticdemand.

As in other regions, resource-seekingFDI into Latin America and the Caribbeanwas stimulated in 2004 by the high prices

Table II.5. Latin America and the Caribbean: country distribution of FDI inflows,by range, 2003, 2004

2003 2004Range Economy a Economy a

More than $10 billion Mexico, and Brazil Brazil and Mexico

$5.0-9.9 billion .. Chile

$1.0-4.9 billion Chile, Cayman Islands, Venezuela, Bermuda, Argentina, Bermuda, Cayman Islands, Colombia, Peru,Argentina, Colombia, Ecuador and Peru Venezuela, Ecuador, Panama and Trinidad and Tobago

Less than $1 billion Trinidad and Tobago, Panama, Jamaica, Dominican Jamaica, Dominican Republic, Costa Rica, El Salvador,Republic, Costa Rica, Uruguay, Honduras, Uruguay, Honduras, Nicaragua, Bahamas, Belize,Nicaragua, Bolivia, Aruba, Antigua and Barbuda, Guatemala, Aruba, Paraguay, Bolivia, Saint Lucia, AntiguaEl Salvador, Bahamas, Guatemala and Saint Lucia, and Barbuda, Anguilla and British Virgin Islands, SaintGrenada, Saint Kitts and Nevis, Barbados, Belize, Kitts and Nevis, Saint Vincent and the Grenadines,Saint Vincent and the Grenadines, Paraguay, Barbados, Guyana, Grenada, Puerto Rico, Dominica, Haiti,Anguilla, Guyana, Dominica British Virgin Islands, Montserrat, Cuba, Netherlands Antilles and SurinameHaiti, Montserrat, Turks and Caicos Islands, Cuba,Suriname and Netherlands Antilles

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex table B.1.a Listed in order of the magnitude of FDI inflows for each respective year.

Figure II.13. Latin America and the Caribbean: FDIflows, top 10 economies,a 2003, 2004

(Bill ions of dollars)

Source : UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex table B.1.

a Ranked on the basis of the magnitude of 2004 FDI flows.

(a) FDI inflows

(b) FDI outflows

20032004

2003

2004

0 1 2 3 4 5

Ecuador

Venezuela

Peru

Colombia

Cayman Islands

Bermuda

Argentina

Chile

Mexico

Brazil

0 1 2 3 4 5

Trinidad and Tobago

Peru

Costa Rica

Jamaica

Colombia

Argentina

Chile

Panama

Mexico

Brazil

17

18

8

11

10

10

65CHAPTER II

of commodities. As discussed below, somecountries have changed their taxes and legislationconcerning non-renewable natural resourceactivities, specifically in the non-oil miningindustry in Chile and Peru, and in the oil industryin Argentina, Bolivia and Venezuela, in order toincrease the State’s share in natural resourcerevenues. So far these changes do not seem tohave had a major effect on FDI in non-oil mining.In 2004, $774 million – more than one-fifth of

global exploration resources in non-oil mining– was invested in Latin American countries(Chaparro 2005). Moreover, significant non-oilmining projects in Argentina, Brazil, Chile andPeru have been announced since 2004 (annextable A.II.3).

In oil and gas, TNCs have held backinvesting in Argentina, Bolivia and Venezuelapending the adoption of new regulations.However, high oil prices and the need for TNCsto maintain their reserve levels in a context ofdwindling exploration opportunities elsewhere,are likely to sustain their interest in the region.As in the case of non-oil mining, significantprojects and investment plans have beenannounced by TNCs in the hydrocarbons industryin Latin America since 2004 (annex table A.II.3).

Agricultural exports from Latin Americaand the Caribbean countries also enjoyedunusually strong growth in 2004. Overseas sales– particularly of soya beans but also of meats– were at record levels in Argentina and Brazil,notably as a result of strong demand from China.Some TNCs (e.g. Cargill (United States) andBunge (United States)), have been positioningthemselves to profit from this export boom.45

In manufacturing, TNCs registered highersales than in 2003 in South America due to theregion’s economic recovery and the growth of

Table II.6. Latin America and the Caribbean: distribution of cross-border M&Asales, by sector and industry, 2003, 2004

(Millions of dollars and per cent)

2003 2004 Growth rate Sector/industry Value % Value % in 2004 (%)

Primary 518 4.3 1 022 4.0 97Agriculture, forestry and fishing 45 0.4 26 0.1 -42Mining 473 3.9 996 3.9 111

Manufacturing 4 294 35.5 7 718 30.5 80Food, beverages and tobacco 1 175 9.7 4 182 16.5 256Wood and wood products 220 1.8 348 1.4 58Oil and gas; petroleum refining 1 490 12.3 1 070 4.2 -28Chemicals and chemical products 192 1.6 631 2.5 229Stone, clay, glass and concrete products - - 634 2.5 -Metals and metal products 964 8.0 195 0.8 -80Electrical and electronic equipment 113 0.9 565 2.2 403Other manufacturing 141 1.2 93 0.4 -35

Services 7 273 60.2 16 544 65.4 127Electricity, gas, and water distribution 334 2.8 190 0.8 -43Hotels and restaurants 97 0.8 387 1.5 297Trade - - 489 1.9 ..Transport, storage and communications 2 731 22.6 8 209 32.5 201Finance 4 003 33.1 6 275 24.8 57Business activit ies 62 0.5 744 2.9 1 099Other services 46 0.4 250 1.0 444

All industries 12 085 100.0 25 284 100.0 109

Source: UNCTAD, cross-border M&As database (www.unctad.org/fdistatistics).

Source: UNCTAD, based on data from Banco Centraldo Brazil.

Figure II.14. FDI inflows by sector inBrazil, 1996-2004(Billions of dollars)

0

5

10

15

20

25

1996 1997 1998 1999 2000 2001 2002 2003 2004

Primary Manufacturing Services

66 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

external demand. Investments by foreigncompanies were the most buoyant in theautomotive, steel, food and beverage, and sugarrefining industries. It was a boom year for thecar industry in MERCOSUR: in Argentina –where the automobile industry had experiencedpoor performance since 1999 – production andexport of vehicles jumped by 54% and 35%respectively (in units) in 2004, while domesticsales doubled. In Brazil, where the scale ofautomobile production is much larger than inArgentina, production, exports and domesticsales rose by 21%, 20% and 11% respectively(figure II.17). Car manufacturers announcedimportant investment projects in 2004, mainlyin Brazil, but also in Argentina, notably export-oriented projects in compact cars (annex tableA.II.4). In Brazil , however, the industry’sexpectations have subsequently been adjusteddownwards, mainly because of the continuedstrength of the country’s currency, relativelyhigh interest rates and declining sales abroadduring the first few months of 2005.46 FDI inthe automobile industry that targeted theMERCOSUR market during the 1990s isshifting towards export-oriented production formarkets outside MERCOSUR (box II.10).

The recovery of United States demandand the devaluation of the currencies in thedollar zone (i.e. currencies which move moreor less in conjunction with the dollar) have alsoincreased the interest of carmakers in investingin Mexico. According to the Mexicanautomotive industry association, carmakers areplanning to invest some $5.5 billion in thecountry between 2004 and 2007.47 In factseveral TNCs have already started, or haveannounced, new projects in the country (annextable A.II.4). The conclusion of an FTA withJapan is also likely to improve Mexico’sposition as a recipient of FDI in the automotiveindustry. This agreement, scheduled forimplementation in spring 2005, is part ofMexico’s strategy of reducing its heavydependence on the United States market. It isexpected to raise Japanese FDI in theautomotive industry to an estimated $1.3 billionper year up to 2015.48

Strong global demand is encouraginginvestment in Brazil’s steel industry. TheBrazilian Steel Industry (IBS) predictsinvestment (foreign and domestic) of $13 billionin 2005-2010, most of it in the form of newoutlays.49

Source: UNCTAD, based on data from Instituto Nacional deEstadística y Censos (INDEC), Argentina.

Note: The steep rise in FDI inflows to the primary sectorin Argentina in 1999 is due to the acquisition of theState-owned petroleum company, YPF (Argentina),by Repsol (Spain) for $15.2 billion.

Figure II.15. FDI inflows by sector inArgentina, 1996-2003

(Bill ions of dollars)

Figure II.16. FDI inflows by sector inMexico, 1996-2004

(Billions of dollars)

Source: UNCTAD, based on Secretaría de Economía deMéxico, Informe Estadistico Trimestral Sobre elComportamiento de la Inversión Extranjera Directaen México, Comisión Nacional de InversionesExtranjeras, www.economia.gob.mx.

Note: The marked increase in FDI inflows to the servicessector in 2001 was due to the acquisition of theMexican bank Banamex-Accival by Citigroup (UnitedStates) for $12.5 billion.

18

-1

0

1

2

3

4

5

6

7

1996 1997 1998 1999 2000 2001 2002 2003

Primary Manufacturing

Services Others

0

5

10

15

20

25

1996 1997 1998 1999 2000 2001 2002 2003 2004

Primary Manufacturing Services

67CHAPTER II

TNCs in the food and beverages industryof Latin America have benefited from growingexports and higher purchasing power in domesticmarkets, with consumers increasingly basing theirbuying decisions on brands, rather than prices,and returning to premium brands. This behaviourhas boosted business for producers of well-knownbranded foods – where TNCs have a strongpresence. Some firms have announced newinvestments,50 while others have been engagedin acquisitions in search of stronger marketposition. In beverages, for instance, the mostnotable deal is the merger between AmBev(Brazil) and Interbrew (Belgium) (mentionedearlier), and in foods it is the acquisition by Arcor(Argentina) of a majority stake (51%) inDanone’s (France) cookie and biscuit activitiesin South America.

Sugar refining in Brazil is becomingattractive to investors mainly because of theshift of car manufacturers in that countrytowards flex-fuel vehicles that run on sugar-cane-based alcohol as well as petrol.51 Foreignand local companies are reported to beplanning investments of some $3 billion inBrazil’s sugar-cane-based ethanol industry.52

FDI in the maquiladora industry inMexico surged in 2004, with a 26% increase,after three consecutive years of decline, asUnited States demand picked up. Maquilaexports were 13% higher than in 2003 andemployment levels rose for the first time since2000, registering a 5% increase. However,there is still some way to go to recover the300,000 jobs that were lost between end 2000and end 2003 (figure II.18). Employmenttrends were uneven across industries. Labour-intensive industries such as textiles andclothing, footwear and toys continued towitness a decrease in employment, while theelectrical and electronic products industryregistered the biggest rise (8% growth).53

Some attribute the upsurge in the electrical andelectronics industry to the return of someenterprises that had moved to China after thatcountry entered the WTO in 2001. Motorola,for example, inaugurated its new plant inNogales in April 2005. Others point to therelocation of some United States firms toMexico in response to the challenge posed byAsian competitors.

In Central America and the Caribbean, FDIin manufacturing is concentrated in labour-intensive activities, mainly in the apparelindustry, where TNCs have set up assembly

operations for exports almost exclusively to theUnited States. Six countries are important exportplatforms in this respect: Costa Rica, theDominican Republic, El Salvador, Guatemala,Honduras and Nicaragua. The removal of textilesand clothing quotas in January 2005 has raisedconcerns about the future of the apparel industryin the six countries.54 Some fear that the impactcould be similar to that of the entry of China intothe WTO in 2001, which, combined with theslowdown in United States demand, led to thestagnation of United States apparel imports fromCentral America (figure II.19) (UNCTAD2005b).55 Competition exists not only withChina, but with other Asian countries such asIndia, Bangladesh and Turkey. The industry couldsurvive if Central American and Caribbean

Figure II.17. Automotive industry in Argentinaand Brazil: production, domestic sales, exports

and imports, 1992-2004(Thousands of units)

Source : UNCTAD, based on Asociación de Fábricas deAutomotores (ADEFA), www.adefa.com.ar/ ;Associação Nacional dos Fabricantes de VeículosAutomotores (Anfavea), www.anfavea.com.br/.

19

92

19

93

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

20

03

20

04

Argentina

0

100

200

300

400

500

Devaluation of peso

Brazil

0

500

1 000

1 500

2 000

2 500

Domestic production Exports

Domestic sales Imports

Devaluation of real

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

68 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Box II.10. MERCOSUR: FDI in the automobile industry is targeting broader exportmarkets

Source: UNCTAD, based on ECLAC 2004b; “Latin America: Industry forecast: Getting up to speed”, Business LatinAmerica, 17 May 2004 (London: EIU); Asociación de Fábricas de Automotores (ADEFA), www.adefa.com.ar/;Associação Nacional dos Fabricantes de Veículos Automotores (Anfavea), www.anfavea.com.br/; UnitedNations Comtrade database; La Razón, www.larazon.com.

During the 1990s, TNCs made large market-seeking investments in the automotive industryin Brazil and Argentina. By the early 2000s, anestimated $20-25 billion was invested – dividedroughly four-to-one between Brazil andArgentina. The economic crises suffered bycountries in the MERCOSUR subregion from thesecond half of the 1990s until 2003 severelyaffected the automotive industry and disruptedinitial strategies aimed at the expandingMERCOSUR market.

The devaluation of the Brazilian real in1999 and of the Argentinean peso in 2002improved the export competitiveness of the twocountries and encouraged TNCs in the automobileindustries to use their capacity increasingly toproduce for export markets outside MERCOSUR.At the same time, TNC producers reorganized

their Latin American production networks:MERCOSUR affiliates specialized in small, low-cost vehicles with high fuel economy directedtowards consumers with lower purchasing power,while Mexican affiliates focused on moreexpensive models, targeting consumers with highpurchasing power, mainly in the United States(ECLAC 2004b).

Bilateral agreements between MERCOSURmember countries and Mexico, which entered intoforce in January 2003, supported this new exportstrategy through the reduction of tariffs andimplementation of import quotas. Significantincreases of automobile exports from Argentinaand Brazil to Mexico have been registered sincethen, making Mexico the main destination ofMERCOSUR countries’ vehicle exports, followedby the United States and Chile.

countries carefully evaluated their competitiveadvantages over the Asian countries (box.II.11)while building a strategy to go beyond themaquila model and diversify their export markets.

In service-related activities, assetdivestments by foreign firms that had begun inthe early 2000s are continuing, for example,Royal Ahold (Netherlands) and Carrefour

(France) in the retail industry as well as Bellsouthand AT&T in the telecom industry have sold partor all of their assets in the region. Thesewithdrawals have given opportunities tocompetitors – including Latin American TNCs(e.g. Chilean retailer Cencosud, the Mexicantelecom company Telmex)56 – to expand. Otherwithdrawals are envisaged in telecom, electricity,gas and water activities.57

Figure II.18. Maquila industry in Mexico, 1997-2004

Source: UNCTAD, based on Instituto Nacional de Estadística, Geografía e Informática (INEGI) of Mexico.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1997 1998 1999 2000 2001 2002 2003 20040

20

40

60

80

100

120

140

160

180

200FDI inflows

Exports

FDI inflows and exports($ billion)

Number of companies and employees(Thousands)

2.5

3.0

3.5

4.0

1997 1998 1999 2000 2001 2002 2003 2004800

900

1 000

1 100

1 200

1 300

1 400Number of employees

Number of companies

FD

Iin

flow

s

Export

s

Com

panie

s

Em

plo

yees

69CHAPTER II

b. Policy developments: some changesin the area of natural resources

FDI has received favourable treatment inmost Latin American countries as part of abroader free-market and liberalization policy putin place in the 1990s. This includes preferentialtreatment through, for instance, special taxregimes,58 debt-to-equity swap mechanisms59 andaccess to investor-State dispute settlementmechanisms.

To a large extent, policy-makers sought totarget a large volume of FDI on the assumptionthat i t would make a vital contribution toeconomic development. This led to the view,shared by a number of experts, that “in recentyears the region’s FDI policies have focusedalmost exclusively on attracting FDI, with noconcern for selecting or channelling it accordingto national developmental priorities. That is, FDIpolicies tended to reflect short-termmacroeconomic priorities much more than therequirements for productive development”.60

The deterioration of the economic situationduring the period 1999-2003, reflected by thestagnation of the regional economy and increasein unemployment and poverty, led to widespreaddisenchantment with the results of the economicreforms related to FDI promotion and

privatization.61 The discontent has in somecases had repercussions at the policy level.In public util i ty services, several recentinitiatives were either cancelled or suspended,such as in water services in Bolivia,telecommunications in Paraguay andelectricity in the Dominican Republic,Ecuador and Peru. In Argentina, therelationship between the Government and theprivatized enterprises – now foreign affiliatesof TNCs – had deteriorated since the end ofthe “convertibility” regime in January 2002.The incentives used in that country to attractFDI during the 1990s turned out to beunsustainable when economic conditionschanged. To address the deepest economicrecession the country had ever known, theauthorities implemented a series of measuresthat proved successful in restoring economicrecovery and growth. However, some of thesemeasures led a significant number of foreignfirms – mainly public utilities – to resort tointernational arbitration (box II.12).

In natural resource activities, social andpolitical pressures, fuelled by the strong rise incommodity prices, are pushing governments insome countries of Latin America and theCaribbean to modify their tax regimes and changeexisting legislation:

• In Argentina , taxes on oil exports wereincreased from 20% to a range of 25-45%,depending on the level of the internationalprice of oil . Moreover, after an energyshortage attributable to insufficientinvestment in the oil industry – entirelyprivatized in the 1990s and mainlycomprising foreign affiliates – the Congressapproved a bill , introduced by theGovernment in October 2004, to create aState-owned petroleum company EnergíaArgentina Sociedad Anónima (ENARSA).62

The latter has formed joint ventures withPetróleos de Venezuela (PDVSA), Lukoil(Russian Federation), Sinopec (China) andBrazil’s Petrobrás to explore offshore areas.

• In Bolivia – where petroleum activity wasprivatized in the 1990s – a new HydrocarbonLaw was approved in May 2005 by both theParliament and the Senate. It increases taxeson oil production from 18% to 50% andrequires producers to accept new contractsbased on State ownership of well-head gasin line with the results of a referendum inJuly 2004.63

Figure II.19. United States imports of appareland textile productsa from selected countries

and regions, 1997-2004(Bill ions of dollars)

Source: UNCTAD, based on data from the United StatesInternational Trade Commission, www.usitc.gov.

a Includes text i les and fabrics (NAICS-313), text i le mil lproducts (NAICS-312) and apparel and accessories (NAICS-315).

b The signatory countries of DR-CAFTA with the United Statescomprise: Costa Rica, the Dominican Republic, El Salvador,Guatemala, Honduras and Nicaragua.

0

5

10

15

20

25

1997 1998 1999 2000 2001 2002 2003 2004

China

Mexico

DR - CAFTAb

Entry of China to WTO

70 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Box II.11. Can the apparel industry in Central America and the Caribbean compete withAsia for the United States market?

Source: UNCTAD, based on IADB 2004, ECLAC 2004b, Quinteros 2004, UNCTAD 2005b.

a There are some exceptions: for example, the Costa Rican apparel industry uses a qualified workforce and isspecialized in niche markets.

b The production-sharing mechanism allows imports incorporating United States-made components to enter theUnited States either free of duty or at reduced duties.

c At the time of writing this report, DR-CAFTA had been ratified by Guatemala and El Salvador and still needs tobe ratified by each of the other parties before it can enter into force.

The high level of competitiveness of Asia’sapparel industry stems not only from lower wages,but also from the reorganization of that industryinto an integrated system of production thatencompasses all phases, from inputs to completedproducts. The integrated system of production inAsia has boosted the development of a strongregional cluster in textiles and apparel. It offersrapid and cheap access to a vast supply ofspecialized inputs for the industry (fibres, yarnsand fabrics) as well as access to diversified exportmarkets. The competitive advantage of the CentralAmerican and Caribbean countries in the industryhas, by contrast, been derived from a combinationof factors, including low wages,a exportprocessing zones and preferential access to theNorth American market – characteristics that makethem well suited to final product assembly(ECLAC 2004b). The apparel industry in CentralAmerica is specialized in catering to a singleexport market – that of the United States. Exportsare, moreover, strongly dependent on aproduction-sharing mechanism.b This mechanismhas led foreign apparel firms operating in thesecountries to use expensive United States inputs,while keeping domestic value added low (ECLAC2004b).

Central American countries have twoadvantages over Asia: geographic proximity to

the United States, which offers the opportunityto deliver goods faster than China or other Asiancountries can do, and to respond quickly tochanges in United States market conditions andspecial demands; and duty-free access to theUnited States market for textile and apparelexports under the United States-Caribbean BasinTrade Partnership Act (CBTPA), provided theyarns, fabrics and threads are imported from theUnited States.

In 2004, Costa Rica, the DominicanRepublic, El Salvador, Guatemala, Honduras,Nicaragua and the United States signed the UnitedStates-Dominican Republic-Central America FreeTrade Agreement (DR-CAFTA).c The commercialpart of this agreement transforms the unilateralUnited States concessions of the CBTPA intopreferential treatment by each party for goodsimported from any other party. It relaxes the rulesof origin by extending the agreement to regionalinputs and making it more flexible for somespecific products; but, generally, it fails to securetariff preferences for exports within the DR-CAFTA region that use cloth and materials fromthird countries outside the region. The latter wouldhave allowed the region to import competitiveinputs, including from Asia, and to compete betterwith Asian final producers no longer restrictedby quotas.

Argentina’s privatization of public utilityfirms is an example of the need for policy-makersto weigh carefully the costs and benefits ofincentives for FDI. At the beginning of the 1990s,a programme to privatize public utility firms waslaunched, which set bidding conditions that madeit necessary for interested local firms to associatewith foreign ones and offered incentives such asa debt-to-equity swap mechanism. Furtherincentives were added shortly after privatization:

some taxes were reduced or eliminated and newclauses were introduced to the contracts in whichutility rates were denominated in dollars andindexed to the United States’ inflation index.During the same decade, Argentina signed 54 BITsto provide security and guarantees for investors.

Problems began to surface when economicconditions in the country deteriorated. Economiccontraction, massive withdrawals of banking

Box II.12. The need to weigh the costs and benefits of incentives to FDI: the experience ofArgentina

/...

71CHAPTER II

Source: UNCTAD, based on ICSID 2005, IISD 2005, Azpiazú 2004, Bouzas and Chudnovsky 2004, Alfaro 2004, “Laespañola Gas Natural Ban retira su demanda contra la Argentina”, Clarín, 15 March 2005; “AES retiró sudemanda en el Ciadi y se acelera el acuerdo”, La Nación, 15 Abril 2005, “, “Acuerdo del Gobierno y Edesurpara subir tarifas”, La Nación, 12 June 2005, and communication from the Mission of Argentina to the UnitedNations office in Geneva.

a Official communications from the Government of Argentina.b The tribunal also decided that after the payment of the compensation CMS will transfer its assets in its Argentinean

affiliate to the Argentinean State, provided the latter makes the payment of an additional $1.1 million. The tribunalgives Argentina a period of one year in which to accept such a transfer (ICSID 2005).

c Section 5 of Chapter IV deals with the “interpretation, revision and annulment of the award”.

Box II.12. The need to weigh the costs and benefits of incentives to FDI: the experience ofArgentina (concluded)

deposits and a rapid decline in internationalreserves forced the Government in January 2002to abrogate the convertibility law that fixed thepeso’s exchange rate at par with the United Statesdollar. The trebling of the value of the dollar inlocal currency that resulted, in the context of deepeconomic recession, led the Government totransform all the dollar-denominated contractsinto national-currency-denominated contracts,including those signed with public utility firms.The periodic adjustments of public utility tariffsbased on foreign inflation indices were alsoeliminated.

In the following months a number of foreigninvestors resorted to arbitration by theInternational Centre for Settlement of InvestmentDisputes (ICSID) and other fora. Indeed, 37 outof the 40 arbitration cases to which the ArgentineGovernment is party (as of June 2005) wereregistered after the 2002 emergency measureswere introduced, and are related, at least in part,to the financial crisis. A majority of these caseswere launched by public utility firms claimingbreach of contract and violation of treatyguarantees provided under BITs, such as fair andequitable treatment or guarantee against (indirect)expropriation.

Argentina has stated that “it has not offeredany guarantee concerning the maintenance of theconvertibility system and in case of devaluationof its currency, because the Government couldnot have assumed an obligation to follow anyspecific economic or exchange policy since it canfreely modify those policies.”a In Argentina’sview, its actions had been rendered necessary byan imminent economic, financial and social crisisin the country, and it thus referred to a state ofnecessity. Argentina has also contended that “theemergency measures adopted by the Governmentare to be considered as economic policy regulatory

measures that do not give right to compensation.They were instrumented through legislative actsof general scope, non-discriminatory, andtherefore applicable to both Argentine and foreignnationals without any distinction. They aretemporary in nature and oriented at the protectionof public welfare interests, with a view tonormalize the life of the country, to guarantee thecontinuity of public utilities and to keep rates forcustomers at an affordable level.”a

At the same time, the Government has beennegotiating gradual tariff increases with privately-owned public utilities provided that internationalclaims are withdrawn. At least one complainant– the energy company Pioneer Natural Resources(United States) – withdrew its complaint in April2005, and negotiations with other energy firmssuch as AES (United States), Gas natural BAN(Spain) and Edesur (Spain) are reported to be atan advanced stage.

An ICSID tribunal rendered a first awardin the long list of pending cases on 12 May 2005.The tribunal ordered Argentina to pay $133.5million plus interest in compensation to CMSb

on the grounds of breach of contract and violationof the BIT between Argentina and the UnitedStates. The tribunal rejected Argentina’s argumentsbased on a state of necessity as well as theinvestor’s contention that it had suffered anindirect or regulatory expropriation of itsinvestment.

At the time of writing this report, it is notknown whether Argentina or CMS will initiateany of the procedures established in Chapter IV,Section 5 of the ICSID Conventionc in relationto this award. Some officials have mentioned,however, that considering the scope of ICSIDarbitration awards, their validity could bechallenged in Argentina’s Supreme Court.

72 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

• In Chile, the Congress approved a law inMay 2005 creating a tax of 5% on theoperating profits of non-oil mining groupswith an aggregate annual output of 50,000tonnes or more of fine copper equivalent.The new tax, effective in January 2006, willbe deposited in a fund to finance innovationand R&D activities generally so as toprepare for the time when mining resourcesare exhausted.

• In Peru the Congress approved a bill tocharge royalties ranging between 1 and 3%on non-oil mining outputs.

• In Venezuela, the Government increasedroyalties on extra-heavy oil from 1% to16.67% in October 2004. Later, in April2005, it announced that 32 oilfield operatingcontracts with foreign oil companies, whichaccount for almost one-quarter of total oilproduction, would be cancelled by the endof the year and renegotiated under newterms. Income taxes and royalty levels willbe higher, and Venezuela’s State-owned oilcompany, Petróleos de Venezuela (PDVSA),will hold a majority share in the ventures.To be allowed even to enter into talks fornew deals, operators may have to paycompensation for underpaying their incometax, which the Government is claiming theyhave been doing since 2000. 64

These policy changes show growingconcern in Latin America and the Caribbeancountries regarding the impact of FDI on theireconomies, in particular in the area of natural

resources. I t does not mean, however, thatopenness to FDI in the region is being reversed.For instance, a number of policy changes thatcan have a favourable impact on FDI also tookplace in these countries in 2004, including a newinvestment promotion regime in Argentina forinvestments in capital goods in manufactures andinfrastructure;65 a new industrial and innovationpolicy in Brazil that gives incentives toinvestments in targeted sectors (ECLAC 2005);measures to end monopolies in mobiletelecommunications in Barbados and in thetelecom sector in Cayman Islands; removal oflimitations to foreign ownership in the transportindustry in Guatemala; and a reduction of thecorporate income tax rate (for both foreign andlocal firms) in Barbados, Mexico and Uruguay.

At the bilateral level, Latin Americancountries signed 12 DTTs and 6 BITs during 2004(figure II.20). Among the latter, the BIT signedbetween Uruguay and the United States was thefirst agreement based on the new United Statesmodel BITs text. The total number of BITs andDTTs involving Latin American countries reached451 and 306 respectively at the end of 2004.

At the regional level, an FTA betweenCentral America, the Dominican Republic andthe United States of America (DR-CAFTA), theFree Trade Agreement between the CaribbeanCommunity (CARICOM) and Costa Rica as wellas one between Mexico and Japan for theStrengthening of Economic Partnership (all three

Figure II.20. Latin America and the Caribbean: number of BITs and DTTs concluded,cumulative and annual, 1990-2004

(Number)

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).

0

10

20

30

40

50

60

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 20040

100

200

300

400

500

BITs (total per year: left scale) DTTs (total per year: left scale)

Total BITs (cumulative: right scale) Total DTTs (cumulative: right scale)

73CHAPTER II

with substantive investment disciplines) wereconcluded. Other agreements with investmentprovisions signed in 2004 include the PartialReach Agreement for Economic, Trade andInvestment Promotion between Argentina andBolivia as well as the Comprehensive EconomicCooperation Agreement between Chile and India.

c. Prospects: growing opportunities

FDI flows to Latin America and theCaribbean are expected to rise further in 2005-2006 as most of the driving forces behind FDIgrowth in 2004 still exist. The macroeconomicenvironment in the region has improved, andeconomic growth is expected to remain robustin 2005 (around 4%) (IMF 2005, UNCTAD2005c). After a prolonged period of economicstagnation (1999-2003), investments are requiredthat will help modernize and expand productioncapacity and to remove infrastructure bottlenecksmainly in energy roads and ports to meet growinginternal and external demand. In addition, theeconomic recovery in Argentina and thesuccessful restructuring of its external debt haveremoved a source of macroeconomic instabilityin the Southern Cone region.

UNCTAD’s 2005 survey (box I.3) alsoshows positive prospects for FDI in LatinAmerica and the Caribbean, though the outlookis less optimistic than for countries in Asia andOceania or South-East Europe and the CIS. Themajority of IPAs in Latin America and theCaribbean, along with two out of five FDI expertsand one out of three TNCs, expect FDI to theregion to increase, while about half the FDIexperts and two out of three TNCs expect it toremain at the same level (figure II.21).

FDI is l ikely to grow unevenly acrosssectors and subregions. In the primary sector,where projects are concentrated in the SouthAmerican countries, FDI inflows should continueto be attracted by relatively high levels ofcommodity prices driven by strong worlddemand. Taxes and legislative changes aimed atincreasing the State’s share in natural resourcerevenues have not prevented TNCs fromannouncing important projects in 2004 and 2005.Higher prices and the entry of new investors seemto be improving the bargaining position ofgovernments. Growing demand for resources suchas oil, copper, iron ore and soybeans is increasingdeveloping-country firm’s interest as well ininvesting in Latin America (as noted in the

previous section on Asia and Oceania). Forexample, high profile visits with publicstatements of large investment plans, and thesignature of several cooperation agreements,accompanied by the actual launching of newprojects, have raised expectations of a substantialincrease in Chinese investments in the region incoming years (box.II.13).

In manufacturing, the Governments ofArgentina and Brazil have shown interest indeveloping supportive policies, with incentivesdirected to specific areas identified as priorities.At the same time, there is risk of a slowdown ininvestment projects in Brazil due to the continuedstrength of the currency and high interest rates.66

In the case of FDI in the maquiladora industriesof Mexico, Central America and the Caribbean,prospects are mixed. Economic growth in theUnited States is expected to register a moderateslowdown, but should nonetheless remain at 3-3.5% in 2005 (IMF 2005, UNCTAD 2005c). Ofgreatest concern to those industries is increasingcompetition from Asian countries. However, asfar as the automobile industry is concerned,investment projects launched or announced in2004 and 2005 in Mexico would guaranteesignificant FDI flows into the industry (and henceinto the manufacturing sector as a whole) in theshort term.67

In services, DR-CAFTA is expected tofacilitate FDI in Central America, mainly byUnited States and Mexican firms, although theratification of the agreement is still uncertain.68

Figure II.21. Latin America and the Caribbean:prospects for FDI inflows, 2005-2006

(Per cent of responses from TNCs,experts and IPAs)

Source: UNCTAD (www.unctad.org/fdiprospects).

Increase Remain the same Decrease

0

20

40

60

80

100

TNCs Experts IPAs

74 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

In the Southern Cone countries, privatizationsare likely to be modest due to the near-completionof the process. However, the consolidation of thesubregion’s economic growth is likely to revivethe interest of foreign investors, particularlyleading Latin American TNCs that would like tocontinue expanding regionally.

As regards FDI outflows from the region,a further increase can be expected in the comingyears. Leading Latin American TNCs areexpected to continue to expand, principally toneighbouring countries and regionally, thoughglobal expansion is also likely to increase. Thisis in line with the growing transnationalizationof firms from developing countries in recentyears.

In conclusion, the recovery of economicgrowth in Latin America, higher demand forcommodities and policy support to manufacturingactivities in some countries are opening up newbusiness opportunities for foreign investment inthe region. These opportunities are somewhatdifferent from those that prevailed during thepeak period for FDI in the 1990s; they are likelyto be more in manufacturing, construction andnatural resources, than in the services sector, andto involve the creation of new assets more than

the acquisition of existing ones. Moreover, theyare expected to engage new actors, such asChinese firms, and to give more prominence toLatin American TNCs. Finally, as most of thedrivers behind the resurgence of FDI in the regionrelate to developments in the Southern Cone, FDIis expected to be more buoyant in South Americathan in Mexico, Central America and theCaribbean in 2005 and beyond.

B. South-East Europeand the CIS: FDI rises forthe fourth year in a row

1. Trends: FDI inflows sharply up

FDI inflows to South-East Europe and theCIS, a new regional grouping of economiesintroduced in this WIR (box I.2), recorded theirfourth year of growth in 2004, reaching an all-time high of $35 billion (figure II.22). Trendsin inward FDI to the two subregions differsomewhat, however, reflecting the influence ofdivergent factors. In South-East Europe, FDIinflows started to grow only in 2003, and withintwo years, led by large privatization deals, they

Box.II.13. China’s new investment interest in Latin America

Source: UNCTAD, based on “Abren la puerta para negocios con China por US$ 20.000 millones”, Clarín, 16November 2004, “Brazil/Argentina: China’s long-term commitments”, Business Latin America (London:EIU), 15 November 2004; “Brazil: Lula’s China commitments”, Business Latin America (London: EIU), 7June 2004, “Brazil: China appeal”, Business Latin America (London: EIU), 17 May 2004, Dumbaugh andSullivan 2005.

a Data from United Nations COMTRADE database.

China’s interest in Latin America is a fairlynew phenomenon that has developed along withthe steady increase of its imports – mostly ofnatural resource products – from the region.China’s imports from Latin America rose morethan fivefold between 2000 and 2004, reaching$20.2 billion; this increased the region’s share intotal Chinese imports from 2.1% to 3.6%.a

The visit of the President of China to Brazil,Argentina, Cuba and Chile in November 2004,accompanied by some 200 Chinese businesspeople, demonstrates the growing interest ofChinese TNCs in Latin America. In a speech tothe Brazilian Congress during this visit, it wasannounced that China would invest $100 billionin Latin America over the next 10 years,

particularly in railways, oil exploration andconstruction projects in Argentina; a nickel plantin Cuba; copper mining projects in Chile; alongwith steel mill, railway and oil exploration projectsin Brazil. This reflects the new Chinese strategyin Latin America of securing access to naturalresources through FDI.

While Chinese companies already ownstakes in minerals operations in Ecuador, Peru andVenezuela, among others, China intends to expandits trade and investment activities in the region.Moreover, the country has signed 14 cooperationprotocols with Brazil and 19 with Venezuela. Inaddition, China and Chile announced in 2004 thatthey would be negotiating a bilateral free tradeagreement.

75CHAPTER II

nearly tripled, to $11 billion. In the CIS, inflowsgrew from $5 billion in 2000 to $24 billion in2004, driven largely by high prices of petroleumand natural gas. FDI inflows into the region areexpected to grow further over the next few years.

Of the 19 countries in the group, 16received higher flows than in 2003. Inflowsremain concentrated in a few economies. In 2004,the top 10 destinations accounted for 95% offlows to the region (figure II.23). The RussianFederation alone, with its large natural and humanresources, accounted for more than one-third ofthe group’s total inflows. The oil economies ofAzerbaijan and Kazakhstan accounted for anotherquarter. The two South-East European countries(Bulgaria and Romania) expected to join the EUin 2007 together accounted for more than one-fifth of the regional total and for more than 70%of the South-East European subtotal.

The distribution of FDI inflows by sizeamong the region’s economies remained stablein comparison with that in 2003: only Romaniamoved to a higher bracket of FDI inflows andSerbia and Montenegro to a lower one ascompared with 2003 (table II.7).

In South-East Europe, as in previous years,the EU candidate countries, Bulgaria andRomania, were the main recipients of inward FDI

in 2004. Romania alone attracted more FDI thanthe five countries on the western side of thesubregion (Albania, Bosnia and Herzegovina,Croatia, TFYR Macedonia, Serbia andMontenegro) together. With the exception ofCroatia – the only upper middle-income economyof South-East Europe and the CIS – the lowlevels of inward FDI reflect GDP per capita levelsthat are even lower than in Bulgaria andRomania, combined with a post-conflict situationthat has had a negative impact on infrastructureand has made potential investors cautious.

In Romania, the record level of inflows ($5billion) was partly a result of the privatizationsale of the oil company, Petrom, to OMV(Austria). Inflows were also important ingreenfield and expansion projects, particularlyin the automotive industry and in services. InBulgaria in 2004, Telekom Austria acquired thetelecom operator MobilTel, while Viva Ventures(United States) took majority control of theBulgarian Telecommunications Company (BTC).The power industry also received majorinvestments in 2004 from Austria, the CzechRepublic and Germany.

The industry composition of FDI inflowsin South-East Europe is affected by these majortransactions (annex table A.II.5). The

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex tables B.1 and B.3.

Figure II.22. South-East Europe and CIS: FDI inflows and their share in gross fixed capitalformation, 1992-2004

(Billions of dollars and per cent)

$bill

ion

0

5

10

15

20

25

30

35

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

0

2

4

6

8

10

12

14

16

18

20

%

CIS

South-East Europe

FDI inflows as a percentage of gross fixed capital formation

76 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Table II.7. South-East Europe and CIS: country distribution of FDI inflows,by range, 2003, 2004

2003 2004Range Economy a Economy a

Above $5.0 bil l ion Russian Federation Russian Federation and Romania

$1.0-4.9 bil l ion Azerbaijan, Romania, Bulgaria, Kazakhstan, Azerbaijan, Kazakhstan, Bulgaria, Ukraine andCroatia, Ukraine and Serbia and Montenegro Croatia

Less than $1.0 bil l ion Bosnia and Herzegovina, Georgia, Albania, Serbia and Montenegro, Georgia, Bosnia andBelarus, Armenia, Turkmenistan, TFYR Herzegovina, Albania, Tajikistan, Armenia,Macedonia, Republic of Moldova, Uzbekistan, Belarus, TFYR Macedonia, Republic of Moldova,Kyrgyzstan and Tajikistan Turkmenistan, Uzbekistan and Kyrgyzstan

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex table B.1.a Listed in order of the magnitude of FDI inflows for each respective year.

manufacturing sector dominated inflows only inRomania in 2003 and 2004.69 The sector alsotook a sizeable share of FDI in Bulgaria, althoughthe share declined in 2004. Within services, tradeand telecommunications played particularlyimportant roles as a result of recent privatizationdeals.

In the CIS, four countries, the RussianFederation, Azerbaijan, Kazakhstan and Ukraine,in that order, together accounted for 93% of thesubregional total of FDI inflows in 2004. In thefirst three countries, FDI was driven by projectsin natural resources (especially petroleum andnatural gas) and related activities,70 while inUkraine (the second largest country in area onthe European continent after the Russian

Federation) it was more broad-based:besides oil companies such as Lukoil(Russian Federation) and RegalPetroleum (United Kingdom), the listof companies with major FDI projectsin 2004 in Ukraine includedmanufacturers of consumer goods,construction materials, retailing andtelecommunications firms (annex tableA.II.5).

In the Russian Federation,petroleum and natural gas extractionattracted large investments from TNCsin 2004, especially in the Russian FarEast island of Sakhalin. Inflows alsorose as some round-tripped Russiancapital returned from Cyprus andLuxembourg.71 In Azerbaijan, acombination of high oil prices andprospects of an imminent opening ofthe pipeline linking the Azeri capital,Baku, to the Turkish Mediterraneanport, Ceyhan, prompted a rise in FDI

in petroleum in 2004.72 In Kazakhstan, a surgein FDI led to a 16% rise in oil and gas outputin 2004. The country attracted both globalpetroleum firms and independent oil companies.73

It also attracted large FDI projects in othernatural resources such as aluminium in 2004.

The industry composition of cross-borderM&As has changed from year to year. In 2003,petroleum refining (part of coke, petroleum andnuclear fuel) alone accounted for 82% of cross-border M&A sales receipts (table II.8). This ismainly due to the acquisition of the Tyumen OilCompany (TNK) of the Russian Federation byBP (reported in WIR03, p. 62). In 2004, servicesaccounted for close to two-thirds of the M&A

Figure II.23. South-East Europe and CIS: FDI inflows,top 10 recipients,a 2003, 2004

(Bill ions of dollars)

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) andannex table B.1.

a Ranked on the basis of the magnitude of 2004 FDI inflows.

2003

2004

0 1 2 3 4 5

Bosnia and Herzegovina

Georgia

Serbia and Montenegro

Croatia

Ukraine

Bulgaria

Kazakhstan

Azerbaijan

Romania

Russian Federation8

12

77CHAPTER II

sales, with telecommunications accounting forthe largest deals.

After two years of growth (2002-2003),FDI outflows from South-East Europe and theCIS declined slightly in 2004. This was due tothe slowdown of outward FDI by Russian TNCs,which alone represent about 99% of the regionaltotal. This slowdown, in turn, is mostly the resultof a changing relationship between theGovernment and the business sector that hasprompted firms to slow down their expansionabroad.

Projects abroad by Russian firms oftentarget other CIS countries: for example, LukoilOil Company signed a $1 billion natural gas dealin Uzbekistan in 2004 to be financed over 35years. Lukoil will own 90% of the joint ventureformed for this purpose.74 Outside the CIS,Norilsk Nickel completed in 2004 the acquisitionof its stake in South Africa’s Gold Fields (WIR04,p. 74). While traditionally Russian outward FDIhas been driven by firms based in naturalresources (chapter I and annex table A.I.11), theindustry base for outward FDI is broadening toinclude other activities such as telecommunications.

2. Policy developments: diversityin policy approaches

FDI patterns in individual South-EastEuropean and CIS countries reflect not onlynatural-resource endowments and other location-specific economic factors, but also diversity inpolicy approaches to inward FDI. In Bulgaria andRomania, the prospect of joining the EU in 2007is prompting rapid adoption of the EU’s acquiscommunautaire , increased efforts towardsimproving the business environment and thecompletion of large privatization deals. OtherSouth-East European countries are followingthese two in varying degrees.

In the CIS, policies relating to FDI andprivatization are diverse. So is the approachtowards the treatment of FDI in natural resources.In the area of privatization, for example, theRussian Federation and Ukraine follow divergentstrategies, despite the fact that in both countriesthe main challenge is to tackle the consequencesof earlier deals, which led to insider ownershipof key resources (Bevan and Fennema 2003,Nureev and Runov 2003, Puffer and McCarthy2003, Shlapentokh 2004).

Table II.8. South-East Europe and CIS: distribution of cross-border M&A sales,by sector and industry, 2003, 2004

(Millions of dollars and per cent)

2003 2004 Growth rateSector/industry Value % Value % in 2004 (%)

Primary 94 0.8 32 0.3 -66.3Agriculture, hunting, forestry and fishing 10 0.1 4 0.04 -57.8Mining, quarrying and petroleum 83 0.7 27 0.3 -67.3

Manufacturing 10 997 88.7 3 827 38.1 -65.2Food, beverages and tobacco 743 6.0 241 2.4 -67.5Textiles, clothing and leather 1 0.01 - - -Wood and wood products 0.2 - - - -Publishing and printing 24 0.2 - - -Coke, petroleum and nuclear fuel 10 177 82.1 3 238 32.2 -68.2Chemicals and chemical products 1 0.01 23 0.2 2228Non-metall ic mineral products - - 167 1.7 -Metals and metal products 48 0.4 156 1.6 228.7Machinery and equipment 3 0.03 - - -Motor vehicles and other transport equipment 0.2 - 1 0.01 419.5

Services 1 304 10.5 6 188 61.6 374.6Electricity, gas and water 26 0.2 851 - 3164Trade 128 1.0 9 0.1 -92.8Hotels and restaurants 4 0.03 - - -Transport, storage and communications 677 5.5 4 919 49.0 626.3Finance 423 3.4 347 3.5 -18.0Business services 46 0.4 30 0.3 -34.0Health and social services - - 2 0.02 -Community, social and personal service activit ies - - 31 0.3 -

All industries 12 395 100.0 10 047 100.0 -18.9

Source: UNCTAD, cross-border M&As database (www.unctad.org/fdistatistics).

78 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

In the Russian Federation, authorities haveadopted a two-pronged approach towards firmsprivatized in the early 1990s. This strategy hasimportant implications not only for inward butalso for outward FDI. The Russian strategy onpost-privatization has, on the one hand, tried toincrease de facto the Government’s influence overthese firms. On the other hand, the authoritieshave used, or are planning to use, direct measuresto take back State control of some key companies.For instance, in June 2005 the Governmentincreased its stake in Gazprom, the country’slargest natural gas producer, from 39.27% to50.01%. In the oil industry, following an auditthat identified $28 bill ion in unpaid taxes,authorities took back control of the coreextraction company of the second largest Russiancorporation – and a large outward investor –Yukos.75

There is a danger that these actions couldsend contradictory signals to foreign investors.On the one hand, the weakening of oppositionto foreign shareholding in local companies(mostly informally) and the direct acceptance offoreign minority shareholding (e.g. BP-TNK) aresigns of opening up. The evolution of the taxsystem towards flat and lower taxes could alsoencourage foreign investors. In 2002, corporateincome tax (“profits tax”) was set at a flat 24%,while the Government eliminated the previouslywidespread use of tax concessions and specialfavourable tax regimes (OECD 2004a, p. 33). Onthe other hand, there are measures that coulddiscourage inward FDI. Liberalization of foreignequity investment in key companies is advancingslowly. Limitations on foreign ownership inGazprom and United Energy Systems had beenoriginally set at 20% and 25%, respectively, inthe late 1990s. These limits are to be raisedgradually. Moreover, foreign ownership couldbe de facto limited to 49% by domesticregulations on natural resources, such as thedecision in February 2005 of the Ministry forNatural Resources of the Russian Federation torestrict new tenders for oil and metal depositsto companies that are at least 51% Russian-owned. This prevents not just foreign affiliatesbut also joint ventures from exploiting new oilreserves in the country. This rule could alsopotentially affect Russian oil firms in which thecombined foreign portfolio and direct ownershipmight reach 50%.

In the fiscal area, “…although the new TaxCode significantly clarifies the roles and powersof tax inspectors and tax bodies, and grants

greatly expanded rights to taxpayers, taxenforcement remains political and often arbitrary”(OECD 2004a, pp. 34-35). In this context, theextension of tax audits from Yukos to the BP-TNK joint venture76 has been interpreted as anegative sign by foreign investors (IIF 2004).In the latest investment climate survey of thecountry, as many as 75% of the firms surveyedconsidered the interpretation of regulations byauthorities as unpredictable (World Bank 2005e,pp. 23 and 246).

In Ukraine, the new Government that cameto power at the end of 2004 seems to be openingits doors wider to foreign investors. In February2005, the authorities decided to revise earlierprivatizations by annulling the results of unlawfulinsider deals and putting the shares of thecompanies concerned on sale again. The list offirms that could be re-privatized this wayincludes key companies such as the steelmakerKryvozyzhstal, the metallurgical conglomerateUkrrudrpom, the Petrovsky Steel Plant, theNikopol Ferroaloys Plant, the Dzerzhinsky MetalPlant, the chemical factory Azot Severodonetskand the Nikolaev aluminium plant.77

The Russian Federation and other CIScountries also diverge with regard to theregulation and treatment of FDI in natural-resource extraction. Azerbaijan, Kazakhstan andTurkmenistan not only apply fewer limits on theforeign ownership of oil and gas, but also levylower taxes and royalties on oil than does theRussian Federation. For instance, in 2004, firmsin Kazakhstan paid $1.5-$2 of royalties per barrelof oil compared with $6-$7 in the RussianFederation, and investors were offered taxstability clauses (Dashevsky and Loukashov2004, p. 13).

With respect to the international frameworkfor investment, South-East European and CIScountries signed 17 new BITs in 2004 (figureII.24) bringing the total number of BITs involvingthis group of countries to 642. This increase wasthe lowest level registered since 1991. In 2004,29 new DITs were concluded bringing the totalto 494.

3. Prospects: continuing growth

FDI inflows to South-East Europe and theCIS are expected to grow further in the nearfuture based on the expectation that, with theircompetitive wages, South-East Europe (especiallythe two countries in the subregion that are

79CHAPTER II

Figure II.24. South-East Europe and CIS: number of BITs and DTTs concluded, cumulativeand annual, 1990-2004

(Number)

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).

expected to join the EU in 2007), and Ukrainefrom the CIS will attract an increasing numberof efficiency-seeking or export-oriented projects.At the same time, high oil and gas prices willcontinue to encourage FDI in the natural-resource-rich CIS countries. In both groups, FDIinflows may be affected positively byimprovements in the business environment.

In South-East Europe (and partly also inBelarus, western Russia and Ukraine in the CIS),the eastward expansion of the EU in 2004 createdmajor transportation and logistical advantages,as these countries became immediate neighboursof the EU. This “new frontier” (UNCTAD 2003a,p. 17) could potentially become a magnet forefficiency-seeking investment. It is not yetcertain, however, if new greenfield projects couldcompensate for the drop in privatization-relatedinflows once the current wave of largeprivatization deals is completed.

Adding to the “new frontier” status of thecountries mentioned are the advantages offeredby low labour costs, which are even lower thanthose of the new EU members that joined the EUin 2004 (figure II.25). Gross wages in Bulgariaand Romania are comparable with those of Indiaand China. However, to exploit this advantagethese South-East European countries would alsoneed to offer similar levels of labour productivity.The forecast that their textile, garment andfootwear industries in 2005 would be negativelyaffected by competition from China (Hunya 2005)suggests that currently this is not the case.

In the natural-resource-rich economies ofthe CIS it is not simply the volume of inward FDIthat will matter in the future, but rather, theirsuccess with diversification into new activities.In this respect, Kazakhstan and the RussianFederation have slightly broader natural resourcebases and downstream activities than doAzerbaijan or Turkmenistan. Prospects fordiversifying FDI inflows away from naturalresources are not necessarily promising, however.What makes diversification difficult is theadverse impact of the “Dutch disease”78 onproduction costs in other industries: as large oiland gas exports lead to a real appreciation of thelocal currency, production costs in manufacturing,expressed in dollars, increase to internationallyuncompetitive levels.

The CIS also includes countries, such asKyrgyzstan, the Republic of Moldova, Tajikistanand Uzbekistan, where GDP per capita iscomparable with that of the poorest countries ofthe world. Some of these countries suffer fromconflict situations and other politicaluncertainties. These conditions make it difficultto overcome marginalization through variousstrategies, including attracting and leveraginginward FDI.

On balance, the prospects for FDI inflowsto South-East Europe and the CIS in 2005 and2006 are deemed positive by FDI experts, TNCsand IPAs alike (box I.3). In all three groups nineout of ten respondents believe that FDI flows to

0

10

20

30

40

50

60

70

80

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 20040

100

200

300

400

500

600

700

800

BITs (total per year: left scale) DTTs (total per year: left scale)

Total BITs (cumulative: right scale) Total DTTs (cumulative: right scale)

80 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Figure II.25. The wage ladder: gross pay per annum in selected economies, 2004(Median, thousands of dollars)

Source: UNCTAD, based on Mercer Human Resource Consulting, “2005 international geographic salary differentialreport”, www.mercerhr.com.

Note: Asian Tigers include Hong Kong (China), the Republic of Korea, Singapore and Taiwan Province of China.

0

5

10

15

20

25

30

Western

Europe

New EU

members

Bulgaria

and

Romania

Russian

Federation

(Moscow)

Asian

Tigers

India China Turkey

0

5

10

15

20

25

30

35

40

Western

Europe

New EU

members

Bulgaria

and

Romania

Russian

Federation

(Moscow)

Asian

Tigers

India China Turkey

(b) Skilled workers

(a) Unskilled workers

the region will increase in 2005-2006 (figureII.26).

A comparison with other surveys is notstraightforward because, with the exception ofthe Russian Federation, other surveys do notmonitor South-East Europe and the CIS.Moreover, surveys looking at the RussianFederation from different angles presentcontradictory results. For instance, on the onehand the A.T. Kearney FDI Confidence Index(A.T. Kearney 2004) noted a decline inconfidence in the Russian Federation in theaftermath of the Yukos case, although consumer-related industries (retail trade and food andbeverages) sti l l seemed to have a positiveoutlook; on the other hand, the latest survey ofJapanese manufacturing TNCs (JBIC 2005) raisedthe ranking of the Russian Federation to the 6thmost promising location for TNCs in the next

three years compared to its 10th position in theprevious survey.

Outward FDI in South-East Europe and theCIS is expected to recover, as the fundamentalreason for Russian firms (the principal outwardinvestors in the region) going abroad – to controlthe value chain of their resources – remainsunchanged, and the State is expected to give thegreen light to foreign expansion once again.

C. Developed countries:uneven performance

Total FDI inflows to developed countriesdeclined by 14%, to $380 billion, in 2004. Sincetheir peak in 2000, inflows to those economiesas a group have plummeted by two-thirds, fallingin some major recipient countries. On the one

81CHAPTER II

hand, such flows rose significantly in Australia,the United Kingdom and the United States, aswell as in all of the ten new EU-accessioncountries now classified as developed countries(box I.2). On the other hand, total flows to theEU-15 countries declined by 40% from their 2003level, due mainly to relatively low economicgrowth rates in that region and to large-scalerepayments of intra-firm credits by foreignaffiliates to their parent firms abroad in somemajor host countries (e.g. Germany, theNetherlands, Sweden). Other developedcountries, such as Israel, Norway andSwitzerland, also recorded lower FDI inflows.Outflows of FDI from the developed countriesincreased modestly in 2004.

1. Trends: a turnaround in manycountries

FDI inflows to developed countriesdeclined from $442 bill ion in 2003 to $380billion in 2004. The decline (14%) was lesspronounced than in 2003 (19%). Eight countriesreported FDI inflows of more than $10 billion(table II.9), and inflows into more than half ofthe developed countries – including the 10 EU-accession countries – increased. This, togetherwith a number of factors discussed below,suggests that FDI inflows to developed countriesmay be bottoming out and that a gradual recoveryis finally under way.

There was a significant rebound in FDIinflows to North America: these nearly doubledin 2004 (figure II.27). This was due to an increase

in inflows to the United States, from $57 billionin 2003 to $96 billion in 2004 (figure II.28),making that country the largest FDI recipientworldwide for the first time since 2001, aheadof the United Kingdom, China andLuxembourg. Reinvested earnings accountedfor most of the increase, rising from $1.5 billionin 2003 to $45 billion in 2004. Net repaymentsabroad of intra-company debt by foreignaffiliates in the United States decreased by44%, so that the inflows due to this componentstood at -$17.8 billion in 2004 as comparedwith -$31.7 bill ion in 2003. Favourableeconomic growth prospects and high corporateprofits contributed to the increase in FDI flowsto the United States. In the finance andinsurance services industry, FDI inflowsincreased to $31.8 bill ion in 2004 due toconsolidation in the industry and to theexpansion of European banks into the United

States market. Spurred by financial deregulationand globalization, European financial firms havebeen looking to new markets; the three largestcross-border M&A deals in 2004 took place inthis industry (annex table A.I.1). Besides market-seeking FDI in services and in manufacturing,the United States attracted FDI in chemicals andelectrical equipment,79 industries that aretypically export-oriented, and benefited from thedecline in the value of the United States dollar.Overall FDI inflows to the United Statesmanufacturing sector reached $19.4 billion in2004, a substantial increase compared with the$0.3 billion of the year before. The main homecountries for FDI in the United States in 2004were the EU countries ($41.4 billion), Canada($31.8 billion) and Japan ($16.1 billion). Incontrast to the FDI upswing in the United States,FDI inflows to Canada in 2004 stagnated (atnearly $7 billion).

FDI inflows to the United States amountedto 0.8% of its (nominal) GDP in 2004. Inflows,however, remained smaller than outflows. Thedeficit in the current account was again mostlyfinanced by portfolio capital inflows. Since 2002,the net balance of FDI inflows and the current-account balance have moved together into thered (figure II.29).

FDI flows into the EU fell by 36% to $216billion. However there were large differencesbetween trends in FDI inflows to the EU-15 andto the ten new EU member countries:

• In the EU-15, total FDI inflows declined by40%, to $196 billion in 2004, the lowest

Figure II.26. South-East Europe and CIS:prospects for FDI inflows, 2005-2006

(Per cent of responses from TNCs, experts and IPAs)

Source: UNCTAD (www.unctad.org/fdiprospects).

TNCs Experts IPAs

Increase Remain the same Decrease

0

20

40

60

80

100

82 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

level since 1998.80 A sharp fall in flows tothree EU-15 countries, Germany,Luxembourg and the Netherlands, aloneaccounted for 95% of the total decline. FDIinflows turned negative in the Netherlandswhere foreign investors reduced their FDIstock by $4.6 billion (compared to inflowsof $19.3 billion in 2003). The downturn wasprimarily due to intra-company debtrepayments81 and to a change in the system

of compilation of balance-of-paymentsstatistics introduced in April 2003 (seeannex B, “Definitions and sources”). Loweconomic growth also contributed to thedecline. FDI inflows into Luxembourg fellby 37%, to $57 billion (less than half itsaverage inflows in 2002-2003), primarilybecause fewer special purpose entities wereestablished.

Figure II.27. Developed countries: FDI inflows and their share in gross fixed capitalformation, 1985-2004

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex tables B.1 and B.3.

Table II.9. Developed countries: country distribution of FDI inflows,by range, 2003, 2004

2003 2004Range Economy a Economy a

More than $50 bil l ion Luxembourg and the United States the United States, the United Kingdom andLuxembourg

$10-49 bil l ion France, Belgium, Spain, Germany, Ireland, Australia, Belgium, France, Spain and Italythe United Kingdom, the Netherlands,Switzerland and Italy

$1-9 bil l ion Austria, Australia, Portugal, Canada, Japan, Ireland, Japan, Canada, Poland, Austria, Finland,Poland, Israel, Norway, Finland, Denmark, Switzerland, the Czech Republic, Hungary, NewNew Zealand, Hungary, the Czech Republic, Zealand, Norway, Israel, Greece, Cyprus, SlovakiaSweden and Cyprus and Portugal

Less than $1 bil l ion Estonia, Slovakia, Greece, Slovenia, Iceland, Estonia, Lithuania, Latvia, Slovenia, Malta, Iceland,Latvia, Malta, Lithuania and Gibraltar Gibraltar, Sweden, the Netherlands, Denmark and

Germany

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) and annex table B.1.a Listed in order of the magnitude of FDI inflows for each respective year.

$b

illio

n

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

0

200

400

600

800

1 000

1 200

0

5

10

15

20

25

%

North America

European Union Other developed Europe

Other developed countries

FDI inflows as a percentage of gross fixed capital formation

83CHAPTER II

In Germany, negative FDI inflows of $39billion were recorded as a result of lowerinflows of equity capital and largerepatriations of intra-company loansresulting from tax changes (box II.14).Investment by private equity funds playeda growing role in FDI inflows to Germany,82

in particular in the chemicals industry. Asin Germany and the Netherlands, FDIinflows to Denmark also turned negative,largely as a result of repatriations of equitycapital caused by the economic slowdownand repayment of cross-border intra-company loans by foreign affil iates of

Danish TNCs. France,83 Ireland84

and Spain,85 countries withrelatively large FDI inflows in therecent past, also experienced asubstantial decline (ranging between37% and 66%) in inflows in 2004.Similarly FDI inflows into Swedenand Austria fell, but to a lesser extent.

Whereas the great majority of EU-15 countries attracted less FDI, theUnited Kingdom became the secondlargest recipient of FDI worldwidein 2004, as inflows surged from $20billion to $78 billion. This was thethird largest FDI inflow ever to thatcountry, exceeded only by thatregistered in the peak years of 1999($88 bill ion) and 2000 ($119billion). Increased flows from theUnited States partly explain thisrise. As a result, the position of theUnited States – which alreadyaccounted for 39% of the totalinward FDI stock of the UnitedKingdom in 2003 – as a leadingsource of FDI in the UnitedKingdom strengthened further.86

Both cross-border M&As andgreenfield investments contributedto the increase. The value of somecross-border M&A deals wasextremely high. For instance,Santander Central Hispano, Spain’slargest bank, bought Abbey Nationalat a price of $16 billion, Europe’sbiggest ever cross-border merger inbanking (annex table A.I.1).

Quarterly and even annual FDIfigures are very volatile. They areoften influenced by a single largetransaction or random movements of

individual components of FDI flows that arenot necessarily related to changes in thefundamental determinants of FDI. Amedium-term examination of the 2002-2004period, for instance, provides a better pictureof the FDI performance of the EU-15countries. France and the United Kingdomreceived relatively high FDI inflows duringthat period (on average $38.6 billion and$41 bill ion per year respectively). TheUnited Kingdom experienced relativelystrong economic growth during these yearsof 3%, which is higher than that in the euroarea (IMF 2005). In France, the Government

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics) andannex table B.1.

a Listed on the basis of the magnitude of 2004 FDI flows.

Figure II.28. Developed countries: FDI flows, top 10economies,a 2003, 2004

(Bill ions of dollars)

(a) FDI inflows

(b) FDI outflows

2003

2004

0 10 20 30 40 50 60 70 80

Japan

Ireland

Italy

Spain

France

Belgium

Australia

Luxembourg

United Kingdom

United States

0 10 20 30 40 50 60 70 80

Italy

Switzerland

Belgium

Japan

Canada

France

Spain

Luxembourg

United Kingdom

United States

2003

2004

96

229

91

119

101

84 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

has been actively promoting FDI inflows inrecent years (WIR04, p. 87). In contrast,Italy and Germany, due to weak economicgrowth and relatively rigid labour markets,attracted considerably less FDI ($16 billionand $13 billion, respectively, on average).Part of Italy’s weak performance may beattributed to structural problems such ashigh labour and energy costs. Othereconomies that performed well over the2002-2004 period were Belgium ($27 billionper year in FDI inflows on average), Spain($30 bill ion) and Ireland ($22 bill ion),although FDI flows have been decreasingfor the latter two countries.

• FDI inflows into the 10 EU-accessioncountries (which were previously classifiedunder Central and Eastern Europe (see boxI.2)) rose by 69% in 2004, to $20 billion,with Poland, the Czech Republic andHungary, in that order, receiving the largestFDI inflows. Reinvested earnings accountedfor more than half of the FDI flows to thesecountries, whereas equity investments innew projects and privatization sales werethe dominant forms of FDI in Slovakia,Latvia and Lithuania (Hunya 2005). Withthe rising FDI inflows, the share of inwardFDI in gross fixed capital formation in the10 new EU countries grew from 11% in

2003 to 16% in 2004 (annex table B.3),which is higher than the EU-15 average. FDIstock in relation to economic size, asmeasured by stock as a percentage of GDP,is also higher for these countries (39%) thanfor the EU-15 (31%) (annex table B.3).

As in the past, the EU-15 countries were themajor investors in the 10 new EU countries.A recent study shows that the largestinvestors in these countries were Germanyand the Netherlands, which togetheraccounted for 40% of the inward stock,followed by Austria and France (Hunya2005). It should also be noted that asignificant share of FDI flows to the newcountries is undertaken by foreign affiliatesoperating in the EU-15.

Lithuania, Latvia and the Czech Republicexperienced the largest increase in inwardFDI flows in 2004 among the 10 new EUmembers. Flows to Lithuania more thanquadrupled (to $773 million); they morethan doubled in Latvia ($647 million), theCzech Republic ($4.5 billion) and Hungary($4.2 billion); and Slovakia ($1.2 billion)received 68% higher inflows than in 2003,mainly due to the privatization of threeelectricity distributors.87 Inflows to Cyprusincreased marginally ($1.1 billion) in 2004.

Figure II.29. Current-account balance, net balance of FDI flowsa and net balance ofportfolio flowsb in the United States, 1990-2004

(Bill ions of dollars)

Source: UNCTAD, based on data from FDI/TNC database (www.unctad.org/fdistatistics) and United States Bureauof Economic Analysis (www.bea.doc.gov).

a FDI inflows less FDI outflows.b Foreign securities of United States-owned assets abroad, less United States Treasury securities, and securities

other than Treasury securities of foreign-owned assets in the United States.

-700

-600

-500

-400

-300

-200

-100

0

100

200

300

400

500

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

Balance oncurrent account

Net balanceof FDI flows

Net balance ofportfolio flows

85CHAPTER II

The 10 new EU countries accounted for only9.4% of FDI inflows to the EU-25 in 2004.Whether their share in EU-25 inward FDIflows will increase in the future remains anopen question. But a number of structuralcharacteristics make them attractivelocations for further FDI (box II.15).

FDI inflows into the other developedcountries shrank by 66% in 2004. Israel, Norwayand Switzerland in particular received lessinvestment. Japan, on the other hand, recorded24% higher FDI inflows in 2004 ($7.8 billion).In January 2003, Japan announced its goal ofdoubling inward FDI within five years. Thiswould require average inflows of more than $15

billion per year, considerably higher than whatJapan has received over the past two years. Inorder to achieve this goal, a large number ofmeasures in five priority areas were proposedin 2004 (WIR04, p. 82); one of the most importantones was the introduction of a measure to allowcross-border equity swaps. However, in 2005,there was a move to delay the legislation thatwould allow this scheme after a controversial dealtook place between Livedoor (Japan) and NipponBroadcasting System. It should also be noted thatmuch of recent FDI in Japan has been in the formof distress funds (funds used to purchasecompanies experiencing substantial financialdifficulties) from foreign institutional investors,

Box II.14. What lies behind the negative FDI inflows to Germany in 2004?

Source: UNCTAD.

a In the same period, the share of equity capital in financing FDI inflows in Germany was 70% and the share ofreinvested earnings -17%. The continued losses (after dividend payments) registered by foreign affiliates, that ledto negative reinvested earnings, can be explained in part by relatively high German taxes on such earnings.

b A recent study of the financing patterns of foreign FDI in Germany found statistically significant effects of theprofitability of foreign affiliates on the volume of intra-company loans (Ramb and Weichenrieder 2005).

In 2004, Germany experienced negative FDIinflows (–$38.6 billion) for the first time since1992. This was caused mainly by a large drop inthe equity capital component of FDI and by a netrepayment of cross-border intra-company loansby foreign affiliates in Germany for the secondyear in a row (box table II.14.1).

Intra-company loans have played asubstantial role in financing FDI in Germany,accounting for an average of about 47% of FDIflows over the past 30 years.a Such loans arerelatively volatile. Their movements depend ona variety of factors related to the financial

management of individual companies. In 2003,the repayment of loans by foreign investors waspartly due to a revision of the German CorporationTax Act (Körperschaftssteuergesetz) that wasintended to encourage foreign companies totransform corporate loans to their Germanaffiliates into equity capital. It should have beenno more than a change in the mode of FDIfinancing, but according to the DeutscheBundesbank, the addition to equity was muchlower than the repayment of credits, which resultedin a net reduction in FDI flows to Germany(Deutsche Bundesbank 2005, p.42). Increasedrepayment of intra-company loans by Germanaffiliates of foreign firms in 2004 (46 billion euro)can also largely be explained by a singletransaction (of an estimated 20 billion euro) wherethe German affiliate of a foreign enterprise in thetelecoms industry used the sales proceeds fromits reduced participation in an affiliate abroad torepay loans to a non-German affiliate of the group(Deutsche Bundesbank 2005, p. 41). Furthermore,the improved profitability of companies locatedin Germany may have motivated repayment ofloans by German affiliates to their parentcompanies abroad.b The low value of the UnitedStates dollar may also have played a role byfacilitating the repayment of dollar-denominateddebt.

Box table II.14.1. FDI inflows to Germanyby financing component, 2002-2004

(Billions of euros)

Equity Reinvested Intra-companyYear capital earnings loans Total

2002 35.9 -7.1 25.1 53.7

2003 40.5 -7.4 -8.8 24.2

2004 21.6 -6.4 -46.2 -31.1

Source: UNCTAD, based on data from DeutscheBundesbank, Balance of Payments Statistics.

86 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Box II.15. EU accession and its impact on FDI in the new member countries

Inward FDI stock in the 10 new EUmember countries at the end of 2004 reached$230 billion. Within the ten years 1995-2004,this stock grew fivefold, nearly twice as fast asworld FDI stock. Heading the list of top hostcountries in the group are relatively largecountries such as Poland ($61 billion in FDIstock), Hungary ($60 billion) and the CzechRepublic ($56 billion). Together they accountedfor more than three-quarters of the total inwardFDI stock of the new EU member countries.Inward FDI stock per capita in the 10 new EUcountries amounted to $3,079 at the end of 2004,and inward FDI stock in relation to nominal GDPreached nearly 39%, as compared with $9,790and 31% for the EU-15 average (box figureII.15.1). On a per capita basis, the small

Mediterranean countries, Cyprus and Malta, leadthe country rankings. Both countries havefollowed market-oriented economic policies fora long time and have reached relatively highincome levels.

There are three main trends emerging inFDI inflows to the new EU countries: first, newEU member States are increasingly attracting FDIinto activities that require higher skills such asprecision engineering, design and R&D (chapterIV). This quite often involves upgrading existingfacilities and focusing on export-orientedmanufacturing, particularly in the automotive andmachinery industries (Hunya 2005).a Second,small and medium-sized enterprises from the EU-15 are beginning to invest in the new EU member

Box figure II.15.1. Inward FDI stock as a percentage ofGDP in the EU-15 and EU-10 accession countries

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics).

States. Prior to 2004, these companies werediscouraged from investing in these countriesbecause of the political and economic risks, andbecause stringent border controls made just-in-time delivery impossible. These obstacles havediminished since May 2004.b Third, consolidationof some industries and restructuring of certainTNC operations are taking place in the new EUmember countries.

The main motives of foreign investors toinvest in the 10 new EU members remain similarto those of the pre-accession phase (WIR03, pp.64-66, WIR04, pp. 75-78). For market-seekinginvestors it is the strong economic growth of newEU member countries in 2004: their real GDPgrew by 5.5%, more than double the EU-15average (IMF 2005); and their favourable growth

prospects continue to be veryattractive. For efficiency-seekinginvestors, competitive unit labourcosts are particularly important. In2000, wages in the then-accessioncountries reached one-fifth of thelevel of the EU-15, while inproductivity there was only a one-to-three difference (WIR04, p. 77).According to one estimate, averagewages in new EU members in 2020will still be 60% lower than the EU-15 average (box table II.15.1).c Inthe new EU member States,corporate taxes are lower than in theEU-15: rates were 20%, on average,for the former compared to 31% forthe latter. However, a simplecomparison of tax rates is notsufficient for assessing the relative

tax burdens in each country (WIR04, p.77). Otherelements (such as the tax base, or specific taxregimes) need to be taken into account.

Additionally, full membership of the EU inMay 2004 implied the adoption of the full bodyof EU laws (the acquis communautaire) thatshould reduce risk premiums for investors(WIR04, p.77), while accession to the customsunion has lowered transaction costs. Access toEU Structural Funds (that are intended for basicinfrastructure development, human resourcedevelopment, competitiveness and enterprisedevelopment, rural development andenvironmental protection) can contribute to animprovement of the business environment

0

600

1 200

1 800

2 400

3 000

3 600

4 200

1995 1996 1997 1998 1999 2000 2001 2002 2003 20040

5

10

15

20

25

30

35

40

%

EU-15 new EU-10

EU-15 as a percentage of GDP new EU-10 as a percentage of GDP

$b

illio

n

/...

87CHAPTER II

a somewhat peculiar feature of inward FDI intoJapan.88 FDI inflows into some smallereconomies outside the North American and EUregions – such as New Zealand and Iceland –remained stable.

FDI flows to Australia increased to a record$43 billion in 2004, resulting from a growth ofequity investment, from $2.3 billion in 2003 to$35.5 billion in 2004, and a significant (56%)rise in M&A deals. These were driven by strongdemand for Australia’s natural resources, theprivatization of State-owned assets andliberalization of the media industry.

There was an impressive surge in FDIinflows from developing countries to the UnitedKingdom and Japan – rising by 120% and 56%respectively during the period 2002-2003. In theUnited Kingdom, investment from Latin Americaaccounted for the bulk of the increase in FDIoriginating from developing countries. In Japan,investment from developing Asia more thanquadrupled during this period. For developedcountries as a group, flows from developingcountries remain volatile, rising and fallingsharply from year to year.

Source: UNCTAD.

a According to one study, foreign affiliates generated 70% of manufactured exports in the Czech Republic, Hungary,Poland and Slovakia in 2001 (Hunya 2004, WIR02). On the other hand, the importance of services in inward FDIoverall continues to rise (annex tables I.4 and I.6).

b Ernst & Young’s European Investment Monitor shows a substantial increase in the number of projects in the newmember States after accession, both in absolute terms and relative to Western Europe.

c It is assumed that the convergence rate, the rate at which the wage gap between the EU-15 and the ten EUaccession countries declines, is 1.5% per year. The convergence rate between rich and poor countries in WesternEurope in the period 1963-2000 was 1.1% (Sinn and Ochel 2003).

d In order to join the European Monetary Union new EU member countries have to fulfil several convergencecriteria such as low inflation rates, low long-term interest rates that reflect low inflation expectations, stableexchange rates and two fiscal criteria (a current deficit lower than 3% of GDP and an outstanding deficit smallerthan 60% of GDP). This convergence process should lead to falling interest rates in these countries.

(WIR04, p.77). In addition, the full membershipin the European Monetary Union envisaged bythe end of this decade is expected to lead tofalling interest rates in the coming years, whichwould improve financing conditions in thesecountries. d

However, despite entry into the EU and theexpected burst of investor interest, risks persistin the new EU member countries. A recent surveyhas shown that corporate investors perceive poorinfrastructure, corruption and the gradual erosionof low-cost advantage as leading threats to thecompetitiveness of the ten new EU members(A.T. Kearney 2004, p.21). EU reforms areexpected to bring infrastructure investments andgive regulatory stability to the EU single market,but the economic and social costs of adjustmentare also expected to be high. Rising incomes mayerode wage competitiveness. EU law will likelyadd a new layer of regulations and mayundermine new members’ relative FDIadvantages in areas such as taxes and labourcosts. These factors could also push investorsfurther East and South outside the new EU.

Box II.15. EU accession and its impact on FDI in the new member countries (concluded)

Box table II.15.1. Convergence ofwage levels in the EU: a projection,

2004, 2020(Average of EU-15=100)

Country 2004 2020

Poland 29 40Czech Republic 25 38Hungary 31 38Slovakia 18 36Slovenia 44a 55Cyprus 48b 61Estonia 20 36Lithuania 23 34Latvia 19 33

EU-15 average 100 100

Source: UNCTAD, based on Rottmann and Jost2004, and Mercer Human ResourceConsulting, 2005 Inter-NationalGeographic Salary Differential Report(www.mercerhr.com).

Note: Under the assumption of a convergencerate of 1.5% per year.

a 2002.b 2001.

88 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

There are some notable changes in thesectoral pattern of FDI in the developedcountries. Overall, the importance of services ininward FDI continues to rise (annex tables A.I.4and A.I.6). The industries in developed countrieswith the largest cross-border M&A deals in termsof value were construction, health and socialservices, and business activities, followed closelyby electrical and electronic equipment, andtextiles and clothing (table II.10 and annex tableA.I.1). Furthermore, the real estate industry hasrecently witnessed an impressive surge in M&As.

FDI outflows from developed countriesincreased by 10% in 2004 to $637 bill ion,stimulated by high economic growth rates andrising corporate profits in many parts of theworld. Such outflows exceeded inflows ofdeveloped countries by $148 billion per annum,on average, during the period 2002-2004, thusmaintaining the dominant position of developedcountries as net providers of FDI. As in the past,the largest share of outflows from developed

countries was directed towards other developedcountries.

In 2004, the United States was by far thelargest source of FDI worldwide, recording itslargest outflows ever ($229 billion), followed bythe United Kingdom ($65 billion), Luxembourg($59 billion) and France ($48 billion) (figureII.28). In addition there was a marked increasein FDI outflows from the new EU membercountries such as Poland (311%), Lithuania(606%) and Latvia (201%). For most developedcountries, FDI outflows exceeded inflows. Thecountries in which FDI outflows exceeded FDIinflows the most were: the United States ($133billion), Canada ($41 billion), Germany ($31billion), Japan ($23 billion), Spain ($36 billion)and Switzerland ($21 billion). The 10 new EUcountries were all net importers of FDI capitalin 2004, as in previous years.

Until the 1970s the vast majority ofdeveloped-country FDI abroad was resource- ormarket-seeking in nature. In the 1980s and 1990s,

Table II.10. Developed countries: distribution of cross-border M&A sales,by sector and industry, 2003, 2004

(Millions of dollars and per cent)

2003 2004 Growth rateSector/industry Value % Value % in 2004 (%)

Primary 6 232 2.5 2 791 0.9 -55Agriculture, forestry and fishing 1 287 0.5 1 205 0.4 -6Mining 4 945 2.0 1 587 0.5 -68

Manufacturing 101 954 41.7 114 187 36.2 12Food, beverages and tobacco 24 746 10.1 17 774 5.6 -28Textiles, clothing and leather 648 0.3 1 511 0.5 133Wood and wood products 2 528 1.0 3 101 1.0 23Printing, publishing and all ied services 11 812 4.8 8 853 2.8 -25Oil and gas; petroleum refining 7 713 3.2 9 110 2.9 18Chemicals and chemical products 21 377 8.7 38 741 12.3 81Rubber and miscellaneous plastic products 1 319 0.5 557 0.2 -58Stone, clay, glass and concrete products 2 652 1.1 4 161 1.3 57Metals and metal products 6 862 2.8 3 947 1.2 -42Machinery 3 829 1.6 6 491 2.1 70Electrical and electronic equipment 4 354 1.8 10 741 3.4 147Motor vehicles and other transport equipment 4 417 1.8 3 082 1.0 -30Measuring, medical and photo equipment; clocks 8 018 3.3 5 815 1.8 -27Miscellaneous manufacturing 1 681 0.7 303 0.1 -82

Services 136 240 55.7 198 872 63.0 46Electricity, gas and water distribution 14 336 5.9 22 848 7.2 59Construction firms 911 0.4 3 138 1.0 245Hotels and restaurants 3 946 1.6 4 103 1.3 4Trade 12 572 5.1 25 476 8.1 103Transport, storage and communications 27 527 11.3 21 909 6.9 -20Finance 44 222 18.1 64 149 20.3 45Business activit ies 20 961 8.6 51 636 16.3 146Public administration 55 - 3 - -95Health and social services 1 085 0.4 2 722 0.9 151Educational services 77 - 67 - -12Community, social and personal service activit ies 10 547 4.3 2 818 0.9 -73

All industries 244 426 100.0 315 851 100.0 29

Source: UNCTAD, cross-border M&A database (www.unctad.org/fdistatistics).

89CHAPTER II

developed-country firms increasingly sought totake advantage of cost differences in differentproduction locations by building up globalproduction networks to produce for regional andworld markets (efficiency-seeking FDI). In recentyears, another kind of trend in FDI fromdeveloped countries has emerged as companiesalso engage in R&D activities abroad (see PartTwo). Most FDI in R&D by developed-countryfirms is targeted to other developed countries.The United States is the largest host country forFDI – both greenfield and M&A – in R&D,followed by the United Kingdom. In the case ofgreenfield FDI in R&D, Ireland and Spain alsofigure as large recipients in addition to Canada,France, Germany and Japan. But lately,developing countries like China and India arebecoming increasingly important as hosts forR&D activities by developed-country TNCs(chapter IV.C).

2. Policy developments: divergingtendencies

Many developed countries have furtherliberalized their FDI rules and continue toconclude bilateral and regional agreements. Thenumber of national regulatory changes in 2004exceeded that in 2003 by 20%, rising from 48to 60. Most of the changes were investor-friendly.The proliferation of BITs and DTTs continued,with 39 BITs and 53 DTTs involving a developedcountry (figure II.30) concluded in 2004. Thisbrought the total number of BITs and DTTsinvolving developed countries to 2,014 and 1,464,respectively, at the end of 2004. Belgium-Luxembourg, Finland, Sweden and Switzerlandwere the most active with respect to BITs,concluding five new BITs each. Despite anoverall attitude that is friendly towards FDI, fearsof job losses and decreasing corporate taxpayments have led to attempts and measures insome developed countries (e.g. the United States)to encourage companies to invest more at home.Others have undertaken a number of reforms. InGermany, for example, several measures wereadopted to reform the labour market.89

Furthermore, in 2004 France and Germanylaunched an initiative to set minimum corporatetax rates in Europe to avoid excessive taxcompetition among EU member States. However,this initiative requires unanimous approval bythe EU members. The corporate income tax wasreduced in a number of EU-15 and other

developed countries such as Austria, Canada,Denmark, Finland, Greece and Portugal (chapter I).

Further liberalization with respect to FDIin real estate was undertaken in a number ofdeveloped countries, including the 10 new EUcountries. For example in Poland, permitrequirements for investment in real estate wereabolished through an amendment to the real estatelaw. This may partly explain the 10% increasein FDI inflows to the real estate industry inPoland in 2004.90 In Germany, the regulation ofreal estate has been partly liberalized, which hasled to the selling of property by public entitiesas a way of reducing the fiscal deficit. Similarly,in Italy the introduction of a new tax regime forreal estate investment funds may have led to somelarge M&A deals in the real estate industry in2004.91 Further deregulation and privatizationof State-owned assets were reported in Canada(petroleum industry),92 Italy (electricity industryand media activities), the Netherlands andHungary (electricity industry) as well as inLithuania (stock exchange).

3. Prospects: positive overall

FDI prospects for developed countries in2005 are favourable both for inward and outwardflows, underpinned by the forecast of continuingrelatively high GDP growth (2.6%), a strongpick-up in corporate profits and a renewedenthusiasm for cross-border M&As (IMF 2005,ECB 2004). The significant increase in cross-border M&As in the first half of the year indeveloped countries could signal higher FDIflows in 2005. The situation will, however, differamong countries and subregions according todifferent growth prospects and risk factors.

For the United States, economic growthprospects for 2005 are encouraging – althoughgrowth in 2005 may prove somewhat weaker thanin 2004. Recent data releases suggest buoyantcorporate profitability, an increase in exportgrowth rates (ECB 2005), strong business andconsumer confidence (IMF 2005), and an increaseof 15% in cross-border M&As transactions in thefirst half of 2005. This may trigger furtherincreases in inward FDI in the United States,although significant imbalances in the economyare a potential concern.

FDI outflows from the United States in2005 may be held back by recent legislation (theHomeland Investment Act passed in November

90 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Figure II.30. Developed countries: number of BITs and DTTs concluded, cumulative andannual, 1990-2004

Source: UNCTAD, BIT/DTT database (www.unctad.org/iia).

0

20

40

60

80

100

120

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

0

400

800

1 200

1 600

2 000

2 400

BITs (total per year: left scale) DTTs (total per year: left scale)

Total BITs (cumulative: right scale) Total DTTs (cumulative: right scale)

2004) that lowers the tax on repatriated foreignearnings of United States firms.93 This law, whichprovides a one-time tax break on corporateforeign profits, is likely to reduce FDI outflowsfrom the United States significantly in 2005,given that over 60% of outward FDI flows (2001-2004) are in the form of reinvested earnings.United States holdings abroad worthapproximately $400-600 billion could potentiallybe eligible for this tax relief and $100-150 billionof them are expected to flow back to the UnitedStates instead of being reinvested or held byforeign affil iates of United States TNCs.94

Indeed, a number of United States TNCs havealready planned to repatriate a significant amountof foreign profits (table II.11), which wouldfinance some M&A deals within the UnitedStates. It would also help finance the UnitedStates trade deficit, estimated to be around $600billion in 2005, and may contribute to astrengthening of the United States dollar.95

For the EU-15, a marginal rise in FDIinflows is expected, partly as a result of anupswing in cross-border M&A activity in the firsthalf of 2005 and healthy corporate profits (IMF2005). For the euro area, there is a consensusamong a number of forecasts that annual GDPgrowth will average 1.2-1.6% in 2005.96 Somecountries such as the United Kingdom and thenew EU members should attract high market-seeking FDI inflows as robust economic growthis expected in 2005 (IMF 2005). Privatization

should also contribute to higher FDI inflows insome large economies.97 On the other hand, somecountries – notably Germany and Italy – areexpected to suffer from low economic growthrates. Nevertheless, according to a recent survey(Ernst & Young 2005), Western Europe is themost attractive region for FDI.

Competitive pressures in some industriesare driving firms, especially in the EU, to seekeconomies of scale and scope through cross-

Table II.11. Expected repatriation ofprofits from United States

affiliates abroad to their parents,selected TNCs, 2005

Profits to berepatriated to

TNCs parent firms

3M 1.0Bristol Myers Group 9.0Coca-Cola 6.1Dell 4.1Eli Lil ly and Company 8.0ExxonMobil -General Electric -IBM 8.0Intel 6.0Johnson & Johnson 11.0Kellogg 1.0Oracle 3.1Pepsico 7.5Pfizer 29.0Procter & Gamble 10.7Schering-Plough 9.4

Source: UNCTAD, based on var ious newspaperaccounts.

91CHAPTER II

border M&As. Thus outflows from EU-15countries in these industries are expected toincrease. In addition, improved corporateprofits are likely to encourage EU firms toexpand into new markets, especially in Asiaand in the new EU member countries. A surveyof German firms by the Deutsche Industrie- undHandelskammertag, for instance, shows that40% of respondent German companies planto continue investing abroad (DIHK 2005a).

For the 10 new EU member States, FDIprospects look good. As of March-April 2005these countries were considered to be, afterWestern Europe, the second most attractivelocations for FDI. This is mainly due to thehigh priority accorded to them by EuropeanTNCs (Ernst & Young 2005, p. 9). Althoughnew EU members continue to show solidgrowth, FDI in these countries is dependenton the health of the European economy as awhole. Consequently, deceleration of growth inthe EU-15 might curtail investments at home andabroad (Hunya 2005).

For Japan the rise in FDI inflows is likelyto continue, supported by economic growth andimproving structural features of the Japaneseeconomy. As far as outflows are concerned, asurvey by JBIC in late 2004 indicated that 47%of Japanese manufacturing TNCs that respondedto the survey plan to strengthen and expand theirforeign activities, while another 46% expect tomaintain their current level of activities over thefollowing three years (JBIC 2005). In the servicessector, for example, Japanese banks are returninggradually to foreign markets by establishingaffiliates abroad for the first time, following acontinuous three-year decline in FDI projects inbanking since 2001. For Australia, privatizationof State-owned assets is expected to boost FDIinflows further.

UNCTAD’s 2005 survey of top TNCs, FDIlocational experts and IPAs (box I.3) shows that60% of TNCs and experts expect FDI inflows toremain the same in 2005-2006 while about one-third of them expect such flows to increase(figure II.31).98 Looking ahead, FDI flows tomajor developed countries have risen in the firstquarter of 2005, indicating favourable FDIprospects for developed countries as a whole. Forexample, FDI flows in the United States, theUnited Kingdom, France, Germany and Australiarose by 81%, 41%, 15%, 109% and 30%respectively.

Notes1 Major revisions have been made to the 2003 data on

FDI inflows to the top host African countries, with thecombined inflows to Angola and Nigeria in that yearrising by up to $6 billion after the revision. Accordingto the revised data, total FDI inflows to Africa were$18 billion in 2003 (annex table B.1).

2 Oil prices, for instance, soared above $50 a barrel, upfrom $22 in 2003. Gold prices rose to above $400 perounce in 2004 as against $280 in 2003, while copperprices rose by 90% (Kitco Bullion Dealers(www.kitco.com)). Prices also rose for diamonds andplatinum.

3 The Royal Dutch /Shell Group of Companies inNigeria, for instance, reported an annual net incomefor the year ending 31 December 2004 of $18.2 billion,38% higher than in the previous year(www.allafrica.com).

4 Algeria, Egypt, the Libyan Arab Jamahiriya, Morocco,Sudan and Tunisia.

5 Burundi, Comoros, Djibouti, Eritrea, Ethiopia, Kenya,Madagascar, Malawi, Mauritius, Mozambique,Reunion, Rwanda, Seychelles, Somalia, the UnitedRepublic of Tanzania, Uganda, Zambia and Zimbabwe.

6 Benin, Burkina Faso, Cape Verde, Côte d’Ivoire,Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali,Mauritania, Niger, Nigeria, Senegal, Sierra Leone andTogo.

7 Angola, Cameroon, Central African Republic, Chad,Congo, Democratic Republic of the Congo, EquatorialGuinea, Gabon and Sao Tome and Principe.

8 Botswana, Lesotho, Namibia, South Africa andSwaziland.

9 Source: Coca Cola Newsletter (www.inboxrobot.com/news/CocaCola).

10 In 2001-2002 FDI flows to Nigeria were, on average,$1.7 billion per year and to Angola $1.9 billion(www.unctad.org/fdistatistics).

11 Egypt’s Orascom is the major telecoms operator inAlgeria (WIR04, pp. 46-47). Also, Kuwait’s National

Source: UNCTAD (www.unctad.org/fdiprospects).

Figure II.31. Developed countries: prospects forFDI inflows, 2005-2006

(Per cent of responses from TNCs,experts and IPAs)

TNCs Experts IPAs

Increase Remain the same Decrease

0

20

40

60

80

100

92 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

mobile telecoms company (AlWatanya) invested $400million there in 2004 (source: Economist IntelligenceUnit, Algeria 2004 Country Report).

12 Source: Economist Intelligence Unit, Morocco 2004Country Report.

13 Information is from the EIU’s country reports(www.eiu.com).

14 Source: MIGA (www.miga.org).15 ATI was established by the Common Market for Eastern

and Southern Africa (COMESA) Summit of Heads ofState in May 2000 and launched in August 2001.

16 In 2001, Japan established categories of products forwhich preference is granted to LDCs, as a result ofwhich about 99% of individual products (some 360items, including all the textile and clothing products)from LDCs are imported duty-free and quota-free.

17 Source: “Sub-Saharan oil growing “force” on worldmarkets”, Mail&Guardian (www.mg.co.za), 6 July2005.

18 Sources: IPAWorld (www.ipaworld.com), 24 June 2004;Mining News (www.miningnews.net), 19 August 2004and www.numsa.org.za.

19 Source: “TLC: Egypt’s Orascom plans new acquisitionsin Italy”, Euro-Mediterranean Network for Culture andSocial Dialogue, 11 July 2005, www.ansamed.info.

20 Following a reclassification, Asia and Oceania(previously Asia and the Pacific) includes a total of61 countries and territories. On the one hand, eightcountries in Central Asia that were included as partof the region in previous WIRs are now reclassifiedunder the CIS. Cyprus, formerly under West Asia, isnow reclassified under the EU (box I.2). On the otherhand, ten additional countries and territories in Oceania(formerly Pacific islands) and Timor-Leste are nowclassified under Asia and Oceania. Data are availablefor 54 countries and territories in the region.

21 Three regulations promulgated by the China SecuritiesRegulatory Commission in 2002 provide proceduralprovisions for the acquisition of listed companies. Inaddition, the “Interim Provisions on the Utilisation ofForeign Investment to Restructure State-ownedEnterprises” adopted in 2002 include provisions forforeign M&As of State-owned enterprises (excludinglisted companies and financial institutions). The“Interim Provisions on Mergers and Acquisition ofDomestic Enterprises by Foreign Investors” adoptedin 2003 include more detailed provisions for theacquisition of domestic firms.

22 Includes China, Hong Kong (China), the DemocraticPeople’s Republic of Korea, the Republic of Korea,Macao (China), Mongolia and Taiwan Province ofChina.

23 The FDI flow data reported by China’s Ministry ofCommerce (MOFCOM), and used by UNCTAD inrecent WIRs, are gathered on a gross basis (recordingonly credit transactions) rather than a net (credit lessdebit) or balance-of-payments basis. Thus divestments,capital withdrawals and repayment of debt to foreignparent firms are not included. Data on inward FDI stockare revised as reported by MOFCOM (see annex B,Definitions and sources, for details).

24 For example, HSBC (United Kingdom) invested $1.7billion for a 20% stake in the Bank of Communication.By the end of 2004, a total of 10 Chinese banks had

foreign ownership (Source: data from China BankingRegulatory Commission).

25 Some recent large investment projects by private equityfunds include: Texas Pacific Group, General Atlanticand New Bridge Capital’s investment in Lenovo ($350million), Carlyle and Prudential Financial’s investmentin China Pacific Life Insurance ($400 million) and NewBridge Capital’s investment in Shenzhen DevelopmentBank ($160 million) (Source: data from variousnewspaper accounts).

26 This is illustrated by the FAW-Toyota ($2.5 billion)and DMC-Nissan ($2 billion) joint ventures.

27 Comprises ASEAN member countries (BruneiDarussalam, Cambodia, Indonesia, Lao People’sDemocratic Republic, Malaysia, Myanmar, thePhilippines, Singapore, Thailand and Viet Nam) andTimor-Leste.

28 Other, similar studies reached the same conclusion.See for instance Cheong 2000 and Chantasasawat etal. 2003.

29 Includes Afghanistan, Bangladesh, Bhutan, India,Maldives, Nepal, Pakistan and Sri Lanka.

30 In September 2002, the Afghan Government passedthe Law on Domestic and Foreign Private Investmentthat includes investor-friendly incentives to attractforeign investment. Wholly owned foreign affiliatesare also allowed to be established. Firms from China,France, Germany, the Islamic Republic of Iran, theNetherlands, Pakistan (Afghan expatriates), Turkey,the United Kingdom and the United States have alreadyinvested in Afghanistan. Major investments during 2004and early 2005 include those by Universal Guardian(United States) in business services, Heidelberger(Germany) in business machines and equipment, HomeEssentials (Hong Kong, China) in consumer productsand a Coca-Cola bottling plant ($40 million). Infinancial services, Standard Chartered Bank (UnitedKingdom), Habib Bank (Pakistan) and Arian Bank(Islamic Republic of Iran) are major foreign-ownedbanks (BBC Morning South Asia, 14 July 2004 andNihon Keizai Shimbun, 21 March 2005).

31 Includes Bahrain, the Islamic Republic of Iran, Iraq,Jordan, Kuwait, Lebanon, Oman, the PalestinianTerritory, Qatar, Saudi Arabia, the Syrian ArabRepublic, Turkey, the United Arab Emirates and Yemen.

32 Including the data from Bahrain, Oman, Saudi Arabiaand the Syrian Arab Republic, where a survey oninward FDI was undertaken for the first time in 2004,with technical assistance from the Economic and SocialCommission for West Asia (ESCWA) and UNCTAD.See, for example, the Saudi Arabian General InvestmentAuthority (SAGIA), “SAGIA initiates first major FDIsurvey in Kingdom”, 14 July 2004 (www.sagia.gov.sa).In June 2005 SAGIA released a report entitled “Foreigndirect investment survey report”, detailing informationon inward FDI (both flows and stock).

33 American Samoa, Cook Islands, Fiji, French Polynesia,Guam, Kiribati, Marshall Islands, the Federated Statesof Micronesia, Nauru, New Caledonia, Niue, NorthernMarina Islands, Palau, Papua New Guinea, Samoa, theSolomon Islands, Tokelau, Tonga, Tuvalu, Vanuatu,Wallis and the Futuna Islands.

34 Data from UNCTAD, FDI/TNC database(www.unctad.org/fdistatistics).

93CHAPTER II

35 In October 2004, for instance, the NationalDevelopment and Reform Commission and the Export-Import Bank of China jointly promulgated a circularto encourage overseas investment projects in thefollowing four areas: (i) resource exploration projectsthat can mitigate the domestic shortage of naturalresources, (ii) projects that can promote the export ofdomestic technologies, products, equipment and labour,(iii) overseas R&D centres that can utilizeinternationally advanced technologies, managerial skillsand professionals, and (iv) M&As that can enhancethe international competitiveness of Chinese enterprisesand accelerate their entry into foreign markets. Apreferential credit policy encourages investment inthese key projects supported by the State.

36 In 2005, for instance, Bank of America signed anagreement to invest $2.5 billion in China ConstructionBank for a 9% stake.

37 As a result, Japanese manufacturers planning to“expand business operations in ASEAN” within thenext two years increased to 57% in the 2004 surveyfrom 54% in the 2003 survey. Source: JETRO, “JETROreleases its survey of Japanese manufacturers inASEAN and India”, Press Release, 6 April 2005,www.jetro.go.jp).

38 See “Ratan Tata to head Investment Commission”,Economic Times, 14 December 2004(www.economictimes.indiatimes.com).

39 In January 2004, Baosteel signed a frameworkagreement with Arcelor and CVRD to build a steel plantin Brazil. The total investment will be $8 billion.

40 A group of Shanghai developers plans to invest over$1.2 billion in a project in Saint Petersburg(www.people.com.cn, 18 October 2004).

41 Given the large sums of “Chinese dollars”, which arestill rapidly accumulating, these and otherdevelopments suggest that China is looking to acquirecorporate equities in the United States, rather thanremaining merely a large holder of United StatesTreasury bonds.

42 In terms of real effective exchange rates, nationalcurrencies appreciated in 2004 in countries like Brazil(4%), Chile (6.9%), Colombia (8.4%), Guatemala(1.9%) and Paraguay (5.1%), but they remained atlower levels than in 2000, except in the case ofGuatemala. Between 2000 and 2004 the five largestdepreciations in national currency occurred in Argentina(55%), Uruguay (37%), Venezuela (30%), Brazil (23%)and Jamaica (16%) (calculations based on data inECLAC 2004a).

43 Interbrew acquired 100% of Braco S.A., a Brazilianholding company with a 52.8% voting interest and21.8% financial interest in AmBev. The operation wasregistered as both inward and outward FDI becausethe former shareholders of Braco S.A. (Brazil) receivedshares of Inbev from Interbrew (Belgium). Inbev isthe new group that resulted from the operation, andis headquartered in Belgium.

44 For instance, the Swiss cement company Holcimacquired the remainder of its Mexican affiliate, HolcimApasco, for $750 million.

45 In 2004, Cargill (United States) completed anacquisition in the meat industry in Argentina for $70million, and announced an acquisition in Brazil for$130 million. It is also spending $200 million in

Argentina for a new soya-processing plant and a privateport to handle exports (Business Latin America, 8March 2004 (London: EIU)). Dreyfus (France), ArcherDaniels Midland and Bunge (both United States) areexpanding their capacities in Argentina (“Argentina:soya’s heady days”, Business Latin America, 23February 2004 (London: EIU)).

46 “Brazilian car parts suppliers cut back”, Business LatinAmerica, 23 May 2005 (London: EIU).

47 Business Latin America, 19 January 2004 (London:EIU).

48 Nihon Kaizai Shimbun, 24 February 2005, and ECLAC2005.

49 Source: “Siderurgia investirá US$ 13 bilhões até 2010”,IBS, www.ibs.org.br. Among foreign investors, Arcelor(Luxembourg), plans to invest $3 billion by 2008, afterhaving invested more than $1 billion in 2004; NipponSteel (Japan) plans to build a fourth high-blast furnaceworth $600 million at Usiminas; and China’s largeststeel producer, Shanghai Baosteel Group, is planningto set up a joint-venture steel mill in Brazil with CVRD,which will involve investments of $1-1.4 billion inits first stage (Business Latin America, 24 May 2004and 13 September 2004 (London: EIU); Arcelor pressreleases, 29 June 2004 and 20 December 2004,www.arcelor.com; “Baosteel Moves To Secure BrazilianIron Ore Sources With JV”, China Business Strategy,4 February 2004, www.china-ready.com.

50 Fonterra (New Zealand) plans to build a new milk-processing plant in Chile and to expand its dairyexports, mostly to Latin America, from its Soproleaffiliate there. Meanwhile, the joint venture of its DairyPartners Americas (DPA) with Nestlé (Switzerland)is expanding its activities from Brazil, Argentina andVenezuela to Ecuador, Colombia and Trinidad andTobago. (“Latin America: Industry forecast: Redeemingbrands”, Business Latin America , 10 May 2004(London: EIU)).

51 Volkswagen, Fiat, General Motors and Ford Motor havelaunched a range of 40 flex-fuel models since the mid-2003. Renault (France) launched its first flex-fuelmodel in November 2004, and PSA Peugeot Citroën(France) will follow suit in June 2005 (“Brazil: refineddrive”, Business Latin America, 13 December 2004(London: EIU)).

52 Source: “Brazil: refined drive”, Business LatinAmerica, 13 December 2004 (London: EIU) and “LatinAmerica: Industry forecast: Trading back-up”, BusinessLatin America, 13 December 2004 (London: EIU).

53 Information from Instituto Nacional de EstadísticaGeografía e Informática (INEGI) of Mexico.

54 In these six countries, the apparel industry accountsfor a significant share of total manufacturingemployment (generating around 500,000 jobs), and hasbeen responsible for most of the growth of theirmanufactured exports since the mid-1980s (IADB,2004).

55 Fourteen textile firms are reported as having alreadyclosed in Guatemala in the first 49 days of 2005, with3,426 job losses (Lapress , 10 March 2005,www.lapress.org).

56 In the retail industry, Royal Ahold sold its assets inArgentina and Brazil, while Carrefour withdrew fromChile and announced in March 2005 its retreat fromMexico. Cencosud (Chile) bought Royal Ahold’s assets

94 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

in Argentina after acquiring in 2003 the company’sassets in Chile, and Walmart (United States) purchasedRoyal Ahold’s Bompreço chain in Brazil. In the telecomsector, Telmex (Mexico) acquired AT&T Latin America,which gave it a region-wide reach in the fixed-linesegment.

57 Electricité de France (EDF) is considering the sale ofits majority stake in Edenor, one of Argentina’s biggestelectricity distributors (“Argentina govt not concernedover EDF’s withdrawal - cabinet chief Messenger”,Yahoo! Finance, 27 April 2005); Worldcom is innegotiations to divest itself of its controlling stake inEmbratel, Brazil’s long-distance telephone company;British Gas (United Kingdom) is in negotiations withEmgasud (Argentina), for the sale of its Argentineanaffiliate Metrogas (“Un grupo argentino, cerca deMetroGas”, Clarín , 5 May 2005); and the watercompany Uragua (Spain), announced in November2004 its intention to leave the Uruguayan market(“Uruguay: Vázquez’s investor nod”, Business LatinAmerica, 18 April 2005 (London: EIU)).

58 For example, in Chile, foreign investors and Chileanswith residence abroad can invest through the ForeignInvestment Statute known as Decree Law 600 thatoffers some tax advantages for foreign investors. Theyare provided with a stable tax horizon. Indeed, thedecree allows investors to lock into the tax regimeprevailing at the time an investment is made (ChileForeign Investment Committee, “FDI in Chile,regulations and procedures”, www.cinver.cl).

59 In Chile, the debt-to-equity swap mechanism waslimited to foreigners or Chileans with residence anddomicile abroad. In Mexico, foreign companies weregiven priority in terms of eligibility for investmentunder the debt-for-equity conversion programme.

60 ECLAC press releases, “Latin America will have todesign and implement better foreign direct investmentpolicies”, 9 January 2002, available at www.eclac.cl,quoting the Regional Seminar on FDI Policies in LatinAmerica: “Evaluating the Old, Contemplating theNew”, jointly organized by ECLAC and UNCTAD, andheld at ECLAC headquarters in Santiago, Chile, 7-9January 2002.

61 Surveys implemented by Latinobarometro in 17 LatinAmerican and Caribbean countries indicate that thegeneral public has increasingly turned against theprivatization process, with the percentage ofrespondents dissatisfied with the process rising from43% in 1988 to 75% in 2004. (LatinoBarometro 1998-2000, 2003, 2004, www.latinobarometro.org).

62 ENARSA will be the vehicle for companies wantingto enter the energy market or to obtain governmentincentives for investing in exploration and production.In May 2005, the Government presented beforeCongress a package of fiscal incentives featuring taxbreaks for hydrocarbon companies that invest inexploration and production. To be eligible for thesebenefits the firms will have to work in partnership withthe new State energy company (“Argentina: officialinvestment push”, Business Latin America, 30 May2005 (London: EIU)).

63 TNCs oppose this law, claiming that it is in violationof their contracts, and they are threatening to take theircase to international tribunals. It is also opposed by

civil society groups (native Indian groups, labourunions, teachers, miners and coca-leaf farmers), whichare pressing for the nationalization of Bolivia’s energyindustry and greater indigenous rights, among otherdemands. The growing tensions led the President toresign in June 2005.

64 Avances de la Nueva PDVSA, 15 April 2005,www.pdvsa.com.

65 To benefit from these fiscal incentives, investmentprojects must be approved by the authorities followingpublic bids. A number of foreign firms such as Repsol-YPF, Peugeot Citroen, General Motors Argentina,Volkswagen Argentina, Cargill and Louis Dreyfus areamong those that won the bids. (“Grandes inversionesen marcha están vinculadas a los subsidios estatales”,Clarín, 15 May 2005).

66 To compensate for the effects of high interest rates anda strong currency, Brazilian officials pledged in May2005 to grant incentives to exporters and softwaremanufacturers to boost medium- and long-term foreignsales and investments (“Lula offers exporters taxbreaks”, Business Latin America, 30 May 2005(London: EIU)).

67 In a 2004 survey by the Japan Bank for InternationalCooperation, for example, Brazil and Mexico wereranked 8th and 10th in the world, respectively, amongthe top destinations of Japanese automobile TNCs forthe next three years (JBIC 2005).

68 DR-CAFTA is currently before the United StatesCongress. Opponents to the agreement are concernedabout its potential to undermine the domestic sugarand apparel industries, the impact on the United Statestrade deficit and the differences prevailing in labourand environmental protection laws between the UnitedStates and the other signatory countries (Bloomberg,3 May 2005, www.bloomberg.com, and EconomistIntelligence Unit, Viewswire, 13 May 2005,www.viewswire.com). The agreement is also opposedby civil society groups in the Dominican Republic andthe Central American countries, where the issues ofgreatest concern include the provisions on investment,services, and government procurement that might leadto or extend privatizations. There are also concernsabout the impact of the free access of United Statesagricultural products to Central American markets onthe Central American agricultural sector, which is thesource of half of local employment.

69 The main activity of the oil company Petrom ispetroleum products and this is registered as part ofmanufacturing.

70 The FDI statistics for Turkmenistan, another natural-resource-rich country of the region, are incomplete andmay underestimate the extent of investment in oil andnatural gas there. Sources other than balance ofpayments indicate that foreign firms in that industryhave invested large sums (“2005 Investment ClimateStatement – Turkmenistan”, Washington, D.C.: UnitedStates Department of State, www.state.gov).

71 In 2004, Cyprus was the largest source of foreigninvestment in the Russian Federation, and Luxembourgwas third (Russian Federation, State Statistical Service,Current Statistical Survey: Quarterly Magazine, No.1 (52), 2005). As noted in WIR00 (p. 65), most FDIcoming from Cyprus is actually round-tripping Russian

95CHAPTER II

capital. See also Pelto et al. 2003. Similarly,Luxembourg is a source of “trans-shipped” FDI(WIR04, p. 69).

72 The strategic importance of the Baku-Tbilisi-Ceyhanpipeline lies in the fact that it is the first alternativeroute outside the Russian Federation for transportingCaspian oil to Western Europe. The construction ofthe pipeline has been accompanied by an intense debateon its environmental and human rights impact (Shelley2005, pp. 107-109).

73 Global firms include such as the BG Group (UnitedKingdom), Agip (Italy), Chevron Corp. (United States),ExxonMobil (United States), Lukoil (RussianFederation) and BP (United Kingdom). Independentcompanies are incorporated and listed abroad, despitethe fact that all of their oil exploration and extractiontakes place in Kazakhstan. Petrokazakhstan (Canada),the largest independent oil company operating inKazakhstan, is the second largest foreign-ownedpetroleum producer there (Dashevsky and Loukashov2004, p. 38). There are other independent oil firms inthe country such as Chaparral Resources (UnitedStates), Nelson Resources (Bermuda) andTransmeridian Exploration (United States), BMB Munai(United States), Aurado Energy (Canada) and EMPS(United States).

74 “OAO Lukoil: oil company, Uzbekistan sign $1 billionnatural gas deal”, Wall Street Journal, 17 June 2004.p. 1.

75 As Yukos could not pay its tax arrears, its assets wereseized and put on auction. At one auction in December2004, the Yuganskneftegaz oil extraction affiliate ofYukos was sold to a financial company, which in turnwas taken over by the State-owned Rosneft companythree days later (“Kremlin-owned firm buys Yukosasset”, Wall Street Journal, 23 December 2004. p. A.3;“Rosneft buys Yukos unit’s mysterious new owner”International Herald Tribune, 24 December 2004, p.13).

76 “TNK-BP faces dollars 87m back-tax bill”, FinancialTimes, 12 November 2004. p. 16. In April 2005, thetax arrears claim on BP-TNK was increased from lessthan $100 million to almost $1 billion (“Putin givesbig oil the cold shoulder”, Fortune, 16 May 2005, p.32.)

77 “Ukraine trims privatisation check”, BBC News, 21February 2005, www.news.bbc.co.uk, and “Daily newsand analysis”, MFK Investment Bank (Kiev), 16February 2005, mimeo.

78 The term “Dutch disease” is named after the effectson the economy of natural gas discoveries in theNetherlands, and is most commonly applied toexchange rate appreciation caused by massive exportsby the natural resource extractive industries, leadingto high production costs (including wages) in othermanufacturing activities.

79 FDI inflows to the chemicals industry more thandoubled to $7.5 billion and they also rose in theelectrical equipment industry, from -$6.5 billion in 2003to $1 billion in 2004. This industry accounted for morethan one-fifth of total United States exports in 2003(data from United States Department of Commerce,www.bea.gov.doc and annex table A.I.1).

80 In 2004, the euro appreciated substantially against theUnited States dollar. This appreciation alone resulted

in a 4% decline in the dollar value of FDI inflows intothe euro-zone countries.

81 Total FDI inflows were negative as the net repaymentof intra-company debt ($13 billion) by foreign affiliatesin the Netherlands was larger than inflows of equityinvestment ($2.8 billion) and reinvested earnings ($5.7billion) combined.

82 Germany became the world’s third largest private equitymarket by value after the United States and the UnitedKingdom in 2004. “German business welcomes theprivate equity “locusts”, Financial Times, 5 May 2005.Carlyle, Kohlberg Kravis Roberts and Goldman Sachsare typical foreign equity investors active in theGerman market. (For a brief description of privateequity companies and their cross-border investments,see chapter I, footnotes 30 and 31).

83 FDI inflows to France fell by nearly half, from $42billion in 2003 to $24 billion in 2004, due primarilyto divestment in equity capital linked to cross-borderM&As and a sizeable reduction in intra-company loans.In 2004, inward equity investment flows to France fellby 67% and intra-company loans (which are recordedin the category “other types of inward investment”)fell by 37%.

84 In Ireland, FDI inflows fell sharply from $27 billionin 2003 to $9 billion in 2004. This is largely explainedby a fall in inward equity investment, by $5.7 billionin 2004, combined with a sizeable decline ($8.8 billion)in reinvested earnings.

85 In Spain, FDI inflows have been declining over thelast couple of years owing to the diminishing impactof a special corporate income tax regime (Law 43/1995,last amendment 2000) of which companies have alreadytaken advantage. Also, Spain’s traditional low-labour-cost advantage, which had successfully attractedmanufacturing investors, might be eroded with theenlargement of the EU to include countries with evenlower labour costs. This may affect FDI inflowsadversely. For example, Samsung withdrew from Spainand relocated its affiliate to lower cost Slovenia.

86 In 2004, 40% and 43% of cross-border M&A sales,in terms of value and number respectively, in the UnitedKingdom were concluded with United States firms/investors (data from United Kingdom, NationalStatistical Office).

87 The Government sold a 49% stake of ZapadoslovenskaEnergetika to Germany’s EON Energie, a 49% stakein Stredoslovenska Energetika to Electricité de Franceand a 49% stake in Yvychodoslovenska Energetika toGermany’s RWE Plus (www.slovakia.org).

88 Out of 88 cross-border M&As completed in Japan in2004, almost one-third were undertaken by either assetmanagement companies (fund managers) or securitybrokers (e.g. Carlyle Group (United States), Lone StarFund (United States), Morgan Stanley (United States).

89 For example, a new immigration law approved in July2004 makes it easier for companies to attract and keephighly qualified foreign employees, and for foreigninvestors to gain permanent resident status in Germanyby investing one million euros and creating ten newjobs.

90 The largest FDI-related investment – $800 million byApollo Rida (United States) in Poland in 2004 – wasin real estate (Polish Information and ForeignInvestment Agency).

96 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

91 For example, Fondo Immobilli Pubblici was acquiredby a United Kingdom Investor group for $1.9 billionand New Real SpA was acquired by Excelsia Otto(Germany) for $1.7 billion in 2004 (annex table A.I.1).

92 In 2004, the Government of Canada sold all PetroCanada shares in a global offer, making this the fifthlargest global privatization of the decade (Departmentof Finance, Canada, www.fin.gc.ca).

93 Under the Act, corporate taxes on dividends to theparent firm are taxed at a one-off effective tax rateof 5.25%, available for one of two tax years, as opposedto the previous rate of 35% under certain conditions.

This is aimed at boosting job creation and R&D in theUnited States.

94 Estimated by Deutsche Bank (www.db.riskwaters.com).95 Financial Times, 31 Jan 2005, p.17.96 European Central Bank, June 2005, p. 68.97 For instance Terna (Italy’s national power grid), Snecma

(France’s national maker of aircraft engines), Electricitéde France (EDF) and Gaz de France (GDF) have goneor are expected to go to initial public offerings in 2005.

98 The survey did not include the 10 new EU accessioncountries.

PART TWO

R&D INTERNATIONALIZATION ANDDEVELOPMENT

INTRODUCTION

Bridging the technology gap betweencountries is necessary to foster sustainableeconomic development. Technology is advancingfaster than ever before. Developing countries thatfail to build capabilit ies enabling them toparticipate in the evolving global networks ofknowledge creation risk falling further behindin terms of competitiveness as well as economicand social development. While internationaltechnology transfer can bring importantknowledge to an economy, that alone is notenough. Using new technologies efficientlyrequires creating additional absorptive capacity,while a continuous effort has to be made to keepup with technical change. This is particularly truegiven the fact that wages tend to rise as a countrydevelops, facilitating the entry of lower costcompetitors in the market. While actions of bothdomestic enterprises and the government areessential to build technology capabilities indeveloping countries, TNCs can also play a role.

One of the main reasons why developingcountries promote inward FDI is indeed to linkup to the global technology and innovationnetworks led by these firms. In terms of creatingnew technology and diffusing it internationally,TNCs are world leaders in many industries. Theyaccount for the bulk of global businessexpenditures on R&D. They dominate newpatents and often lead innovation in managementand organization. Establishing links with theirinnovation and production networks can helpcountries enhance their technological capabilitiesand enable them to compete better in internationalmarkets.

Technological capabilities are difficult toacquire. The rapid pace of technical change andthe growing importance of science-basedtechnologies in many industries call for moreadvanced and diverse skills and intense technicaleffort. These require better infrastructure, not the

least in information and communicationstechnologies. They also require strong supportinginstitutions as well as stable and efficient legaland governance systems. Moreover, they requireaccess to the international knowledge base,combined with a strategy to leverage this accessfor the benefit of local innovation systems. Thecumulative forces that are increasing the gapbetween countries with respect to innovationperformance make the role of policy increasinglyimportant at all levels – national and international.

The manner in which TNCs allocate theirR&D activities internationally is significant inthis context. R&D is among the leastinternationalized functions of TNCs.Traditionally, when R&D internationalizationtook place, both home and host countries werefound in the developed world. To the extent thatTNCs undertook R&D in developing countries,they did so almost exclusively to adapt productsand processes to local conditions. These stylizedfacts have begun to change.

These changes manifest themselves inseveral ways. First , the degree of R&Dinternationalization by firms is rising in all keyhome countries as part of the overall trendtowards the offshoring of services (WIR04).German TNCs, for example, set up more foreignR&D units during the 1990s than they did duringthe preceding 50 years (Ambos 2005). Second,R&D internationalization is now growing fastestin some host developing countries, notably inAsia. Third, the drivers of R&Dinternationalization are changing. The processis no longer driven only by the need for localadaptation or to tap into established knowledgecentres. In response to increasing competition,TNCs now relocate segments of R&D so as toaccess foreign pools of research talent, reduceR&D costs and speed up the process oftechnology development. Fourth, R&D in some

developing countries now goes well beyond localadaptation and involves complex stages of R&Don a par with work undertaken in the developedeconomies. Fifth, developing-country firms arealso setting up R&D units abroad. These trendshave become apparent only in the past few yearsand are likely to continue.

This new phenomenon is partly expectedand partly unexpected. It is expected in two ways.First, in most cases R&D undertaken abroadsupports production. As TNCs increaseproduction in developing countries, some R&D(of the adaptive kind) can be expected to follow.Second, R&D is a form of service activity. Manyother services are fragmenting in a processwhereby certain segments are located in countrieswith lower wages and appropriate skills. It is notsurprising that R&D is following suit. Indeed,the survey of Europe’s largest firms conductedin 2004 by UNCTAD and Roland Berger showedthat all service functions – including R&D – arenow candidates for offshoring (WIR04). It isunexpected in that R&D is a service activity withvery demanding skill, knowledge and supportneeds — traditionally only met in developedcountries with strong national innovationsystems. Moreover, R&D is taken to be the least“fragmentable” of economic activities becauseit involves knowledge that is strategic to firms,and because it often requires dense knowledgeexchange (much of it tacit) between users andproducers within localized clusters. A home-country bias in R&D activities “reflects thelinguistic and geographic constraints imposed byperson-embodied exchanges and transfers of tacitknowledge” (Patel and Pavitt 2000, p 218).

The extent to which developing countriesconnect with the internationalizing R&Dnetworks of TNCs depends in particular on thestrength of their national innovation systems. Thisin turn is dependent on policies, the quality ofinstitutions (including both organizations and therules governing innovation activities), the qualityof human resources and the production andinnovative capabilities of enterprises. Innovation

reflects an intense interaction between firms andother actors in the public and private sectors.Innovation in developing countries is often carriedout on the shop floor, in process or productengineering, quality control, procurement,distribution and overall management. However,a significant part also involves technical effortin R&D laboratories separated from production.R&D-based innovation is greater the moreadvanced, fast changing and large-scale thetechnology involved, but it is needed even if itdoes not aim to push forward frontiers ofknowledge.

Part Two of WIR05 reviews recent trendsin the internationalization of R&D by TNCs. Itbegins in chapter III by looking at the linksbetween R&D, innovation and development, andconsiders the levels of innovative capabilitiesamong countries around the world. Large gapsin this area prevail between countries — gapsthat limit the ability of many of them to take partin the global networks of knowledge creation anddiffusion. Addressing these gaps is a majordevelopment challenge; it is also essential toensure that the internationalization of R&D byTNCs benefits larger parts of the world.

Chapter IV identifies the main players(firms and countries) in the R&Dinternationalization process. The analysis isconfined to R&D due to data constraints, but,where available, other qualitative informationrelated to innovation, notably in services, is alsoconsidered. Chapter V discusses the changingdrivers and determinants of R&Dinternationalization. Chapter VI reviews theimplications of R&D internationalization for hostand home economies, recognizing the difficultiesinvolved in assessing the impact of thisphenomenon. The last two chapters (VII andVIII) focus on policy implications at the nationaland international levels. They place particularemphasis on the need to promote interactionbetween TNCs and domestic players (firms andinstitutions) in national innovation systems.

100 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

A. Innovation matters forall countries

Innovative activity and capabilities areessential for economic growth and development.A recent report identifies science, technology andinnovation as essential to achieving theMillennium Development Goals (UN MillenniumProject 2005, Sachs and McArthur 2005). Thisis true for the industrialized countries that areat the technology frontiers, as well as fordeveloping countries that need to catch up interms of technology.

Given the large gap between the developedand developing countries in terms oftechnological advancement, the latter continueto rely heavily on technology transfer from theformer in their development process. However,sustainable economic development requires thatcountries do more than simply “open up” andpassively wait for new technologies to flow in.It demands active, continuous technological effortby enterprises, along with government policiesthat help firms attract technologies, use themeffectively and innovate. Technology requiresefforts to absorb and adapt; it has strong “tacit”elements that cannot be embodied in equipmentor codified in instructions or blueprints. Tacitknowledge can only be transferred effectivelyif the recipient develops capabilities to learn andincorporate the knowledge. It must seek newinformation, experiment with the technology, findnew ways of organizing production and train itsemployees in new skills. It involves not just theenterprise itself but also interaction with otherfirms and institutions.

The development of technologicalcapabilities has always been necessary for theeffective use of new technologies; all the moreso today. Greater openness to trade and capital

��������

INNOVATION, R&D AND DEVELOPMENTflows does not reduce the need for localtechnological effort – on the contrary.Technologies are changing more rapidly, fallingtransport costs and liberalization are intensifyingcompetition, and TNCs are seeking locations withstrong capabilit ies to produce efficiently.Moreover, i t is not just export-orientedmanufacturing that needs to be competitive;manufacturers selling to domestic markets haveto compete against imports. Export-orientedservices and primary activities need to use newtechnologies to remain competitive in worldmarkets. The development of new capabilitiesapplies to both technical functions andmanagerial ones: organizational and marketinginnovation is as important as technical innovationto growth and competitiveness (Teece 2000).

Technological innovation means theintroduction of new products, processes orservices into the market.1 Innovation does notnecessarily mean pushing the frontiers ofknowledge, particularly in a developing-countrycontext. Rather, innovations can be new to theuser but not necessarily new to the world.2 Thenature of innovation – and of required capabilities– varies greatly between activities according totheir technological complexity, the creation ofnew technology being at one extreme and the useof existing technologies at the other.3 Figure III.1shows an illustrative pyramid, with the leastcomplex technological functions (in terms ofinnovative efforts) at the base, and the mostdemanding ones at the top.4 While thesecategories are generic activities in all threesectors – primary, manufacturing and services– they can be adapted to different technologiesto take account of particular machinery, process,product and organizational characteristics.

• The starting point is the acquisition of basicproduction capabilities to absorb and useexisting technology. This sounds easy but

102 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

it is not, at least in order for capabilities tomatch relevant global best practice and foractivity that goes beyond simple assembly.Reaching internationally acceptable levels ofproduction efficiency and quality in complexactivities is very demanding. Manyenterprises fail to do this, even after yearsof operation, unless they invest sufficientlyin collecting information, creating new skillsand developing appropriate managementstructures.

• Absorption and adaptation of technology areparticularly challenging if conditions aresignificantly different from those at the originof the technology, and if local support andsupply structures are weak.

• Adaptation, in turn, can grow into significanttechnological improvement and technologicallearning, with systematic efforts made toimprove product and process performance.At this stage, many firms start monitoringinternational technological trends andselecting those technologies that can feed intotheir own efforts.

• Finally there is the frontier innovation stage,when firms design, develop and test entirelynew products and processes.

Research and development (R&D) is onesource of innovation (box III.1). In the early stagesof technological activity, enterprises need not setup formal R&D departments. As they mature,however, it becomes increasingly desirable tomonitor, import and implement technologies. R&Das a distinct activity may appear as early as thesecond level of complexity, where multifacetedtechnologies are involved or if local conditionsdemand significant adaptation. In a developingcountry, such R&D is feasible once the operationis fairly large scale and the necessary technicalskills are available. The role of formal R&D thengrows as the firm attempts significanttechnological improvements to introduce newproducts or processes. Firms that reach the highestlevel in the pyramid need not, however, be frontierinnovators (technological “leaders”) – their R&Dmay build on or improve upon innovations doneelsewhere (technological “followers”). Aspecialized unit not involved in routine technicalor production work is needed to monitor newdevelopments outside the firm or country, assesstheir significance for the firm and master, adaptand improve on existing technologies.5 FormalR&D becomes an essential part of the

Figure III.1. Stages of technology development by innovation effort

Source: UNCTAD.

FRONTIERINNOVATION

Create new technologies:as leader or follower

TECHNOLOGYIMPROVEMENT &

MONITORING

Change products and processes, plant layout, productivitymanagement and quality systems, procurement methods andlogistics to adapt technology to local or export-market needs.

This is based on in-house experimentation and R&D as well ason search and interactions with other firms and institutions

SIGNIFICANT ADAPTATION

BASIC PRODUCTION

Train workers in essential production and technical skills; reach plant design capacityand performance levels; configure products and processes; set up essential quality

management systems; institute supervisory, procurement and inventory managementsystems; establish in-bound and out-bound logistics

Improve products, processes and skills toraise productivity and competitiveness,

based on own R&D, licensing, interactionswith other firms or institutions

103CHAPTER III

technological learning process, especially forcomplex and fast moving technologies.

Empirical studies suggest a directrelationship between R&D and growth.6 Thelong-term impacts on economic growth of publicR&D and business R&D have been found to bestrong and significant (Guellec and vanPottelsberghe 2004a). Business R&D undertakenin other countries also plays an important role.Moreover, increased domestic business R&Daccentuates the positive impact of both publicand foreign business R&D. In other words,business R&D (either domestic or foreign-funded) has both a direct impact on a country’seconomic growth and an indirect one throughimproved absorption of the results of public R&Dand R&D performed in other countries.

Enterprises are the principal agents ofinnovation today, but they do not innovate andlearn in isolation. They rely on intricate (formaland informal) links with other firms and withpublic research institutions, universities and other

knowledge creating bodies like standards andmetrology institutes. In undertaking innovation,they react to government policies on trade,competition, investment and innovation. Theyseek human resources for innovation from theeducation and training system, and they drawupon the financial system for funding innovativeefforts. The complex web within whichinnovation occurs is commonly referred to as the“national innovation system” or NIS (Nelson1993, Lundvall 1992b).

Most of the NIS li terature focuses onfrontier invention in industrialized countries,rather than on mastery and adaptation oftechnology that take place in developingcountries. However, the innovation systemconcept is just as relevant for the latter (UNIDO2003, Edquist and McKelvey 2001). Mostlearning, mastery and adaptive activity requiresclose and continuous interaction with otherenterprises l ike suppliers, subcontractors,competitors and consultants, as well as with otheractors such as public R&D institutes, universities,

R&D is only one component of innovationactivities, but it represents the most developed,widely available, and internationally comparablestatistical indicator of industrial innovationactivities.

According to international guidelines, R&D(also called research and experimentaldevelopment) comprises creative work“undertaken on a systematic basis in order toincrease the stock of knowledge, includingknowledge of man, culture and society, and theuse of this stock of knowledge to devise newapplications” (OECD 2002b, p. 30).

R&D involves novelty and the resolutionof scientific and technological uncertainty. Itincludes basic and applied research along withdevelopment (United States, NSB 2004):

• Basic research. The objective of basic researchis to gain a more comprehensive knowledgeor understanding of the subject under studywithout specific applications in mind. Inindustry, basic research is defined as researchthat advances scientific knowledge but doesnot have specific immediate commercialobjectives.

• Applied research. The objective of appliedresearch is to gain the knowledge orunderstanding to meet a specific, recognizedneed. In industry, applied research includesinvestigations to discover new scientificknowledge that has specific commercialobjectives with respect to products, processes,or services.

• Development. Development is the systematicuse of the knowledge or understanding gainedfrom research directed towards the productionof useful materials, devices, systems ormethods, including the design and developmentof prototypes and processes.

For data collection purposes, the boundarybetween R&D and other technological innovationactivities can be found in pre-productiondevelopment activities (OECD 2002b). Inpractice, however, it is difficult to make thedistinction. In technology-intensive industriesdistinguishing between “research” and“development” is especially difficult since muchof the R&D work conducted involves closeinteraction between researchers in both the privateand public sectors, often also including closecollaboration with customers and suppliers (BIAC2005, Amsden and Tschang 2003).

Box III.1. Definition of R&D

Source: UNCTAD and Moris 2005b.

104 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

the metrology, standards, testing and quality(MSTQ) system, small and medium-sizedenterprise (SME) extension services, venturecapital funds and export marketing or traininginstitutions. A good supportive institutionalinfrastructure is therefore important for effectiveinnovation. Incentive structures that fosterentrepreneurship, risk-taking and innovation atthe firm, industry and university level are alsoimportant.

As the internationalization of productiondeepens and communication costs decline, eachNIS increasingly draws on knowledge createdin other systems. Rapid technical progress andthe rising costs and risks of innovation forceinnovators to seek centres of scientific excellenceinternationally. Global production networks – inwhich TNCs play the leading role – link togetherthe productive activities that underly innovation.Parent companies are instrumental in suchnetworks, providing the initial technology to theiraffiliates and helping them absorb, adapt andsubsequently upgrade it . As a result , theinnovation systems of more and more countriesare becoming interlinked in a global network in

which technological activity is international andinformation networks span the world.

From an economic developmentperspective it is becoming increasingly importantto take part in this international exchange. Thosecountries that are in a position to do so stand abetter chance of accessing new technologies atan early stage, as well as commercializinginnovations developed in their own NIS. However,the capabilities needed for participating areunequally distributed among countries (seebelow), which increases the risk of a furtherwidening of already large development gaps.

While there are different ways for countriesto participate in the international exchange ofinnovation (box III.2), WIR05 focuses on the roleof TNCs in this process, with special emphasison the internationalization of R&D. As notedabove, R&D is not always necessary forinnovation. Due to data limitations, however, theanalysis in Part Two is confined to this particulartype of innovative activity. The next two sectionsdescribe the global allocation of R&D and ofinnovative capabilities. Subsequent chapters

Box III.2. Different ways of internationalizing innovation

There are three main categories ofinnovation internationalization (box table III.2.1).In the first category, national enterprises and TNCsas well as individuals are engaged in theinternational commercialization of technologydeveloped at home. The second category relatesto domestic and international technical and

scientific collaborations among private and publicinstitutions, including domestic firms and TNCs,universities and research centres. Internationalinnovation by TNCs is the third category. TheTNC is the only institution that, by definition,can control and carry out within its boundariesthe process of innovation across the globe.

Source: UNCTAD.

Box table III.2.1. Taxonomy of internationalization of innovation

Category Actors Forms

International exploitation Profit-seeking (national and • Exports of innovative productsof nationally produced transnational) firms and • Cession of licenses and patentsinnovations individuals • Foreign production of innovative goods internally

designed and developed

International techno- Universities and public • Joint scientific projectsscientific collaborations research centres • Scientific exchanges, sabbaticals

• International flows of studentsNational and transnational • Joint ventures for specific projectsfirms • Production agreements with exchange of technical

information and/or equipment

International generation TNCs • R&D and other innovative activities both in homeof innovations and host countries

• Acquisitions of existing R&D units or greenfieldR&D investment in host countries

Source: adapted from Archibugi and Michie 1995, Narula and Zanfei 2004.

105CHAPTER III

focus on the internationalization of R&D, thetrend towards increased R&D by TNCs indeveloping countries, the driving forces behindthis phenomenon, potential impacts and policyimplications.

B. Global R&D trends

1. R&D is geographicallyconcentrated

Between 1991 and 1996, global R&Dspending increased from $438 billion to $576billion (an average annual growth of 4.4%; annex

table A.III.2). The momentum of R&D spendingcontinued throughout the late 1990s and thebeginning of the new millennium. By 2002 it hadrisen to $677 bill ion,7 corresponding to anaverage annual growth rate of 2.8% since 1996.

R&D expenditure is geographicallyconcentrated. In 1996 and 2002, the ten largestspenders accounted for more than 86% of theworld total, with their share marginally increasingover that period (table III.1). Eight of them aredeveloped countries, of which the United Statesreported by far the largest amounts in both years.Only two developing countries are among the topten: China and the Republic of Korea.

Table III.1. The 10 leading economies in R&D and business R&D spending,1996 and 2002

(Ranked by their 2002 values, billions of dollars)

Total R&D Business R&D

Rank Economy 1996 2002 Rank Economy 1996 2002

World 575.6 676.5 World 376.3 449.81 United States 197.3 276.2 1 United States 142.4 194.42 Japan 138.6 133.0 2 Japan 92.5 92.33 Germany 52.3 50.2 3 Germany 34.6 34.84 France 35.3 32.5 4 France 21.8 20.65 United Kingdom 22.4 29.3 5 United Kingdom 14.5 19.66 China 4.9 15.6 6 Korea, Republic of 9.9 10.47 Korea, Republic of 13.5 13.8 7 China .. 9.58 Canada 10.1 13.8 8 Canada 5.9 7.99 Italy 12.6 13.7 9 Sweden 6.6 a 7.3 b

10 Sweden 8.8 a 9.4 b 10 Italy 6.7 6.6Total 495.8 587.6 Total 334.7 c 403.4Share in world (%) 86.1 86.9 Share in world (%) 88.9 89.7

Developing economies, Developing economies,South-East Europe and CIS 44.5 57.1 South-East Europe and CIS 20.4 31.9

1 China 4.9 15.6 1 Korea, Republic of 9.9 10.42 Korea, Republic of 13.5 13.8 2 China .. 9.53 Taiwan Province of China 5.0 6.5 3 Taiwan Province of China 2.9 4.04 Brazil 6.0 4.6 e 4 Russian Federation 2.6 3.05 Russian Federation 3.8 4.3 5 Brazil 2.7 1.9 e

6 India 2.1 3.7 b 6 Singapore 0.8 1.27 Mexico 1.0 2.7 7 Mexico 0.2 0.8 b

8 Singapore 1.3 1.9 8 Turkey 0.2 0.49 Turkey 0.8 1.2 9 Hong Kong, China 0.2 d 0.3

10 Hong Kong, China 0.7 d 1.0 10 Chile 0.1 0.2Total 39.1 55.4 Total 19.7 31.5Share in developing economies, Share in developing economies,South-East Europe and CIS (%) 88.0 97.0 South-East Europe and CIS (%) 96.4 98.7

Source: UNCTAD, based on annex table A.III.2.a 1995.b 2001.c In 1996, Switzerland was the 10th largest spender on business R&D ($5.7 billion). Thus, the total of the top ten in

that year was $340.4 billion.d 1998.e 2003.

106 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

The growth in global R&D is partly dueto increased expenditures by the largest spenders.Between 1996 and 2002, the growth in the R&Dexpenditure of the United States (5.8% per year)was twice as high as the world average. Canadaand the United Kingdom also showed fastexpansion during that period. The expendituresof China rose at an average annual rate of morethan 20% during the same period. This dynamismcontrasts sharply with the trends of France,Germany and Japan, where R&D expendituresactually contracted in dollar terms.8

The combined share of developingeconomies, South-East Europe and the CIS inglobal R&D spending is on the rise, althoughfrom a very low level. In 1991 they accountedfor only 2.5% of the world total (annex tableA.III.2). By 1996 their share had reached 7.7%,and by 2002 it had increased further to 8.4%(figure III.2). This increase was concentratedmainly in South, East and South-East Asia (tableIII.2), which accounted for a dominant andgrowing share in R&D expenditure outsidedeveloped countries (more than two-thirds in2002). With the exception of West Asia, the shareof all other subregions in the grouping droppedbetween 1996 and 2002. The decline was the mostpronounced in Latin America and the Caribbean,the share of which shrunk from 21% to 16% ofthe total for the countries included in table III.2.Africa’s share also declined from 2.2% to 1.9%.

The concentration of R&D expendituresoutside developed countries is high and rising.The ten largest R&D spenders of the developingeconomies, South-East Europe and the CIS in2002 accounted for 97% of all R&D in theseeconomies (table III.1). Reflecting the dynamicsof South, East and South-East Asia, six of thetop ten are from these subregions. In the majorityof these economies, R&D expenditure grew fastduring the period. Double-digit annual growthrates were recorded for China, India and Mexico.R&D expenditures contracted in dollar terms onlyin Brazil.

In today’s world economy, enterprises(private and State-owned) account for the lion’sshare of global R&D. In 1991, they spent $292billion on R&D (annex table A.III.2). Thatamount increased to $376 billion in 1996 and$450 billion in 2002 (figure III.2). In other words,in each of these years enterprises wereresponsible for two-thirds of global R&Dspending; the remaining one-third was accounted

for by governments, higher education institutionsand non-profit private entities.

While the overall share was stable at theglobal level, the share of business enterprises intotal R&D expenditure varied considerably byregion and country (figure III.3). In the Triad– Japan, the United States and the EU – the shareof enterprises was above 60% in 2002. Between1996 and 2002 this share rose in Japan and theEU but not in the United States. In developingAsia, the share of enterprises rose rapidly overthat period, reaching a level similar to that ofthe EU by 2002 (62%). Conversely, the share ofenterprises in Latin America and the Caribbeanwas low and even declined in 1996-2002 (from37% to 33%).9

Reflecting the dominant role of enterpriseR&D in global R&D, the geographical patternsof the former show various similarities with thoseof the latter. R&D in the business sector isconcentrated, just like total R&D. Both in 1996and in 2002, the ten largest spenders on businessR&D accounted for about 90% of the world total,their share marginally increasing over that period(table III.1). The list of the largest business R&Dspenders is identical with that of the largest totalR&D spenders; only the rankings vary. In a slightcontrast to the global picture of total R&D, inbusiness R&D only the spending of France, Italyand Japan declined in dollar terms in 1996-2002.

The share of developing economies, South-East Europe and the CIS in global business R&Dspending is lower than in total R&D spending,reflecting a greater reliance on government R&Din these economies. Their share in the formerreached only 5.4% in 1996 and 7.1% in 2002(figure III.2). The top ten positions in terms ofbusiness R&D among the developing economies,South-East Europe and the CIS differ from thosefor total R&D only because data are not availablefrom India, and the tenth place is thus taken byChile (table III.1). Six of the ten economies arefrom South, East and South-East Asia. Anotherfeature of the list of the largest businessenterprise R&D spenders among the developingcountries is i ts very high geographicalconcentration (the share of the largest ten is 99%of the group total in 2002), reflecting in part alack of data reporting on business R&D in themajority of developing economies.

An output-based assessment of globalinnovation activities confirms the patternsobserved above. Whereas developed countries

107CHAPTER III

Figure III.2. Gross expenditure on R&D (GERD) and business enterpriseR&D (BERD), by country group, 1996 and 2002

(Billions of dollar)

Source: UNCTAD, based on annex table A.III.2.

Table III.2. Developing economies,South-East Europe and CIS:

distribution of R&D, by region(Per cent)

Region 1996 2002

South, East and South-East Asia 63.5 70.1Latin America and the Caribbean 21.1 16.0South-East Europe and CIS 11.2 9.6West Asia 2.0 2.4Africa 2.2 1.9Total developing economies,South-East Europe and CIS 100.0 100.0

Source: UNCTAD, based on annex table A.III.2.

in 2003 still accounted for 83% of all foreignpatent applications to the United States Patentand Trademark Office (USPTO), the share ofdeveloping countries and South-East Europe andthe CIS has risen particularly fast. Between theperiods 1991-1993 and 2001-2003, it jumpedfrom 7% to 17% (annex table A.III.3). The annualaverage number of applications from thesecountries increased from around 5,000 to almost26,000 between the two periods. South, East andSouth-East Asia showed by far the greatestdynamism, followed by South-East Europe andthe CIS. Two economies (Taiwan Province ofChina, Republic of Korea) accounted for four-fifths of the total. They were followed distantlyby India, China, Singapore, Hong Kong (China),the Russian Federation and Brazil. Asia accountsfor more than 95% of the patents granted in the

United States to recipients from developing,economies South-East Europe and the CIS. Theshare of patent applications from Latin Americaand Africa, on the other hand, fell from alreadylow levels between the two periods (see alsosection IV.B.4).

2. R&D by industry

Manufacturing firms have long conductedthe bulk of business sector R&D in developedeconomies. In the United States, for instance,they accounted for 60% of company-funded R&Din 2001, with mining and extraction contributingonly 0.5%, transportation 0.9% and utilities andconstruction 0.3% (United States, NSB 2004).However, the services sector also contributedsignificantly, with trade and other servicestogether contributing 38% (see below). Withinmanufacturing, industries vary greatly in R&Dintensity. For example, the OECD dividesindustries into four groups: high technology;medium-high technology; medium-low technologyand low technology (table III.3).10 The table isbased only on the intensity of R&D; it does notnecessarily depict the nature of the R&Dconducted.11

R&D in services has traditionally beenneglected in the literature, perhaps because ofthe assumption that services do not innovate orare primarily users of innovation inmanufacturing (Howells, 2000; Tether 2004).

GERD

531619

0

100

200

300

400

500

600

700

1996 2002

Developing economies, South-East Europe and CIS

Developed countries

BERD

418356

0

100

200

300

400

500

600

700

1996 2002

Developing economies, South-East Europe and CIS

Developed countries

44

531

57

619 20

356

32

418

108 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Services do innovate in the broader sense in bothprocesses (organizational change) and products(new services), but much of this innovation doesnot involve formal R&D. Data on this aretherefore scarce, which makes empirical analysisdifficult. This may be changing, however, as aresult of new information and communicationtechnologies (ICTs) and their growing role inservice industries. The telecommunications andcomputer service industries have been investingin R&D for some time, and a new industry is nowemerging that provides R&D services tomanufacturers on a contractual basis (Tether2002).

Data on services R&D are patchy.Published sources cover only a few industrialized

Figure III.3. Share of enterprise R&D in total R&D by country/region, 1996 and 2002(Per cent)

Source: UNCTAD, based on annex table A.III.2.

Table III.3. Classification of manufacturing industries by R&D intensity

Industry category R&D intensity Industries

High technology >5% Aircraft and spacecraft; pharmaceuticals; office, accounting and computing equipment;radio, television and communications equipment; medical, precision and optical instruments

Medium-high 1.5-5% Electrical machinery and apparatus not elsewhere classified; motor vehicles, trailerstechnology and semi-trailers; chemicals excluding pharmaceuticals; railroad equipment and transport

equipment not elsewhere classified; machinery and equipment not elsewhere classifiedMedium-low 0.7-1.5% Coke, refined petroleum products and nuclear fuel; rubber and plastic products; othertechnology non-metallic mineral products; building and repair of ships and boats; basic metals;

fabricated metal products, except machinery and equipmentLow technology <0.7% Manufacturing, not elsewhere classified, and recycling; wood, pulp, paper, paper

products, printing and publishing; food products, beverages and tobacco; textiles,textile products, leather and footwear

Source: United States, NSB 2004, Table 6-1.

Note: R&D intensity is direct R&D expenditures as a percentage of production (gross output).

countries up to 2000. However, they suggest thatservices R&D is rising in most economies, butthat its share in total R&D varies greatly. Severalcountries showed substantial increases in servicesR&D from the early 1980s to the late 1990s; forinstance, the shares of services in company-funded R&D increased by about 5 percentagepoints in France and Italy and 13 percentagepoints in Canada and the United Kingdom(United States, NSB 2004). The United Statesled the industrialized economies in terms ofservices R&D (box III.3). Interestingly, the R&Dintensity of services (R&D as a percentage ofsales) was higher than for manufacturing, thoughit also varied greatly by activity.

0

20

40

60

80

Japan United States European

Union

Developing

Asia

South-East

Europe and

CIS

Latin America

and the

Caribbean

1996 2002

109CHAPTER III

3. Capability needs and benefitsdiffer across activities

The efforts and capabilities required tomaster, adapt and create technologies, and thusto undertake R&D, differ. At the industry level,clothing manufacture is usually less complex inthe range and depth of technical skills orinformation needed than making semiconductors.Within complex industries, technical processesmay differ according to the speed of change andin the effort needed to create new generationsof technology: steel technology today is morestable and less demanding in product innovationthan electronics. Within any industry there canbe differences according to product: in textiles,for instance, yarn spinning, a capital- and scale-intensive activity, requires more advancedtechnical skills than clothing manufacture.Finally, there are differences by function for anygiven product. In clothing, sewing is easier thandesigning new fashion products or managing aninternational supply chain.

There is a similar hierarchy of technicalcomplexity in services, though it may be moredifficult to define than in manufacturing. As notedin box III.2, some services now performconsiderable R&D (indeed, the only output ofcontract research firms is research anddevelopment). Others do not conduct muchformal R&D but innovate in terms of productdevelopment (e.g. new financial services bybanks or new packages by tour operators) andmanagement practices. In broad terms, serviceactivities and functions can be ranked by the levelof skills required – formal (education levels) orinformal (employee training). In export-orientedservices, for instance, the bottom end may includesome call centres while the top end representsadvanced R&D (WIR04).

Different types of R&D also yield differentbenefits in terms of adding value, learning, skillcreation, productivity improvement, marketgrowth and spillovers to other activities (chapterVI). Complex R&D activities generally call for,and so create, more advanced skills andknowledge than simple ones; they also yieldhigher value added. Activities associated withrapid technical progress offer better prospectsfor future productivity increase and enjoy fastergrowth than other activities.12 Within atechnology, advanced functions like design anddevelopment (as compared to basic production)provide higher value added and so higher wages.As innovation moves into higher functions, the

NIS itself grows stronger and permits greaterinnovation in a more diverse range of activities.

The deepening of the industrial structurefrom simple to complex activities, and ofinnovative activities from simple to advancedfunctions, is a natural result of economicdevelopment, but accelerating and facilitatingthe process often requires active policies.13 Thisapplies not only to manufacturing but also toprimary production (with the advent ofbiotechnology and genetic modification inagriculture), infrastructure and services(particularly those IT-based ones that areundergoing rapid offshoring, analysed in WIR04).

The R&D hierarchy for the manufacturingsector depicted above is actually a goodrepresentation of the industrialization process.Most developing economies start modernmanufacturing with the simplest (low R&D)technologies: textiles, clothing, food-processingand wood products. Some move up the scale intoheavy process industries (metals, petroleumrefining) and metal products, providing basicintermediates. A few go on to become efficientusers of “medium-high” technologies, makingmore advanced intermediate and capital goods(chemicals, automobiles, and industrialmachinery). Even fewer develop competitivecapabilities in high-technology industries likeaerospace, micro-electronics or pharmaceuticals.

There is an important exception to thisdepiction, of special interest to this analysis. The“fragmentation” of production (i.e. the relocationof processes or functions across countries byTNCs to take advantage of differences inproduction and communication costs and skills)allows some countries without a strong R&D baseto leapfrog to production in high-technologyindustries l ike electronics (Arndt andKierzkowski 2001, Lall and Zhang 2004).14

While developing countries generally start at thelowest level of technical complexity – finalassembly – it is possible for them to move upthe innovation ladder in electronics, taking onmore demanding functions, handling moreadvanced equipment and making the morecomplex products.15 For such science-basedindustries as biotechnology and some ICT-relatedindustries, there may be limited need to locatethe R&D activity in close proximity toproduction. As noted by one observer (Reddy2000, p. 174): “because of their science base eventheoretically trained personnel, with little or noindustrial experience, can be employed for R&Dfunctions in new technologies.”

110 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Service enterprises in the United Statessharply increased their R&D spending and theirshare of total industrial R&D after the mid-1980s.Before 1983, service industries accounted for lessthan 5% of total industrial R&D; by 2002, theirshare reached 43%. The total value of R&D byservices was $82 billion compared to $109 billionfor manufacturing in 2002.

The amount of R&D by firms in serviceactivities varied greatly (box table III.3.1). Theleading performers were trade, scientific R&Dservices, software and computer systems design.With a combined R&D of $63 billion, theyaccounted for 77% of R&D by service firms.

The R&D intensity of service firms (R&Das a percentage of sales) is higher than that formanufacturing firms, though it also varies greatlyby activity (box table III.3.2).

However, the classification of firms underservice categories has to be treated with care.Companies are classified under various serviceactivities on the basis of payroll, and theclassification may be misleading as a result. Thisis particularly true of “trade”. Thus, firms with ahigh payroll in sales and marketing are classifiedunder “trade”, and may include manufacturers withhigh marketing payrolls or diversified industrialconglomerates. One example of misclassification(noted by NSF) is that over $1 billion of biotechR&D in 2001 appears to have been performed by

Box III.3. Services sector R&D in the United States

trading companies, when it is likely to have beenperformed by manufacturing companies.

Firms in software and computer systemsdesign and related services jointly spent $21billion on R&D in 2002, raising their share oftotal United States company-funded R&D from4% in 1987 to 12% in 2002.

Scientific R&D services, the leaders in R&Dintensity in 2001, are provided by companies thatperform R&D for other firms on a contractualbasis, mainly in manufacturing. R&D by thesefirms more than doubled during 1997-2001,showing both the rising pace of innovation andthe growing willingness of manufacturers tooutsource R&D previously kept in-house(Jankowski, 2001).

Health-care services are tightly linked tothe high-technology pharmaceutical industry.Firms in these services have traditionally donerelatively little R&D, but there was a sharpincrease in 2002. The financial services andinsurance industry, along with broadcasting andtelecommunications, does very little. However,formal R&D may not be the best way to measureinnovation in these industries, as they areconstantly designing and introducing newproducts and processes.

Box table III.3.1. R&D spending bynon-manufacturing activities in the

United States, 2002(Millions of dollars)

Total non-manufacturing 81 824

Mining, extraction and support activities app. 700Utilities app. 100Construction 164Trade app. 25 000Information 17 870Transportation and warehousing app. 300Newspapers, periodicals, books and databases 614Software 12 927Broadcasting and telecommunications app. 1 600Other information services app. 2 600Finance, insurance and real estate 1 903Architecture, engineering, related services 4 159Computer systems design, related services 11 983Scientific R&D services 13 034Other professional and scientific services 1 182Management of companies and enterprises 148Health-care services app. 4 200Other app. 900

Source: United States, NSF (forthcoming), tables A-2, A-3.Note: Approximate (app.) figures are based onR&D funded by industry; data on federal fundingof R&D are suppressed for confidentiality reasons,so that total R&D spending is also suppressed.

Box table III.3.2. R&D intensity: companyand other (non-federal) R&D funds as % of

net sales in R&D-performing firms

2001 2002All industries 3.8 3.6

Manufacturing 3.6 3.2Non-manufacturing 4.0 4.1Scientific R&D services 36.5 17.6Software 19.3 21.4Computer systems design, related services 16.5 14.3Management of companies 7.8 7.6Trade 6.2 5.0Architectural, engineering, related services 5.2 5.3Health-care services 4.1 15.1Newspapers, periodicals, books, databases 2.7 2.8Transportation and warehousing 2.4 0.5Construction 1.4 0.6Mining, extraction and support 1.3 3.2Finance, insurance and real estate 0.7 0.6Broadcasting and telecommunications 0.5 0.7Information 4.4 4.0

Source: United States, NSF (forthcoming), table A-27.

Source: UNCTAD, based on information provided by NSF.

111CHAPTER III

Some countries (Singapore amongdeveloping countries, Ireland among developedones) have managed such upgrading rapidly;China appears set to follow suit. In other words,provided they have the absorptive capacity andappropriate policies and institutions in place,developing countries can take advantage offragmentation to move up the technology ladder,both across activities and within them. Thefragmentation of functions is proceeding evenmore rapidly in some services, as communicationcosts fall dramatically due to new informationand communications technologies (WIR04) .However, taking advantage of the potential offragmentation requires countries to createknowledge and build local capabilities. As shownin the next section, the gap between theinnovative capabilities of countries is very wide.

C. The innovationcapability gap

1. Measuring innovationcapabilities

In order for countries to connect withglobal networks of knowledge creation as wellas to attract and benefit from R&D by TNCs, acertain basic level of innovative capabilities isneeded. However, countries vary greatly in thisrespect, and in many cases the gaps betweencountries have been growing over time. In orderto il lustrate the current situation, WIR05introduces a new measure of national innovationcapabilities: the UNCTAD Innovation CapabilityIndex (UNICI). The UNICI measures two criticaldimensions: (i) innovative activity (theTechnological Activity Index) and (ii) the skillsavailability for such activity (the Human CapitalIndex). As it is not possible to measure nationaltechnological activity or skills directly, theindices use proxies. Since the data available evenfor the proxies are not complete (caveats arenoted below) the indices should be interpretedwith caution and seen mainly as broad indicators(box III.4).16

National innovative activity can bemeasured by its inputs or outputs. On the “input”side, the usual measures are R&D expendituresand/or employment. R&D is a narrow measureof innovation effort in that it does not captureinformal technological effort; at the same timeit is rather broad in that it includes defence and

basic research that may not be relevant to thetypes of company R&D important for the presentanalysis.17 Still, R&D data are the only onesavailable on a comparable basis across countries,and they provide an indicator of technical effortin complex activities (where the absorption oftechnologies requires formal R&D). As R&Dexpenditure data are more limited than R&Dmanpower data for a given year, only the latterappear in the index.

Innovation “outputs” are often proxied bypatents (national or international) and scientificpublications.18 Data on patents taken out in theUnited States are singled out as they indicate thatthe innovation has reached a comparable levelof novelty and is commercially valuable.19

Patents are a better indicator of invention thanof innovation, since they do not capture thecommercial utility of the discovery; scientificpublications are further removed from the market,though they do show the knowledge base onwhich technological activities depend.

The human resource base for technologicalactivity is generally measured by educationalenrolment. Enrolment data do not capturedifferences in the quality and relevance of theeducation; neither do they reflect skilldevelopment by learning on the job or otherforms of employee training. Moreover, theavailable enrolment data are patchy and, in somecountries, out of date. Again, they are the onlydata available for benchmarking skills and theydo indicate differences in the education base onwhich technological capabilities are built.

These measures have to be normalized byeconomic size (say, population) to make themcomparable across countries. However, where theabsolute size of technological effort or skilledresearchers matters (i.e. where there are minimumcritical mass effects), it is also important tocompare total values for economies. This isparticularly relevant for the cross-border locationof R&D (chapter V).

The components and variables of theUNICI are shown in table III .4. The threecomponents of the Technological Activity Indexare weighted equally while those making up theHuman Capital Index are assigned differentweights to capture the greater importance of high-level skills for innovation. The UNICI iscalculated for 117 countries for the years 1995and 2001. The starting year, 1995, was selectedso as to include a large number of economies inSouth East Europe and the CIS.

112 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

The Technological Activity Index is shownin annex table A.III.4, with countries divided intofour roughly equal groups. Its ranks were stablebetween 1995 and 2001 (with a correlationcoefficient of 0.955). However, some countrieschanged ranks significantly. At the lower levelsthe changes generally arose from small shifts inone component, and so are difficult to interpret.At the higher levels they appear to be moreclearly related to changes in technological effort.It should be noted that the Index does not capturethe absolute size of the technological activities

in each country, thus biasing the Index againstcountries like China or India with large ruralpopulations, combined with large values for R&Dspending. To the extent that the internationalizationof R&D is affected by the absolute size oftechnological activity rather than its innovationintensity per capita, it is important to look at thisfactor as well (chapter V).

The Human Capital Index could becalculated for 119 countries.20 The countries aregrouped into three sets (annex table A.III.5). Most

Various attempts have been made tobenchmark national competitiveness andinnovation, separately or together (all analystsaccept innovation to be a vital ingredient ofcompetitiveness).a A recent survey of many of themain indices found that they have several elementsin common (Archibugi and Coco 2005).b All havevariables for innovation inputs (R&D effort,measured by R&D spending or personnel), outputs(patents, nationally or in the United States) andhuman capital (different measures of educationenrolment). Some also use scientific and technicaljournal articles, and some include variables forinfrastructure (power and ICT). UNDP uses theseinfrastructure variables to capture technologydiffusion (power for traditional technology andICT for modern technology). The Rand indexincludes GDP per capita along with the numberof universities and R&D institutions per capita.Some of these variables, like infrastructure, appearto be only remotely related to innovation; others,like GDP per capita, appear too broad to capturedifferences in technological capability.

The index which is probably closest to theUNICI is the Knowledge Index used by the WorldBank (www.worldbank.org/kam). However, whilethe Knowledge Index encompasses 14 dimensionsof knowledge capacities, the UNICI focuses oninnovation capacity, drawing on a smaller set ofvariables. The UNICI weightings (especially withregard to human capital) are also different.

Broader competitiveness indices like theone calculated by the World Economic Forum(published in its annual Global CompetitivenessReport) include subjective perceptions on thequality of innovation institutions, the strengthof intellectual property protection, theaggressiveness of local enterprises in absorbingtechnology and the uniqueness of local productinnovations.c These qualitative variables are notalways reliable, however, as respondents fromdifferent countries may use different standardsto answer the questions.

A merit of the UNICI is that it is basedentirely on quantitative variables, and uses onlythose that are direct measures of technologicalactivity and technical human capital. Thetechnological activity component of the indexuses R&D manpower,d patents taken out in theUnited States and scientific and technicalpublications (all deflated by population). TheHuman Capital Index uses literacy rates as thebroadest indicator of skills, secondary enrolmentsas an indicator of workforce skills and tertiaryenrolments as an indicator of high level skills.The components of the Technology Activity Indexare not weighted, but those of the Human CapitalIndex are: higher levels of education are assignedhigher weights because they are considered moreimportant for technical and managerialinnovation.e

Source: UNCTAD.

a See Archibugi and Coco 2004, IMD various years, Lall 2003, United States, NSB 2004, Porter and Stern 2001,UNIDO 2003, UNDP 2001, WEF various years.

b They discuss the UNDP index, their own ArCo index, the index developed by Lall and Albaladejo, 2002 and theRand index (Wagner et al. 2001).

c For a detailed critique see Lall 2001b.d The R&D manpower data were available for a larger number of countries than data for R&D spending.e A simple weighting scheme of 1 for literacy, 2 for secondary enrolment and 3 for tertiary enrolment is used.

Box III.4. Comparing the UNCTAD Innovation Capability Index with other indices

113CHAPTER III

developed and some transition economies are inthe leading group; this group also has fourdeveloping economies: the Republic of Korea,Taiwan Province of China, Argentina andUruguay in that order. As with the TechnologicalActivity Index, the Human Capital Index is stableover time, with a correlation coefficient of 0.973between 1995 and 2001. Again, the absolute sizeof the skills availability is not captured by theindex but is of importance for the internationalallocation of R&D internationalization (chapterV). The technology and skill indices are highlycorrelated (coefficients of 0.910 in 1995 and0.889 in 2001), though technological effort andskill formation do not always go together.

2. The UNCTAD InnovationCapability Index

The UNCTAD Innovation Capability Index(UNICI) consists of the unweighted averages ofthe two indices mentioned above. Countries aredivided into three groups: high, medium and low(table III.5). The high capability group in theUNICI comprises all developed countries(including the new EU members) as well as fourdeveloping and four South-East European andCIS countries (all from Europe). Three of the fourdeveloping economies are from South-East andEast Asia; the fourth (Argentina) is from LatinAmerica. The Asian ones combine strongtechnological and skill performance, whileArgentina is weak in technology but somewhatstronger in skills. The economies in transitionare in the top group mainly because of their skillbase – their technological performance isrelatively weak, with only one (the RussianFederation) in the high innovation group.

Table III.4. Components of the UNCTAD Innovation Capability Index

Indices Components Weights attached

Technological Activity Index R&D personnel per mill ion population All 3 components have equal weightsUnited States patents granted per mill ion populationScientific publications per mill ion population

Human Capital Index Literacy rate as % of population Weight of 1Secondary school enrolment as % age group Weight of 2Tertiary enrolment as % of age group Weight of 3

UNCTAD Innovation Technological Activity Index Both indices have equal weightsCapability Index Human Capital Index

Source: UNCTAD.

The “medium” capability group containsother South-East European and CIS economiesas well as most resource-rich and newlyindustrializing economies (including China andtwo sub-Saharan African economies, South Africaand Mauritius). The “low” capability group hasall the South Asian economies, one from South-East Asia (Indonesia), most sub-Saharan Africaneconomies and the remaining countries of LatinAmerica, West Asia and North Africa. Therankings are in line with received knowledgeabout national capabilities. If some economies(like India) seem misplaced, the explanation liesin the use of total population as the deflator;while this is the correct way to construct theindex, i t can be misleading when minimumcritical mass is important.

The unweighted regional averages for theUNICI are shown in table III.6. The developedcountries are well in the lead, albeit with a slightdecline in the average score. This does not meanthat they are investing less in skills or innovation,but rather, that other countries are spendingrelatively more. The new EU members improvedtheir scores during the period studied,approaching the levels of developed countries.The South-East and East Asia subregions are theclear leaders among developing regions, and theiraverage score combined has improved over time.The West Asia and North African subregions alsoimproved their performance, and overtook LatinAmerica and the Caribbean, which had adeteriorating score between 1995 and 2001. SouthAsia also shows a lower score over time, mainlybecause of weaker technological performance byPakistan and declining human capitalperformance by Sri Lanka. Sub-Saharan Africaimproves its average score marginally but stilllags behind all other regions.

114 World Investment Report 2005: Transnational Corporations and the Internationalization of R&DT

ab

le I

II.5

. T

he

UN

CT

AD

In

no

va

tio

n C

ap

ab

ilit

y I

nd

ex

Hig

hM

ediu

m

Low

199

5

20

01

1

995

2

001

19

95

2

001

1Sw

eden

0.

957

Swed

en

0.97

940

Uzb

ekis

tan

0.60

5Jo

rdan

0.

595

79M

ongo

lia0.

321

Sri L

anka

0.

317

2Fi

nlan

d 0.

947

Finl

and

0.97

741

Hon

g K

ong

(Chi

na)

0.59

3G

eorg

ia0.

593

80Tu

nisi

a0.

302

Bot

swan

a0.

315

3C

anad

a 0.

947

Uni

ted

Stat

es

0.92

742

Cyp

rus

0.58

1C

hile

0.

576

81In

dia

0.28

7A

lger

ia0.

312

4U

nite

d St

ates

0.

946

Den

mar

k 0.

926

43C

hile

0.

581

Cyp

rus

0.56

682

Bol

ivia

0.28

3Vi

et N

am0.

295

5A

ustr

alia

0.94

4N

orw

ay

0.92

344

Slo

vaki

a0.

580

Uzb

ekis

tan

0.56

483

Hon

dura

s0.

279

Indi

a 0.

285

6D

enm

ark

0.93

4A

ustr

alia

0.92

045

Sou

th A

fric

a0.

579

Hon

g K

ong

(Chi

na)

0.56

384

Ecu

ador

0.27

9El

Sal

vado

r 0.

279

7N

orw

ay

0.92

9C

anad

a 0.

907

46A

rmen

ia0.

574

Leba

non

0.55

585

El S

alva

dor

0.27

6Zi

mba

bwe

0.27

88

Uni

ted

King

dom

0.

914

Uni

ted

King

dom

0.9

0647

Cos

ta R

ica

0.55

5R

oman

ia0.

554

86In

done

sia

0.27

6M

oroc

co0.

277

9N

ethe

rland

s 0.

912

Bel

gium

0.89

448

Latv

ia

0.55

4S

outh

Afr

ica

0.54

887

Bot

swan

a0.

264

Indo

nesi

a 0.

261

10B

elgi

um0.

911

Net

herla

nds

0.88

849

Rom

ania

0.55

4B

razi

l0.

529

88N

amib

ia0.

261

Ken

ya0.

260

11Ja

pan

0.90

6Ja

pan

0.88

550

Leba

non

0.53

8A

rmen

ia0.

526

89A

lger

ia0.

257

Syr

ian

Ara

b R

ep.

0.24

612

Fran

ce

0.90

2N

ew Z

eala

nd0.

879

51K

azak

hsta

n0.

521

Kaz

akhs

tan

0.52

590

Om

an0.

234

Om

an0.

232

13G

erm

any

0.88

9S

witz

erla

nd0.

877

52K

uwai

t0.

515

Uru

guay

0.50

691

Par

agua

y0.

233

Dom

inic

an R

epub

lic0.

221

14N

ew Z

eala

nd0.

874

Icel

and

0.87

653

Vene

zuel

a0.

504

Kyr

gyzs

tan

0.50

092

Syr

ian

Ara

b R

ep.

0.22

5N

amib

ia0.

218

15S

witz

erla

nd0.

871

Taiw

an P

OC

0.86

554

Mol

dova

, R

ep.

of0.

497

Thai

land

0.

488

93Vi

et N

am0.

218

Par

agua

y0.

213

16A

ustr

ia0.

852

Fran

ce

0.86

355

Sau

di A

rabi

a0.

496

Sau

di A

rabi

a0.

476

94N

icar

agua

0.21

2N

icar

agua

0.17

917

Taiw

an P

OC

0.85

2A

ustr

ia0.

852

56B

ahra

in0.

485

Egyp

t 0.

474

95K

enya

0.17

3H

ondu

ras

0.17

418

Icel

and

0.83

5G

erm

any

0.85

057

Qat

ar0.

471

Kuw

ait

0.47

396

Gua

tem

ala

0.16

5N

iger

ia0.

157

19Ire

land

0.

829

Kore

a, R

ep. o

f 0.

839

58Ta

jikis

tan

0.45

4C

osta

Ric

a0.

472

97Pa

kist

an

0.16

0Ta

nzan

ia, U

nite

d R

ep. o

f0.

145

20Ko

rea,

Rep

. of

0.82

1Sp

ain

0.81

959

Mex

ico

0.45

4M

exic

o0.

469

98Za

mbi

a0.

150

Gha

na0.

143

21Sp

ain

0.81

4Ire

land

0.

814

60Ph

ilipp

ines

0.

452

Mal

aysi

a 0.

467

99N

iger

ia0.

137

Uga

nda

0.14

022

Isra

el0.

808

Isra

el0.

804

61Eg

ypt

0.44

9B

ahra

in0.

466

100

Cam

eroo

n0.

133

Paki

stan

0.

137

23R

ussi

an F

ed.

0.79

7S

love

nia

0.80

162

Per

u0.

448

Vene

zuel

a0.

460

101

Gha

na0.

131

Gua

tem

ala

0.13

524

Italy

0.78

1R

ussi

an F

ed.

0.78

863

Turk

ey

0.43

0P

eru

0.42

510

2C

ôte

d’Iv

oire

0.12

9C

amer

oon

0.13

425

Esto

nia

0.77

4Es

toni

a 0.

775

64B

razi

l0.

421

Phili

ppin

es

0.42

310

3B

enin

0.11

8M

adag

asca

r0.

133

26Be

laru

s 0.

770

Sing

apor

e 0.

748

65Th

aila

nd

0.41

3M

oldo

va,

Rep

. of

0.41

310

4M

alaw

i0.

109

Yem

en0.

130

27S

love

nia

0.76

3Ita

ly 0.

746

66Ja

mai

ca0.

394

Qat

ar0.

403

105

Ban

glad

esh

0.10

9C

ôte

d’Iv

oire

0.12

728

Gre

ece

0.73

3P

ortu

gal

0.74

667

Mal

aysi

a 0.

393

Jam

aica

0.39

510

6S

eneg

al0.

105

Ban

glad

esh

0.12

129

Ukr

aine

0.72

8Li

thua

nia

0.74

268

Mau

ritiu

s0.

390

Col

ombi

a 0.

393

107

Uga

nda

0.08

1Za

mbi

a0.

115

30Si

ngap

ore

0.71

9G

reec

e0.

737

69K

yrgy

zsta

n0.

370

Mon

golia

0.39

110

8Ta

nzan

ia, U

nite

d R

ep. o

f0.

080

Ben

in0.

106

31Po

land

0.

717

Pola

nd

0.73

270

Col

ombi

a 0.

367

Turk

ey

0.39

010

9Ye

men

0.07

9M

alaw

i0.

105

32H

unga

ry

0.70

5H

unga

ry

0.72

571

Dom

inic

an R

ep.

0.35

7B

oliv

ia0.

367

110

Mau

ritan

ia0.

070

Sen

egal

0.10

133

Por

tuga

l0.

704

Ukr

aine

0.70

572

Chi

na

0.35

4Tu

nisi

a0.

365

111

Mad

agas

car

0.06

5M

aurit

ania

0.06

834

Geo

rgia

0.68

3La

tvia

0.

705

73Zi

mba

bwe

0.35

1Ta

jikis

tan

0.36

211

2D

jibou

ti0.

057

Erit

rea

0.05

435

Bul

garia

0.67

1Be

laru

s 0.

697

74Ir

an, I

slam

ic R

ep. o

f0.

349

Chi

na

0.35

811

3H

aiti

0.05

5E

thio

pia

0.05

136

Lith

uani

a 0.

665

Cze

ch R

ep.

0.69

075

Uni

ted

Ara

b E

mira

tes

0.34

6Ir

an, I

slam

ic R

ep. o

f0.

346

114

Erit

rea

0.04

7H

aiti

0.04

637

Cze

ch R

ep.

0.64

8A

rgen

tina

0.68

576

Jord

an

0.33

9U

nite

d A

rab

Em

irate

s0.

326

115

Eth

iopi

a0.

046

Moz

ambi

que

0.03

038

Arg

entin

a0.

640

Bul

garia

0.66

577

Sri L

anka

0.

336

Mau

ritiu

s0.

323

116

Ang

ola

0.02

2D

jibou

ti0.

028

39U

rugu

ay0.

617

Slo

vaki

a0.

626

78M

oroc

co0.

324

Ecu

ador

0.31

911

7M

ozam

biqu

e0.

018

Ang

ola

0.01

9

So

urc

e:

UN

CT

AD

.N

ote

:T

he

co

mb

ine

d i

nd

ex

is a

n u

nw

eig

hte

d a

vera

ge

of

the

Te

chn

olo

gic

al A

ctiv

ity

Ind

ex

an

d t

he

Hu

ma

n C

ap

ita

l In

de

x.

115CHAPTER III

Table III.6. Regional unweightedaverages for the UNCTAD

Innovation Capability Index

Region 1995 2001

Developed countries (excl. thenew EU members) 0.876 0.869The new EU members 0.665 0.707South-East Europe and CIS 0.602 0.584South-East and East Asia 0.492 0.518West Asia and North Africa 0.348 0.361Latin America and the Caribbean 0.375 0.360South Asia 0.223 0.215Sub-Saharan Africa 0.157 0.160

Source: UNCTAD.

Each of these three indices is highlycorrelated with income. In a regression analysis,the log of per capita income “explains” 75% ofthe variation in the Technology Activity Indexin 2001, 66% of the variation in the HumanCapital Index and 74% of the variation in theUNICI. As expected, technological activity, skillsand incomes reinforce each other. The causalconnections between the three are highlycomplex, and there are many possible feedbackloops. For example, more technological activityleads to higher incomes, and higher incomesallow countries to invest more in innovation.However, it can be argued that the main causallink is likely to run from innovative activity and

skills to incomes, and that innovative activityrequires more advanced skills.21

Still, the indices do not rise uniformly withincome levels. As the scatter diagram shows,there is a large variation around the regressionline for the UNCTAD Innovation Capability Index(figure III.4).22 Countries above the line havehigher incomes than predicted by their innovationindex value (i.e. scoring lower on the index thanpredicted by their incomes); those below the linescore higher on the index than predicted by theirincomes. Hong Kong (China) has the lowestcomposite innovation score in relation to its percapita income (presumably earning high incomefrom service activities that do not requiresignificant technological effort), followed bysome small resource-rich economies. At the otherend of the spectrum, various economies intransition have high composite scores relativeto income, a result, as noted above, of theirrelatively strong performance in skill creation.

To sum up, there are large gaps betweencountries in terms of technological activity andhuman capital. The gap is not just between thedeveloped and developing countries, but alsowithin the developing and transition economies.In the developing world, innovative capabilitiesare highly skewed, with South-East and East Asiaat the high end and sub-Saharan Africa at the lowend of the spectrum. Within South-East and East

Figure III.4. Relationship between the UNCTAD Innovation Capability Index andlog per capita GDP, 2001

Source: UNCTAD.

2

3

4

5

6

7

8

9

10

11

12

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Innovation capability index

EthiopiaMalawi

Kenya MongoliaKyrgyzstan

Georgia

UzbekistanUkraine

BelarusLithunia

Djibouti

Guatemala

NamibiaMauritius

Botswana

Bahrain

Kuwait

Hong Kong,China

Cyprus

Italy

Singapore Ireland

Switzerland

Viet Nam

China

India Tajikistan

Bulgaria

RussianFederation

Japan

Sweden

Norway United States

Multiple R: 0.86R Square: 0.74Adjusted R Square: 0.74n= 113

Pe

rca

pita

GD

P(in

log

s)

116 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Asia, the three leaders (the Republic of Korea,Taiwan Province of China, Singapore) are wellahead of the other economies. Transitioneconomies have large reservoirs of skills inrelation to their income levels but seem to lagin technological effort.

While the Index suffers from the inevitableproblems of finding the appropriate measures fortechnological effort and human capital, its useof hard statistics provides intuitively plausibleresults:

• Innovative capabilities differ greatly acrosscountries, and the ranks are quite stable overthe period considered. It is proving difficultfor countries at the bottom to improve theirposition over time; there are cumulativeforces at work that seem to reinforce theadvantages of the leaders. It also suggeststhat significant change takes time to achieve.

• However, some countries have improvedtheir ranking. Thus, while developedcountries dominate the “high” group in theUNICI, that group also includes fourdeveloping economies and four economiesin transition.

• The three leading developing economieshave participated vigorously in the globalproduction and innovation system, but eachdid so using different means to accesstechnologies and build domesticcapabilities.23 Each invested heavily ineducation and skills development, sincesustained progress in either strategy requireshighly skilled human capital. Mostfundamentally, in each case access to globaltechnologies and to foreign markets wascritical to sustained growth and upgrading.

• The main strength of the economies intransition, particularly those in Europe, liesin their human capital, rather than intechnological activity, suggesting that thereis scope for using the former to enhance thelatter.

• South Asia and sub-Saharan Africa lagbehind the other regions in innovation and,more particularly, in human capital creation.

What are the implications of theseobservations? The first , of course, is thatinnovative capabilities affect countries’ abilityto develop and raise l iving standards. In aglobalizing world with rapid technical change,strong and growing innovative capabilities are

essential to economic progress. This is as trueof resource-based economies as of others, andit applies as much to services and agriculture asit does to manufacturing. As technologicalprogress proceeds at an accelerating pace, andas the competitive pressure on firms intensifies,the demands made on countries’ capabilities rise.This makes it more important than ever beforeto seek ways to bridge the gaps that exist.

Second, innovative capabilities are directlyrelevant to the location of internationally mobileR&D – the theme of WIR05. TNCs seeking R&Dsites overseas look for adequate supplies ofqualified technical manpower and innovativeactivity (chapter V). This is not to say that theseare the only factors at work in their choices.Attracting global R&D, whether conducted in-house by TNCs or outsourced to local serviceproviders, also needs such conditions as a stableand conducive investment climate, capable localfirms, adequate ICT and other infrastructure, andintellectual property protection. But innovationcapabilities – of the right quality and at the rightcost – are clearly the conditio sine qua non.

Third, innovative capabilities also affectthe scope for host-country benefits frominternationalized R&D (chapter VI). The qualityof R&D that is internationalized depends on localcapabilities. The same applies to the resultingexternalities, in terms of how much local firmsand institutions are able to absorb and learn fromexposure to best practice R&D techniques andskills. Whether or not R&D deepens over time,and how far it spreads over different activities,are almost entirely a function of the strength ofthe local skill and innovation system.

Finally, a word of caution. Nationalinnovative capabilities as measured above canbe misleading where minimum critical massconsiderations apply. While deflatingtechnological effort and skill formation by thesize of the economy is the right way to calculatea capability index, it skews the result againstcountries that have a large pool of employableskilled manpower with diverse skills, even withlow rates of skill creation at the national level.Thus the absolute size of the stock of educatedpeople has to be taken into account whenconsidering the determinants of R&D location.This explains the relatively modest positions inthe UICI rankings of China and India, twosignificant players in the recent increase in R&Dinternationalization by TNCs (chapter IV).

117CHAPTER III

D. Conclusion

There is a co-evolution of economicdevelopment and technological complexity (byactivity and function). The higher levels of skillsand technological capabilities that accompanydevelopment permit countries to shift into moreadvanced activities and functions. More advancedactivities and functions, in turn, yield highervalue added, and allow countries to remaincompetitive despite higher wages. While this isa natural feature of the development process,countries can improve their innovativecapabilities by appropriate policy interventions(chapter VII).

To summarize the main features ofinnovation highlighted above:

• Innovation is essential for economicdevelopment. Although in today’sglobalizing world economy developingcountries can obtain new technology fromother, more developed countries, they haveto learn and innovate in order to use newtechnologies efficiently. As countries moveup the development ladder and undertake morecomplex activities they need to upgrade theirtechnological capabilities and undertakemore advanced forms of innovation.

• The ways in which innovation takes placecan be diverse, but an important source ofinnovation is through R&D. Formal R&Dbecomes essential at a certain stage,certainly in manufacturing, and increasinglyin some kinds of modern services andagriculture.

• Enterprise innovation involves interactionswith other firms and institutions: technologydevelopment is a systemic process. Giventhe externalities, coordination problems andpublic goods (basic research, testing,metrology) inherent in this process,government involvement is vital particularlyin the early stages. In fact, withoutappropriate industrial , technology andeducation policies, R&D in the businesssector is unlikely to take off (chapter VII).

• Business R&D is geographically andsectorally concentrated. While the bulk isundertaken in developed countries, R&D insome developing countries – especially indeveloping Asia – is expanding particularlyfast. Most R&D takes place inmanufacturing, but it is also growing in theservices sector.

• Technological advances worldwide,especially in ICT, have created newopportunities for developing countries toparticipate in global knowledge networksonce they have the necessary capabilities.At the same time, minimum entry levels arerising in terms of the capabilities required.The cumulative nature of capabilitybuilding, together with scale andagglomeration economies, means that thesuccessful early starters can continue pullingahead of latecomers that are unable to reachthe minimum entry levels. Policyintervention is necessary to reverse thistrend.

• Innovation – and especially R&D –increasingly needs constant access tointernational knowledge. All “lateindustrializers” tapped technical knowledgeand skills from the early starters, though indifferent ways. While there are various waysto link up with global knowledge networks,inward and outward FDI in R&D is perhapsthe most direct way in which a country canconnect with centres of knowledge in othercountries.

• National innovation systems are becomingincreasingly interdependent. The absenceof local capabilities can effectively limitinteraction between one system and the restof the world, and thereby condemn thesystem in question to isolation from themainsprings of technical change andcompetitiveness.

The extent to which developing countriescan link up with global networks of learning andknowledge creation depends on their nationalinnovative strengths. These strengths differgreatly, and the UNCTAD Innovation CapabilityIndex shows that gaps between countries tendto persist over long periods. While the earlystages of development necessarily have to involvenurturing indigenous innovative capabilities inthe public as well as in the private sector, TNCscan play a role in strengthening an NIS (chapterVI). But foreign affil iates do not alwaysundertake high-level technological activities inhost countries. Many developing economies havelong had FDI in resource extraction,manufacturing and services without foreignaffiliates doing R&D. What is new is that thetrend is for more TNCs to spread R&D to somedeveloping countries, to a degree and in waysnot seen before. The next two chapters map thisprocess and discuss the factors that drive itsinternationalization and location.

118 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Notes1 According to the so-called Oslo Manual:

“Technological product and process (TPP) innovationcomprise implemented technologically new productsand processes and significant technologicalimprovements in products and processes. A TPPinnovation has been implemented if it has beenintroduced on the market (product innovation) or usedwithin a production process (process innovation). TPPinnovations involve a series of scientific, technological,organisational, financial and commercial activities.”(OECD 1997a, p. 31).

2 A large body of “evolutionary” literature on technologyargues that there is no essential difference betweenabsorbing, adapting and improving technologies andcreating entirely new technologies (Nelson and Winter1982, Metcalfe 1995). There is also a growing literaturein this tradition which analyses technological activityin developing countries, see, e.g. Bell and Pavitt 1993,Dahlman et al. 1987, Katz 1987, Ernst et al. 1998,Lall1992 and 2001a, Nelson 1990, Radosevic 1999,UNIDO 2003.

3 Several authors have classified technical functions byinnovativeness. See, for instance, Bell and Pavitt 1993,Hobday 2001, Figueiredo 2001, Ernst et al. 1998, Lall1992.

4 A more detailed classification of functions by levels oftechnical complexity is provided in annex table A.III.1.

5 Even in developed countries, much R&D (Cohen andLevinthal 1989 estimate it at about half) is of this type;R&D has “two faces”: learning and innovation.

6 For a survey, see Guellec and van Pottelsberghe 2004a.7 Data for at least one year’s total R&D spending over

the period 1996-2002 are available for 93 economies,including all the major R&D performers (annex tableA.III.2). Additionally, partial data are available from57 economies on business enterprise spending on R&D.

8 In national currencies, however, R&D expendituresincreased somewhat in these economies as well.

9 Data on business expenditures on R&D are notavailable for African countries.

10 For updated versions, see Hatzichronoglou 1997 andUnited States, NSB 2004.

11 For example, it is possible that low-technologyindustries engage in more complex or fundamentalresearch than do high-technology sectors.

12 Data on 70 economies that account for 97% of globaleconomic activity show that high-technologymanufacturing output grew at 6.5% per annum over1980-2001, while other manufacturing output grew at2.4% (United States, NSF 2004).

13 Government efforts to tap such technologicaldifferences date back to the beginnings of industrialpolicy in the 15th century (Reinert 1995). Countrieshave long tried to move their productive structures fromactivities with decreasing returns to those withincreasing returns – initially from primary productionto manufacturing and later, within manufacturing, fromlow- to high-technology activities. In modern economictheory, conditions of diffuse externalities withcoordination problems and other market failures leadto multiple equilibriums, and so require coherentgovernment intervention to move from low to highgrowth equilibriums (Hoff and Stiglitz 2001).

14 Other high-technology activities (e.g. in aerospace,precision instruments and pharmaceuticals), may not

be suited to fragmentation because of security concerns,specific skill needs, continuous processes of productionor scale economies.

15 Foreign technology and R&D facilities can also beacquired through outward FDI.

16 The UNCTAD Innovation Capability Index draws onthe World Bank (2004) for data on literacy rates,tertiary enrolment rates, technical publications, R&Dand general data on population and GDP; UNIDO(2003) for enrolments in technical subjects; theUNESCO website (www.unesco.org) for researchersin R&D and enrolments at primary and tertiary levels;the USPTO website (www.uspto.gov) for patents inthe United States; the Eurostat website (europa.eu.int/comm/eurostat) for R&D data; and the RICYT website(www.ricyt.org) for R&D in Latin America.

17 Even formal R&D data are deficient. Many developingcountries do not collect or publish them, or they providevery outdated information. Some data may not conformto internationally accepted definitions of whatcomprises R&D. For the purposes of industrialinnovation, the most important variable in R&Dinternationalization, the best measure would be R&Dconducted by enterprises. However, data on thiscomponent of R&D are even scarcer in developingcountries than on total R&D, and this measure was notused here for this reason.

18 Some studies also use total factor productivity (TFP)to measure the “output” of innovation. However,comparable TFP data are difficult to obtain and theresults are subject to severe methodological andinterpretational problems at the national level.

19 While there are potential biases associated with theuse of USPTO data, it is the least biased indicator(Dernis et al. 2001). Data on Triadic patents – takenout at the European Patent Office, the Japanese PatentOffice and the USPTO – can reduce the “home bias”,and may capture the most commercially valuablepatents (since taking them out involves substantialcosts). However, the number of Triadic patents isrelatively small (around 44,000 compared to some180,000 for USPTO patents) (OECD 2004b). They mayalso be biased against developing-country firms thattend to focus on patenting in the United States, whichis the largest export market for many of them.

20 This was two more than the Technological ActivityIndex, but the extra two were dropped for the combinedIndex.

21 There might be a reverse causality between per capitaincome and the UNICI. Richer countries are better ableto support education and innovation. In addition,countries with oil resources consistently display ahigher per capita income than the UNICI would predict.At the same time, in poorer countries, it is likely thata higher human capital index leads directly to higherincome, which in turn leads to higher technologicalcapabilities and a higher value in the UNICI.

22 Only countries deviating significantly from the lineare mentioned in the chart.

23 Singapore relied heavily on FDI and insertion into theproduction (and later, R&D) networks of developed-country TNCs (chapter V); the other two have reliedmore on arm’s length technology transfers by TNCs,using original equipment manufacture (OEM) contractsand licensing as well as developing local technologicaland R&D capabilities (Lall 2001a).

��������

R&D BY TNCs AND DEVELOPINGCOUNTRIES

TNCs are playing a major role in globalR&D, not only through activities in their homecountries but also increasingly abroad. Theinternationalization of R&D is not a newphenomenon. What is new is its faster pace inrecent years and its spread to developingcountries (albeit to only a few, mainly in Asia).Moreover, R&D activities in developing countriesare no longer aimed at adapting technologies tolocal conditions only; they increasingly involve“innovative” R&D, including developingtechnologies for regional and world markets. Atthe same time, TNCs from developing countriesare themselves investing in R&D abroad,primarily in order to access advancedtechnologies and research capabilit ies indeveloped countries, as well as to adapt productsto new markets and tap sources of specializedexpertise in other developing countries. Thischapter maps these trends.

A. TNCs are dominantR&D players

TNCs account for a major share of globalR&D. Indeed, with $310 billion spent in 2002(United Kingdom, DTI 2004), the 700 largestR&D spending firms of the world – of which atleast 98% are TNCs1 – accounted for close to half(46%) of the world’s total R&D expenditure andmore than two-thirds (69%) of the world’sbusiness R&D (annex table A.III.2).2 Given thatthere are an estimated 70,000 TNCs in the world(annex table A.I.8), this is a conservativeestimate. It confirms earlier findings that in themid-1990s TNCs already accounted for a verylarge share of the R&D expenditure of the Triad(Gassmann and von Zedtwitz 1999).3

In fact, the R&D spending of some largecorporations is higher than that of many countries.In four TNCs (Ford Motor, Pfizer, DaimlerChryslerand Siemens), R&D spending exceeded $6 billionin 2003 (table IV.1). In another two (Toyota Motorand General Motors), it surpassed $5 billion. Byway of comparison, in developing economies,South-East Europe and the CIS as a group, totalgross expenditure on R&D (GERD) came close toor exceeded $5 billion in 2002 (the latest availableyear) only in China, the Republic of Korea, TaiwanProvince of China and Brazil, in that order (tableIII.1). Even in large economies, such as India,Mexico and the Russian Federation, it remainedwell below the $5 billion mark. The same is truefor such small, developed and R&D-intensivecountries as Austria, Denmark and Finland (figureIV.1).

Over 80% of the 700 largest R&D spendingfirms come from only five countries: the UnitedStates, Japan, Germany, the United Kingdom andFrance, in that order (table IV.2). Only 1% of thetop 700 are based in developing countries orSouth-East Europe and the CIS (table IV.1),although several have moved up the ranks sincethe late 1990s (United Kingdom, DTI 2004).Almost all these firms come from Asia, notablyfrom the Republic of Korea and Taiwan Provinceof China (table IV.2), while only one is fromAfrica and two are from Latin America.

The 700 largest R&D spenders areconcentrated in relatively few industries. In 2003,more than half of them were in three industries(IT hardware, automotive and pharmaceuticals/biotechnology) (table IV.3).

Within each industry, the two largest R&Dperforming firms were responsible for very highshares. The two most concentrated industries

120 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Table IV.1. The top 20 firms, by R&D expenditure in the world andin developing economies, South-East Europe and CIS, 2003

(Millions of dollars)

World Developing economies, South-East Europe and CIS

World R&D World R&Drank Corporation Home economy spending rank Corporation Home economy spending

1 Ford Motor United States 6 841 33 Samsung Electronic Republic of Korea 2 7402 Pfizer United States 6 504 95 Hyundai Motor Republic of Korea 7343 DaimlerChrysler Germany 6 409 110 LG Electronics Republic of Korea 6124 Siemens Germany 6 340 178 Taiwan Semiconductor Taiwan Province of China 3425 Toyota Motor Japan 5 688 219 PetroChina China 2656 General Motors United States 5 199 255 Accenture Bermuda 2287 Matsushita Electric Japan 4 929 258 Korea Electric Power Republic of Korea 2278 Volkswagen Germany 4 763 267 KT Republic of Korea 2199 IBM United States 4 614 298 Marvell Technology Bermuda 197

10 Nokia Finland 4 577 300 POSCO Republic of Korea 19611 GlaxoSmithKline United Kingdom 4 557 317 Petroleo Brasileiro Brazil 18312 Johnson & Johnson United States 4 272 328 SK Telecom Republic of Korea 17213 Microsoft United States 4 249 337 China Petroleum & Chemical China 16714 Intel United States 3 977 348 Winbond Electronic Taiwan Province of China 15815 Sony Japan 3 771 349 Embraer Brazil 15816 Honda Motor Japan 3 718 350 United Microelectronics Taiwan Province of China 15717 Ericsson Sweden 3 715 486 Pliva Croatia 9918 Roche Switzerland 3 515 516 Sasol South Africa 9119 Motorola United States 3 439 518 AU Optronics Taiwan Province of China 9120 Novartis Switzerland 3 426 585 Hyundai Heavy Industries Republic of Korea 77

Source: UNCTAD, based on United Kingdom, DTI 2004.

Figure IV.1. R&D expenditure by selected TNCs and economies, 2002(Billions of dollars)

Source: UNCTAD, based on annex table A.III.2 and United Kingdom, DTI 2004.

3.5

3.7

3.7

3.8

4.0

4.3

4.3

4.3

4.3

4.4

4.4

4.5

4.5

4.6

4.6

5.4

5.4

5.5

5.7

5.9

6.3

6.5

6.8

4.8

7.2

0 1 2 3 4 5 6 7 8

Motorola (United States)

Johnson & Johnson (United States)

India (2001)

Intel (United States)

Microsoft (United States)

Volkswagen (Germany)

Russian Federation

Matsushita Electric (Japan)

Denmark

GlaxoSmithKline (United Kingdom)

IBM (United States)

Austria

Finland

Toyota Motor (Japan)

Brazil (2003)

Pfizer (United States)

Israel (2001)

General Motors (United States)

Belgium

Siemens (Germany)

DaimlerChrysler (Germany)

Switzerland (2000)

Taiwan Province of China

Spain

Ford Motor (United States)

0.2

0.3

0.4

0.5

0.7

0.7

0.7

0.9

1.0

1.1

1.2

1.4

1.9

2.7

2.9

3.0

3.0

3.1

3.1

3.1

3.1

3.2

3.3

3.4

3.4

0 1 2 3 4

Egypt (2000)

Thailand

Argentina

Chile

Malaysia

Hungary

South Africa

Czech Republic

Hong Kong, China

Poland

Turkey

Ireland

Singapore

Mexico

Novartis (Switzerland)

Hitachi (Japan)

Philips (Netherlands)

NTT (Japan)

Hewlett-Packard (United States)

Honda Motor (Japan)

Ericsson (Sweden)

Cisco Systems (United States)

Aventis (France)

Nokia (Finland)

Sony (Japan)

TNCs Economies

121CHAPTER IV

were telecommunications (because of NTT) andsoftware and computer services (because ofMicrosoft and IBM). The industry compositionof the top R&D spenders varies by region (UnitedKingdom, DTI 2004, p. 5). Those inpharmaceuticals and health, electronics and ICTaccount for more than two-thirds of the R&Ddone by United States-based firms. German firmsare concentrated in chemicals and engineering(64%), while Japanese firms are concentrated inelectronics, ICT, engineering and chemicals(90%).

In sum, TNCs dominate global businessR&D. A few countries, generally the largest R&Dspenders, account for a major share of businessR&D. Within those countries a relatively smallnumber of enterprises dominate R&D activity.Most R&D is conducted by firms in the ICT,automotive and pharmaceutical industries.

B. R&D by TNCs isinternationalizing

R&D is among the least internationalizedsegments of the TNCs’ value chain; production,marketing and other functions have moved abroadmuch more quickly. However, some R&D hasbeen undertaken abroad for a long time. In someform, R&D internationalization may date backto the earliest days of FDI; TNCs have alwayshad to adapt technologies for selling in hostcountries, and in many cases some R&D has beennecessary for this purpose (Safarian 1966, Brash1966). There have also been cases of

internationalization of basic research.In the years after the Second World War,Monsanto Chemicals (United States)expanded its centre for basic researchin New Port, United Kingdom. EssoPetroleum Company’s (United States)laboratories in the United Kingdom alsoperformed basic research, andpioneered, among other inventions, anew synthetic lubricant for high-speedjet aircraft (Dunning 1958, p. 169).Firms from small developed homecountries have conducted innovative(“asset-seeking”) R&D abroad in otherdeveloped countries in order to tapother centres of innovation andovercome the constraints of theirdomestic economy (such as relativelysmall and/or specialized pools ofknowledge and skills). Although the

Table IV.2. Home economies of the 700 largest R&D spending firms

of the world, 2003(Number of companies and per cent)

Number Percentage of largestEconomy of firms 700 R&D spenders

United States 296 42.3Japan 154 22.0Germany 53 7.6United Kingdom 39 5.6France 35 5.0Switzerland 20 2.9Sweden 15 2.1Republic of Korea 10 1.4Denmark 8 1.1Taiwan Province of China 8 1.1Netherlands 8 1.1Canada 7 1.0Belgium 6 0.9Finland 6 0.9Italy 6 0.9Spain 4 0.6Bermuda 3 0.4Norway 3 0.4Austria 2 0.3Australia 2 0.3Brazil 2 0.3China 2 0.3Ireland 2 0.3Israel 2 0.3Luxembourg 2 0.3Croatia 1 0.1Greece 1 0.1Hong Kong, China 1 0.1Liechtenstein 1 0.1South Africa 1 0.1Total 700 100.0

Source: UNCTAD, based on United Kingdom, DTI 2004.

Table IV.3. Industry breakdown of the 700 largestR&D performing firms, 2003

(Per cent)

Share of 700 Share of twocompanies’ R&D largest spenders

Industry expenditure within the industry

IT hardware 21.7 13Automotive 18.0 21Pharmaceuticals and biotechnology 17.5 18Electronic and electrical 10.4 31IT software and computer services 6.3 44Chemicals 4.8 23Aerospace and defence 3.9 35Engineering 2.9 20Telecommunications 2.2 58Health-care products and services 2.2 33Others 8.2 ..

Source: UNCTAD, based on United Kingdom, DTI 2004.

122 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Table IV.4. Global employment, R&D employment, andR&D expenditures of United States TNCs, by domestic and

overseas components, 1994, 1999, 2002

R&D R&DTotal R&D R&D expenditures per employment

employment employment expenditures R&D employee intensitya

Item (Thousands) ($ mill ion) ($) (%)

1994Total 24 273 727 103 451 142 338 3.0Domestic operations (United States parent companies) 18 565 625 91 574 146 565 3.4Overseas operationsb 5 707 102 11 877 116 441 1.8

1999Total 30 773 770 144 435 187 505 2.5Domestic operations (United States parent companies) 23 007 647 126 291 195 255 2.8Overseas operations b 7 766 124 18 144 146 915 1.6

2002Total .. .. 159 119 .. ..Domestic operations (United States parent companies) .. .. 137 968 .. ..Overseas operations b .. .. 21 151 .. ..

Source: UNCTAD, adapted from Moris 2005a, based on United States, National Science Foundation 2004.a R&D employment intensity refers to the share of R&D employment in total employment.b Majority-owned foreign affiliates.

internationalization of R&D has lagged behindthat of other activities, the share of foreign R&Din the total is rising steadily.

R&D between countries can be linked inseveral ways, involving flows in both directionsand several types of actors. Through FDI, TNCscan set up new foreign affil iates or acquireexisting firms that are already conducting R&Din host countries. Greenfield investments aremore common than acquisitions of localenterprises with R&D capacity, though exceptionsexist in countries with strong local firms(Brockhoff 1998, van Boehmer 1995, Håkansonand Nobel 1993a). TNCs can also contract R&Dto service providers in host countries withoutacquiring an ownership stake. In some activities(such as in software or pharmaceuticals in India),arm’s length contracts with local enterprises orresearch laboratories are increasingly common.Internationalization of R&D can also take theform of contracts between two non-transnationalfirms that are located in different countries.Finally, enterprises in two or more countries canenter into alliances to conduct R&D jointly.

1. A growing share of TNCs’ R&D isperformed abroad

Despite difficulties in data gathering, theavailable evidence gives a reasonable picture ofthe R&D being carried out by TNCs abroad.

Patterns vary significantly according to homecountries, as illustrated by the United States,Sweden, Japan and Germany, but the trend isclear: a growing share of R&D is undertakenabroad.

In the United Kingdom, the United Statesand some smaller European countries, TNCsstarted internationalizing R&D on a large scalein the 1980s and this trend was accelerated inthe 1990s.4 R&D expenditures by majority-owned foreign affiliates of United States TNCsincreased every year from 1994 to 2002 (exceptin 2001), reaching a record $21 billion in 2002.This level represented 13.3% of those TNCs’ totalR&D, up from 11.5% in 1994 (Moris 2005a).5

In terms of employment, 16% of the R&Dworkers of United States TNCs were in foreignaffiliates in 1999, up from 14% five years earlier(table IV.4).6 Following the international trend,Swedish TNCs have also expanded their R&Dactivities abroad over time. Between 1995 and2003, R&D spending by the largest SwedishTNCs increased modestly, from $5.1 billion to$5.8 billion (table IV.5),7 but the share of R&Doutside Sweden shot up from 22% to 43%.

In other home countries such as France,Germany, Italy, Japan and Spain, internationa-lization of R&D started much later, sometimesfocusing more on licensing than on FDI.8 TheR&D expenditure of Japanese TNCs abroad rosefrom $1.9 billion to $3.3 billion during the period

123CHAPTER IV

1995-2002 and its share in total Japanese R&Ddoubled from 2% to 4% (figure IV.2). Data fromother home countries (e.g. Germany, box IV.1)are less comprehensive, although they are alsoindicative of the growing internationalization ofR&D.

A number of surveys confirm the increasedinternationalization of R&D. One such surveyfounds that firms steadily increased their R&Dspending abroad from 15% of their total R&Dbudget in 1995 to 22% in 2001 (Roberts 2001).Other recent studies also pointed to a trendtowards increasing R&D abroad by TNCs fromthe Triad, especially European TNCs (Edler etal. 2002, von Zedtwitz and Gassmann 2002).9

A survey undertaken by UNCTAD fromNovember 2004 to March 2005 of the world’slargest R&D investors (box IV.2) suggests thatthe pace of R&D internationalization may beaccelerating (section F). The average firm in theUNCTAD survey spent 28% of its R&D budget

Table IV.5. R&D expenditures of the 20 largestSwedish TNCs, 1995-2003

(Billions of dollars)

Item 1995 1997 1999 2001 2003

Total R&D expenditure by Swedish TNCs 5.07 6.06 5.45 5.86 5.81R&D in Sweden 3.97 3.90 3.13 3.36 3.34R&D abroad 1.11 2.17 2.31 2.50 2.47In developing countries and economies in transition 0.03 0.07 0.10 0.15 0.18Foreign share (%) 22 36 42 43 43

Source: UNCTAD, based on ITPS 2003 and 2005, and additional information providedby ITPS.

abroad in 2003,10 including in-house expenditure by foreignaffiliates and extramuralspending on R&D contracted toother countries (figure IV.3).The share of R&D workersabroad in total R&D employeeswas similar.11 Within this globalpicture, significant differencesexist in the degree ofinternationa-lization of R&Dof the various countries oforigin (figure IV.3). Japaneseand Korean TNCs displayed

the lowest share of foreign R&D (15% and 2%,respectively; figure IV.3). North American TNCswere also below the average (24%). Conversely,European TNCs had high levels of R&Dinternationalization (41% on average).12 WithinWestern Europe, companies from France, theNetherlands, Switzerland and the UnitedKingdom had the most internationalized R&Dactivities on average.

Due to the small size of the sample in theUNCTAD survey, only tentative conclusions canbe drawn concerning industry-wide variations.The chemical and pharmaceutical industries werethe most internationalized in terms of R&D(figure IV.4). The relatively low level ofinternationalization of R&D in the electronicsand electrical industry (compared to chemicalsand pharmaceuticals) partly reflects the strongpresence of Japanese firms in that industry.Interestingly, the IT hardware industry’s levelof R&D internationalization was more

Figure IV.2. R&D expenditure by Japanese foreign affiliates abroad and its share in thetotal R&D spending of Japanese TNCs, 1986-2002

(Billions of dollars and per cent)

Source: UNCTAD, based on Japan, METI various issues.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

1986 1989 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 20020.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

R&D by Japanese affiliates abroad (left scale)

Share of foreign affiliates in total R&D (right scale)

$b

illio

n

%

124 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Box IV.1. Foreign R&D affiliates of German TNCs

The number of foreign affiliates establishedor acquired abroad by German TNCs that carryout R&D as a primary or secondary business issmall but growing, as is the outward FDI stockattributed to them (box table IV.1.1). Between 1995and 2003 this stock rose from $43 million to $891million, while employment by those affiliates grewfrom 2,000 to 11,000 during the same period. TheR&D spending of German TNCs abroad rose by130%, to $12 billion within the six-year periodfrom 1995 to 2001.

Of the German TNCs, Siemens alone spentmore than $6 billion on R&D in 2003 (table IV.1),accounting for about 7% of its sales (Sorg 2005).In 2004, of the 45,000 R&D employees of thecompany, 49% worked outside Germany. Thenumber of R&D personnel in developing countriesgrew from 800 in 1994 (2% of the company total)to 2,700 (6%) in 2004, located in seven countries:Brazil, China, India, Malaysia, Mexico and SouthAfrica (Sorg 2005).

A survey of 49 German TNCs accounting fortwo-thirds of Germany’s privately funded R&Dspending in that country, undertaken in 2000,concluded that internationalization of GermanR&D was the “phenomenon of the 1990s” (Ambos2005, p. 401). In the 1990s, German firms

established as many overseas R&D sites as in theprevious 50 years combined. In 2000, the TNCssurveyed already had 134 R&D laboratories abroad(idem, p. 397). More than half of the foreignlaboratories in pharmaceuticals, electronics andsemiconductors spent more than 20 million peryear, while those laboratories in the chemical andmachinery industries generally had budgets of lessthan 5 million.

Source: UNCTAD.

Box table IV.1.1. German R&D-relatedFDI abroad, 1995-2003

FDI stock in Number EmploymentR&D foreign of R&D of R&D foreign

affiliates abroad foreign affiliatesYear ($ million) affiliates (Thousand)

1995 43.2 20 21996 83.8 25 21997 133.8 31 31998 199.6 55 51999 467.7 59 62000 647.7 89 92001 630.0 105 102002 934.3 73a 112003 891.4 78 11

Source: UNCTAD, based on Deutsche Bundesbank,unpublished data.

a Break in the series, not directly comparable withprevious year.

Box IV.2. Explanatory note on the UNCTAD survey on R&D internationalization

Between November 2004 and March 2005,UNCTAD conducted a survey aimed at establishingthe current patterns of internationalization of R&Dby the largest private R&D spenders. Thepopulation basis for the survey was the R&DScoreboard published by the United KingdomDepartment of Trade and Industry (DTI). Of the700 top R&D spenders, UNCTAD contacted theleading 300 firms, which account for more than85% of all R&D by the top 700. In addition, allcompanies in the DTI Scoreboard that were fromdeveloping, South-East European and CISeconomies were invited to participate in the surveyeven if they fell outside the top 300. This broughtthe number of questionnaires sent out to 316.

The response rate was 22% of the sample or68 companies. The relatively low response ratewas due to the fact that many firms are unwillingto participate in such surveys as they consider

information concerning their R&D activities toostrategically sensitive to be disclosed.

Some potential shortcomings should be bornein mind. First, the reporting of R&D may notalways be done in the same way due to differentnotions of what R&D entails. Second, somerespondents may have omitted smaller R&Dactivities. Third, the United States isunderrepresented, although some of the largestUnited States R&D investors participated in thesurvey.

The industrial composition of the sample isbroadly similar to that of the DTI R&DScoreboard: IT hardware, automotive,pharmaceuticals, electronic and electrical andchemicals are 5 of the 6 main R&D investingindustries. The software and computer servicesindustry was underrepresented, mainly due to alow response rate by United States companies.

Source: UNCTAD.

125CHAPTER IV

pronounced in terms of R&D employees abroadthan in terms of expenditure – possibly indicatingthat R&D abroad is undertaken with a view toreducing labour costs. The opposite was the casefor the automotive industry – possibly suggestingthe greater importance in that industry of market-seeking motives for foreign R&D.

2. The growing role of foreignaffiliates in host-country R&D

The increasing internationalization of R&Dby TNCs is also reflected in the growing roleplayed by foreign affiliates in the R&D activitiesof many countries. In 1993, the R&D expenditureof foreign affiliates in host countries worldwide– the operations equivalent of inward FDI inR&D – amounted to about $29 billion (i.e. 10%of global business enterprise spending on R&D)(figure IV.5). Within a decade, by2002, that spending had more thandoubled to $67 billion or 16% ofglobal business R&D.13 This growthwas more than twice as fast as thatof global spending by enterprises onR&D, spending that grew by about49% over the same period.

The share of foreign affiliatesin host-country R&D varies bycountry. In 2003, it exceeded 50%in Ireland, Hungary and Singapore(figure IV.6), and 40% in five othercountries (Brazil , the CzechRepublic, Sweden, the UnitedKingdom and Australia in

descending order). Conversely, it remainedunder 10% in the Republic of Korea,Japan, India,14 Chile and Greece.

The share of foreign affiliates in thebusiness R&D of developed countries isclose to the world average and has beengrowing gradually, from 11% in 1996 to16% in 2002 (annex table A.IV.1). In thefour new EU members for which data wereavailable, the share of foreign affiliateswas already above the world average in1996 (17%) and increased further, to 41%,by 2002.15 In the developing countries forwhich data are available, the share offoreign affil iates rose faster than indeveloped countries (from 2% in 1996 toclose to 18% in 2002, annex table A.IV.1).

In fact, more than two-thirds of the 30countries for which data were availableexperienced a rise in the share of foreignaffiliates in business R&D after 1995, and thisrise was larger in developing countries (figureIV.6).16 In the new EU member countries, as wellas in Sweden and the United Kingdom, the shareof foreign affiliates also rose rapidly as localhigh-technology firms were taken over by foreignTNCs17 and new R&D facilities were located inthese economies. The high share of foreignaffiliates in the new EU member countries reflectsnot only the rising degree of penetration byforeign TNCs but also the low level of domesticR&D efforts (both total and business R&D; seealso chapter III).

The large number of majority-ownedforeign affiliates with R&D as their main activity(2,600 in 2004)18 reflects the spread of the R&Dactivities that TNCs are conducting outside their

Source: UNCTAD survey.

Figure IV.3. Degree of R&D internationalization byhome region or country in the UNCTAD survey,

2004-2005(Per cent)

Source: UNCTAD survey.

Figure IV.4. Degree of R&D internationalizationby industry, 2004-2005

(Per cent)

0

10

20

30

40

50

Electronics Pharmaceuticals Chemicals IT hardware Automotive Others

0

10

20

30

40

NorthAmerica

WesternEurope

Republic ofKorea

Japan Weightedaverage

126 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

home base (figure IV.7). Close to 70% of theseaffiliates are located in the Triad, but the mapalso indicates the presence of such activities invarious developing economies, especially inAsia.

3. Growing use of strategicalliances

Another indication of a rise in theinternationalization of R&D is the expansion ofcooperative arrangements, such as strategicalliances, in R&D (Dunning and Narula 2005,p. 130). Since the 1980s firms have increasinglysought to undertake R&D activities throughcollaborative efforts, as evidenced by informationfrom the MERIT/CATI database,19 whichcontains data on nearly 10,000 strategictechnology alliances of 3,500 parent companiesfor the period 1960-1998 (Hagedoorn 2002).Growth was steady in the early years of thisperiod and accelerated from the 1980s onwards.Although collaborative activity in R&D is nota new practice – economic units havecollaborated for decades – it has evolvedincontestably towards direct strategic uses(Narula 2003, p. 110). The relative share of non-equity (contractual) partnerships in the totalnumber of strategic alliances increasedconsiderably over the same period. Thegeography of strategic alliances was dominated

by intra-North Americanpartnerships, followed by EU-North America and intra-EUalliances (Hagedoorn 2002).

Data for a more recentperiod (1991-2001) show adoubling of new internationaltechnology alliances, from 339 to602, and a growing dominance ofnon-equity forms withinalliances.20 Indeed, while thenumber of non-equity alliancesincreased from 265 in 1991 to 545in 2001 (i.e. in more than 90% ofthe alliances) the number ofequity-based partnerships declinedfrom 74 to 57. United States firmscontinued to participate in a largemajority of strategic alliances,although their share in the total ofsuch alliances declined from 80%in 1991 to 73% in 2001. At the

same time the participation of non-Triad firmsincreased from 4% to 14%.

Between 1991 and 2001, the industrycomposition of alliances shifted strongly frominformation technologies (whose share droppedfrom 54% to 28%) to pharmaceuticals andbiotechnology (whose share increased from 11%to 58%). In the latter, there is a strong incentivefor TNCs to form strategic alliances with othercompanies in the industry as well as withacademic institutions, as no single company couldpossibly develop excellence in all the areas ofresearch that may be required to develop a newdrug. Moreover, there are strong pressures onpharmaceutical companies to reduce drugdevelopment costs and to share the risksinvolved.

C. The emergence ofdeveloping economies aslocations for TNCs’ R&D

Developed countries remain the main hostlocations of foreign R&D activities by TNCs,21

but there is a clear trend towards locating moreR&D activities to developing economies, South-East Europe and the CIS. This is confirmed byavailable national statistics as well as bycorporate surveys and case studies. The kind ofR&D being undertaken by TNCs in developing

Figure IV.5. R&D expenditure by foreign affiliates, basedon a sample of 30 economies, value and share in

business R&D, 1993-2002(Billions of dollars and per cent)

Source: UNCTAD, based on annex table A.IV.1.

0

10

20

30

40

50

60

70

80

1993 1994 1995 1996 1998 1999 2000 2001 20020

2

4

6

8

10

12

14

16

18

Value ($ billion, left scale) Share (%, right scale)

$bill

ion

%

127CHAPTER IV

countries is also changing. While i t hastraditionally involved mainly product or processadaptation to meet local market demands, recentdevelopments suggest that some developing,South-East European and CIS markets areemerging as key nodes in the global R&Dsystems of TNCs. At the same time, the extentto which developing countries participate in thesesystems varies considerably, and large parts ofthe developing world remain de-linked.

1. TNCs are expanding R&D todeveloping locations

Data on overseas R&D by TNCs fromthe United States show a decline in the share ofsome developed countries during the pastdecade.22 In 1994 developed countries accountedfor 92% of overseas R&D expenditures by UnitedStates TNCs (table IV.6), but by 2002 their sharehad dropped by 8 percentage points due to a

Source: UNCTAD’s calculations, based on national sources and data provided from the OECD AFA database.

Note: In Argentina, Chile, Israel, the Republic of Korea and Mexico, the R&D expenditure of United States-owned affiliateshas been used as a proxy for the R&D spending of all foreign affiliates. In India, the share of foreign affiliatesin total R&D spending has been used as a proxy for their share in business R&D spending.

Figure IV.6. Trends in R&D spending by foreign affiliates, selected economies, 1995-2003(Per cent)

Share of foreign affiliates in business R&D, selectedcountries, 2003 or latest year available

1.6

3.4

3.4

3.6

4.5

10.6

14.1

15.0

15.9

19.0

19.1

19.4

20.7

22.1

23.2

23.7

24.7

27.3

28.1

30.9

32.5

33.0

34.8

41.1

45.0

45.3

46.6

47.9

59.8

62.5

72.1

Korea, Rep. of (2002)

Japan (2001)

India (1999)

Chile (2002)

Greece (1999)

Turkey (2000)

United States (2002)

Finland (2002)

Average (2002)

Slovakia

Poland

France (2002)

Israel (2001)

Germany (2001)

Argentina (2002)

China

Netherlands (2001)

Spain

Thailand

Portugal (2001)

Mexico (2001)

Italy (2001)

Canada

Australia (1999)

United Kingdom

Sweden

Czech Republic

Brazil

Singapore

Hungary

Ireland

Change over 1995

4.4

0 25 50 75 -15 0 15 30 45

4.8

Netherlands (1997)

Average

Greece

United States

Finland

Korea, Rep. of

India

Japan

Singapore

France

Mexico

Canada

Ireland

China (1998)

Poland (1997)

Argentina (1996)

Germany

Australia

Portugal (1999)

Israel

Slovakia

United Kingdom

Sweden

Czech Republic

Hungary

0.8

0.6

1.0

1.3

1.8

2.0

2.2

2.3

3.2

5.1

5.4

5.7

8.8

8.9

9.1

13.0

14.0

15.1

15.4

26.0

25.8

40.7

Italy (2001)

Thailand (2003) ..

..

10.8

Chile

Turkey (1997)

Spain

Brazil (2000)

-4.2

-2.7

-0.1

-10.5

128 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Fig

ure

IV

.7.

Wo

rld

wid

e l

oc

ati

on

of

ma

jori

ty-o

wn

ed

fo

reig

n a

ffil

iate

s e

ng

ag

ed

in

R&

D,

20

04

So

urc

e:

UN

CT

AD

, b

ase

d o

n t

he

Wh

o O

wn

s W

ho

m d

ata

ba

se (

Du

n &

Bra

dst

ree

t).

No

te:

On

th

e b

asi

s o

f 2

,60

3 m

ajo

rity

-ow

ne

d f

ore

ign

aff

ilia

tes

en

ga

ge

d i

n R

&D

.

1-

34

-1

21

3-

27

28

-5

75

8-

81

129CHAPTER IV

strong decline in the shares of the EU (by 11percentage points) and Japan (by 3 percentagepoints). Not all developed economies have beenlosing ground, however. Rapid growth wasobserved in Canada and Israel and there wassome growth in Switzerland.

The shares lost by developed countrieswere picked up by developing economies, almostexclusively in Asia. China, Singapore, HongKong (China), Malaysia and the Republic ofKorea were among the main gainers of R&Dshares. As a result , the role of developingcountries as a whole increased, from 7.6% to 13.5%.

Table IV.6. R&D expenditure abroad by majority-owned foreign affiliates ofUnited States parent companies, by selected region/country, 1994–2002

(Millions of dollars)

Share of totalYear (%)

Region/economy 1994 1995 1996 1997 1998 1999 2000 2001 2002a 1994 2002

Total 11 877 12 582 14 039 14 593 14 664 18 144 20 457 19 702 21 151 100.0 100.0Developed economies 10 975 11 891 13 152 13 510 13 545 16 113 17 791 16 720 17 844 92.4 84.4

of which:Canada 836 1 068 1 563 1 823 1 750 1 681 2 332 2 131 2 345 7.0 11.1EUc 8 271 8 852 9 386 9 691 10 058 11 900 12 472 11 578 b 69.6 58.8Switzerland 191 242 190 230 223 231 286 392 405 1.6 1.9Israel 96 97 169 208 141 389 630 726 889 0.8 4.2Japan 1 130 1 286 1 333 1 089 962 1 523 1 630 1 507 1 433 9.5 6.8Australia 230 287 409 369 290 294 349 286 329 1.9 1.6New Zealand 7 9 16 18 15 9 8 10 6 0.1 -

Developing economies 902 691 886 1 082 1 119 2 031 2 637 2 982 2 855 7.6 13.5Developing Asia 408 283 318 393 336 1 400 1 949 2 391 2 113 3.4 10.0

of which:China 7 13 25 35 52 319 506 b 646 0.1 3.1Hong Kong, China 51 55 38 82 66 214 b 289 b 0.4 b

India 5 5 9 22 23 20 b b 80 - 0.4Indonesia 5 9 6 5 4 1 2 3 3 - -Korea, Republic of 17 29 34 41 29 101 143 157 167 0.1 0.8Malaysia 27 21 23 32 30 161 218 b b 0.2 b

Philippines 14 23 14 12 10 31 40 48 50 0.1 0.2Singapore 167 63 88 73 62 426 551 755 589 1.4 2.8Taiwan Province of China 110 61 75 84 55 122 143 139 70 0.9 0.3Thailand 3 5 5 5 4 7 13 18 22 - 0.1

Latin America and the Caribbean 477 389 546 663 748 613 663 562 b 4.0 3.2e

of which:Argentina 21 22 42 43 56 26 38 43 24 0.2 0.1Brazil 238 249 346 437 446 288 253 199 306 2.0 1.4Chile 2 15 6 7 6 4 11 8 6 - -Colombia 8 9 9 12 11 6 10 11 10 0.1 0.1Costa Rica 2 2 2 4 6 2 b 4 7 - -Mexico 183 58 121 126 191 238 303 248 284 1.5 1.3Venezuela 17 25 9 11 14 40 22 24 42 0.1 0.2

West Asia and North Africa 15 19 21 26 35 18 25 29 b 0.1 b

Sub-Saharan Africa 15 19 21 26 35 18 25 29 b 0.1 b

of which:South Africa 14 17 18 22 30 14 21 24 b 0.1 0.1

Economies in transitiond 5 18 36 48 79 54 83 38 68 - 0.3

Source: UNCTAD, adapted from Moris 2005a, based on data from United States Department of Commerce, Bureau ofEconomic Analysis, Survey of U.S. Direct Investment Abroad, www.bea.gov/bea.

a Estimates for 2002 are preliminary.b Withheld to avoid disclosing operations of individual companies. Note that due to undisclosed data, shares do not add

up to 100%.c EU covers 12 countries for 1994 and 15 countries thereafter.d Including new EU members.e Based on data for countries listed below.

Note: Data are for majority-owned foreign affiliates of United States parent companies. Majority-owned affiliates arethose in which the combined ownership of all United States parents is more than 50%.

130 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Expenditures on R&D by affil iates ofUnited States TNCs in developing economies areconcentrated mostly in five countries: China,Singapore, Brazil, Mexico and the Republic ofKorea in that order. They accounted for 70% ofthe total R&D expenditure of United States TNCsin developing countries in 2002. In contrast,Taiwan Province of China and India attractedrelatively small amounts of their R&D. India, amajor site for foreign R&D in recent years,accounted for only a small share of R&Dspending by United States TNCs until 2002according to official data, although more recentlythis share has risen.

In Latin America and the Caribbean, Braziland Mexico have accounted for around 80% ofR&D expenditures by United States TNCs in theregion since 1994. In absolute terms, their growthhas been modest compared to that in the majorAsian economies, and the relative importance ofLatin America and the Caribbean in the R&D ofUnited States TNCs has fallen. Venezuela is arelatively significant host for United States TNCs’R&D, much of it concentrated in the petroleumindustry. South Africa accounted for virtually allof the R&D by United States TNCs in Africa overthe same period.

The rising share of developing economiesis also noticeable in R&D employment by UnitedStates TNCs. Their share grew faster than thatof developed countries over the period 1994-1999although the EU still dominates. In particular,the share of R&D employment in developing Asiadoubled from 4.1% in 1994 to 8.1% in 1999(United States, NSF 2004). This figure is likelyto increase further judging from data on R&Dexpenditures, which shows the share ofdeveloping Asia rose from 7.7% to 10% between1999 and 2002 (table IV.6).

In 1999, the latest year for which R&Demployment data are available,23 the number ofscientists and engineers employed full time forcarrying out R&D for United States TNCsreached 770,300 (i.e. 3% of the total workforceof these firms in 1999). About 123,500 of them– or 16% – worked abroad in majority-ownedforeign affiliates of those TNCs (table IV.7).Close to 16% of these employees abroad wereemployed in developing countries.

The R&D intensity of employment stillremains low in developing economies comparedto the developed countries. Among the developingeconomies, only Singapore and the Republic ofKorea reached an R&D intensity similar to that

of developed countries (table IV.7). R&Dexpenditures per R&D employee in the foreignaffiliates of United States TNCs reached$146,915 in 1999, 26% up from 1994. Between1994 and 1999 R&D expenditures per R&Demployee increased at double digits in alldeveloping host regions except Latin America.

The selection of developing countries aslocations for R&D is gaining momentum inEurope as well. In the foreign R&D activities ofSwedish TNCs (table IV.5), the share ofdeveloping countries and economies in transition(including the new EU members) increasedrapidly, from 2.7% in 1995 to 7.2% in 2003. Asurvey of 1,554 German enterprises conductedin 2005 by the Deutsche Industrie- undHandelskammertag, the umbrella organization forGerman chambers of commerce, found that whileforeign R&D units were most frequently locatedin other EU States, about a third of respondentsconducted R&D in new EU member countries,South-East Europe or the CIS and 28% in Asia(DIHK 2005b).

In Japan, surveys carried out by the JapanBank for International Cooperation (JBIC)confirm the trend that Japanese companies arechanging their R&D strategies to become moreinternational (table IV.8). The overall number of“R&D bases”24 set up by the firms covered inthe surveys increased by 70%, to 310, between2000 and 2004, and that of “R&D bases” indeveloping countries more than tripled, to 134.The increase was most pronounced for China:its share of all R&D units rose from 7% to 22%between 2000 and 2004.

Official statistics do not necessarily capturethe rise of developing-country locations over thepast 2 to 3 years. Recent company surveys,however, indicate that the trend has gainedmomentum. In a 2004 survey, 70% of theresponding firms stated that they alreadyundertook R&D abroad, and that more R&D hadrecently been allocated to locations outside thedeveloped countries (EIU 2004a). Similarly,recent information on new greenfield andexpansion FDI projects involving R&D indicatesa surge of developing destinations and service-related R&D (OCO Consulting, LOCOmonitordatabase).25 Of the 1,773 FDI projects in R&Dworldwide for which information has beencollected for the period 2002–2004, the majority(1,095) were undertaken in developingeconomies, South-East Europe and the CIS.Developing Asia and Oceania alone accounted

131CHAPTER IV

for close to half of the world total (861projects). These data also suggest that themajority of new jobs created in greenfield FDIprojects related to R&D also went todeveloping countries, mostly to India andChina.

More than 90% of the above-mentionednew FDI projects involving R&D wereinitiated by TNCs from developed countries.The United States was the top source country,accounting for almost half of the world total,followed by the EU-15 and Japan. However,developing-country TNCs are also becomingmore active in this area (see also section E).Of the 160 projects carried out by developing-economy TNCs, 151 originated in Asia,mainly in India, the Republic of Korea,Taiwan Province of China, China andSingapore, in that order.

A matrix of the home and host countriesof R&D projects (table IV.9) reveals that the“traditional” pattern of developed-country

Table IV.7. R&D employment by majority-owned foreign affiliates ofUnited States TNCs by region/economy, 1999

(Thousand employees and per cent)

Total R&D R&D Total R&D R&D employment intensity employment intensity

Region/economy (Thousand) (%) Region/economy (Thousand) (%)

All economies 7 765.8 123.5 1.6 Thailand 102.3 0.1 0.1

Developed economies 4 378.9 96.2 2.2 Latin America andof which: the Caribbean 1 536.4 9.0 0.6Canada 1 004.2 7.9 0.8 of which:European Union 3 167.4 80.8 2.6 Argentina 93.8 0.3 0.3Japan 207.3 7.5 3.6 Brazil 348.8 5.4 1.5Israel 33.0 2.6 7.9 Chile 43.6 a b

Colombia 43.9 0.1 0.2Developing economies 2 702.7 19.2 0.7 Costa Rica 25.3 a b

Developing Asia 1 021.1 10.0 1.0 Mexico 780.8 2.7 0.3of which: Venezuela 63.2 0.4 0.6China 252.4 2 0.8Hong Kong (China) 93.8 1.2 1.3 West Asia and North Africa 19.2 - -India 62.2 0.2 0.3Indonesia 61.6 a b Sub-Saharan Africa 126 0.2 0.2Korea, Republic of 46.1 1.0 2.2 of which:Malaysia 119.1 a b South Africa 55 0.1 0.2Phil ippines 78.1 0.5 0.6Singapore 114.8 2.6 2.3 Unspecified 684.2 8.1 1.2Taiwan Province of China 71.3 0.9 1.3

Source: United States Bureau of Economic Analysis, Survey of U.S. Direct Investment Abroad, annual series, www.bea.gov/bea.

a Less than 50 employees.b Withheld to avoid disclosing operations of individual companies.

Note: R&D employment intensity is R&D employment as a percentage of total employment. EU comprises the 15 membersin 1999.

Table IV.8. R&D bases of Japanesemanufacturing companies, by host region,

2000-2004(Number of R&D bases)

Host region 2000 2001 2002 2003 2004

NIEs 16 15 30 21 25ASEAN-4 10 18 21 18 29China 13 19 28 29 67Other Asia 2 2 2 3 6North America 88 84 92 88 108Latin America 2 1 1 0 4EU-15 44 47 70 48 60Central and Eastern Europe 1 1 3 3 3South-East Asia and Oceania - 4 6 6 8Other countries 1 2 3 - -Total R&D bases 177 193 256 216 310

Source: UNCTAD, based on JBIC (various years), Survey Reporton Overseas Business Operations by JapaneseManufacturing Companies (Tokyo: JBIC).

Note: ASEAN-4 consists of Indonesia, Malaysia, thePhilippines and Thailand.

NIE (newly industrializing economies) consists of HongKong (China), the Republic of Korea, Singapore andTaiwan Province of China.

132 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

TNCs investing in other developed countries(well documented and analysed in the literature;von Zedtwitz 2005) accounted for less than one-third of the new R&D projects in 2002-2004.Meanwhile, the “modern” type of R&Dexpansion (developed-country TNCs investingin developing countries, South-East Europe andthe CIS) has become significant (almost three-fifths of the cases). Examples include Intel’sR&D laboratories in China and India (box IV.3),IBM’s R&D in India, Microsoft’s researchlaboratory in China and Fujitsu’s developmentcentre in Malaysia.

In turn, the other patterns of R&D-relatedFDI (“catch-up”, whereby TNCs fromdeveloping economies conduct R&D indeveloped countries with the aim of catchingup with developed-country TNCs; and“expansionary”, whereby a TNC from a

Box IV.3. Intel’s R&D network in developing countries

Intel has over 20,000 R&D employeeslocated in more than 30 countries. Some of thefacilities are owned by the parent firm whileothers are managed in collaboration withuniversities or through venture-capitalinvestments in technology-intensive companies.

Intel’s R&D investments in developing andSouth-East European and CIS economies,especially in China, India and the RussianFederation, are growing faster than elsewhere.That expansion is motivated by the availabilityof an educated and skilled workforce withspecific competencies in relevant areas. In thesecountries, Intel owns laboratories that conductkey research in a variety of fields; it has alsosigned a series of collaboration agreements withuniversities.

Intel China Research Centre (ICRC) inBeijing was established in 1998 as the company’sfirst research lab in the Asia-Oceania region.ICRC has conducted applied research in the areasof human computer interface, computerarchitecture, future workloads and compilers andruntime. In early 2005, it had a staff of 75researchers, most of whom hold a PhD or an MScfrom Chinese universities. Among the researchinnovations that have emerged from ICRC areOpen Research Compiler, developed jointly withthe Chinese Academy of Science; Audio Visual

Speech Recognition, a system using computervision to assist speech recognition; andMicrophone Array and audio signal processingtechnology. A second Chinese R&D laboratorywith over 150 employees is operating in Shanghaideveloping software for Intel.

The Intel India Design Centre in Bangaloreemploys more than 800 employees and deliverssoftware solutions to the company. Incomparison, the Nizhny Novgorod (RussianFederation) software development centre is hometo 340 specialists and engineers who aredeveloping software tools and applications forIntel.

Cooperation with universities abroad is animportant aspect of Intel’s global strategy. TheIntel Research Council, an internal group oftechnical experts, awards university researchgrants worldwide for projects in key areas. A finalvector of Intel’s global strategy is Intel Capital,Intel’s strategic investment programme. Itsmission is to make and manage financiallyattractive investments that support Intel’sstrategic objectives. Its overseas presence grewfrom less than 5% of the value of the deals in1998 to about 40% in 2003. Of these overseasinvestments, about half were in companies basedin Asia (including Japan) and the rest in Europe,Israel and Latin America.

Source: UNCTAD, based on information provided by Intel in March 2005.

Table IV.9. Greenfield FDI projects in R&D, 2002-2004(Number of projects)

Host economy South-East

EuropeHome economy Developed Developing and CIS Total

Developed “Traditional” “Modern”612 953 40 1 605

Developing “Catch-up” “Expansionary”63 97 2 162

South-East “Catch-up” “Expansionary”

Europe and CIS 3 3 - 6

Total 678 1 053 42 1 773

Source: UNCTAD’s calculations, based on the LOCOmonitordatabase (c lass i f icat ion draws on von Zedtwi tz2005).

133CHAPTER IV

developing country invests in R&D in anotherdeveloping country to support either second-generation technology transfers or other localbusiness activities) together accounted for lessthan one-tenth of the total.26 Samsung’s(Republic of Korea) laboratories in Europe, andAcer’s (Taiwan Province of China) laboratoriesin the United States are examples of the “catch-up” type of R&D-related FDI, while Acer’s R&Dlaboratory in China and Huawei’s R&D centrein Bangalore illustrate the “expansionary” type(see also section E).

UNCTAD’s survey of the largest R&Dspenders among TNCs (referred to above)

confirms the growing importance of developing-country locations. Although the majority of theR&D conducted abroad is in other developedcountries (the United States and the UnitedKingdom being the two top destinations), anumber of developing countries were alsomentioned by the 68 respondents. The currentlocation of their foreign R&D efforts indeveloping countries was reported as being,among others, China (3rd global destination),India (6th), Singapore (9th) and Brazil (11th)(figure IV.8).27 Also, notably, a large numberof other developing-country R&D locations (14economies) were indicated by at least one of the

Figure IV.8. Current foreign locations of R&D in the UNCTAD survey, 2004(Per cent)

Source: UNCTAD survey.

Note: Countries mentioned by two respondents include: the Czech Republic, Hungary, Indonesia, Malaysia, Mexicoand Portugal. Economies mentioned by one respondent include: Argentina, Bulgaria, Denmark, Estonia, Greece,Hong Kong (China), Morocco, the Philippines, Saudi Arabia, Slovakia, South Africa, Turkey and Viet Nam.

4.4

4.4

4.4

4.4

5.9

5.9

5.9

7.4

7.4

7.4

7.4

7.4

8.8

8.8

11.8

13.2

13.2

14.7

17.6

19.1

19.1

25.0

29.4

35.3

35.3

47.1

58.8

0 10 20 30 40 50 60

Thailand

Republic of Korea

Israel

Austria

Taiwan Province of China

Poland

Ireland

The Netherlands

Russian Federation

Norway

Finland

Australia

Switzerland

Sweden

Belgium

Spain

Brazil

Italy

Singapore

Germany

Canada

India

Japan

France

China

United Kingdom

United States

Developed countries Developing economies

South-East Europe and CIS

134 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

respondents. In South-East Europe and the CIS,the Russian Federation and Bulgaria were theonly target economies mentioned.28

The companies responding to the UNCTADsurvey also answered questions related tointernational non-equity collaboration in the areaof R&D. The most frequently mentioned locationfor such arrangements was again the UnitedStates, followed by the United Kingdom. Chinawas in third place ahead of Germany, France andJapan. A roughly equal share of the respondingcompanies had R&D collaboration withcounterparts in the Russian Federation and inIndia. Other developing and South-East Europeanand CIS economies mentioned included:Argentina, Brazil, Mexico, Morocco, Singapore,Taiwan Province of China and Tunisia. A recentsurvey of 104 TNCs (EIU 2004a) has also foundthat Europe and Asia are the most commonlocations of R&D (indicated by 34% and 30%of the respondents), followed by North America(17%).29

2. Foreign affiliates in patenting bydeveloping economies

The role of TNCs in the R&D activities ofa country can also be gauged from measuresrelated to the output of R&D activities. Theanalysis in this section draws on information fromthe United States Patent and Trademark Office(USPTO).30 As noted above (chapter III), thenumber of patent applications to the USPTOfrom developing economies and countries inSouth-East Europe and the CIS has risendramatically in recent years (albeit from a lowbase), primarily due to increased researchactivities in Asia and Oceania (annex tableA.III.3). A detailed analysis suggests that foreigncompanies play an important role in thepatentable outputs of these countries, with someimportant exceptions.

In order to assess the role of TNCs it isnecessary to distinguish between the “inventor”and the “assignee” of a patent. According to thepatent law of the United States, the applicant ina patent application must always be the inventor.Therefore, patents are granted to an inventor ora group of inventors, but not to institutions.However, many patents or patent applications areassigned (i.e. transferred) to those other than the

inventor(s), usually to institutions. The assigneethen becomes the legal owner of the patent.31

The number of USPTO patents granted toinventors resident in the economies included intable IV.10 increased more than fourfold between1993 and 2003.32 The table shows that for theperiod of 2001-2003, many patents granted toinventors resident in these economies wereassigned to entities (typically TNCs) based inother countries. Patents assigned to foreignersmay be the output of R&D outsourced by foreignTNCs to scientists in the listed economies or theoutput of R&D conducted by inventors employedby foreign affiliates in these economies. Thusthe share of patents assigned to foreigners in thetotal number of patents granted to residents ina country can be seen as an indicator of the roleof foreign TNCs in the innovation activities ofthe economies (e.g. Guellec and vanPottelsberghe de la Potterie 2001, 2004b).

By this measure, foreign companies playeda very small role in the patents granted by theUSPTO to inventors in the Republic of Korea andTaiwan Province of China during the period2001-2003; only 4% of them were assigned toforeigners (table IV.10). However, in most othereconomies in the table — including Brazil, China,India and the Russian Federation — a large shareof the patents were assigned to foreign entities— ranging from 25% in Saudi Arabia to 86% inKenya.33

While TNCs thus appear to own a largeshare of USPTO patents granted to inventors indeveloping economies and South-East Europe andthe CIS, the number of patents that are ownedby foreign affiliates located in these economiesis generally small. USPTO data show that mostpatents assigned during the period 2001-2003 toentities in the economies listed in table IV.11were owned by domestic enterprises or, in someeconomies, by public institutions, but only rarelyby foreign affiliates. Only in Bulgaria and Brazildid foreign affiliates account for more than 20%of all patents assigned.34 In India and Cuba,public research institutions accounted for thelargest shares (68% and 84% respectively) ofthose countries’ totals.35 Public researchinstitutions in Singapore, the Russian Federationand Ukraine also receive a significant proportionof the patents assigned by the USPTO.

135CHAPTER IV

Table IV.10. United States Patent and Trademark Office patents granted toresidents of selected developing economies and countries in

South-East Europe and CIS, 2001-2003(Number of patents and per cent)

Patents granted Patents assigned to The shareto residents foreign institutions of (b) in (a)

Region/economy (a) (b) (%)

AfricaSouth Africa 428 126 29Egypt 32 21 66Kenya 21 18 86

Asia and OceaniaTaiwan Province of China 20 414 889 4Republic of Korea 12 195 482 4China 1 543 979 63Singapore 1 485 669 45Hong Kong (China) 2 069 692 33India 1 022 409 40Malaysia 281 207 74Turkey 101 71 70Thailand 208 116 56Phil ippines 108 92 85Saudi Arabia 64 16 25Indonesia 108 69 64

Latin America and the CaribbeanBrazil 524 220 42Mexico 409 215 53Argentina 202 70 35Bahamas 47 36 77Bermuda 22 12 55Cuba 21 - -Chile 54 27 50

South-East Europe and CISRussian Federation 956 654 68Ukraine 131 98 75Bulgaria 34 16 47

Source: UNCTAD, based on information from the USPTO patent database.

Note: The patent count in tables in this section includes all types of patents, i.e. utility, design as well as plantpatents. Column (a) lists the number of patents where at least one inventor is from a developing economyor a country in South-East Europe or the CIS. Column (b) lists the number of patents in (a) that are assignedto foreigners (usually institutions).

In sum, with the important exceptions ofthe Republic of Korea and Taiwan Province ofChina, foreign companies play a significant rolein the innovation activities of those developingeconomies and countries in South-East Europeand the CIS that have expanded their patentingactivities in the United States during the pastdecade. A large share of all patents granted toinventors in these economies is assigned toowners abroad, notably TNCs. However, sincefew foreign affiliates are owners of patents inthese countries it would appear that TNCs tendto centralize the ownership of patents atheadquarters.

D. Features of R&Dundertaken in developing,South-East European and

CIS markets

1. Industry composition of R&D byTNCs in developing countries

The industry composition of R&D byforeign affiliates differs by region and economy.For instance, three-quarters of R&D by UnitedStates affiliates located in Asia (excluding Japan)

136 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

were in computers and electronic productsindustries in 2002 (figure IV.9, see also annextable A.IV.2). In India, over three-quarters ofaffiliates’ R&D expenditures ($61 million) werein non-manufacturing industries in 2002,compared to only about 20% in 1999, probablyreflecting a focus on software development inthat country. On the other hand, chemicals andtransportation equipment combined accountedfor over half of all R&D by foreign affiliates ofUnited States TNCs located in both Brazil (figureIV.9) and Mexico (Moris 2005a). These patternsare different from that of the aggregate for allhost countries, in which transportation equipmentwas the top industry, followed by computers andelectronic products, with chemicals andpharmaceuticals in third place (figure IV.9, annextable A.IV.2).

Overseas R&D by German TNCs showssimilar patterns. In the electronics andsemiconductor industries, both industries witha high percentage of production abroad, Asia wasan above-average location in 2000, while R&Dby the German chemical and pharmaceuticalTNCs was heavily skewed in favour of NorthAmerican locations. The remaining industriesappeared to focus on Europe (Ambos 2005, p. 400).

The industry composition of recentgreenfield R&D projects in 2002-2004, for whichinformation was available, also shows a highshare of information technologies (IT) andsoftware in new projects in developing countries(39%), which may indicate a gradual shift ofR&D towards services and in particular IT. 36

IT is gaining importance within R&D because,in more and more TNCs, the share of software

Table IV.11. United States Patent and Trademark Office patents assigned toinstitutions in selected economies by the type of assignee, 2001-2003

(Number of patents)

Region/economy Domestic firms Foreign affiliates Public institutions Total

Africa South Africa 153 7 7 167 Egypt 3 - 4 7Asia and Oceania Taiwan Province of China 11 621 118 947 12 686 Republic of Korea 9 829 562 761 11 152 Hong Kong (China) 1 251 89 87 1 427 Singapore 610 41 144 795 India 177 2 379 558 China 408 18 49 475 Malaysia 43 5 1 49 Saudi Arabia 35 - 4 39 Thailand 36 - 2 38 Indonesia 27 - 4 31 Turkey 24 - - 24Latin America and the Caribbean Brazil 191 54 9 254 Bermuda 140 30 - 170 Mexico 101 6 12 119 Bahamas 54 - - 54 Argentina 27 5 1 33 Cuba 3 - 16 19 Chile 15 - 2 17 Panama 14 1 - 15 Uruguay 3 - - 3South-East Europe and the CIS Russian Federation 126 - 37 163 Ukraine 8 - 3 11 Bulgaria 7 2 - 9

Source: UNCTAD, based on information from the USPTO patent database.

Note: When patents are assigned to an individual, they are counted as “domestic firms”. The classification of assigneesis according to the Who Owns Whom database and other sources. The Who Owns Whom database gives informationon the “Ultimate Parent”. Foreign affiliates are those firms whose ultimate parent is in a different country.

137CHAPTER IV

development is taking up an increasing part ofthe R&D budget.37

2. Types of R&D

R&D carried out by TNCs in developingcountries can be categorized in various ways (boxIV.4). One relates to the types of R&D undertakenby TNCs’ affiliates in host countries, reflectingthe different technological functions assigned toforeign affiliates. The foreign affiliates mayundertake:

• Adaptive R&D;

• Innovative R&D linked to production forlocal or regional markets;

• Global innovative R&D for new productsor processes, or for basic research; and

• Technology-monitoring R&D.

There can be many varieties of adaptiveR&D, ranging from basic production support tothe upgrading of imported technologies. Not allTNC production abroad gives rise to formal R&D(as a distinct operation separate from routineengineering or initial plant design). Muchdepends on the size and growth of the localfacility, the differences between local conditionsand those for which the technology was designed,and the availability of local technical skills. Theextent to which adaptive R&D evolves intoinnovative R&D depends even more on the

Figure IV.9. Industry composition of R&D by majority-owned foreign affiliatesof United States TNCs, 2002

(Per cent)

a. All host countries b. Asia (excluding Japan)

c. Brazil

Source: UNCTAD, adapted from Moris 2005a, based on data from United States Bureau of Economic Analysis, Surveyof United States Direct Investment Abroad, www.bea.gov/bea.

Notes: Data are preliminary estimates. PST refers to professional, scientific, and technical services. Data for transportationequipment for affiliates in Asia (excluding Japan) and Brazil are for 2001. Data for PST services for affiliatesin Japan and Asia (excluding Japan) are for 2001.

Chemicals4%

PST2%

Machinery1%

Otherindustries

13%

Transportationequipment

2%

Computers andelectronic products

78%

Chemicals23%

PST6%

Other industries15%

Machinery3%

Transportationequipment

28%

Computers andelectronic products

25%

Transportationequipment

30%

Otherindustries

28%

Machinery9%

PST1%

Chemicals22%

Computers andelectronic products

10%

138 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

availability of suitable technical skills along withsupplier R&D capabilities (where this feeds intothe R&D done by an affiliate) and institutionalsupport (for testing or other specialized work).Innovative R&D for local or regional marketscan evolve into global innovative R&D when thehost economy is able to meet even more stringentskill and institutional needs. However, thisevolution is not the only way for TNCs to launchR&D in developing countries. Some developingcountries are attracting “pure” TNC R&D, notrelated to production (either for the domesticmarket or export-oriented). Technologymonitoring units are another example of R&D.The main roles of technology monitoring unitsare to keep abreast of technological developmentsin foreign markets and to learn from leadinginnovators and consumers there (Roberts 2001).

It is difficult to quantify R&D accordingto the types identified above (the data are toolimited). However, one study, undertaken in 1999on 209 R&D performing firms from the Triad(Roberts 2001), found that the establishment ofworldwide centres of excellence for a particulartechnology or discipline was the primary functionof overseas R&D; it varied between a high of47% for Western European TNCs and a low of25% for Japanese firms (Roberts 2001, p. 30).Adaptation for local markets was a close secondin Japan and the United States, and a distantsecond in Western Europe. Regional technicalsupport activities and basic and/or appliedresearch in other countries held third and fourthplaces respectively. In developing countries,while most R&D has traditionally been of anadaptive nature, recent trends suggest that moresophisticated activities are also expanding. A2004 survey found that 22% of the respondents

Overseas R&D by TNCs is a multifacetedactivity. For instance, it can be analysed in termsof the nature of the activity undertaken or interms of the motives for undertaking R&Dabroad. According to these two criteria, thetypologies overlap considerably and distinctionsare not always easy to draw; moreover, over timethe distinctions can become increasingly blurredas R&D units evolve.

The following provides illustrations of thetwo typologies based on the nature of the R&Dactivity and on TNC motivations. Despite the factthat these two taxonomies are drawn from a largebody of literature that has focused almost entirelyon R&D by foreign affiliates in developedcountries, they can also be applied, in most cases,to the developing countries that are emerging inthe global R&D landscape.

Based on the nature of technologicalactivity in foreign affiliates: This typologydivides foreign affiliates doing R&D into fourbroad types (sometimes with sub-categories) onthe basis of the kind of R&D undertaken (Pearce1989, Nobel and Birkinshaw 1998, von Zedtwitz2005).

• Local adapters: These are “market-seeking”R&D units for absorbing and adaptingtechnologies, essentially to support productand process engineering departments in

Box IV.4. Taxonomy of R&D by foreign affiliates

making existing technologies work moreefficiently in new environments. They are alsovariously called “support units” and“technology transfer units”.

• Locally integrated laboratories: Also called“indigenous technology units” and“international independent laboratories”, theseare more advanced than local adapters and arecapable of independent innovation aimedprimarily at local (and perhaps regional)markets. The units remain linked to localproduction and are usually a natural evolutionfrom adaptive R&D.

• International technology creater: This is themost advanced type of innovative activity byforeign affiliates and places them on an equallevel with core innovating centres in the homecountries and in other developed countries.Also known as “internationally interdependentlaboratories” or “global technology units”,these facilities can do both research anddevelopment, and their output is typicallyaimed at global exploitation by the parentcompany. They may evolve out of locallyintegrated laboratories, and so retain tight linkswith production in the host economy, or theymay be set up independently of localproduction to tap local innovation clusters andskills.

• Technology scanning or monitoring unit: Thisis normally a “business intelligence” functionundertaken by an “asset-seeking” R&D unit

/...

139CHAPTER IV

were already conducting some applied researchin overseas developing markets (EIU 2004a).

The following analysis looks at the salientfeatures of TNC-controlled R&D in developingcountries, beginning with the region where themagnitude of the phenomenon is the highest. Itstresses that Asia has taken the lead amongdeveloping countries not only in terms of thenumber of projects and jobs created but also interms of the types of R&D undertaken, includinginnovative R&D for local and global markets.Indeed, some R&D activities in some Asiandeveloping countries in particular are now takingon a more sophisticated role within the globalR&D networks of TNCs. The analysis ofdeveloping Asia is followed by those of LatinAmerica and the Caribbean, and Africarespectively. An analysis of the economies intransition of South-East Europe and the CIS, andof the former economies in transition of the newEU members38 is added at the end of the section

because R&D-related FDI in those countries hasgrown fast, and in some respects the features ofthese economies with regard to skills and wageadvantages are similar to the ones offered byvarious developing countries at comparableincome levels.

a. Asia and Oceania: dynamic trends

The rise of developing Asia and Oceaniahas been the most dramatic development in theglobal landscape of R&D. Some economies inthe region have been able to capture a broadrange of R&D functions from TNCs, includinginnovative R&D and basic research. For example,electronics firms in Taiwan Province of Chinaare attracting the outsourcing of complete productdesign (Engardio and Einhorn 2005). While mostdeveloping host economies do not offer theadvanced design and production capabilities ofTaiwan Province of China, the kind of work they

Box IV.4. Taxonomy of R&D by foreign affiliates (concluded)

Source: UNCTAD, based on the literature cited.

a Based on Archibugi and Iammarino 2002, Le Bas and Sierra 2002, Edler et al. 2002, Gassmann and von Zedtwitz1999, Gerybadze and Reger 1999, Kuemmerle 1997, Medcof 1997, Nobel and Birkinshaw 1998, Pearce 1989 and1999, Reddy 2000, Ronstadt 1977, Voelker and Stead 1999, von Zedtwitz 2005, and von Zedtwitz and Gassmann2002.

under the headings above, but in the absenceof a separate R&D facility, scanning can alsobe done by another department of the TNC.

Based on TNC motivation: This typologygroups affiliate R&D activities by thetechnological objectives of the parent company(Le Bas and Sierra 2002). Four types emerge:

• Technology-seeking FDI in R&D: The TNCseeks to offset areas of weakness in the home-country innovation system by setting up R&Dfacilities or acquiring local innovators incountries with complementary strengths. Anumber of R&D-related M&As in the UnitedStates in biotechnology, electronics andpharmaceuticals are of this type. Developing-country firms with technological ambitionsalso undertake such R&D investments oracquisitions.

• Home-base (or asset-) exploiting FDI in R&D:This essentially corresponds to the adaptivecategory in the typology above, where themain functions of the R&D are to absorb andadapt technologies transferred by the parent

company so that the TNC can effectivelyexploit its technology assets.

• Home-base (or asset-) augmenting FDI inR&D: This is where TNCs undertake R&D intechnologies in which they are strong at homeand where the host country also has strengths.This has been called “strategic asset-seekingR&D” by TNCs. It aims not only to accessforeign technological assets but also to capturethe externalities created by host-countrytechnology clusters (Dunning and Narula1995). The distinction between this andtechnology-seeking FDI is not very strong,especially in the case of developed countries,as it hinges on an evaluation of the relativestrengths of home- and host-countryinnovation systems.

There are other ways to classify foreignR&D. It is possible to categorize it, for example,by the organizational strategy of TNCs and bytheir R&D management practices. However, forthe purposes of analysing the impact ondeveloping countries, the relevance of thesetaxonomies is more limited.

140 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

conduct can also be quite sophisticated. Contractmanufacturers like Flextronics (Singapore), forinstance, set up R&D bases in some countriessuch as India and China in 2004 in order toprovide state-of-the-art product developmentservices (Engardio and Einhorn 2005).Meanwhile, pharmaceutical companies areseeking to cut the cost of bringing new drugs tothe market by collaborating with biotech firmsin India. Thus the dividing line between the kindof R&D that is suited for expansion in developingcountries and that which is best kept at home –or in developed as opposed to developingcountries – has become blurred.

China and India have been the mainbeneficiaries of this trend. Of the 885 R&D-oriented greenfield FDI projects announced inthe region in 2002-2004, three-fourths (723) wereconcentrated in these two large economies. InChina, some 700 foreign-affiliate R&D centreshad been established by the end of 2004 (boxIV.5). In India, more than 100 TNCs haveestablished R&D facilities.39 Microsoft launchedits sixth global research centre in Bangalore inearly 2005 after opening one in Beijing in 1998.Other such Microsoft R&D centres outside theUnited States are located in the United Kingdomand the Republic of Korea. In the case ofMotorola (box IV.6), 6 of i ts 19 main R&Dcentres are located in developing countries: fivein Asia (China, India, the Republic of Korea,Malaysia and Singapore) and one in Brazil. Thenumber of large pharmaceutical TNCs that havea research presence in India in particular isgrowing fast. Astra-Zeneca inaugurated a largefacility for research on tuberculosis in 2003 andsubsequently expanded it to includepharmaceutical development. Pfizer startedclinical research in India in 1995 and added abiometrics unit in 1998 along with a formulationdevelopment group in 2004. In addition, as ofJune 2005, Eli Lilly,40 Sanofi-Aventis, Novartisand GlaxoSmithKline had clinical research unitsand Novartis and GlaxoSmithKline hadbiometrics centres in India (Mukherjee 2005).

FDI in R&D in Asia and Oceania flows notonly to very large countries like China and Indiabut also to other, smaller, economies in theregion. Data on greenfield projects in 2002-2004show that at least 16 other Asian economiesreceived R&D-oriented FDI during the periodof observation. Within this group, East and South-East Asian economies, especially Hong Kong(China), Malaysia, the Philippines, the Republic

of Korea, Singapore, Taiwan Province of China,and Thailand, frequently appear on the radarscreen of TNCs.

Those economies that traditionally havehad a considerable presence of foreign affiliatesin local innovation (e.g. Singapore) also have alarge share of business R&D (figure IV.6). Overthe past decade more than 100 TNCs, includingRolls Royce, Motorola, Philips, GE, Delphi, EliLilly, Hewlett-Packard, Matsushita, Sony, 3M andDaimlerChrysler, have located R&D laboratoriesin Singapore (Toh 2005, pp. 11-12). Morerecently, pharmaceutical TNCs such as Aventis,Merck, GlaxoSmithKline and Wyeth have set upR&D facilities there (annex table A.IV.3). InThailand, the size of FDI in R&D was small overthe period of 1995-1999 averaging $4.1 millionper annum, although it accounted for an importantpart of business R&D (Intarkumnerd andSittivijan 2005, pp. 4-5). By the period 2000-2004 both business R&D and R&D by foreignaffiliates had increased substantially (the latterto $34 million per year).41 The industry focusof R&D-related FDI in Thailand, too, shiftedbetween the two periods, from metals and non-metal-working industry to machinery,transportation equipment (led by Japanese TNCssuch as Toyota; box IV.7) and electricalappliances (especially hard disk drives).

The share of foreign affiliates in R&Dexpenditure in the Republic of Korea is still low(figure IV.6). It is only recently that TNCs havestarted investing in R&D in that country, in partas a response to more active government policiesthat welcome and encourage such FDI (chapterVII). As of December 2004 a total of 140 foreign-affiliate research institutes had been opened, 61of which were established after 2000 (Republicof Korea, Ministry of Commerce, Industry andEnergy 2005). Most foreign research institutesare now using their facilities to develop newproducts and processes, and in some recent casesthey are performing innovative R&D activitiesfor global innovation and production (box IV.8).

Some of the development work conductedin Asia is world-class, such as chip design in thesemiconductor industry. This industry was oneof the earliest to globalize production indeveloping countries, and has been among thefirst to move advanced design to selecteddeveloping economies including the Republic ofKorea, Taiwan Province of China, and, morerecently, to China, India, Singapore and Malaysia(annex to chapter V). Asia is not only undertaking

141CHAPTER IV

Box IV.5. The boom in R&D-related FDI inflows in China

R&D-related FDI inflows in China havesurged in recent years. The accumulated R&Dinvestment of TNCs in China had reachedapproximately $4 billion by June 2004 (estimatedby the Ministry of Commerce), while the numberof foreign-affiliate R&D centres, registeredaccording to the eligibility criteria in place sincethe year 2000, reached 700 by the end of 2004.Although the first TNC R&D centre dates backto 1993, most of the known projects are recent(established after China’s accession to the WTOin December 2001).

Most foreign-affiliate R&D centres arewholly-owned by their parent companies,although some of them are joint ventures (suchas the one established by Lenovo and Intel in2003). The majority of these centres still focuson adaptive innovations for the Chinese market.However, some do innovative R&D that is closelyintegrated with TNCs’ global innovationnetworks, and thereby target global markets.

R&D-related FDI inflows have beenconcentrated in technology-intensive industriessuch as ICT, automotive and chemicals

(according to the data of the Beijing MunicipalBureau of Statistics). The ICT industry, inparticular, has witnessed a boom in R&Dinvestment by TNCs (box table IV.5.1). Motorola(see also box IV.6), one of the largest foreigninvestors in China, had set up 15 local and globalR&D centres in China by the end of 2004, withseveral others under construction. In addition toMotorola, major R&D investments have beenmade by Microsoft, Nokia, GE (box table IV.5.1)as well as IBM, Siemens, Nortel, Dupont, GeneralMotors, Honda, Hitachi and Toshiba, to mentiononly a few (Sigurdson 2005a, p. 2).

Foreign-affiliate R&D centres in China areconcentrated in large cities with strongtechnological bases and skilled human resources,particularly in Beijing and Shanghai (box figureIV.5.1). At the end of 2004, 189 centres werelocated in Beijing alone, with almost 60% of themin the ICT industry. Many of them followed onthe footsteps of IBM, which established itswholly-owned R&D centre there in 1995. Withinthe capital, the Haidian District (whereZhongguancun Science Park is located) is home

Box table IV.5.1. Selected foreign affiliate R&D centres in the electronicsand ICT industries of China, as of 2004

Numberof R&Dcentres

Company in China Location Features

General 1 Shanghai • China Technology Centre, opened in Shanghai in 2003, is the thirdElectric global R&D centre of the company after those in the United States and India.

• Invested $640 million and centralized its previous by existing R&D units in China.• 500 R&D engineers (planned to increase to 1,200 in 2005).

Microsoft 5 Beijing • Invested $130 million.Shanghai • Microsoft Research Asia (MRA), established in 1998, is the company’s basic

research facility in the Asia and Oceania region and the fifth largest researchcentre in the world.

• MRA employs over 170 researchers.

Motorola 15 Beijing • The first TNC R&D centre in China (set up in 1990).Shanghai • Total of 1,300 R&D engineers.Tianjin • Invested $300 million in R&D in China until 2001.Suzhou • Motorola China Research Institute (MCRDI) was established in 1999.Nanjing • Will invest $500 million in a new R&D centre in Beijing.Chengdu

Nokia 5 Beijing • Nokia China R&D Centre, established in 1998, employs 300 R&D engineers.Shanghai • Hangzhou R&D Centre, established in 1998, employs 180 R&D engineersHangzhou (will increase to 400).

Source: UNCTAD, based on company press information./...

142 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

more chip-related R&D; the levels of complexityof this R&D are also on the rise. A few firms fromthe Republic of Korea and Taiwan Province ofChina, and to a lesser extent from China andIndia, now develop cutting-edge technology.

In sum, the range of R&D activitiesundertaken by or for TNCs in Asia, mainly ininformation technology and pharmaceuticals, issurprisingly wide:

“Today, the likes of Dell, Motorola, andPhilips are buying complete designs ofsome digital devices from Asiandevelopers, tweaking them to their ownspecifications, and slapping on their ownbrand names. It’s not just cell phones.

Asian contract manufacturers andindependent design houses have becomeforces in nearly every tech device, fromlaptops and high-definition TVs to MP3music players and digital cameras…While the electronics sector is furthestdown this road, the search for offshorehelp with innovation is spreading tonearly every corner of the economy...[Boeing] is working with India's HCLTechnologies to co-develop software foreverything from the navigation systemsand landing gear to the cockpit controlsfor its upcoming 7E7 Dreamliner jet.Pharmaceutical giants such asGlaxoSmithKline and Eli Lilly are

to 40 universities and 130 research institutes andis the capital city’s R&D hub.

In Shanghai, over 140 TNC R&D centreshave been established, of which 91 are in thePudong New District. In addition, the Guangdongand Jiangsu provinces had accounted for 28%and 19% of the accumulated FDI inflows ofChina until 2003 (estimated by the Ministry ofCommerce) and are home to more than 100

Box figure IV.5.1. Location of foreign-affiliate R&D centres in China, 2004(Numbers)

Source: UNCTAD.

foreign-affiliate R&D centres. Some otherregional economic centres in other coastalprovinces such as Hangzhou in Zhejing province,Qingdao in Shandong province and Dalian inLiaoning province have also attracted importantforeign-affiliate R&D centres (box figure IV.5.1).Finally, TNCs have also set up some R&D centresin a limited number of inland cities such as Xi’anand Chengdu.

Box IV.5. The boom in R&D-related FDI inflows in China (concluded)

>10>50>100

Chengdu

Xian

Guangzhou Dongguan

Shenzhen

Hangzhou

ShanghaiSuzhou

Nanjing

Tianjin

Beijing

Dalian

Qingdao

143CHAPTER IV

teaming up with Asian biotech researchcompanies in a bid to cut the average$500 million cost of bringing a new drugto market” (Engardio and Einhorn 2005,pp. 52-53).

b. Latin America and the Caribbean:limited R&D but with potential

TNCs have so far located only limitedR&D in Latin America and the Caribbean. FDIthere is rarely in R&D-intensive activities, andwhen it is, i t mainly remains confined toadaptation of technology or products for localmarkets, called “tropicalization”42 in the LatinAmerican context (Cimoli 2001). Foreignaffiliates play a relatively large role in businessenterprise R&D in Brazil and Mexico, moderatein Argentina and low in Chile (figure IV.6).

Employment data for the majority-ownedforeign affiliates of United States TNCs showthat, while the share of Latin America and theCaribbean in 1999 was about 20% of theworldwide total employment in such foreignaffil iates, the share of the region in R&Demployment of foreign affiliates was only 7%(table IV.7).43 Most of this is in two countries:Brazil and Mexico (table IV.6).

In Brazil , adaptive R&D dominates,although some change has been noted in thestrategies of some TNCs since the late 1990s.They include Brazilian affiliates in their strategyof globalization of R&D, upgrading theirtechnological activities and giving them newR&D responsibilities (Costa 2005). This hasoccurred mainly in the auto parts and automotiveindustries (box IV.9) as well as in the electronicsindustry. In these industries some TNCs havereversed previous downsizing of local R&Dactivities,44 following their loss of market shareeither locally or regionally (Costa 2005, Queirozet al. 2003, Furtado et al. 2003, Consoni andQuadros 2003, Galina 2003). Thepharmaceuticals industry displays a differentpattern: few pharmaceutical TNCs do R&D inBrazil , despite the availabili ty of localcapabilities and public laboratories (Costa 2005,Furtado et al. 2003).

In Mexico foreign affiliates are activemainly in assembly work, relying on their parentcompanies for most R&D activities. Innovationin export-oriented TNCs appears to be confinedto organizational and marketing activities ratherthan product and process technology (AbdelMusik 2004). A study of Mexico’s Baja Californiaelectronics and automotive manufacturing clusterconcluded that more than a quarter of the plantssurveyed were engaged in R&D, one-fifth did

Box IV.6. Motorola’s R&D network

Telecommunications equipment manu-facturer Motorola (United States) is the world’s19th largest R&D spending firm (table IV.1). Asof end 2004 it operated major R&D centres (thosewith over 100 R&D staff) in 19 countriesworldwide: two in North America, six in the EU-15, one in Poland, three in other developedcountries, six in developing countries, includingBrazil, China, India, the Republic of Korea,Malaysia and Singapore, as well as one in theRussian Federation (box figure IV.6.1).

The first overseas R&D centres wereopened in 1950 in Canada and the UnitedKingdom, followed by various other Europeanlocations in 1960. Motorola began conductingR&D in developing countries fairly early, withoperations in Singapore and Malaysia alreadyin place in 1970. Most R&D centres concentrate

on product development rather than on research.The latter is conducted in only five countries,three of them developed (the United States, theUnited Kingdom and Israel) and two of themdeveloping: India and China.

The R&D activities of Motorola in Chinaillustrate well the interaction between a TNC witha global network of R&D centres and a wide-ranging host-country R&D structure includingbusiness and government R&D units (Sigurdson2005a). Motorola has also entered into a numberof collaborative research agreements with localuniversities, which also explains the broadpresence of its R&D centres in the country.Motorola originally focused on manufacturingin China. In the early 2000s, the companyincreased its R&D activities in China to be closerto the local market and to be more cost-efficient.

/...

144 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Mo

toro

lah

ea

dq

ua

rte

rs(U

nite

dS

tate

s)

Esta

blis

he

d1

92

9

R&

De

mp

loye

es

12

60

0:

R&

Din

ten

sity

42

%:

Ire

lan

dE

sta

blis

he

d1

98

0R

&D

em

plo

ye

es

37

0:

R&

Din

ten

sity

74

%:

Ca

na

da

Esta

blis

he

d1

95

0R

&D

em

plo

ye

es

23

0:

R&

Din

ten

sity

33

%:

Bra

zil

Esta

blis

he

d1

99

0R

&D

em

plo

ye

es

16

0:

R&

Din

ten

sity

7%

:

Isra

el

Esta

blis

he

d1

96

0R

&D

em

plo

ye

es

90

0:

R&

Din

ten

sity

28

%:

Po

lan

dE

sta

blis

he

d1

99

0R

&D

em

plo

ye

es

30

0:

R&

Din

ten

sity

75

%:

De

nm

ark

Esta

blis

he

d1

99

0R

&D

em

plo

ye

es

13

0:

R&

Din

ten

sity

52

%:

Fra

nce

Esta

blis

he

d1

96

0R

&D

em

plo

ye

es

30

0:

R&

Din

ten

sity

60

%:

Spa

inE

sta

blis

he

d1

98

0R

&D

em

plo

ye

es

20

:1

R&

Din

ten

sity

40

%:

Un

ite

dK

ing

do

mE

sta

blis

he

d1

95

0R

&D

em

plo

ye

es

70

0:

R&

Din

ten

sity

19

%:

Ind

iaE

sta

blis

he

d1

99

0R

&D

em

plo

yee

s1

35

0:

R&

Din

ten

sity

64

%:

Ch

ina

Est

ab

lish

ed

19

90

R&

De

mp

loye

es

13

00

:R

&D

inte

nsi

ty1

2%

:

Ru

ssia

nF

ed

era

tion

Est

ab

lish

ed

19

90

R&

De

mp

loye

es:

24

0R

&D

inte

nsi

ty6

0%

:

Au

stra

liaE

sta

blis

he

d1

97

0R

&D

em

plo

yee

s2

30

:R

&D

inte

nsi

ty3

9%

:

Sin

ga

po

reE

sta

blis

he

d1

97

0R

&D

em

plo

yee

s4

30

:R

&D

inte

nsi

ty2

0%

:

Japa

nE

sta

blis

he

d1

96

0R

&D

em

plo

yee

s1

30

:R

&D

inte

nsi

ty7

%:

Re

pu

blic

of

Ko

rea

Est

ab

lish

ed

19

90

R&

De

mp

loye

es

45

0:

R&

Din

ten

sity

45

%:

Ma

lays

iaE

sta

blis

he

d1

97

0R

&D

em

plo

yee

s5

50

:R

&D

inte

nsi

ty7

%:

Ge

rma

ny

Esta

blis

he

d1

96

0R

&D

em

plo

ye

es

20

0:

R&

Din

ten

sity

7%

:

Sour

ce: U

NC

TAD

, bas

ed o

n in

form

atio

n an

d da

ta p

rovi

ded

by M

otor

ola.

Box IV.6. Motorola’s R&D network (concluded)B

ox f

igu

re I

V.6

.1.

Mot

orol

a’s

R&

D n

etw

ork

, 20

04

145CHAPTER IV

product design, more than one-tenth haddeveloped a patent and more than one-third hadISO 9002 Certification (Gerber and Carillo 2002).An example of R&D for global markets is foundin the automotive industry of Mexico. Forinstance, Delphi Automotive (United States) has

a technical centre in Ciudad Juárez employing3,000 people, half of whom are engineersdesigning auto parts for global use. Examplesof R&D for the regional market can be found inthe country’s banking industry (BBVA of Spain).

Box IV.7. Thailand in Toyota’s global R&D network

Source: UNCTAD, based on company interview conducted on 4 May 2005.

a The other overseas R&D centres are in the United States, Europe and Australia (see box figure IV.7.1).

Box figure IV.7.1. Toyota’s global R&D network, 2005

Toyota Motor Corporation founded itsfourth overseas R&D centre – and the first onein a developing country (box figure IV.7.1) – inThailand in August 2003.a The “Toyota TechnicalCenter Asia Pacific (Thailand)” was officiallyopened in May 2005. Toyota has invested 1.1billion baht ($27 million) into this centre so far.During the two-year preparation for opening,almost all locally recruited engineers andscientists were sent to Japan for a training periodof 6 to 12 months.

When it first opened, the “Toyota TechnicalCenter Asia Pacific (Thailand)” employed 275persons (including 32 Japanese), of which 250were engineers and technicians (2% of Toyota’sglobal R&D staff). The centre has both a regionalmandate for Asia (excluding China) and a globalone to carry out R&D for the parent corporation.

It is in charge of projects in basic research,technology development, research on marketconditions and design, along with testing andevaluation.

Thailand was chosen as a location forToyota’s Asian R&D centre for various reasons.The existence of a manufacturing and salesaffiliate there was an important consideration,although there is no equity or administrative linkbetween the two units. Other reasons includegood local infrastructure, political stability,favourable geographical location, a skilled labourforce and favourable government policies(including incentives). In the area of policies,outstanding issues include the eventual exemptionfrom customs duties of materials (such as motorvehicles) imported for testing, and the provisionof full licences for test-driving.

Thailand

275

Toyota Technical Center Asia PacificEstablished: 2003R&D employees:

Australia

100

Toyota Technical Center Asia PacificEstablished: 2003R&D employees:

United StatesToyota Techinical Center USA, Inc.Established: 1977R&D employees: 700

Japan

Head Office Technical CenterEstablished: 1954R&D employees: 12,000

Toyota Motor Corporation

BelgiuToyota Motor Engineering &

mManufacturing

Europe Technical CenterEstablished: 1987R&D employees: 300

146 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Box IV.8. Innovative R&D by foreign affiliates in the Republic of Korea: Microsoft,Siemens and Philips

Source: UNCTAD, based on information provided by the Ministry of Information and Communication of the Republicof Korea, Investment Opportunities in Broadband IT Korea, 2004, www.mic.go.kr, www.investkorea.org.

The Republic of Korea has recently attractedinnovative R&D centres from a variety of majorTNCs.

In March 2005, United States software makerMicrosoft opened its Mobile Innovation Lab at theheadquarters of its Korean affiliate, MicrosoftKorea, in Seoul to develop technology for wirelessdevices. The company is committed to creatingsoftware programmes for next-generation mobiledevices. Microsoft has plans to invest up to $30million in this R&D centre over the next threeyears, and employ 30 researchers.

Siemens, the German electronics andinformation communications corporation,announced in June 2004 that it would invest $119million in the Republic of Korea over five years.The investment was intended to establish a forwardbase of information communications and network

equipment in the country and develop productsfor the world market. Siemens had invested $45million by early June 2004 and had bought a 38.7%share of Dasan Networks (Republic of Korea),making it that company’s largest shareholder.Siemens is developing Dasan Networks into anR&D centre and distributes communications andnetwork equipment to world markets, includingthose in Europe, the United States and Asia.

In 1999, Royal Philips Electronics of theNetherlands acquired a 50% share of LGElectronics’ LCD (liquid crystal display) divisionfor $1.6 billion. The new joint-venture companyplans to invest a total of $10 billion and build theLG-Philips Plant on a 408-acre site in Paju,Gyeonggi Province by 2006. Along with theproduction lines, LG-Philips plans to set up anR&D centre to develop technology for nextgeneration TVs.

Box IV.9. General Motors in Brazil: from tropicalization to global innovative R&D

Source: UNCTAD, based on company interview.

General Motors has an important R&D centreat its Sao Caetano plant in southern Brazil.Established in the 1960s as a small unit to adapt(“tropicalize”) GM autos and parts to Brazilianconditions, it became a large laboratory by the endof the 1980s, focusing on a variety of projectsdirected at the host-country market. By the late1990s, GM Brazil had accumulated technicalexpertise in designing local versions of GM modelssuch as the Opel Corsa sedan, the Corsa pickupand the Astra sedan. The continuous building upof the product development engineering team andlocal infrastructure permitted GM Brazil to gofurther through engagement in the Blue Macawproject, origin of its Celta model.

After 1996 the Brazilian automotive regimebecame increasingly open to parts imports whilestill protecting the assemblers with fiscaladvantages and import tariffs. GM responded tothose policies by streamlining its manufacturingprocess, whereby suppliers co-located theirproduction of sub-assemblies for GM cars at theassembly plant in Rio Grande do Sul, therebyreducing GM’s inventory holdings.

Concomitantly, GM also changed themandate of its Brazilian R&D centre from local

to international: GM Brazil was assignedresponsibility for designing a new vehicle forglobal sales (the Meriva minivan). Instead offollowing the usual strategy of car makers, whichconsisted of designing a partial derivative of analready existing model, GM Brazil was givenresponsibility for a more complex project called“global derivative” consisting of designing a newvehicle for global rather than local application(Consoni 2004).

These additions to GM Brazil’s portfolio ofactivities have meant expanded product andprocess development for both local and globalapplications. About 1,000 technical and hourlyemployees are now engaged in productdevelopment in Brazil, and about 500 in processengineering R&D work. The value of this activityis not large when considering GM’s global R&Dactivities, although it has increased theresponsibility and autonomy of the Brazilian R&Dteam significantly. Today, GM in Brazil competeswith other GM affiliates in the United States,Europe and Asia for the right to design and buildnew vehicles and to carry out other core activitiesfor the global company.

147CHAPTER IV

c. Africa: generally marginal in R&Dby TNCs

In Africa the R&D component of FDI isoverall very small. With a few exceptions suchas Kenya, Morocco and especially South Africa,R&D by TNCs is virtually absent. This is partlybecause of weak domestic R&D capabilities(chapter III) and, in many cases, the lack ofinstitutional mechanisms that provide incentivesfor investors to devote resources to R&D(Oyelaran-Oyeyinka 2004a). This does notnecessarily mean that innovation per se is absentfrom Africa but rather that such innovation isundertaken outside R&D laboratories.

In the South African auto industry – inwhich all assemblers are wholly or partly ownedby their respective parent companies from Japan,Europe or the United States – firms spend 2.5%of their total sales on R&D (UNCTAD 2003b,p. 16). This is generally carried out incollaboration with the South African Bureau ofStandards (SABS) and the engineering facultiesof some of the leading universities.45

Collaboration between SABS and the automotiveforeign affiliates has led to the establishment of

the EuroType Test Centre, a state-of-the-artlaboratory that has made South Africa one of theworld leaders in testing engines and catalyticconverters. In the South African aerospaceindustry, BAE Systems of the United Kingdomcontracted Aerosud South Africa as an exclusivesupplier of leading-edge wing components forthe Airbus A320 jetliners.46 In health care,Innovex, a South African affiliate of Quintiles(United States), offers contractual services forclinical testing, health economics, marketing andsales.

North Africa provides some recentexamples of FDI in R&D. Morocco has attractedR&D centres, especially in software andelectronics: SQLI (France) set up an R&Dplatform in the country in 2003, EolaneElectronics Manufacturing Services (France)opened an R&D centre in the country in 2004next to its manufacturing and distribution unit,and STMicroelectronics has had a chip designCentre in Casablanca since 2000 (box IV.10). Inthe automotive industry, Pininfarina/Matra (Italy)opened a 60-person R&D centre in Morocco in2004, together with a test circuit. Other NorthAfrican countries are less targeted by R&D,

Box IV.10. STMicroelectronics’ design and software centre in Rabat

Source: STMicroelectronics.

a The presence of the seventh largest semiconductor producer in the world (49,000 employees worldwide) in Moroccodates back to 1952. Operations in Morocco were expanded in 1979 to carry out subsystem development, and againin 1997 to create a state-of-the-art “back-end” assembly and test plant.

In 2000 STMicroelectronics (registered inthe Netherlands and headquartered inSwitzerland) located parts of its design activitiesin Morocco.a The Rabat Design Centre is partof a global network of 16 advanced R&D centresand 39 design centres in the Czech Republic,France, Germany, India, Italy, Morocco, Tunisia,the United Kingdom and the United States.Within this network, the primary mission of theRabat Design Centre is to develop advancedsystem-on-chip products for digital TVs, DVDplayers and flat-screen displays, along withdigital still and video cameras. The Rabat Centrecurrently employs 170 people, scheduled to growto 700 by 2009.

In addition, the firm has established atraining centre, the first of its kind in the country,to train teachers and students from engineeringschools and to provide them with the necessary

syllabus to enable them to make a valuablecontribution to the innovation needs of thesemiconductor industry. In 2001 it launched itsfirst cooperative activity with the MohammedV-Agdal University in Rabat, which includedscholarships, exchange programme andsponsorship of microelectronics courses. It alsoestablished a design centre at the MohammadiaSchool of Engineers, within the Mohammed V-Agdal University.

STMicroelectronics chose Morocco as thelocation for the design centre for several reasons.These included a favourable educational andcommunications infrastructure, the availabilityof a rich pool of engineering talent, the proximityof Europe and competitive costs. Rabat waschosen specifically for its schools and universitiesthat train engineers specialized in the computer/IT domain.

148 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

though in Algeria the Jordanian pharmaceuticalfirm Hikma opened an R&D centre at its localfactory in 2003, while Novell (United States)entered into a strategic alliance with Net-Skills,a local software firm (Marseille Innovation andANIMA 2005).

The rest of the R&D-related FDI in Africamirrors the resource-based orientation of thecontinent, focusing on petroleum exploration andexploitation and agriculture. In the petroleumindustry, a number of TNCs47 conducted someR&D in Algeria, Egypt, Morocco, the LibyanArab Jamahiriya and Tunisia in 2004.48 Inagriculture, the United States-based Agro-Management Group developed pyrethrum flowersin Uganda, for the international market.49 Kenyais also home to selected agricultural R&Dprojects carried out by and for TNCs and theiraffiliates (box IV.11).

d. A comparison with economies intransition

In the former transition economies that arenow new EU member countries, foreign affiliateshave become important R&D players since the

mid-1990s (figure IV.6, box IV.12). This hashappened partly through the early acquisition offlagship firms carrying out R&D such as ŠkodaAuto in the Czech Republic in 1991 andTungsram in Hungary in 1990. In those instancesthe new owners decided to transform the localR&D laboratories of the acquired affiliates intospecialized corporate R&D centres. The majorityof the R&D privatized laboratories acquired byforeign investors in the acceding new EU membercountries managed to adapt to the newenvirornment of increased competition fromimported technologies. An UNCTAD survey ofprivatization through FDI carried out in 199950

found that in the two years following theprivatization deals, R&D expenditure increasedby 13.6% in the sample firms (Kalotay and Hunya2000, p. 53).51

In the new EU member countries, R&D byforeign affiliates has also expanded throughgreenfield projects. Of the 108 R&D projectsinitiated in the new EU, South-East Europe andthe CIS taken together in 2002-2004, 66 wereregistered in the new EU member countries, withthe Czech Republic, Hungary and Poland takingthe lead. Information on key R&D affiliates in

Box IV.11. R&D by TNCs in agriculture: Kenya

Source: UNCTAD.

a CGIAR, ASTI Database (www.asti.cgiar.org/expenditures.cfm), and Beintema, N. and Phillip G. Pardey (2001).“Slow magic: agricultural R&D a century after Mendel”, ASTI Initiative, IFPPI, mimeo.

b The share of private firms in Kenyan agricultural R&D may be higher, because the original sample was based oninformation available on only three firms.

c The non-profit Donald Danforth Plant Science Center is a partnership organization of the Monsanto Company andvarious United States-based academic research institutions.

Kenya is not a major player in global R&D.In agriculture, which generates a large share ofits export earnings, R&D expenditures representedonly slightly more than 1% of the total fordeveloping countries in 2000.a Moreover, theprivate sector accounted for only 3% of totalagricultural R&D expenditure in Kenya that year.b

There are however several agricultural/horticultural or related firms, including TNCs,conducting some form of R&D in Kenya. Theknown cases of R&D by TNCs in Kenya havefollowed different strategies. Some have decidedto conduct in-house R&D. Examples include DeRuiter’s, Regina Seeds, Fourteen Flowers (theNetherlands), Del Monte (United States) and

Kordes & Söhne (Germany). Other TNCs suchas East African Breweries (United Kingdom),Monsanto (United States) and Syngenta(Switzerland), have opted for collaborativearrangements with local and foreign partners. TheKenyan Agricultural Research Institute (KARI)carries out research on barley on behalf of theEast Africa Breweries, and works for Syngentato develop insect-resistant maize for Africa.Monsanto’s involvement in Kenyan R&D is moreindirect, as its project initiated originally in directcollaboration with KARI and the InternationalService for the Acquisition of Agri-techApplications has been transferred to its UnitedStates non-profit partner, Donald Danforth PlantScience Center.c

149CHAPTER IV

Box IV.12. R&D by foreign affiliates in the Czech Republic

Source: UNCTAD, based on Srholec 2005.

As in most new EU member countries, theCzech R&D system underwent a majortransformation during the transition fromcentrally planned to market economy. In thisprocess, foreign affiliates have become importantplayers in the national R&D system, accountingfor nearly 47% of business expenditure on R&Din 2003 (figure IV.6) and for 30% of businessR&D employment in 2002.

R&D activity of foreign affiliates istypically related to the presence of manufacturingplants in the country, although this trend mightbe changing as a consequence of severalgreenfield projects that have been attracted intostrategic services recently. In pure R&D activities(stand-alone R&D laboratories, ISIC 73) foreignaffiliates play a limited role, accounting only for6.3% of employment in 2002. The R&D servicesindustry received only 0.1% of total FDI inflow

until the end of 2002 (more than 80% of whichcame from Germany).

In manufacturing, most of the businessR&D is concentrated in medium-technologyindustries such as automobiles, which accountedfor 68.2% of manufacturing-related R&D in2002. Automotive production has a long traditionin the Czech Republic with Škoda Auto, takenover by Volkswagen in the early 1990s, as themain showcase. Foreign affiliates in theautomotive industry are committed to the long-term upgrading of their overseas R&D, as theirpatenting record and their cooperation agreementswith universities and R&D laboratories indicate.This contrasts with the case of electronics,another significant FDI recipient in the CzechRepublic. Activities in that industry are drivenprimarily by local cost advantages, with limitedinvestment in overseas R&D. In fact, in thisindustry the R&D intensity of foreign affiliatesis substantially lower than that of domestic firms.

these three countries in 2004 suggests adominance of EU-15 investors, although theUnited States, Japan and some developingeconomies (India, the Republic of Korea) are alsoamong the home countries. Most of theseaffiliates are linked to manufacturing sites, andhence are mainly in the automotive andelectronics industries (including spare partsproducers and telecom equipment manufacturers).Various affiliates on the list have “innovative”R&D mandates for regional or global markets.

In South-East Europe as well , foreignaffiliates have gained a prominent role in R&D.In Romania, for example, Automobile Dacia(affiliate of French Renault) and Petrom (nowaffiliate of Austrian OMV) were the two largestR&D spenders in the country in 2003. InBulgaria, Bulgarian Telecom (65% owned byViva Ventures, United States) was the secondlargest R&D spender in the same year.

In the CIS, and the Russian Federation inparticular, the entry of TNCs in R&D hasremained at a low level and in most cases islimited to alliances or other contractualarrangements. Boeing (United States), Pratt &Whitney (United States), Airbus (France/

Germany/Spain/United Kingdom) and Dassault(France) have been actively cooperating with theR&D institutes and laboratories of the Russianaerospace industry and the Russian Academy ofSciences since the early 1990s (Ivanova 2004,p. 151). For example, one of the leading RussianR&D centres, the Zhukovski CentralAerohydrodynamics Institute, has contributed toR&D on the Hermes air space system and theDASA Hypersonic vehicle, on commercialtransporter A3XX and on Boeing’s 757 and 777aircraft (Ivanova 2004, p. 152). Outside theRussian Federation, Antonov, the leadingUkrainian aviation firm, signed in 2002 contractsto modernize Chinese aircraft in cooperation withShanxi Aircraft Industry based on earlier Antonovdesigns (Yegorov 2004, p. 159).

R&D on a basis other than contractual tiesis less frequent in the CIS. As a whole, there wereonly 30 greenfield R&D projects reported in theLOCOmonitor database for the CIS in 2002-2004,of which the Russian Federation alone accountedfor 27. Compared to the science and technologybase in the Russian Federation that number issmall but could grow rapidly in the near future.One of the largest of the foreign-affiliate R&D

150 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

centres of the Russian Federation was opened byIntel in 2000 (box IV.3). In another case, theEuropean Aeronautic Defence and SpaceCompany, the parent firm of Airbus (EADS,headquartered in the Netherlands), opened a 30-employee engineering centre in Moscow in 2003together with the Russian Federation’s KaskolGroup, an aerospace and defence conglomeratethat controls the MiG producer in NizhnyNovgorod.52

E. Developing-countryTNCs are also expanding

R&D abroad

Another new trend whereby developingcountries are connecting to global knowledgenetworks is the emergence and fast growth offoreign R&D activities by TNCs from developingeconomies. As the phenomenon is very recent,the top R&D spenders of developing countriesare still relatively small (section A and tableIV.1). However, some – almost all from Asia –have moved up in ranking on the list of thelargest R&D-spending firms since the late 1990s.Moreover, the expansion of their R&D appearedto be on a relatively large scale in 2002-2004(table IV.9).

Some developing-country TNCs such asthe IT company, Ingenuity Solutions (Malaysia),have targeted the knowledge base of developedcountries such as the United States, wheninvesting in R&D abroad. Similarly, Bionova ofMexico acquired DNA Plant Technology of theUnited States in 1996 and, as a more recentexample, the Singaporean firm Cordlife, acquiredCytomatrix (United States) in 2004.

There are also examples of South-SouthFDI in R&D. A number of firms from Malaysia,the Republic of Korea, Singapore, and Thailandhave set up R&D activities in India relatedspecifically to software development (Reddy2000, pp. 97-103). In 2003 Samsung Electronics(Republic of Korea) announced plans to openR&D centres in China, India and the RussianFederation; LG (Republic of Korea) has expandedits R&D activities into India; and BogasariInternational (Indonesia, food processing) choseSingapore, in part due to the country’s favourableR&D incentive schemes for foreign investors.

The following section examines the casesof Chinese, Indian and Korean TNCs, which areamong the most active developing-country firmsestablishing R&D activities abroad.

A recent study of large Chinese TNCsfound that they operated 77 R&D units at the endof 2004, including a surprisingly high 37 unitsabroad (von Zedtwitz 2005). Of these foreignR&D units, 26 are located in developed countries,predominantly in the United States (11) andEurope (11), mostly serving as listening postsor in product design roles.53 The remaining 11units, located in developing countries, aretypically small in size (e.g. just a handful ofpeople in a small technology outpost in Pakistanand the Islamic Republic of Iran).54 Two ChineseTNCs, Huawei55 and Haier,56 are illustrative ofthe trend of R&D units being located mainly indeveloped countries. Other Chinese companiesfrom the electronics industry, such as ZTE andUTStarcom, have also established R&D centresin India aimed essentially at offshore softwaredevelopment.

Indian TNCs are also globalizing theirR&D, focusing mainly on serving their customersin specific regional markets. The leading softwarefirms have all invested abroad, mostly indeveloped countries. For example, Infosys,Wipro, Birlasoft (part of Aditya Birla Group) andHCL Technologies have operations in the UnitedStates. They are also moving into selecteddeveloping-country locations where they havemajor customers, especially China, South-EastEurope and the CIS.57 Some Indian softwareR&D affiliates are located in other developingregions (e.g. Tata has invested in Uruguay) aswell as in new EU member countries (Hungary).Indian firms in other industries such aspharmaceuticals and chemicals are also investingin R&D abroad (box IV.13).58

TNCs from the Republic of Korea startedestablishing R&D affiliates abroad only in the1990s. In 2005, a survey carried out by the KoreaIndustrial Technology Association identified 60foreign R&D centres owned by Korean firms.The United States was the main target of suchinvestment (17 R&D centres) followed by China(15), Japan (7), the Russian Federation (5) andGermany (5). The majority of R&D centres inChina (12 of the 15) have been operating since2000. Some of the Korean firms investing abroadin R&D also figure prominently on the list of the

151CHAPTER IV

700 largest R&D-spending companies of theworld (table IV.1): these include SamsungElectronic (33rd in world ranking and the largestR&D spender in the developing world),59

Hyundai Motor (95th) and LG Electronics(110th).

F. Prospects

In sum, TNCs are dominant players inglobal R&D, and their R&D is being increasinglyinternationalized, including in developingcountries. The trend towards the greaterinvolvement of developing countries in the R&Dactivities of TNCs is l ikely to accelerate,although, to date, the majority of developingcountries remain excluded from this phenomenon.Whether R&D activities will spread to a growingnumber of developing countries remains an openquestion, and will largely depend on the policiespursued by these countries (chapter VII).

In the UNCTAD survey of the world’slargest R&D-spending TNCs, as many as 69%of the responding firms stated that their shareof foreign R&D is set to increase; only 2%indicated the opposite, while the remaining 29%expected the level of internationalization toremain unchanged (figure IV.10).60 Themomentum appears to be particularly strongamong companies in Japan and the Republic ofKorea, which have so far been less aggressivein terms of R&D internationalization. Nine outof ten Japanese companies in the sample andabout 80% of the Korean firms planned toincrease their foreign R&D, while 61% ofEuropean firms indicated similar intentions. Thisfinding is corroborated by information providedby the Government of Japan: 95% of Japaneseaffiliates abroad plan either to expand their R&Dactivities (17%) or to maintain them (78%) at thesame level as before, regardless of their location(Japan, METI 2004).

Source: UNCTAD, based on information provided by Alexandria Carbon Black in March 2005.a Carbon black is a key raw material input mainly for the manufacture of tyres and other rubber products.

Box IV.13. Alexandria Carbon Black: Indian FDI in R&D in Egypt

The Aditya Birla Group is one of India’s topTNCs. It has 72,000 employees worldwide andmanufacturing units in Australia, Canada, China,Egypt, Indonesia, Malaysia, the Philippines andThailand. In 1994 the company established theAlexandria Carbon Black (ACB) factory in Egypt.Owing in part to continuous product and processinnovation, the ACB plant has grown to becomeone of the world’s largest single carbon blackplants.a It employs 300 persons in Egypt, 25 ofwhom work in its R&D centre.

The ACB plant has a sophisticated R&Dcentre with the latest analytical equipment. Thecentre has, among other things, developed a keygrade of carbon black for providing criticalproperties to the final product. Other innovationsinclude manufacturing process improvements toimprove quality and increase efficiency, utilizationof information technology to computerizeprocesses, innovations in the area of packagingand environment management, as well as adoptingtotal quality management and total productivemaintenance.

The R&D centre provides various forms oftechnical support to domestic enterprises. Localcompanies can use the centre’s analytical

equipment, and it also provides training toemployees of local companies. The trainingincludes best practices in quality management,how to use sophisticated analytical equipment,statistical quality control tools and totalproductive maintenance. In order to upgrade theskills of the employees of its suppliers, thecompany also offers technical and managerialsupport. Some development work (e.g. relatedto improvements in raw material and packaging)has also been done in partnership with suppliers.Six major partnerships with suppliers have beenforged in the areas of packaging, raw materialsand manufacturing of sophisticated equipment.As a founding member of the RegionalGeographical Committee of the Petro-ChemicalArea, ACB also helps the adoption of bestpractices by local companies.

The R&D centre is closely collaboratingwith the parent company’s Fundamental ResearchInstitute in India. The Aditya Birla Groupprovides significant support to ACB in a numberof areas, and members of ACB’s technical teamfrequently travel to other carbon black units ofthe group to exchange experiences and learn fromthe others.

152 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Source: UNCTAD survey.

Figure IV.10. Prospects of TNCs locatingR&D abroad, 2005-2009

(Per cent of responses)

A further shift towards some specificdeveloping, South-East European and CISmarkets is also expected (figure IV.11). In theabove-mentioned UNCTAD survey, for instance,China was the R&D destination mentioned mostoften, followed by the United States. In thirdplace was India, another significant newcomerlocation for R&D. The Russian Federation wasalso among the top 10 target locations for R&Dexpansion. Other developing economies that werementioned as candidates for further R&D by atleast 2% of the companies were the Republic of

Korea, Singapore, Taiwan Province of China andThailand. However, only a few respondentsindicated possible plans for expanding R&D inLatin America and Africa. Another survey (EIU2004a) reached similar conclusions, with the top10 destinations for R&D expansion includingthree developing economies: China for R&Dexpansion (in first position), India (3rd) andBrazil (6th); and three others in the followingranks: Hong Kong, China (13th), Mexico andSingapore (both14th) (EIU 2004a).

* * *This chapter has examined the dominant

role of TNCs in global R&D along with the riseof some developing countries as locations chosenfor TNC-led R&D. It has also analysed the shiftsin the industry composition and the mandates ofthe R&D carried out abroad, especially indeveloping countries. In particular it has shownthat R&D in some developing countriesincreasingly involves “innovative” activities. Ithas found that TNCs from developing countriesare also investing in R&D abroad. The nextchapter examines the drivers and determinantsof the internationalization of R&D by TNCs, withthe aim of determining the implications fordevelopment (chapter VI) and deriving somepolicy lessons (chapter VII).

Increase69%

Decrease2%

Stay the same29%

153CHAPTER IV

Figure IV.11. Most attractive prospective R&D locations in the UNCTAD survey,2005-2009

(Per cent of responses)

Source: UNCTAD survey.

Developed countries

Developing economies

South-East Europe and CIS

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

1.5

2.9

2.9

2.9

2.9

2.9

4.4

4.4

4.4

4.4

5.9

8.8

10.3

13.2

14.7

29.4

41.2

61.8

0 10 20 30 40 50 60

Viet Nam

Turkey

Tunisia

Sweden

Spain

South Africa

Romania

Poland

Norway

Morocco

Mexico

Israel

Ireland

Czech Republic

Brazil

Australia

Thailand

Republic of Korea

Malaysia

Italy

Belgium

Taiwan Province of China

Singapore

Canada

The Netherlands

Germany

France

Russian Federation

United Kingdom

Japan

India

United States

China

154 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Notes1 Some pharmaceutical firms with no identified foreign

affiliates pursue their internationalization throughstrategic alliances with TNCs. For example, CellGenesys is in a technology alliance with Novartis(Switzerland). The latter is also a 5% shareholder ofthe former. Human Genome Sciences (United States)has strategic alliances with GlaxoSmithKline (UnitedStates), Takeda (Japan), Schering-Plough (UnitedStates), Sanofi-Synthelabo (France), Merck (Germany)and the Pharmaceutical Division of the Kirin Brewery(China). In another case, ICOS (United States) is a 50%owner of the Lilly ICOS joint venture formed with EliLilly (United States) for the global distribution of thedrug Cialis.

2 In 2003, the R&D expenditure of the 700 largestspenders rose further, by more than 5%, to $327 billion.

3 In Sweden, the top 20 TNCs accounted for up to three-quarters of the total R&D expenditure in the late 1980s(Håkanson and Nobel 1993a). In Germany, only 49firms accounted for two-thirds of the privately fundedR&D spending in the late 1990s (Ambos 2005, p. 398).

4 Zander, 1994, Håkanson and Nobel 1993a, Pearce 1989,Dalton and Serapio 1995, von Zedtwitz and Gassmann2002.

5 R&D expenditure data are for R&D activity regardlessof the source of funding. The R&D data from the UnitedStates Bureau of Economic Analysis (BEA) definesR&D to include basic and applied research in scienceand engineering as well as the design and developmentof prototypes and processes. R&D expenses includewages and salaries, taxes, materials and supplies,depreciation, amortization, and allocated overhead andindirect costs, but exclude capital expenditures. R&Dexpenses also exclude routine product testing andquality control conducted during commercialproduction, geological and geophysical exploration,market research and surveys, and legal work pertainingto patents. BEA data used here exclude banks and otherdepository institutions. However, data on thedistribution of overseas R&D in terms of basic, appliedand development expenditures, along with their costcomponents (e.g. labour, equipment, taxes) are notavailable. Expenditure data are in current dollars (Moris2005a). For further information and surveymethodology, see http://www.bea.gov/bea/di/usdscrpt.htm.

6 R&D employment data from the United States BEASurvey of U.S. Direct Investment Abroad are availableonly every 5 years from benchmark surveys. The latestavailable data are for 1999.

7 In local currency, total R&D spending increased from36 billion Swedish kronor to 47 billion Swedish kronor.

8 Granstrand 1999, Sachwald 2004a, Archibugi andMichie 1995, Archibugi and Iammarino 2002, Molero1998.

9 Roberts (2001) and Edler et al. (2002) surveyed 209Triad firms each; von Zedtwitz and Gassmann (2002)conducted a total of 290 interviews (over the period1994-1998).

10 In order to eliminate the distortions caused by under-and over-representation, this has been calculated asa weighted average of responses using the regional

distribution of the 316 questionnaires for weighting.Due to the over-representation of Western Europe inthe responses, the unweighted average would have been34%.

11 Not all firms answered both questions.12 Previous studies (Roberts 2001, Edler et al. 2002, von

Zedtwitz and Gassmann 2002), while finding that theWestern European firms were the mostinternationalized, also noted that their lead over theUnited States TNCs was small. In the Edler et al. 2002survey (p. 158), the European firms were estimatedto spend one-third of their R&D budget abroad in 2001,followed closely by the North American firms (32%),and only very distantly by the Japanese firms (11%).In Roberts’ (2001) survey, Western European firms wereestimated to spend 35% of their R&D budget abroad,followed by the North American firms (33%) and theJapanese firms (10%). The discrepancy with theUNCTAD survey is due to the fact that the survey byRoberts treated intra-European and intra-NorthAmerican R&D flows as domestic.

13 These are estimates based on data from 30 economies,which accounted for 99% of global business R&D in2002. For more details, see the note in annex tableA.IV.1.

14 The presence of India in this group may be surprising.The low share of foreign affiliates in total R&Dspending in India may be due to various factors. Oneis that the latest statistics available are only for 1999(i.e. the period before the take-off of many largeprojects). A second reason may relate to the definitionof R&D: India specializes in software development,an industry that is not always categorized as R&D instatistics. Finally, many of the projects started in Indiahave been of a non-equity nature, and hence are notreflected in FDI.

15 The share of foreign affiliates in the R&D of thetransition economies of South-East Europe may beequally high, while that of the CIS is probably low.

16 Historical data were missing for two economies: Italyand Thailand.

17 Such as in the case of the merger between Sweden’sAstra and the United Kingdom’s Zeneca, the acquisitionof the United Kingdom’s Celltech by Belgian UCB,or the takeover of Škoda Auto by Volkswagen in theCzech Republic and Tungsram by GE in Hungary.

18 These foreign affiliates are engaged in commercial,physical and educational research (Standard IndustrialClassification (SIC) code: 8731), commercialeconomics and biological research (SIC code 8732),non-commercial research (SIC code: 8733) and testinglabs (SIC code: 8734) as their main activity.

19 Maastricht Economic Research Institute on Innovationand Technology, Cooperative Agreements andTechnology Indicators (MERIT CATI) database.

20 Source: MERIT-CATI database.21 Prior studies concluded similarly that R&D activities

were not equally distributed around the world andtended to reside mostly in developed countries(Gassmann and von Zedtwitz 1999, Meyer-Krahmerand Reger 1999, Schmaul 1995, Archibugi andIammarino 2002).

22 Information in this discussion related to the UnitedStates is based in part on a background paper prepared

155CHAPTER IV

by Francisco Moris (Moris 2005b) for WIR05.23 Surveys are conducted at five-year intervals. The results

for 2004 are not yet available.24 R&D bases are key nodes of R&D, typically regrouping

various affiliates. Hence the number of bases is lowerthan the number of foreign affiliates.

25 LOCOmonitor collects, validates and crosschecks real-time information on new (greenfield) and expansionFDI projects worldwide. Both announced and realizedFDI projects are included. Each project identified iscross-referenced against multiple sources and thecompany website. Full global data collection startedin 2002. Each FDI project is classified into one “key”business function (out of a list of 17, including R&D)and, if applicable, into additional business functions(following the same categorization). As a result, thenumber of projects whose “key” business function isR&D is smaller (1,489 over the period 2002-2004,annex table A.I.3) than the number of projects for whichR&D is “any” business function (1,773 over the sameperiod of time). The data presented in this Report referto the second, broader definition of R&D. The usualcaveat on completeness and accuracy of informationapplies.

26 The source of these categories is von Zedtwitz 2005.27 The Edler et al. 2002 survey concluded in a similar

way (pp. 159-160) that North America and WesternEurope were the most attractive target regions forforeign R&D, while Japan’s attractiveness for R&Dcarried out by TNCs from abroad was well below thecountry’s science and technology potential. Amongthe developing regions and South-East Europe and theCIS, the “Asian Tigers” were mentioned by 23% ofthe firms surveyed. “Eastern Europe” (12%) and LatinAmerica (10%) were far less important, while Africawas hardly mentioned.

28 Bulgaria was mentioned by only one respondent. Therest of South-East Europe and the CIS did not appearon the investment map for R&D.

29 Respondents indicated only regions and not individualcountries.

30 For the analysis of the innovatory activities indeveloping countries, USPTO data are preferred overnational patent data and those of other developedcountries, since they are regarded as providing a morecomparable and representative measure of suchactivities (chapter III).

31 USPTO glossary, www.uspto.gov/main/glossary/index.html.

32 The total number of USPTO patents granted increasedby 70% in the same period.

33 For some patents, the USPTO database does notidentify any assignees. In such cases, it is assumedthat the inventor(s), to whom the patent is granted,remains the legal owner.

34 In 2003, 17 patents (13% of that year’s total) weregranted to the Brazilian affiliate of Johnson & Johnson(United States), and five patents to the Brazilianaffiliate of Dana Corporation (United States) forinstance.

35 In India, the Council for Scientific and IndustrialResearch was the most important institute with 324patents.

36 Data from the LOCOmonitor database.

37 For Ericsson (Sweden), over the past 40 years, R&Din telecommunications equipment production hasshifted from hardware to software. Today, the companyis spending 85% of its R&D budget on softwaredevelopment (Goldstein and Hira 2004).

38 In the new United Nations classification, the eightformer Central and Eastern European economies intransition that joined the EU in 2004 are shown as partof the developed-country group, under the categoryof the EU-25 (box I.2). For analytical purposes,especially when drawing conclusions from the lessonsof transition, their experience is shown here togetherwith that of South-East Europe and the CIS.

39 These TNCs include Caterpillar, Cisco Systems,DaimlerChrysler, Du Pont, General Electric, GeneralMotors, Hewlett Packard, IBM, Intel, Lucent,Microsoft, Motorola, Oracle, Philips, SAP and TexasInstruments. For instance, GE’s John F WelchTechnology Center in India, with an investment of $80million and 1,600 employees, is the company’s firstand largest R&D centre outside the United States(LOCOmonitor database).

40 The R&D centre of Eli Lilly is its largest researchfacility in Asia and the third largest in the world.

41 Estimates by the Board of Investment of Thailand. Analternative source of information, the R&D/InnovationSurvey of the National Science and TechnologyDevelopment Agency for the year 2003, has estimatedthe R&D expenditure of majority-owned foreignaffiliates to be about $40 million (about 28% of thetotal R&D expenditure of the private sector) in thatyear (Intarkumnerd and Sittivijan 2005, pp. 5-6),indicating that the Board of Investment may haveunderestimated the R&D expenditure of local firms.

42 The term “tropicalization” has been used in particularto denote the adaptation of automotive products to thelocal conditions and climate of Brazil (Kuntz 1999).

43 By comparison, the corresponding figures for foreignaffiliates in developing Asia were 13% and 8%.

44 This happened with the car makers Ford andVolkswagen, and the telecom equipment supplierAlcatel (Costa 2005, p. 6).

45 At the University of Stellenbosch, for example,important work has been done on emission control andengine testing in collaboration with regulatory bodiesin the EU.

46 Source: BusinessDay (www.bday.co.za/bday).47 They include Burlington Resources, Amerald Hess

Corporation, ConocoPhilips, Anardarko and OccidentalPetroleum Corporation (Oxy) from the United States,and Woodside from Australia, BG Group from theUnited Kingdom, Repsol from Spain and Edison fromItaly.

48 Their R&D focuses on integrated sedimentology,geochemistry, seismic interpretation, petrophysics,reservoir engineering and petroleum geology research(narg.web.mcc.ac.uk/home.html).

49 www.roncoconsulting.com/post-conflict/uganda.html.50 The survey, conducted from January to June 1999,

reviewed the pre- and post-privatization performanceof 23 major companies selected from seven countries,of which five became new EU members in 2004 (theCzech Republic, Hungary, Latvia, Poland and Slovenia)and two are candidates for accession (Croatia and

156 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Romania). The combined asset value of these largeenterprises at the moment of their privatizationexceeded $5 billion — 8% of the inward FDI stockof the seven countries in 1999 (Kalotay and Hunya2000, p. 52).

51 Unpublished data of the Hungarian Central StatisticalOffice on the performance of foreign affiliates in 1992-1998 (reported in Kalotay 2000, p. 165) confirm therising trend of R&D: over the period of observation,the R&D expenditure of foreign affiliates in Hungaryincreased from $6.3 million to $96.5 million, raisingthe R&D intensity of these firms (measured as apercentage of total sales) from almost nil to 0.4% oftotal sales.

52 EADS holds a 51% share in the venture. Komarov,Alexey, “EADS East Airbus-trained Russian engineers,data exchange network in place”, Aviation Week &Space Technology, 159,6, 11 August 2003, p. 54.

53 Japan, with only two Chinese R&D units, seems to besomewhat underrepresented in the sample, probablydue to the small sample size. However even in thecomplete database of 776 international R&D units,Japan has only 55 or approximately 7% of total foreignR&D laboratories (von Zedtwitz 2005).

54 One exception is Huawei’s software laboratory inBangalore (550 engineers in 2003, expected to growto more than 2,000 by 2005). The value of thatinvestment was almost $100 million, or about 7% ofHuawei’s overall R&D activities.

55 In addition to Bangalore, Huawei has also investedin Stockholm (Sweden), Moscow (Russian Federation)and Dallas (United States).

56 Haier operates ten small-scale research units abroad,which focus on technology monitoring and other R&Dactivities.

57 Jointly with GE for instance, TCS has established anR&D centre in Hangzhou, the capital city of Zhejiangprovince in China. Other top Indian IT services playerssuch as Infosys, Satyam and Wipro have also investedin China.

58 For example, in 2003 the pharmaceutical firm Ranbaxy(India) set up a new plant in Abu Dhabi that will alsoconduct R&D.

59 The operations of Samsung Electronics are particularlyR&D-intensive, accounting for 8% of revenues in 2003.Ten of its 16 R&D centres are located abroad (China,India, Israel, Japan, the Russian Federation, the UnitedKingdom, and the United States). Its global R&Dnetwork develops new technologies in digital media,telecommunications, digital appliances andsemiconductors. The company also carries out jointR&D projects through strategic alliances with Sony,IBM, Hewlett-Packard and Microsoft.

60 Similar observations were made in another recentsurvey (EIU 2004a), in which more than half of therespondents were planning to increase their overseasR&D investment. And a DIHK survey conducted in2005 found that nearly 20% of German companiesplanned to move R&D jobs abroad in the next threeyears (DIHK 2005b).

The expansion of R&D by TNCs in somedeveloping countries reflects changes in thedrivers and determinants of R&Dinternationalization. In view of increasedcompetitive pressures, shorter product life cyclesand the need to innovate more at lower costs,firms are compelled to search for new ways oforganizing their R&D. At the same time, somedeveloping-country governments have been ableto vastly improve the supply of relevant skills– often costing much less than comparative humanresources elsewhere. R&D internationalizationis not confined to TNCs from developedcountries; developing-country firms are alsosetting up R&D activities abroad to access theseforeign markets and centres of excellence.

This chapter analyses these trends fromthree perspectives: the changing drivers of R&Dinternationalization; the locational determinants;and factors affecting the mode of R&Dinternationalization. The annex to this chapterpresents a case study of the expansion of chipdesign in Asia.

A. What drives theinternationalization of

R&D?

R&D is one of the least mobile of TNCactivities; there are several reasons for i tslocational “stickiness” (Lall 1979). The complexand tacit nature of advanced technical knowledgemakes it difficult and costly to fragment R&Dand to locate the different segments in differentplaces. Researchers often need face-to-faceinteraction to exchange information and ideas.Moreover, research skills tend to develop in acumulative manner, so that centres that start earlyoften retain or increase their lead; history shows

��������

DRIVERS AND DETERMINANTSthat “centres of excellence” in technologies tendto survive for long periods. R&D also hasextensive spillovers – ideas and people flowbetween innovating firms, with significantsynergies – creating strong cluster oragglomeration advantages. Where reputablepublic research institutes and universities arepresent as part of the cluster, the advantages ofa particular location are even greater.

These factors tend to anchor innovativeactivity in specific locations or clusters withinan economy, mostly in the home country (Pateland Pavitt 1991). However, recent trends in R&Dinternationalization suggest that these factors arechanging, leading to greater dispersion of R&Dactivities (box V.1). Although many TNCinnovators sti l l keep their core innovationactivities in one location, most large companies,particularly those with multi-plant operations anddiverse products, now have dispersed R&D units.What determines whether TNCs locate these unitsat home or abroad?

In general, TNCs prefer to retain R&D athome when the costs of communicatingknowledge across national borders are high.These costs rise with geographical, economic,cultural and linguistic distance (Fisch 2003, Jonesand Teegen 2001).1 Moreover, TNCs are reluctantto locate R&D abroad when they want to maintaingreater control over the innovation process andits outcome. Due to the risk of technologyleakage, they are also reluctant to place R&D inlocations where there are weak intellectualproperty rights (IPR) regimes. The size of thefirm and the industrial structure also matter.Larger TNCs tend to have more far-flungoperations as well as greater experience andorganizational skills, thus finding it easier to setup R&D overseas. Small firms may have a greaterneed to tap into foreign R&D centres, but often

158 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

lack the organizational resources to set up andmanage dispersed R&D systems. Oligopolisticindustries, with a small number of competingTNCs, may have firms trying to match each other’sR&D activities in a kind of herd reaction.

Adaptive R&D to support foreign productionand customize technologies to local conditions hasbeen the main form of R&D abroad (see alsochapter IV). Even today, local adaptation remainsthe dominant type of foreign R&D undertaken byTNCs (Edler et al . 2002, OECD and BelgianScience Policy 2005, Roberts 2001, Ambos 2005).But even local adaptive R&D in a foreign affiliateis economical only under certain conditions(Voelker and Stead 1999). The host economy mustbe sufficiently different from the home economyto make a major adaptive effort necessary; thescale of operations (a large domestic market orproduction aimed at export markets) must besizeable enough; and the host country must possessthe necessary human resources and institutionalframework. TNCs from developing countries alsoundertake adaptive R&D abroad. For instance,

Huawei Corporation of China has set up a largeR&D facility in Bangalore, India, to undertakesoftware design, while Indian software companieslike Infosys and Satyam have set up developmentcentres in China to adapt products to the localmarket.

Technology sourcing or monitoring is anincreasingly important reason for TNCs to placeR&D facili t ies in countries with centres ofexcellence that can serve as monitoring outpoststo keep track of new technological developments(e.g. Cantwell and Janne 1999, Kuemmerle 1999,Patel and Vega 1999, Roberts 2001, Le Bas andSierra 2002). Such R&D internationalization aimsat augmenting the technological assets of theparent company. This is why many electronics andinformation technology firms have establishedR&D facili t ies in Silicon Valley andpharmaceutical R&D units cluster around Boston.Technology sourcing and monitoring have alsobecome important drivers for R&Dinternationalization by enterprises from developingcountries (chapter IV, von Zedtwitz 2005).2

Enterprises practically always launch R&Dnear the headquarters and/or their main productionfacilities. The first step towards internationalizingR&D is to disperse it from one location to several,which involves overcoming the inherent costs oftransferring tacit knowledge and coordinatingresearch over distances. Firms have to weighseveral internal and external factors before decidingwhether to keep R&D centralized or to disperseit.

Internal factors concern scale economies inR&D, the need for close interaction between R&Dand other corporate functions, along with the desireto control and manage the R&D process fromheadquarters (Gertler 2003, Fisch, 2003). Ingeneral, where R&D involves high minimuminvestment in equipment and personnel, or requiresgeographical proximity to headquarters or the mainproduction plant in order to be effective, there isa strong case for centralization. The case isstrengthened if communication costs are high and

Box V.1. The case for dispersing R&D from a centralized base

the company lacks the managerial andorganizational skills to handle dispersed units.

However, centralization of R&D can alsogenerate costs. Facilities over a certain size maylose flexibility and lose contact with parts of thefirm located elsewhere.a Moreover, somedecentralization is inevitable in a multi-plant firmto the extent that the R&D conducted is supportingproduction – production that is itself dispersed. Newcommunication technologies and managementpractices are reducing the transaction costs ofmanaging dispersed R&D units. In addition, newresearch methodologies permit greater codificationof scientific knowledge and standardization of someR&D work, which facilitates the dispersal of R&Dunits (Patel and Pavitt 1991, Prencipe et al. 2003).

External factors affecting R&D location arethe relative availability and cost of technical skillsand knowledge institutions and the proximity ofinnovation clusters (Carrincazeaux et al. 2001,Cantwell and Janne 1999, Porter and Stern 2001).

Source: UNCTAD.

a There is also a need to separate research from development (von Zedtwitz and Gassmann 2002). Science-orientedresearch may have to be separated from engineering-oriented development work to improve efficiency. This isparticularly the case in industries where product development is highly science-based, as in pharmaceuticals andbiotechnology.

159CHAPTER V

A study of over 200 TNCs from the UnitedStates, Europe and Japan identified nine reasonsfor internationalizing R&D (Edler et al. 2002).The three most important motives for the samplefirms were to adapt foreign technologies to localmarkets, to access skilled research personnel andto learn from foreign lead markets andcustomers.3 The four motives of mediumimportance were to take advantage oftechnologies developed by foreign companies,to keep abreast of foreign technologies, to supportlocal production and to comply with local market-access regulations and pressures. Finally, the twoleast important motives were to take advantageof public R&D programmes in host countries andto evade an inappropriate R&D environment athome. This survey was conducted at the end ofthe 1990s and related to R&D offshoring in otherdeveloped countries. It more or less confirmedwhat previous studies of R&Dinternationalization had found (Mariani 2002,Jones and Teegen 2003, Roberts 2001).

The recent expansion of R&D outside theTriad (chapter IV) suggests that a new set ofdrivers – the cost and the availability of researchmanpower – has become increasingly important.Rising R&D expenditures, along withintensifying pressures to cut costs and to bringproducts quickly to the market, are forcing TNCsto look for ways to do research more quickly,outsource non-core work (see next section) andlocate R&D in countries with low-cost and amplescientific manpower. This becomes even moreimportant when companies fail to find a sufficientnumber of skilled people in their home base,especially in science-based activities. Forexample, it has been reported that the EuropeanUnion lacks 700,000 scientists and engineersneeded to meet its target of devoting on average3% of GDP to R&D.4 A study of R&D in Asiaconcluded that:

“[o]ne main reason for offshoreoutsourcing is that very often there isn’tenough talent in the company’s own homecountry... the personnel available forspecific tasks does not have the sufficientqualifications, where programmers andscientists from countries such as India dohave the right qualifications and skills tomatch the outsourcers’ needs” (Frost andSullivan 2004, p. 8).

As the internationalization of manufacturingproduction and IT-based services reveals its costadvantages, firms are starting to apply the same

principles to innovation. Many companies acceptthat, all else being equal, the cost and availabilityof researchers are now important drivers forinternationalizing R&D, particularly in industriesrelying on new technologies. A survey of foreigncompanies’ R&D activities in India noted thatfor companies in conventional technologyindustries, proximity to manufacturing and to theIndian market were the two main motives forundertaking R&D in India (Reddy 2000).5

Conversely, for companies in new technologyindustries availability of R&D personnel and lowcosts of doing R&D topped the list. Moreover,for this category of companies a shortage of R&Dpersonnel in the developed countries wasperceived as a relatively important driver,whereas it was unimportant for companies inconventional industries. This observation is inline with the dominance of electronics, ICT andsoftware industries among the globally orientedR&D labs that have been established in variousAsian economies in the past decade (chapter IV).

Other recent surveys and media reportsconfirm the growing relevance of cost reductionand the importance of accessing talent poolsabroad:

• A survey of German companies found thatthe lower cost of R&D manpower abroadwas the second most important reason, afterproduction support, to locate R&D abroad(DIHK 2005b).

• A survey of 104 senior executives notedthat: “[in] industries where a constant streamof high-tech innovations is crucial tosurvival, companies will go wherever theymust to access top R&D talent. A total of70% of executives in the survey see theability to exploit pools of skilled labour asa very important or crit ical benefit ofglobalized R&D, making this a moresignificant driver than cost control or thedesire to accelerate innovation cycles” (EIU2004a, p. 2). Moreover, more than half saidthat lower costs were an important benefitof globalized R&D. Cost benefits came fromcheaper labour and lower land and officerents, as well as from favourable taxregimes.

• Cost reduction has been identified as oneof the main drivers of expanding TNC R&Din China (Armbrecht 2003).

• In a survey of product engineeringcompanies in California conducted by theIndian company, Wipro Technologies, thetop reasons for outsourcing were to reduce

160 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

the time it takes from product developmentto sales (“time-to-market”), as well asoverall R&D costs.6

• The need for cost reduction has also beenan important driver for the offshoring ofchip design to Asia (Ernst 2003, see alsoannex to this chapter).

Cost advantages derived from conductinginnovative R&D in developing countries can besignificant. A recent report on the pharmaceuticalindustry compared the cost structures of Indiawith those of developed countries (GoldmanSachs 2005). It concluded that the cost of clinicaldevelopment in India was 45%, drugmanufacturing 30%, and R&D related to drugdiscovery only 12.5% of the corresponding workconducted in a developed country.

While costs matter, the expansion ofinnovative R&D in Asia has also been driven byvarious supply-oriented factors. Concerted effortson the part of many of the countries in that regionhave increased the supply of skills, notably inthe areas of science and engineering. In somecases, researchers, engineers and managers ofthe diaspora have returned to their home countriesand brought with them new capital, skills,networks and their reputation. Policyinterventions include new incentives to promoteR&D, more effective IPR regimes, improvedpublic research activities and the establishmentof science and technology parks (chapter VII).For some industries such as electronics, the factthat manufacturing activities have already beenglobally organized is making it easier – andsometimes even necessary – to disperse R&Dactivities internationally. It is no coincidence thatEast and South-East Asia are over-representedamong the “winners” in export competitivenessin the same product areas in which TNCs arescaling up their R&D work in the region.7

Finally, it is important to consider a fewtechnical and organizational advances that arereducing the constraints to the cross-borderexchange of knowledge and compelling firms tointernationalize their R&D (Zanfei 2000, Ernst2003). First, liberalization and technologicalprogress have made competition more intense,forcing TNCs to invest more in R&D withoutallowing costs to spiral out of control. Companiesthat are unsuccessful in curbing developmentcosts tend not to be rewarded by the stock market.Thus they look for more economical ways ofboosting innovation. Second, advances in ICTs

allow for faster, cheaper and denser informationexchange across long distances. Third, in “newtechnology” industries the proximity to basicscience makes it possible for countries that havean ample supply of scientists and engineers tohost R&D work of TNCs, even if their industrialexperience is otherwise lacking (Reddy 2000).Fourth, the “modularization” – or finerspecialization of the R&D process into separateactivities – of some types of R&D is allowingfirms to fragment the development process (ofproducts and services) to raise efficiency and cutcosts (Baldwin and Clark 2000).

In summary, most R&D internationalizationis driven by the need to adapt products andprocesses to local markets. However, the needto tap into foreign centres of excellence andsource foreign technology is gaining inimportance, especially in the case of R&D setup in developed countries. But to understand theexpansion of innovative R&D units in somedeveloping countries, it is necessary to considera complex mix of driving forces encompassingdemand factors, supply factors and variousenabling factors. For TNCs, especially in newtechnology industries, developing economiesoffer new opportunities to reduce costs, accessskills that are not readily available at home insufficient supply or at attractive costs, and speedup the development of new goods and services.

B. Host-countrydeterminants of R&D

location

Given the pressures inducing TNCs tointernationalize R&D and the factors making thispossible, what determines where TNCs locateR&D in the developing world? The global mapof R&D shows that its spread is uneven. R&D inhost developing countries is mainly concentratedin Asia and in a few large economies in LatinAmerica and the Caribbean. The present sectionrelies on survey evidence from developedcountries and qualitative evidence fromdeveloping ones. The picture that emerges isfairly clear and persuasive.

While some basic determinants arecommon, different types of R&D (chapter IV)– adaptive R&D, innovative R&D linked toproduction for local/regional markets, globalinnovative R&D for new product/processdevelopment or basic research, and technology

161CHAPTER V

monitoring – are attracted by different factors.The general investment climate – comprising, forexample macroeconomic and social stability,security, transparency, administrative rules andregulations – is as important for R&D locationas it is for FDI in general. Similarly, the type ofR&D that may be attracted depends on theeconomic structure of the location, including theindustrial structure, market size and growth,culture and language, natural resourceendowments, living conditions and physicalinfrastructure. Most of these factors are“created”, rather than natural, assets and thereforecan be altered through government intervention.Hence, host-country policies play a significantrole in determining a country’s abili ty toparticipate in the international restructuring ofR&D activities by TNCs (chapter VII).

Adaptive R&D is typically closely relatedto production and involves the adaptation ofimported technologies. This is the dominant formof R&D by foreign affiliates in Latin Americaand in Africa (chapter IV). The location of suchdevelopment work is determined by the need tosupport production and adapt technologies, tobe near customers, to cooperate with localpartners, to access markets, to improve the local“image” of a company, to launch a productsimultaneously, to facilitate rapid scale-up inmanufacturing and to overcome protectionistbarriers against imports (von Zedtwitz andGassmann 2002, p. 584). The larger the hostmarket, the greater the need for local adaptationof goods and services. As national marketsbecome regionally more integrated, somecountries may become the preferred base foradaptation, not only for the local market but forthe region as a whole. In this case, appropriateskills and other aspects of the national innovationsystem (such as the technical and economicinfrastructure, proximity to suppliers/keycustomers) become more important. Dependingon the industry, adaptive R&D needs technicaland engineering skills that are specialized in thetechnologies used in production. Cost factors arelikely to be of secondary importance.

Innovative R&D has emerged as a featureof some foreign affiliates in parts of South, Eastand South-East Asia as well as in some transitioneconomies (chapter IV). Internationalization ofsuch R&D for global markets is driven by thesearch for advanced skills in relevant areas ofscience-based technologies. Such R&D work canbe intended for regional or global markets andis determined primarily by the quality of the

national innovation system (NIS). In China,adaptive R&D has evolved into more advancedforms of innovation, with the local market servingas a test-bed for new products for regional oreven global markets (Sigurdson 2005b; chapterIV). The precise features of a host country thatare needed to attract innovative R&D depend onthe industry and activity involved. Keydeterminants in host developing countries forattracting innovative R&D include a large poolof scientific and technical manpower, a wellfunctioning NIS featuring strong public researchinstitutions, science parks and an adequate systemof IPR protection, and government incentives(von Zedtwitz and Gassmann 2002, Reddy 2000,Toh 2005).

The availabili ty of the right kinds ofscientific and engineering skills is probably themost critical factor in attracting innovative R&D,especially in new, science-based technologyindustries. The importance of researchers andscientists covering a broader range of disciplinesis not new. What is new is that competitivepressures are forcing companies to pay greaterattention to wage costs and availability ofscientists and engineers in large numbers. Withwage rates for skilled researchers in developing-country R&D locations significantly lower thanthose in developed countries, the attractivenessto TNCs is compelling. But wages per se are notthe main location determinant. TNCs value theability to set up a research facility rapidly andtap into an existing knowledge centre where theycan find skilled researchers (often in thehundreds) at short notice. This gives a “criticalmass” advantage to countries that combine lowwages with good education systems that turn outlarge numbers of well-trained researchers. Astheir low ranking in the UNCTAD InnovationCapability Index (chapter III) shows, China andIndia are not the most attractive locations in termsof human resources normalized by populationsize. However, when TNCs need to recruitresearchers in large quantities, these countriesoffer a growing body of skilled people at lowcost.

The global distribution of tertiaryenrolments has changed dramatically (box V.2).8

Developing Asia has emerged as the main sourceof new university graduates, and this trendappears to be continuing. This is one of the mainreasons why, for example, a growing number ofTNCs are turning their attention to China andIndia for innovative R&D work. China isexpanding its tertiary education system at an

162 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

unprecedented rate.9 The total number of studentsenrolled in tertiary education increased to more than19 million in 2003, a 100% increase over 2000.10

It has been estimated that the accumulatednumber of university graduates in China couldexceed 120 million by 2020 (Sigurdson 2004).If realized, this expansion would pose acompetitive challenge to other countries,developed and developing. India is expandingmore slowly and the tertiary enrolment rate isrelatively low (at around 10% of the age group),but the absolute numbers are large. MeanwhileLatin America, a richer region overall , lagsbehind in enrolments of engineers and scientists.This further constrains its R&D performance,

inducing a significant number of its researchersto seek work in North America.

Of course, not all tertiary students arecandidates for work in the R&D labs of TNCs.A recent analysis of the supply of skilled peoplein various developing countries and economiesin transition (including the new EU members)found that only a small proportion of potentialjob candidates in “degree specific” occupationswere qualified for work in TNCs (McKinseyGlobal Institute 2005).11 The research, which wasbased on interviews with human resourcemanagers in 83 TNCs, found large differencesamong the countries investigated. For example,while 50% of engineers in Poland and Hungary

In 2000/01, developing countries accountedfor 62% of global tertiary enrolments overall, andfor 52% in technical subjects (pure science,engineering and mathematics and computing).Transition economies (including new EU members)accounted for 16% and 20%, and developedcountries (excluding the 10 new EU members) for22% and 28% respectively. Box figure V.2.1 showsthe number of total and technical tertiary enrolmentsacross developing regions. Box figure V.2.2 displaysthe shares of technical tertiary enrolments by region.The first figure also separates the main outliers fromthe totals of each subregion: China in South-Eastand East Asia, India in SouthAsia, and South Africa in sub-Saharan Africa. The data ontechnical enrolments areparticularly important for R&Dlocation as these are the primaryskills involved in such work.

In tertiary technicalenrolments, China, the RussianFederation and India led theworld, ahead of the United States(which had the highest number oftotal tertiary enrolments in 2001)(annex table A.V.1). The Republicof Korea was fourth in the worldin technical enrolments, which isimpressive for a country of only47 million people.a Indonesia,Mexico and Brazil followedamong developing countries,Ukraine and Poland amongtransition economies, and

Germany and Japan in the developed world. BothGermany and the United States saw a decline inthe total number of tertiary students, while thenumber in Japan increased.

In tertiary technical enrolments, Chinaaccounted for 50% of the total for South-East andEast Asia in 2001; it had more students than thewhole of Latin America and the Caribbean (LAC)and sub-Saharan Africa combined. India accountedfor 90% of the total for South Asia; it was slightlybehind LAC as a whole but ahead of West Asia,North Africa and sub-Saharan Africa together. SomeAfrican countries have also expanded their tertiary

Box figure V.2.1. Total and technical tertiary enrolmentsacross developing regions, 2000-2001

(Thousands)

Source: UNCTAD, based on annex table A.V.1.Note: South-East and East Asia 2 excludes China. South Asia 2 excludes

India. SSA 2 is sub-Saharan Africa excluding South Africa.

Box V.2. Tertiary enrolments by region and country

/...

0

2 000

4 000

6 000

8 000

10 000

12 000

14 000

South-East

and EastAsia 2

China LAC WestAsia and

NorthAfrica

SouthAsia 2

India SSA 2 SouthAfrica

Total tertiary

Technical tertiary

163CHAPTER V

were suitable to work for TNCs, thecorresponding number for India was about 25%,and for China and the Russian Federation only10%. The results underline the need to focus notonly on quantity but also on quality in educationprogrammes.

The agglomeration of R&D activity in aspecific part of a country often reflects theconcentration of skilled manpower in thatlocation. For example, most software companiesin India are located in the five states that accountfor nearly half the diploma-granting technicalinstitutions in that country as well as for two-thirds of all diplomas awarded by private traininginstitutions (D’Costa 2003 p. 216). In China,Beijing, Shanghai, Guangzhou and Shenzhenaccount for 85% of all R&D units set up byforeign companies in China, mainly because theyare close to local universities and researchinstitutions (Zhang 2005; box IV.5). Some 50TNC R&D organizations have been set up in theZhongguancun area of Beijing (Zhang 2005).

Source: UNCTAD, based on annex table A.V.1.Note: Transition economies here comprise South-East Europe and the CIS as well as the Czech Republic, Estonia,

Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia.a In the Republic of Korea, as of 2004, 40% of people in the age group of 25-34 years were university graduates.

Every year the country produces some 70,000 engineering graduates, which is equal to the number produced in theUnited States (KICOS 2004).

While the absolute number of skilledpeople plays an important role in R&D location,it is nevertheless possible for small economieswith high levels of technical skill to attract globalR&D as long as they also have a large TNCpresence in technology-intensive activities andcan offer specialized R&D competence. Ireland,Singapore and Hungary are good examples ofsmall newcomer countries that have attracted alarge TNC research presence.12 By the sametoken, countries with large skill pools may notattract much TNC R&D if other conditions arenot met, as is the case for Japan and the RussianFederation.

An important structural determinant ofinnovative R&D location is the strength of acountry’s NIS (see also chapters VI and VII). TheNIS includes knowledge institutions (R&D labsand universities as well as standards, quality andmetrology institutes) and other R&D performingenterprises (local or foreign), along with aninstitutional framework for R&D and innovation.A strong NIS, where knowledge institutions have

education system rapidly, but from low levels. Forexample, in the United Republic of Tanzania thenumber of technical students increased from 1,000to 6,000 between 1990 and 2000; in Ghana thecorresponding rise was from 2,000 to 14,000; and

Box figure V.2.2. Shares of global technical tertiary enrolments(Per cent of total)

in Egypt from 70,000 to 290,000. However, thenumber of people with tertiary education remainsvery small in most of Africa.

Box V.2. Tertiary enrolments by region and country (concluded)

0

5

10

15

20

25

30

35

40

South-Eastand

East Asia

Developed

economies

Transition

economies

South Asia LAC West Asia

and North

Africa

SSA

1990 1998 2001

164 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

tight links with production enterprises and otherfirms that perform world class R&D, is a majordraw to TNCs looking for new R&D locations.The presence of dynamic science parks can bean additional attraction to R&D that requiresinteraction with a diverse range of firms andinstitutions (chapter VII). Basic research callsfor an even stronger NIS, featuring scienceinstitutions that are able to produce world-classresearch and publications and undertake contractresearch work for industry.13

The IPR regime is also part of thisframework. Its role in attracting R&D by TNCstends to differ by industry and type of R&D.14

Adaptive and production support R&D may notrequire strong IPR protection, but it may beessential for other types of R&D (box V.3).15

Do government incentives help attractR&D by TNCs? The question is importantespecially in light of the increased use of R&Dincentives around the world (section VII.C). Ingeneral, incentives are effective only when other,

Box V.3. IPR regimes and R&D location

IPR regimes are often mentioned as a factorthat might influence the location of TNC R&D.However, the evidence is mixed. Surveys suggestthat the role of IPR regimes in attracting FDI ingeneral may be limited, but that it is an importantfactor for R&D-related FDI. Protection ofintellectual property generally improves theenvironment for innovative R&D, but its rolevaries by industry (Maskus 2005). For industriesin which technologies are easy to imitate, IPRprotection may be essential for attractinginternational R&D; for other industries it maybe a less important factor.

A study of IPR protection and FDI, usinga sample of 94 firms from the United States, 45firms from Japan and 35 from Germany foundthat IPR protection was not a critical locationaldeterminant for most types of FDI, but that it didaffect R&D-related investments. The percentageof firms stating that IPR protection is importantwas particularly high in the chemicals andpharmaceuticals industry (Mansfield 1994 and1995).

Econometric analysis of United States TNCsfound that IPR protection was a significantdeterminant of where foreign R&D activities wereperformed, but not a significant factor betweendifferent developing-country locations (Kumar1996). It even suggested that a strong IPR regimecould discourage TNCs from undertaking R&Din developing countries.a However, another studyfound that R&D spending by the affiliates ofUnited States TNCs increased after IPR reformin host countries (Branstetter et al. 2004). Thisstudy also noted that the level and rate of changeof non-resident patenting increased in the post-IPR-reform period, while there was nocorresponding reaction in resident patent filings.

Some developing countries like Brazil,China and India have attracted significant amountsof FDI in R&D; despite being perceived as havingrelatively lax IPR regimes. There are four mainreasons why IPR protection may have a limitedimpact on the location of TNC R&D:

• R&D may be conducted for a completelydifferent market. For example, it has beennoted that IPR issues for TNC R&D labs inChina are mostly handled in the home countryas these labs work on technologies aimed atworld markets (Zhao 2004). Since a patentgives its assignee a monopoly on bothproduction and sales, the TNC can protect itsintellectual property by obtaining patents inthe countries for which the product wasdeveloped rather than in the country where theR&D is undertaken.

• A technology may be highly firm-specific andthus of limited value to others. For example,if different technologies developed by a firmare complementary to one another and can onlybe used jointly, a particular innovation in thehost economy may have little value on itsown.b TNCs may structure their internationalR&D activities so that a foreign affiliate ina country with weak IPR protection undertakesonly R&D with strong complementaryelements.

• TNC R&D in a host economy may deal withtechnologies that are too advanced for localcompetitors to copy and use commercially.

• Certain types of technology involve tacit anduncodifiable elements that are difficult foroutsiders to imitate without intimateknowledge gained by working with thatspecific technology.

/...

165CHAPTER V

more important determinants are in place. Byreducing costs, government incentives mayinduce TNCs to expand or deepen their R&Dactivities. However, if the necessary skills andresearch capabilities are lacking, incentives mayinduce firms merely to re-label routinetechnological activities and report them as R&D(chapter VII). Indeed, countries with ample andlow-cost scientific skills are likely to attractinternational R&D without offering incentives.

A diverse industrial structure, withtechnologically complex activities, is likely toprovide clusters with the skills and linkedsuppliers and buyers that can support innovativeR&D. Countries with strong technologicalspecialization tend to attract TNC R&D in similarareas, and TNCs tend to internationalizeinnovative (asset-augmenting) R&D tocomplement their strengths (Patel and Vega 1999,Le Bas and Sierra 2002, Guellec and vanPottelsberghe 2001 and 2004b).

The fact that developing Asia has emergedas the production base for many globally orientedindustries (WIR02) has also led some TNCs toconduct more R&D in the region so as to becloser to their actual manufacturing activities (seeannex to this chapter). In Malaysia, some foreignaffiliates in electronics have obtained a mandatefrom their parent companies to design, develop,manufacture and market products for globalmarkets. This has allowed them to undertake allstages of innovation. Toyota’s decision to placeone of its global R&D labs in Thailand waslikewise facilitated by the presence of a relativelystrong automotive cluster in that country (boxIV.7). In the case of India, proximity tomanufacturing has been an important driver forR&D by foreign affil iates in “conventionaltechnology” industries, but not in new technologyindustries (Reddy 2000).

Finally, R&D with the aim of monitoringor sourcing technology is mainly drawn tocountries boasting world class clusters oftechnological and industrial activity (Porter and

The design of IPR regimes may play a lessdirect but nevertheless important role. Forinstance, providing effective means of IPRprotection may act as a signalling device tointernational investors. Strengthening the regimemay show that the country is willing to “play bythe rules” and provide a hospitable investmentclimate. Internationalized R&D often involvesactivities where strong protection matters:pharmaceuticals and software – the two majorareas of TNC R&D in India – are good examples.c

For recent R&D investments in developing Asiaby pharmaceutical companies such as Roche andGlaxoSmithKline the question of IPR protectionwas a key consideration.d

The role of IPR protection must of coursebe assessed not only from the perspective of

attracting FDI in R&D. For example, manyeconomies have taken advantage of their weakIPR regimes to build up indigenous technologicalcapabilities. Imitation, copying and reverseengineering have been important sources oflearning in much of East Asia. However, in thecases of the Republic of Korea (Kim 2003) andTaiwan Province of China, they have subsequentlybecome innovators rather than imitators of newtechnology, and now need more effective IPRregimes to promote domestic innovation. At thisadvanced stage of their development, IPRprotection is important for both local andinternational R&D. Even countries at lower levelsof technology development like China and Indiaare fostering local innovation and may benefitfrom stronger IPR protection (Lall 2003).

Source: UNCTAD.

a Sanyal 2004 reached the same conclusion.b For example, Microsoft Research Asia developed AutoMovie for Movie Maker, Mobile HTML Optimizer for Front

Page and the Ink Parsing technology for tablet PCs. These were considered major contributions to the Microsoftproducts, but alone they are of little value to potential imitators (Zhao 2004).

c It may be noted that India as of 1 January 2005, introduced the possibility to patent pharmaceutical products, reflectingobligations under the TRIPS Agreement. According to the Ministry of Commerce and Industry, this is intended tohelp the Indian pharmaceutical industry protect the results of its rising R&D efforts (www.pib.nic.in/release/).

d “Eastern rebirth of the life sciences”, Financial Times, 10 June 2005.

Box V.3. IPR regimes and R&D location (concluded)

166 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

A recent study on R&D investment bymajor TNCs in China, conducted for theIndustrial Research Institute in the United Stateshighlights some of the perceived advantages oflocating industrial R&D in China, many of whichare the result of government policies (Armbrecht,2003):a

• The supply of talented manpower exceedsdemand, at least by foreign firms;

• Universities and research institutes are eagerto get funding from private firms;

• The possibility of entering into IPRagreements with top Chinese universities;

• A large number of high-technology parks;• Incentives; and• The potential for cost reduction across all

stages of the R&D value chain.

The study emphasized that while costsavings matter, TNCs expand R&D in Chinaprimarily for strategic reasons: to tap the vastpool of talent and ideas and to stay abreast ofcompetitors in the increasingly sophisticatedmarkets of China and Asia. It predicted a furtherincrease in TNC R&D in China and argued thatthe focus of these R&D labs would shift fromsupport and adaptation to full-scale R&D workusing China’s emerging technologies and talentpools.

Box V.4. Why are companies setting up R&D in China?

The following taxonomy describes theevolution of TNC R&D in China (box tableV.4.1). “Satellite” R&D laboratories, the leastdeveloped type, have relatively low strategicimportance for the companies and are vulnerableto budget cuts by TNC headquarters, while“contract” R&D laboratories show verticalspecialization within global innovation networks.Within the latter, China’s role is presentlyconfined to the provision of lower-cost skills,capabilities and infrastructure. While denseinformation flows link these labs with R&D teamsat headquarters and at other affiliates, knowledgeexchange remains tightly controlled and unequal.The highest stage – (more) “equal partnership”laboratories – is comprised of TNCs’ R&Dfacilities that are charged with a regional or globalproduct mandate. For these labs, barriers toknowledge exchange are lower and are eventuallyexpected to give way to mutual knowledgeexchange.

Satellite and contract laboratories stilldominate TNC R&D in China (von Zedtwitz2004, Gassmann and Han 2004, Li and Zhong2003), but there are examples of (more) equalpartnership arrangements, especially in thedevelopment of China’s alternative standards inmobile telecommunications, open source softwareand digital consumer electronics (Ernst andNaughton 2005).

Source: UNCTAD.

a The membership of the Industrial Research Institute includes more than 240 leading global manufacturing TNCsthat perform over two thirds of the industrial R&D in the United States.

Box table V.4.1. Taxonomy of TNC R&D laboratories in China

Satellite laboratories • Act as listening post to detect ideas, incentives and innovationsthat reflect local market characteristics

• Adapt existing products and processes• Are vulnerable to budget cuts

Contract R&D • Exploits lower cost skills, capabilities and infrastructure• Implements a specific module of a global research project• Closely interacts with R&D teams at headquarters and at other

affiliates• Requires tight mechanisms to control IPR leakage• Has dense information flows, but unequal knowledge exchange

(More) equal partnership • Full integration into TNC R&D strategy• Centre has regional or global product mandate• No barriers to fully-fledged knowledge exchange

Source: UNCTAD, based on Walsh 2003 and Ernst 2005.

167CHAPTER V

Stern 2001). Technology sourcing R&D isundertaken predominantly in developed countries.A study of the pharmaceutical industry in Europeand the United States noted that Europeanpharmaceutical TNCs were more likely to set upsuch R&D in the United States than vice versa,possibly reflecting the size and profitability ofthe United States market, i ts scientificcompetence and the close links there between

industry and university research (Ramirez 2003).

Many factors thus interact to determine theattractiveness of a site for FDI in R&D (see box

V.4 for the case of China, box V.5 for the caseof India), but the effective functioning of acountry’s NIS is critical (chapter VII). Most ofthe countries in Asia that have successfullyattracted R&D by TNCs have applied deliberate

TNCs performed R&D in India already inthe 1970s, but it was then limited to adaptationor product development for the Indian market.Such R&D was conducted mainly in response togovernment regulations and to certain uniquecharacteristics of the Indian market. Since the mid-1980s the scope and characteristics of TNC R&Dhave changed.

Starting with Texas Instruments (1986) insemiconductor design, followed by Astra (1987)in biopharmaceuticals, more TNCs have set upglobally oriented R&D units in India – mostlywithout local links to manufacturing activities. The1990s saw the entry of TNCs in diverse industries:for example Motorola (telecommunicationssoftware), Microsoft (computer operatingsystems), STMicroelectronics (semiconductordesign), Daimler-Benz (avionics systems) andPfizer (biometrics). Since 2000, other entrantsinclude Intel (semiconductor design), GE (e.g.aircraft engines, white goods and medicalequipment) and Pfizer (veterinary medicines).

These TNCs were attracted for severalreasons (Reddy 2000), the most important beingthe availability of qualified scientists andengineers.a For instance, in 2004, more than340,000 students were admitted to bachelor degreeeducation in engineering.b India annually producesabout 120,000 chemists and chemical engineers.c

A second attractive feature is the existence ofinternationally reputed R&D institutes such as theIndian Institute of Technology, Indian Instituteof Science, Indian Institute of ChemicalTechnologies and Centre for Drug Research. Manyof the TNC R&D units in India collaborate withthese institutes and several TNCs that do not havean R&D presence in India outsource R&D tothem.

Thirdly, several Indian firms have becomeglobal players and are forming R&D alliancesor subcontractual relationships with other TNCs.The Indian software companies TCS, Wipro andInfosys, for example, have alliances withEricsson, Nokia and IBM. Similarly, Indianpharmaceutical companies, such as Dr. ReddyLaboratories and Ranbaxy, have R&D allianceswith Novo Nordisk, Novartis andGlaxoSmithKline.

In a survey conducted at the end of the1990s, the availability of R&D personnel wasranked by TNCs as the most important reason forlocating R&D in India (4.12 out of 5) (Reddy2000). For TNCs in new technology industriesthis factor was even more important (4.31),followed by low costs of performing R&D inIndia (3.25). Conversely, for conventionalindustries, proximity to manufacturing (4.56) andto the Indian market (4.06) were more importantreasons. Government incentives were relativelyunimportant for both groups of companies (1.78).

The use of English as the business languageand medium of instruction for technical andmanagerial education in universities is an addedbenefit. It facilitates communication of technicalspecifications and requirements between TNCheadquarters and their Indian R&D units. Ingeneral, as regard the IPR regime, the first IndianPatent Law was enacted as early as 1856. Inresponse to obligations under the WTO Agreementon Trade-related Aspects of Intellectual PropertyRights (TRIPS Agreement), the Patent Act of1970, which offered only limited protection toinventions in certain industries, has also beenreplaced and the revised IPR regime is now incompliance with the international regulatoryframework.

Source: UNCTAD, based on Reddy 2005.

a See also “Silicon subcontinent: India is becoming the place to be for cutting-edge research, New Scientist, 19February 2005. “Prescription for change: A survey of pharmaceuticals,” The Economist, 18 June 2005.

b See www.nasscom.org (accessed 21 June 2005).c See “Prescription for change: A survey of pharmaceuticals,” The Economist, 18 June 2005.

Box V.5. Why TNCs set up R&D in India

168 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

policies to strengthen their innovation systemsand create an environment that is conducive tosuch investment.

C. How to internationalizeR&D

Once a firm decides to carry out R&Dabroad, it has to make some choices: betweeninternal and external modes of operations abroad,(i.e. whether to conduct the R&D at an affiliateor outsource it to an independent firm); and forinternalized R&D, between establishing agreenfield facility and acquiring or merging witha host-country firm.

1. R&D outsourcing is growing

R&D internationalization can take the formof in-house work within foreign affiliates oroutsourcing to independent local firms orresearch institutions in a host country. A companyusually opts for keeping an activity in-housewhen strict control of that activity is crucial,when high transaction costs are involved, or whenproprietary knowledge and information issensitive, tacit, expensive to produce, complexor idiosyncratic yet easy to replicate (Dunning1989). Moreover, the more strategic the servicefunction is, and the closer i t is to the corecompetence of a firm, the less likely it is to beoutsourced to unrelated firms. R&D functionsgenerally meet these criteria and therefore couldbe expected to be kept in-house.

Still, R&D outsourcing to foreign locationsis growing within developed countries and is nowcommon in some industries such aspharmaceuticals. Basic research has long beencontracted out to public laboratories anduniversities; the recent trend is for other formsof research (traditionally performed in-house bymanufacturing or service firms) also to be farmedout (Jankowski 2001, Engardio and Einhorn2005). R&D services provided on a contractualbasis constitute one of the fastest growing serviceindustries in some developed countries, led bythe United States.16 As noted in chapter IV, R&Dwork is also increasingly being outsourced tofirms in developing countries, especially in Asia.

What drives firms to outsource R&D? Themain forces are the rising costs and risks of R&D,the growing complexity of innovation (calling

for more diverse skills, knowledge andequipment) and intensifying competitive pressureto bring out new products more quickly (Howells1997, Roberts 2001, Engardio and Einhorn 2005).New research methodologies that make tacitknowledge more codifiable also facili tatecontracting R&D to other firms. The same appliesto software that standardizes research and testingprocesses. By specializing in these activities,which often require expensive equipment andskills, contract R&D firms are able to reapeconomies of scale and scope while offeringcustomized products to firms – rather l ikecontract manufacturers in electronicsmanufacturing (WIR 2000). Their customers canreduce in-house laboratory staff and equipmentwhile speeding up the process without losingcontrol of core innovation.17

In some industries, product developmentis becoming so complex and multidisciplinarythat firms with different specializations arerequired to handle the different stages (Pavitt1999). This makes outsourcing these stages notonly more attractive but also, in some cases,necessary (see annex to this chapter). In thoseindustries, no firm, not even a global marketleader like IBM, can mobilize all the resources,capabilities and knowledge it needs internally.In-house creation of new knowledge andcapabilities needs to be supplemented by externalknowledge sourcing. The increased dependenceon external sources of technology is among themost important changes in technologymanagement in recent years, especially in newtechnology industries (Roberts 2001). In someindustries there are pressures to reduce in-housebasic and applied research in order to focusprimarily on product development and theabsorption of external knowledge (Chesbrough2003, Arora et al. 2000). This externalization ofinnovation does not stop at the national border– firms increasingly tap sources of knowledgeoverseas (Ernst 2002). Thus,

“the speed, complexity, andmultidisciplinary nature of scientificresearch, coupled with the increasedrelevance of science and the demands ofa globally competitive environment, have… encouraged an innovation systemincreasingly characterized by networkingand feedback among R&D performers,technology users, and their suppliers and

169CHAPTER V

across industries and nationalboundaries” (United States, NSF 2004,volume I, p. IV-36).

The transformation of IBM (box V.6)shows that in an “open innovation system”, boththe source and the use of knowledge can beexternal to the company. A firm can create ideasfor both external and internal use, accessing ideasfrom the outside as well as from within. Firmscan move to an open innovation system becauseof the increased mobility of knowledge(Chesbrough 2003).

There are similar trends in thepharmaceutical industry. The cost of bringing anew drug to market was around $800 million in2004, rising to $1.7 billion if commercializationcosts were included.18 Firms see outsourcing asone way to reduce these costs. They currentlyoutsource about 26% of their drug discovery and

development; this could rise to 36% by 2008.Over 20% of the $5 billion annual expenditureson new drug development was paid to contractR&D companies and this share was set toincrease (Malek 2000).

The growing number of R&D providersalso facilitates outsourcing. The privatization ofpublic research laboratories and increasing costpressures on universities in many countries hasinduced companies to enter the market and setup spin-offs. Some large manufacturing firmshave hived off their research arms intoindependent companies. In addition, newentrepreneurs with specialist knowledge, data,skills or equipment have also entered the market.

R&D outsourcing has its limits. Firms areunwilling to outsource the core of theirtechnological advantage: contract R&D cannotreplace all in-house R&D (Narula 1999, Engardio

Starting in 1964, when IBM bet its future onthe development of the 360 product family as theglobal standard for mainframe computers, it pushedvertical integration to the extreme. It internalizedpractically all stages of the value chain: it developedthe basic components, assembled them intosubsystems, designed systems out of thesecomponents, manufactured the systems at its ownfactories, distributed and serviced the systemsthemselves, and even handled the financing of thesystems (Flamm 1988, Ferguson and Morris 1993,Campbell-Kelly and Aspray 1996).

Over time, IBM abandoned this strategy. Therecession of the early 1990s had exposed theweaknesses of the “closed” system of innovation.For the first time since 1946 the companyexperienced three years of declining revenues,shrinking profit margins, and even losses in 1991-1993 (Lazonick 2005, p. 38). In response IBMtransformed itself from a hardware producer to asupplier of integrated solutions, with the objectiveof leveraging its broad portfolio of intellectualproperty (IP), not only to exclude rival firms butalso to generate new and highly profitable sourcesof growth.

Source: UNCTAD, based on Ernst 2005.

a The share of R&D in IBM’s sales fell from an annual average of 9.8% during 1983-1992 to an average of 6.1%during 1994-2003 (IBM annual reports). Goldstein and Hira (2004) document IBM’s decline among the world’stop 50 R&D spenders.

IBM had to go beyond its own R&D andfind the best technologies wherever they existed,combining them into integrated solutions. Animportant facilitator was the adoption of openstandards in a variety of areas, including the Linuxoperating system and the Java programminglanguage. IBM realized that it could no longerexercise tight control over its componenttechnologies, as specialized knowledge was spreadacross companies and countries. This led to asubstantial decline in its in-house R&D intensity.

Furthermore, the focus of IBM’s innovationmanagement shifted towards aggressive licensingof intellectual property. Since 1993 IBM hasemerged as the leader in United States patentapplications, up from 9th position in 1990(Lazonick 2005, p. 40). Licensing of technologyhas been much more profitable for the companythan sales of products in some areas. Its licensingrevenues grew from $30 million in 1990 to $1billion in 1998, generating more than 10% of itsnet profits, and to $1.9 billion by 2001. IBM alsoused its status as the leading patent holder in theUnited States to develop a new market forintegrated solutions.

Box V.6. From closed to open innovation: the case of IBM

170 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

and Einhorn 2005). Too much outsourcing canlead to a firm’s loss of knowledge (and goodresearchers) and can create powerful competitorsfor the outsourcing firm. Another aspect is thatIPRs may not always be enforceable, even withthe most efficient legal systems. Managing andintegrating R&D among different firms, withdifferent work cultures and languages, can beextremely difficult. A distinction is emergingbetween “mission critical” R&D, kept in-house,and “commodity” R&D, which can be contractedout efficiently without damaging thecompetitiveness of the company. As stated by thehead of Motorola in an interview: “You have todraw a line: core intellectual property is aboveit, and commodity technology is below”.19

These distinctions are, however,changeable. R&D outsourcing is evolving rapidly.Enterprises may start by contracting out“commodity” R&D. If this succeeds, they mayrealize the benefits of greater specialization andlearn how to manage better the contractual andintegration process. With time they may developtrust in their collaborators and establish durableknowledge networks. This process can continue,pushing back the limits of what is acceptable atany given time. The emergence of newmethodologies and competitive pressures mayaccelerate the push. Box V.7 lists the maindeterminants of R&D outsourcing.

Another way of externalizing R&D workis to establish a strategic alliance withcompetitors, suppliers or clients. Data show thatas of 2001 (the last year for which data areavailable), cross-border R&D alliances hadproliferated (chapter IV). To some extent thedrivers for strategic R&D alliances resemblethose that have led to increased outsourcing ofR&D activities. Alliances can be seen as a wayof sharing the risk involved in R&D, accessingcomplementary proprietary assets and copingwith situations where patenting may not be aneffective option (Dunning and Narula 2005, p.133). R&D alliances tend to emerge when partnercompanies share complementary capabilities, andthese alliances create a greater degree ofinteraction between the partners’ respective pathsof learning and innovation (Mowery et al. 1998,Cantwell and Colombo 2000, Santangelo 2000).Another reason to form an alliance in the areaof R&D is to explore new technologicaldevelopments more rapidly than what would bepossible independently. Strategic alliances mayhere provide “an attractive organizational formfor an environment characterized by rapid

innovation and geographical dispersion in thesources of know-how” (Teece 1992, p. 20).

2. Greenfield versus acquisition

If a company opts for the internalized routeto R&D internationalization, it still needs todecide whether to set up a new “greenfield”activity or to acquire one that already exists. Thepreferred mode here depends on several factors,including the purpose of the R&D, theavailability of suitable targets, the competitivesituation and other features specific to theindustry. Greenfield investment tends to dominatein R&D expansion abroad (chapter IV).

Greenfield entry is the most common modewhen setting up adaptive R&D abroad, as suchR&D is closely attached to the productionactivity. However, if for example a companyacquires a production unit with the aim ofadvancing its market position in the host-countrymarket, some R&D activities may be includedin the transaction. Such takeovers havecontributed to the higher level of R&Dinternationalization of many companies (vonZedtwitz and Gassmann 2002). In this situationthe R&D strategy of the acquiring firm, as wellas the quality of the R&D work taken over, willinfluence whether or not R&D is centralized andmoved to the parent company (or to a sistercompany), or whether it remains and perhapsexpands in the host country (see also chapter VI).

In the case of technology-sourcing (orasset-seeking) FDI in R&D, acquisition maysometimes be the only way to access a foreigntechnology (or other attractions such as brandnames and government contacts). Studies offoreign affiliates of Japanese TNCs have foundthat acquired units tend to have higher R&Dintensity than greenfield establishments, possiblysuggesting that technology sourcing has been animportant driver for the acquisitions (Belderbos2003).

If the sourcing strategy involves theestablishment of a listening post in a foreigncentre of excellence, many firms may prefer toset up a local company from scratch. In order tochannel knowledge effectively to the parent, theR&D unit in the host economy needs to be wellintegrated with the rest of the TNC.

Most takeovers of R&D activities havebeen undertaken in developed countries. This isnot surprising, as the number of target R&D units

171CHAPTER V

can be expected to be considerably larger in thesecountries. This also resembles the patternprevailing for cross-border M&As in general(WIR2000). The higher the level of innovativecapabilit ies in companies considered foracquisition, the more attractive the M&A optionbecomes. The predominance of developedcountries in this area may also reflect similaritiesin specialization between firms in the home andhost countries. TNCs seeking to invest in R&D

abroad are more likely to choose acquisition iflocal firms with strong and similar competenciesare available.

Finally, industry-specific features influencethe choice of entry mode. A more concentratedmarket structure (globally or in any given market)may induce TNCs to acquire one of the leadplayers. Indeed, many mega mergers that havetaken place in the pharmaceutical and automotive

Source: UNCTAD.

Box V.7. The determinants of the make/buy decision in R&D

The following are the main determinantsof whether a firm chooses to maintain R&D in-house or outsource it.

• The tacit nature of the knowledge and theextent of coordination needed. Segments ofR&D where knowledge is highly tacit may bekept in-house if the cost of transfer andcoordination is significantly higher than thepotential benefits from outsourcing. However,the “separability” of processes may rise asknowledge becomes more codified, researchmethodologies evolve, technologies becomestandardized and coordination becomes easier.

• The degree of outsourcing of manufacturing.As companies specialize in core activities andoutsourced production, there may be a parallelincrease in the need for external sourcing ofinnovation.

• The significance of the R&D to the company’score advantages. Critical activities will notbe outsourced so as to protect competitiveness,core skills and the company’s reputation forinnovation. The costs of losing an innovativeedge may be huge for a market leader. The linebetween critical activities and others will,however, vary according to corporate strategy,the IPR regime and the level of trust betweenthe principal and the contractors.

• The need for specialized skills and equipment.Where product innovation becomes verycomplex and modular, involving a broad rangeof skills and expertise (as in semiconductordesign), it becomes impractical for a singlefirm to undertake R&D for all stages andfunctions. Product innovation then has to be“vertically disintegrated” among severalenterprises (Ernst 2003).

• The increasingly multidisciplinary and multi-technology nature of innovation. “The

increasing cross-fertilisation of technologiesacross disciplines and resultant broaderportfolio of competences has becomefundamental to the competitiveness oftechnology-based firms” (Narula 2001, p. 366).This is particularly true of manufacturingprocesses where several technologies interact,leading to a need to find external sources ofknowledge and innovation.

• The need for expensive routine engineeringand testing. This is a significant incentive foroutsourcing, particularly where the facilitiesneeded are capital-intensive. Outsourcing thenbecomes a way to cut fixed costs and reducerisk.

• The need for rapid innovation. In several fast-moving technologies, competitive successdepends on the ability of firms to get products(or modifications) rapidly onto the market. Theavailability of contract research facilities thatcan respond at short notice is a majoradvantage.

• The need to cut costs. In many consumer goodsindustries like electronics, lead firms have toprovide and constantly update a whole rangeof products. For example in the case of digitalcameras, “to get shelf space at a Best Buy orCircuit City often means brand-namecompanies need a full range of models, froma $100 point-and-shoot digital camera with 2megapixels, say, to a $700 8-megapixelmodel… competition can reduce hit productsto cheap commodities within months. So theymust get out the door fast to earn a decentmargin… Such pressures explain outsourcing’sgrowing allure. Take cell phones, which arebecoming akin to fashion items. Using a pre-designed platform can save 70% ofdevelopment costs off a new model.” (Engardioand Einhorn 2005, pp. 56-57).

172 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

industries have been motivated by a desire toachieve synergies in marketing and distribution,but also in R&D work. In industries characterizedby oligopolistic competition, there may bestrategic motives for firms to acquiretechnological assets of rivalling firms in a bidto pre-empt other firms (WIR2000). The M&Aroute is more attractive where speed in accessingthe technology or innovative strengths in a hosteconomy is an important consideration.

* * *

To sum up, the main driver for R&Dinternationalization by TNCs remains the needto adapt products and processes to conditions inhost-country markets. However, the recentincrease of R&D by TNCs in selected developingcountries, especially in Asia, is driven by acomplex set of factors:

• pull factors , such as a growing market,availability of large talent pools atfavourable costs and developing Asia’semergence as a global production base insome industries;

• push factors, such as shortages of skills inspecific categories in home countries, risingcosts and complexity of R&D, greater

competitive pressure that forces TNCs toinnovate more without increasing costs;

• policy factors, such as host-country effortsto strengthen their NISs, to invest ineducation and to use targeted investmentpromotion and incentives;

• enabling factors, including advances in ICT,investment and trade liberalization, all ofwhich make it easier for firms to restructuretheir operations internationally, while at thesame time adding competitive pressure onfirms to do so.

As a result , this new form of R&Dinternationalization can be seen as a logical nextstep in the increasingly globalized productionsystems of TNCs. The process greatly resemblesthe kind of international restructuring that hastaken place in export-oriented manufacturing(WIR02) and services (WIR04) where TNCs seekto improve their competitiveness by exploitingthe different locational advantages of countries.In the annex to this chapter the case of thesemiconductor industry is used to illustrate howthe interaction of the various factors has led tothe growth of chip design in Asia. As noted inthe next chapter, this trend offers importantbenefits to countries that are affected, but mayalso give rise to concerns.

173CHAPTER V

Chip design is a good example to illustratethe complex interaction of factors currentlyfavouring the expansion of innovative R&D indeveloping countries (Ernst 2003, 2005a). Chipdesign not only creates the greatest value in theICT industry while requiring highly complexknowledge, it also involves a generic technologythat affects a large number of user industries,including high-value services. The chip industrywas one of the earliest to globalize productionand it has been one of the most dynamic in worldtrade. Now it appears that design anddevelopment work in this industry is followingon the heels of manufacturing by moving towardsAsia.

Chip design has recently moved fromcentres of excellence in the United States, Europeand Japan to sites in some developing countries,notably in South-East and East Asia. Frompractically nothing during the mid-1990s, thisregion’s share of semiconductor design reachedaround 30% in 2002 (iSuppli 2003, p. 21). South-East and East Asia are now the fastest growingmarkets for electronic design automation tools,expanding by 36% in the first quarter of 2004compared to 5% for North America (which has60% of the world market), 4% for Europe, and-2 % for Japan (EDA Consortium 2004).Developing Asia is not only undertaking morechip-related R&D, but also the levels ofcomplexity are rising in terms of the line-widthof process technology (measured in nanometres),the use of analogue and mixed-signal design(substantially more complex than digital design),the share and type of system-level design (e.g.system-on-chip) and the number of gates usedin these designs.

This section explores the main driversbehind the offshoring of chip design, drawingon interviews with 60 companies and 15 researchinstitutions in the United States and Asia involvedin designing integrated circuits, as well assystems (Ernst 2005). The sample includes globaland regional carriers of chip design in Asia,including specialized research institutes and ninestrategic groups of firms that participate in global

Annex to chapter V

THE RISE OF CHIP DESIGN IN ASIA: A CASE STUDY

design networks.20 With the exception of someChinese companies, all the sample firms areTNCs.21 Their design activities are concentratedin a handful of clusters in Taiwan Province ofChina (Hsinchu and Taipei), the Republic ofKorea (Seoul), China (Beijing, Shanghai,Hangzhou, Suzhou, Shenzhen), India (Bangalore,Hyderabad, Noida/New Delhi), Singapore andMalaysia. The TNCs interviewed emphasized thediversity of functions performed by their Asiandesign centres, from routine (engineering support,adaptation, l istening posts for “technologymarketing”) to highly strategic tasks (globaldevelopment mandates for specific IT products,components and services). The tasks assigned toa design centre depend on its locationalcharacteristics, especially on the quality of theregional and national innovation systems.

The expansion of chip design in Asia hasbeen the result of the synergistic effects of pullfactors, policy factors, push factors and enablingfactors.

1. Pull factors

The cost of employing a chip designengineer in Asia is much lower than in the UnitedStates – typically only 10-20% of the cost inSilicon Valley (table V.1). But this is not the onlypull factor; demand factors are equally important.TNCs need to locate design near the rapidlygrowing Asian markets for communications,computing and digital consumer equipment inorder to interact with the lead users of newproducts. China is already the world’s largestmarket for telecom equipment (wired andwireless) as well as a critical test bed for thethird- (3G) and next-generation wirelesscommunication systems. It is also among the mostdemanding markets for computing and digitalconsumer equipment. As most of the equipmentis produced in China, the country has becomethe world’s third largest market forsemiconductors, generating substantial demandfor chip design. To the extent that China succeedsin setting alternative standards for 3G mobilecommunications, the need for undertaking chipdesign locally may increase to address the

174 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

specific requirements of such standards. In thiscontext all major global system companies inmobile communication systems are expandingtheir Asian chip design centres to establish theirown designs as de facto standards in the region.

Table V.1. Annual cost of employing a chip design engineer, 2002

(Dollars)

Location Annual costa

United States (Silicon Valley) 300 000Canada 150 000Ireland 75 000Republic of Korea <65 000Taiwan Province of China <60 000India 30 000China (Shanghai) 28 000China (Suzhou) 24 000

Sources: UNCTAD, based on PMC-Sierra Inc., Burnaby,Canada (for Silicon Valley, Canada, Ireland,India) cited in Ernst 2005.

a Including salary, benefits, equipment, office space andother infrastructure.

2. Policy factors

Policies cover a wide range of factors, suchas incentives, regulations, infrastructure andeducation – all designed to attract R&D and otherTNC innovative activities, including chip design,to particular locations (Ernst 2005, Armbrecht2003, von Zedtwitz 2004, Walsh 2003).22 TNCsinterviewed expressed concern about obscure andunpredictably changing regulations in some Asiancountries as well as weak IPR regimes.23

In terms of their home-country designactivities, Asian firms interviewed acknowledgedthat policies had played a powerful catalytic rolein building the critical infrastructure, supportingindustries and design capabilities that allowedthem to invest in and upgrade chip design (seealso chapter VII).24 The progress in chip designhas owed much to concerted efforts by bothgovernments and leading companies to establishnew sources of innovation and global standards.In telecommunications, the four leading playersin the Republic of Korea (Samsung, SK Telecom,KT, LG) are all trying to become major platformand content developers for complex technologysystems, especially in mobile communications.These efforts build on considerable capabilitiesaccumulated in public research labs (like the

Electronics and Telecommunications ResearchInstitute, ETRI), as well as in R&D labs of thechaebol, to develop complex systems. China’sattempt to develop an alternative 3G digitalwireless standard has created a powerfulincentive to expand Asian electronic designactivities.25 Thus government procurement hasbeen a powerful tool in driving innovation.

3. Push factors

A number of factors in developed countriesare also greatly contributing to pushing firms toexpand chip design in Asia. Three such pushfactors can be distinguished:

• Changes in the methodology andorganization of chip design;

• More outsourcing and multiple designinterfaces; and

• Changing skills requirements.

a. Changes in design methodologyand organization

Since the mid-1990s growing pressures toimprove design productivity, combined withincreasingly demanding performance features ofelectronic systems, have produced turmoil in chipdesign methodology.26 So-called “system-on-chipdesign” combines “modular design” 27 anddesign automation to move design from theindividual component on a printed circuit boardcloser to “system-level integration” on a chip(Martin and Chang 2003). A key driver behindthese changes has been a widening productivitygap between design and fabrication. While theproductivity of chip fabrication grew at an annualcompound rate of 58% from the 1980s until 1998,that of chip design reached only 21% (SIA 1999).

Chip design is also becoming increasinglycomplex. First , progress in manufacturingtechnology (“miniaturization”) has made itpossible to fabricate millions of transistors ona single chip. This increased complexity needsto be matched by a dramatic improvement indesign productivity (ITRS 2004, pp. 13-14).Second, the convergence of digital computing,communication and consumer devices has raisedthe requirements for essential features ofelectronic systems – they need to become lighter,thinner, shorter, smaller, faster and cheaper, aswell as more multifunctional and less power-consuming. These features are expected to

175CHAPTER V

continue to improve. At the same time companiesare forced to speed up time-to-market as productlife cycles have been reduced to only a fewmonths for some products. Time compression istherefore key in designing chips for such systems.

These changes in methodology haveincreased complexity at two levels of chip design:on the chip (“silicon”) and on the “system”.28

With growing design complexity, verifying at anearly stage whether the design can be used toproduce chips at acceptable yield andperformance has become critical. Some 60-70%of all system-on-chip hardware design time nowgoes into verification, leaving only 30-40% foractual device development. This has inflated thecost of design. For instance, the overalldevelopment cost for complex system-on-chipdesign can be as high as $100 million, a cost levelfew design companies and chip users can afford.

b. More outsourcing and multipledesign interfaces

Until the mid-1980s, system companies andintegrated device makers did almost all their chipdesign in-house. Since then system-on-chipdesign has fostered vertical specialization inproject execution, enabling firms to disintegratethe design value chain and disperse itgeographically. This has given rise to complex,multilayered global design networks with variableconfigurations, depending on the needs of aspecific project (box V.8).29 Until the early 1990s,design networks retained a relatively simplestructure. Over t ime, however, verticalspecialization increased the number and varietyof network participants, business models anddesign interfaces, bringing together design teamsfrom companies that drastically differed in size,market power, location and nationality.

A possible network might be comprised ofthe following players: a Chinese system companyfor the definition of the system architecture; anelectronic manufacturing supplier from TaiwanProvince of China; a United States integrateddevice manufacturer; a European “siliconintellectual property” firm; design houses fromthe United States and Taiwan Province of China;foundries from Taiwan Province of China,Singapore and China; chip packaging companiesfrom China; tool vendors for design automationand testing from the United States and India; anddesign support service providers from variousAsian locations.

Box V.8. Global design networks: the keyplayers

Three layers can be distinguished in global designnetworks:

• The network core encompasses five strategicgroups of firms: the system company, whichdefines the concept, but may well outsourceeverything else. The system-on-chip design maytake place within the “system company”, anintegrated device manufacturer, or a fablessdesign house (or a combination of these).a

Finally, chip fabrication and assembly, may beoutsourced to specialized suppliers.

• A secondary layer of the design network consistsof suppliers of tools (for electronic designautomation, electronic design automation;verification; and chip testing), silicon intellectualproperty licensors and design implementationservices.

• The third layer may involve system contractmanufacturers (both electronic manufacturersservices and original design manufacturers).

Source: Ernst 2005.

a Fabless companies do not manufacture their ownsilicon wafers. Rather, they concentrate on the designand development of semiconductor chips.

Vertical specialization within designnetworks has transformed the structure and thecompetitive dynamics of the globalsemiconductor industry. It has also increased theorganizational complexity of the networks. Atypical system-on-chip design team now needsto manage at least six types of design interfaceswith: system designers, sil icon intellectualproperty providers, software developers,verification teams, electronic design automationtool vendors and foundry services (fabrication).These design communities are rarely located inthe same place, which makes coordinationdifficult. As design teams become larger andgeographically dispersed, more formal interfacesare necessary for effective communicationbetween them.

With product life cycles often as short asa few months, system design requirements keepchanging rapidly. Communication problemsbetween hardware and software designers areparticularly serious. Hence proximity and face-to-face contact become critical: global designnetworks increasingly need to locate in Asia thosechip design stages that closely interface withlocal companies in mobile communications and

176 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

digital consumer electronics. As most of theworld’s leading chip contract manufacturers(“foundries”) are in Asia, this creates powerfulpressures to locate important stages of chipdesign in this region. New processes and changesin design methodology require closer interactionbetween designers and process engineers.

c. Changing skills requirements

Geographic proximity (in the establishedcentres of excellence in the United States orEurope) has sometimes been a disadvantage fordesign projects that require a large number ofcontributors with diverse knowledge sets andcapabilities. For TNCs involved in chip design,it has become costly to bring together a largegroup of diverse design communities in onelocation and keep them there. This is anotherreason for TNCs to offshore chip design to Asia.

Meanwhile, skill requirements and workorganization are growing in importance as pushfactors. Some TNCs interviewed expressedconcern that the supply of scientists andengineers in the United States and Europe isinadequate. As noted above, some Asiangovernments have pursued policies that increasethe availabili ty of well-educated engineers,scientists and managers. Engineers in some Asiancountries are trained to use the latest tools andmethodologies, and the main electronicsexporting countries in Asia have also set uptraining institutions dedicated to chip design.These efforts are especially advanced in Indiaand East Asia.

The expansion of chip design in Asiaappears also to have been influenced by aperceived inflexibility on the part of designengineers in the United States and Europe toadapt to a more structured (“automated”) workorganization (termed “innovation factory”). TNCshave likewise sought to lower design costs byincreasing the workloads and capping the designengineers’ salaries, which rose rapidly during theboom of the 1990s. Cost considerations clearlyfavour design work in Asia.

4. Enabling factors

Finally, new ICTs facili tate theinternationalization of chip design. Coordinatingspecialized design networks in Asia verticallycan involve high communication costs becauseof geographical distance combined with

differences in levels of development andeconomic institutions (labour markets, educationsystems, corporate governance, legal andregulatory systems as well as IPR protection).New ICT-enhanced information management hashelped reduce such costs, codify knowledge,enable remote control and allow more knowledgeto be shared via audio-visual media.

A second enabling factor is the spread of“transnational knowledge communities”, such asprofessional peer group networks, along withAsia’s large diaspora of skilled migrants and “ITmercenaries”. These networks help share complexdesign knowledge and provide experience andlinks with markets and financial institutions.

* * *

In sum, in the case of chip design acombination of pull, push, policy and enablingfactors is creating a compelling case for TNCsto shift more of their design work to Asia. Thetrend is still at an early stage but is set to deepen.Over the past few years all interviewed TNCsmade substantial investments in chip design inAsia and are planning further expansion.

Notes1 “The establishment of international R&D networks and

the management of transnational R&D projects are non-trivial and risky endeavours. The principal challengesare imposed by physical distance among R&D units,as well as between R&D units and corporateheadquarters. Distance impacts communication in termsof frequency and quality, raises transaction costs, andintroduces principal-agent related difficulties” (vonZedtwitz and Gassmann 2002, p. 570).

2 For example, the Chinese automobile manufacturer,Dongfeng Motors, has established listening posts inthe United States, Germany, United Kingdom andFrance for the purpose of being close to majorcompetitors and their technological bases (von Zedtwitz2005).

3 Similar conclusions were drawn in another study ofthe largest R&D spenders. Adapting products to localrequirements, learning from foreign lead markets andcustomers, keeping abreast of foreign technologies,and gaining access to skilled researchers and new talentwere the major reasons for internationalizing R&D(Roberts 2001).

4 “Innovative Asia: how spending on research anddevelopment is opening the way to a new sphere ofinfluence”, Financial Times, 9 June 2005.

5 Conventional technologies included chemicals,pesticides, fertilizers, pharmaceuticals, engineering,hygiene and health-care products, and branded

177CHAPTER V

consumer goods. New technologies includedelectronics, ICT, software, biotechnology and solarenergy (Reddy 2000).

6 “Wipro: R&D budgets falling, interest in globaloutsourcing rising”, Information Week, 1 April 2005( w w w . i n f o r m a t i o n w e e k . c o m / s t o r y /showArticle.jhtml?articleID=160401375).

7 For a review of changes in the export competitivenessof countries, see WIR02.

8 See annex table A.V.1 for data by country.9 China’s tertiary enrolment rate rose from 5% of the

age group in 1995 to over 20% in 2004.10 According to China, Ministry of Education 2004.11 The professional groups included engineers, finance

and accounting specialists, analysts, life scienceresearchers and professional generalists.

12 Proximity to regional markets has been the mostimportant factor attracting foreign R&D activities toSingapore. The second most important factor, however,has been the availability of personnel that can besourced freely within the country and from abroad (Toh2005, p. 16).

13 Public research institutes are traditionally averse tosuch contract work and have to be restructured,upgraded, and given “hard budget” constraints tochange their orientation in order to respond to theshorter-term, practical needs of industry. This has beenaccomplished in India (chapter VII).

14 The connection between IPR regimes and the broadercategory of FDI is ambiguous.

15 See also chapter VII for a discussion of how developingcountries may use IPR systems to benefit more fromTNCs’ internationalization of R&D.

16 In 2001, the United States contract R&D industry spent$14.2 billion on R&D (about 7% of total industrialR&D and 20% of services R&D). Its R&D spendinghas been growing very rapidly, doubling over the period1998-2001 (United States, NSF 2004). In the UnitedKingdom, the contract R&D industry accounted for£428 million of R&D in 2000, up from £142 millionin 1992 (Morgan 2002). In 2000, contract R&Daccounted for 22% of services R&D in the UnitedKingdom, about one-third in Canada, Germany andSweden, 65% in Italy and 77% in the RussianFederation (United States, NSF 2004).

17 As noted in a study of DuPont’s outsourcing ofchlorofluorocarbons (CFC) research: “DuPont mayhave outsourced $5 million of the $400 million it spenton CFC research, but the company saved that amountmany times over by not doing the research in-house”(Paul 1998, pp. 1-2).

18 See report by Ernst and Young at http://www.ey.com/g l o b a l / c o n t e n t . n s f / I n t e r n a t i o n a l /Progressions:GlobalPharmaceuticalReport2004.

19 See Engardio and Einhorn 2005, pp. 53-54.20 These are system companies; integrated device

manufacturers (IDMs); providers of electronicmanufacturing services and design services (the so-called ODMs, or “original-design manufacturers”);“fabless” chip design houses; “chipless” licensors of

“silicon intellectual properties” (SIPs); chip contractmanufacturers (“foundries”); vendors of electronicdesign automation tools; chip packaging and testingcompanies; and design implementation serviceproviders.

21 Interviews were conducted with both parent companiesand foreign affiliates of firms from the United States,Taiwan Province of China and the Republic of Korea,while for Chinese and Malaysian firms, interviews wereconducted only with parent companies. In China thesample included State-owned enterprises, collectiveenterprises and private technology firms.

22 Most firms refer to aggressive incentives implementedin China. For example, in 2002-2003 chips designedby foreign and domestic companies in China wereeligible for a 14% value-added tax (VAT) tax rebate,which lowered the effective tax rate to 3% from thenominal VAT of 17% on sales of imported anddomestically produced chips. This policy created anartificial cost advantage for domestically designedchips, and was later abandoned.

23 More research is needed, however, on whether and howweak IPR regimes prevent TNCs from upgrading theirdesign labs in Asia, or if other motivations overridethese concerns.

24 This supports earlier findings in the literature. See,for example, Shen 1999, Lu 2000, Naughton and Segal2002, Mathews and Cho 2000, Hobday 1995, Ernst,Ganiatsos and Mytelka 1998, Ernst and O’Connor 1992,Ernst 1994 and 2000.

25 The TD-SCDMA standard was developed by DatangTelecom, a Chinese State-owned enterprise, and theResearch Institute of the Ministry of InformationIndustry, with technical assistance from Siemens. Toaccelerate implementation, Datang has formed a seriesof collaborative agreements: a joint venture with Nokia,Texas Instruments, the Korean LG group and Taiwaneseoriginal design manufacturing suppliers; a joint venturewith Philips and Samsung; and a licensing agreementwith STMicroelectronics that will provide the Chinesecompany with access to critical design building blocks(Ernst and Naughton 2004).

26 “Design methodology” is the sequence of steps bywhich a design process will reliably produce a designas close as possible to the design target whilemaintaining feasibility with respect to constraints.

27 In “modular design”, “parameters and tasks areinterdependent within units (modules) and independentacross them” (Baldwin and Clark 2000, p. 88).

28 “Silicon complexity” refers to malfunctions that resultfrom the growing scale and density of the circuit andthe introduction of new materials or designarchitectures. “System complexity” on the other handincreases with the transition to system-level designwith “exploding” multiple functions, an in smart phones(ITRS 2002, pp. 82-83).

29 For instance, designing an embedded micro-controllerfor a mobile handset requires a different global designnetwork configuration than the design of a graphic chip.

A. New developmentopportunities in the

making

R&D is among the highest value-addedactivities undertaken by firms. Its internationa-lization affects the allocation of knowledge andhuman resources across countries and createslinks between domestic actors and the R&Dactivities of TNCs. It deepens technology transfer– from simply transferring the results ofinnovation to transferring the innovation processitself. Until recently however, with the exceptionof some production support and adaptive R&Dfor local markets, FDI in R&D has been out ofthe reach of most countries outside the Triad. Thenew trend of TNCs setting up global R&Dfacilities in some developing countries is stillin its infancy, but it is important. It has significantlong-term implications for host and homecountries alike (table VI.1).

Internationalization of R&D can benefithost developing countries in several ways. It canserve as a training ground by providingchallenging, high-skill jobs to scientists andengineers. It can create new research skills andthereby help enhance human resources in a hostcountry. It can bring in new knowledge andresearch know-how, and it can generateknowledge spillovers to domestic enterprises andother organizations, thus stimulating an R&Dculture in a host economy. Growing R&Dcompetence can, in turn, help host countries moveup the value chain and into new areas of dynamiccomparative advantage. In an increasinglytechnology-based setting this can be of immensebenefit.

��������

DEVELOPMENT IMPLICATIONS

This does not mean that all developingcountries are able to seize these opportunitiesand reap the benefits; TNC R&D is going torelatively few countries (chapter V). Nor doesit mean that all its development benefits willmaterialize automatically. There are potentialcosts. The net outcome depends crucially on thetype of R&D involved and on the economiccontext, including the host country’stechnological capacities, and policy andinstitutional framework (chapter VII).

Overseas investment in R&D also haseconomic implications for TNCs and their homecountries. R&D in developing countries canenhance the innovative and productive efficiencyof TNCs by allowing them to combine theirtechnological strengths with foreign assets. Theymay be able to acquire new technological assetsand thereby enhance their global competitiveness.The home economy may benefit from increasedexports, reverse technology transfers andimproved R&D efficiency of their firms.Developing home economies can reap similarbenefits; indeed, the benefits to them may be evenhigher, because their enterprises can tap intoglobal innovation centres by establishing an R&Dpresence.

At the same time, R&D internationalizationmay trigger concern in home countries. Some fearthat, as lead firms expand their production andR&D activities abroad, the R&D of related andsupplier TNCs may follow, thus leading to a“hollowing-out” effect. As firms restructure theirR&D activities internationally, some knowledgeworkers may have to shift to new jobs, whichcould involve adjustment costs associated withcreating new skills and employmentopportunities. The entry of new locations as

180 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

potential hosts for mobile R&D activities alsoputs greater pressure on all countries to ensurethat their national innovation systems (NISs) arecompetitive.

The implications of R&D internationa-lization for both host and home countries dependprimarily on the extent to which it affects nationalinnovative capabilities. The NIS approach isuseful in examining the implications (Freeman1987, Lundvall 1992b, Nelson 1993). It is basedon the assumption that innovation and technologydevelopment result from complex interactionsbetween enterprises, universities and researchinstitutes, and government agencies. EnterpriseR&D is an important, but not the only, componentof the NIS: the ability of companies to innovateis intrinsically linked to the system in which theyoperate. Figure VI.1 provides a schematicdiagramme of an NIS. In its traditional form theNIS comprises only domestic actors. Howeverthe boundary, components, and interactionsbetween the main actors, change as R&D by FDIbecomes integrated into the NIS (Liang 2004),opening up a new channel through whichresources and learning can take place. Ifsuccessful, this can help transform a traditionalinnovation (science and technology) system intoone in which enterprises play a more importantrole. Since the TNCs that locate R&D overseasare often those engaged in high-technologyactivities like software, electronics and lifesciences, this may also help host countries to shiftinto these knowledge-intensive, fast growingindustries.

Different types of R&D (adaptive,innovative, technology-sourcing) have differentimplications for the NIS of host countries.Implications may also vary according to the mode

through which the TNC internationalizes its R&D– whether by means of FDI (greenfieldinvestment or acquisition), strategic alliances orsubcontracting (outsourcing). Each mode createsconnections to international knowledge networks,but the impacts on home and host countries differ.The impact also depends on the level of economicdevelopment of the host and home countries.There will be little or no impact on developingcountries that lack the basic production andadaptive capabilities needed for new productdevelopment (chapter III). On the other hand,innovative R&D by TNCs can enable countrieswith some manufacturing capabilities to climbthe value chain within existing industries andenter new industries. And it may help the moreadvanced developing countries to move fromdevelopment-oriented work to applied researchand eventually to basic research.

It is difficult to measure the impacts ofR&D internationalization by TNCs. Conceptually,the implications for home and host countries canbe examined in terms of their effects on thestructure and performance of their NIS, humanresources, knowledge spillovers and industrialupgrading. Broader effects (e.g. on income and,education) are also important but are beyond thescope of this report. The causal links betweenR&D internationalization and such aspects asproductivity in home and host countries, exportcompetitiveness and economic growth are hardto measure. The data are limited and mostly relateto developed countries. The phenomenon is stilltoo new in developing countries to allow a fullassessment, and the experience of developedcountries may not offer valid insights since thedrivers of R&D internationalization vary toomuch in the two cases (chapter V).

Table VI.1. Potential implications of R&D internationalization by TNCs

Potential benefits Potential costs

Host country Improved structure and performance of the NIS Downsizing of existing local R&D or losingContribution to human resource development control of technology(R&D employment, training, support to higher Unfair compensation for locally developed

education, reverse brain drain effects) intellectual propertyKnowledge spillovers Crowding out in the labour market, potentialContributions to industrial upgrading harm to basic research

Technology leakageRace to the bottom and unethical behaviour

Home country Improved overall R&D efficiency “Hollowing out” of domestic R&D baseReverse technology transfers and spillovers Disappearance of certain R&D jobsMarket expansion effects Technology leakage

Source: UNCTAD.

181CHAPTER VI

Figure VI.1. National innovation systems and FDI in R&D:a schematic diagram

Source: UNCTAD, adapted from Liang 2004, p. 171.

R&D spillovers – one of the key potentialbenefits – are particularly difficult to measure,and while more tangible indicators of knowledgecreation or dissemination such as innovations,patents or citations exist, they are imperfectmeasures.1 Finally, the counter-factual questionraised in the chapter is: what are the implicationsof R&D internationalization by TNCs ascompared with a situation in which suchinternationalization did not take place? Thisanalysis does not aim to compare the implicationsof R&D through TNCs with those of R&D byother actors. Rather, i t seeks to provide anassessment based mainly on case studies andconceptual analysis.

The following sections review the evidenceof the impact of TNC activities in R&Dinternationalization. Section B considers potentialhost-country implications, while section Cfocuses on implications for home countries.

Section D concludes with a discussion of thepossible implications for countries that are notparticipating in the R&D internationalizationprocess.

B. Implications for hostcountries

1. Effects on the structure andperformance of an NIS

R&D-related FDI leads to structuralchanges in the host-country NIS (figure VI.1).Foreign affiliates conducting R&D become a partof the enterprise segment of the NIS and interactto varying degrees with local firms, science andtechnology (S&T) institutions and governmentagencies, adding to the complexity of the system.

NISof the home country

International production system(global R&D network) of TNCs

NISof the host country

GovernmentNon-firm

institutions

Local firms

Foreignaffiliates

ForeignTNCs

182 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

They provide channels of resource-sharingbetween the TNCs and the host country, affectinglearning and innovation in the latter. As TNCsallocate more R&D resources to the localeconomy, the NIS becomes increasingly linkedwith the global R&D network of the TNC andwith corresponding innovation systemselsewhere.2

Enterprises are a core component of anNIS. In most developed countries they are themain innovators and the main implementers ofnew technologies in production. However, indeveloping countries, enterprises generallyperform little R&D; the bulk of it is done inuniversities and government research institutesand is often de-linked from the productive sector.This weakens the economic impact of R&D onefficiency, growth and competitiveness.

R&D-related FDI can help overcome thisabsence of an innovative enterprise sector – acommon weakness of developing-countryinnovation systems. Over time it is essential forenterprises to become lead R&D performers andfor other knowledge institutions to supplemententerprise effort by undertaking basic research,applied research under contract and othertechnical services. TNCs bring well-developedmethodologies and skills for conducting R&D.They also create demand for related services fromlocal firms. For example, the business researchculture introduced by foreign affiliates in Indiacontributed to the development of some high-technology industries there. Texas Instruments,the first TNC to be allowed to establish a whollyowned software affiliate in India in 1986, notonly inspired other TNCs to set up operationsin India but also spurred the growth of theindigenous software and business servicesindustry.3 The influx of Texas Instruments andother foreign investors opened new jobopportunities for Indian researchers in theinterface between science and business.

Foreign affiliates usually maintain closetechnological linkages with the parent companyand with sister companies. In a survey of 37TNCs with R&D activities in India in 1995, allforeign affiliates conducting R&D (in both newand conventional technologies) had linkages withthe parent firms’ R&D in their home countriesand 81% of R&D units in “new technologies”(mainly ICT, software and biotechnology) hadlinkages with parent firms’ R&D worldwide(Reddy 2000). These intra-firm linkages are achannel through which foreign R&D resources

(financial, human and knowledge) can enter ahost country NIS and potentially diffuse furtherto other actors. These resources may be veryexpensive to purchase in the market – in somecases they may not be available at all.4 Thusintra-firm linkages are potentially of greatimportance for the upgrading of the localinnovation system.

However, the transfer of R&D resourcesbetween a parent TNC and its affiliates does notautomatically lead to a diffusion of theseresources within the host economy. Linkagesbetween TNCs and domestic business entities arevital, and they only arise if the domestic firmshave sufficient innovative capabilit ies. Ineconomies such as the Republic of Korea andTaiwan Province of China, the upgrading fromassembly to design, development and researchwas mainly based on domestic efforts rather thanon the presence of foreign affiliates (chapter IV),although domestic enterprises and researchinstitutes interacted with TNCs in other ways.

In other economies, the relationshipsbetween foreign affi l iates and domesticenterprises are a core factor in the innovationsystem. Through such linkages, the transfer ofresources can be channelled to local companiesand so help improve their R&D efficiency. SomeR&D activities of foreign affil iates areundertaken in direct collaboration with host-country firms. Vertical linkages related to R&Dbetween foreign affiliates and their suppliers(WIR01) are particularly likely to generatespillovers because of the high degree ofknowledge intensity and uncertainty of suchactivities. The outsourcing of R&D to local firmsis another form of linkage. As R&D becomesincreasingly complex, these linkages maybecomes so important that they lead to thecreation of formal partnership whereby the scopefor learning and spillover benefits expandsfurther. The likelihood of partnerships increaseswhen companies have some complementarycapabilities (Mowery et al. 1998, Santangelo2000).

There are also potentially importantimplications from horizontal interactions betweenforeign affiliates and competing domestic firms.R&D by foreign affiliates adds R&D resourcesto host-country industrial clusters and may inducelocal firms to undertake more R&D to competebetter. It may also show local competitors howto conduct R&D more effectively. The basiccondition for this beneficial impact is the

183CHAPTER VI

existence of a competitive and innovativedomestic enterprise sector; this can ensure thatlocal firms rise to the challenge posed by foreignaffiliates rather than being crowded out by them(see also sections VI.B.4 and VII.D).

Foreign affil iates also interact withknowledge institutions such as local universitiesand public research institutes that undertake basicor applied research, produce R&D manpower andprovide technical services to firms (chapter VII).Foreign affiliates may collaborate with theseinstitutions (e.g. by providing financial supportand conducting joint research projects) (boxVI.1). Such collaboration can also benefit theR&D of other enterprises by raising the researchcapabilities of knowledge institutions, bringing

them into contact with industrial work andpromoting spin-offs.

Finally, by affecting the structure of theNIS and reallocating resources to moreproductive R&D, FDI in R&D may help enhancethe overall efficiency of enterprise R&D in a hostcountry. For example if the NIS initially has astrong focus on basic research, the entry offoreign affil iates conducting adaptive orinnovative R&D could help activate underutilizedknowledge potential (Manea 2002, Manea andPearce 2001). R&D efficiency can also beimproved if R&D by foreign affiliates is bettermanaged, better equipped, and directed to morecommercially feasible projects than that of otherenterprises in an NIS. The most positive impact

Box VI.1. Collaboration between foreign affiliates and local universities: selectedexamples

Source: UNCTAD, based on company information.

The following are some examples of R&Dcollaboration between TNCs and local universitiesin host countries.

Microsoft Research Asia partners withacademia and governments throughout the AsiaPacific region to foster innovative research,advance education and promote science andengineering. It pursues collaboration with localuniversities and relevant organizations throughfour avenues: research collaboration, curriculuminnovation, talent fostering and science exchange.In research collaboration it has established jointresearch labs at Tsinghua University, ZhejiangUniversity, Harbin Institute of Technology, HongKong University of Science and Technology andUniversity of Science and Technology of Chinato cooperate with Asian academia. It conductstheme-based project funding to help research inspecific areas.

Intel had more than 250 sponsored researchprojects under way at various internationaluniversities in early 2005. Its teacher trainingprogramme, launched in 2000, has offered trainingto more than 2 million classroom teachers in 30countries, and the company collaborates withministries of education or other governmententities to adapt the curriculum in some countries.

Seagate Technology in Thailand cooperatedwith Khon Kaen University to open the KhonKaen-Seagate Cooperation Research Laboratoryfor applied R&D in recording-head manufacturing

technology. The lab uses system level technologyand a systems research approach to broadenstudents’ knowledge and expertise. The lab willalso be a shared resource for both Seagate staffand students of Khon Kaen University who willbe working together on projects. Cooperationbetween the industrial sector and universitiesoffers opportunities to develop further and drivefuture growth in the hard disk drive and otherrelated industries in Thailand.

In Brazil, the University of Campinas in SaoPaulo collaborates with a number of foreignaffiliates in R&D. More than 250 partnershipagreements with private companies and 60agreements with public companies have beenestablished at the university to date. Amongparticipating foreign affiliates are Ericsson forthe development of technology of fibreglass foroptical amplifiers and Motorola for thedevelopment of professional capabilities inelectronics-related areas. Other agreementsinvolve foreign affiliates of Aventis, Bayer,Compaq, Hewlett-Packard, IBM, Monsanto,Novartis, Roche and Tetra Pak (UNCTADforthcoming e).

In Rabat, Morocco, STMicroelectronics hasestablished a training centre to train teachers andstudents from engineering schools and to providea syllabus that will help them contribute toinnovation activities in the semi-conductorindustry.

184 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

on the NIS structure and efficiency may beachieved if the foreign affiliates initiate projectsthat would otherwise not have been carried outbut that contribute to enhancing the specificstrengths of the local NIS (Pearce 2004). However,such benefits are not automatic; local innovativecapacities are among the most importantdeterminants of their extent and diffusion. Theability to make commercial use of results generatedthrough R&D in a host country depends on factorsthat can be influenced by government actions(chapter VII).

2. Human resource implications

While a good supply of highly skilled humanresources can attract FDI in R&D (chapter V), FDIin R&D can also help in the development of suchresources. TNCs generally have the most advancedcapabilit ies for conducting R&D, and theiraffiliates can make significant contributions bytransferring people with the necessary skills andmethodologies to host countries. In addition, theycan play a part in strengthening local humanresources through in-house training, supportinglocal education and collaborating with localuniversities. They can also facilitate a “reversebrain drain” by attracting back skilled nationalsworking abroad.

Increased R&D employment. R&Demployment by foreign affiliates is growing fast.For example, majority-owned foreign affiliates ofUnited States companies increased their R&D staffby more than one-fifth during the period 1994-1999 (chapter IV). Most of these jobs were createdin developed rather than developing countries, butthe rate of growth in the latter has been evenhigher especially since 1999. There was a rapidincrease in R&D employment in foreign affiliates,e.g., in China, the Czech Republic, India andSingapore, and recent survey data on FDI projectsand TNC strategies suggest there will be furtherincreases (chapter IV).

In China, for example, Motorola establishedthe first foreign-invested R&D centre in 1990 andhas so far hired a total of 1,300 engineers (boxIV.6). Philips has some 700 R&D staff, which isset to increase to 1,300 over the next two years.GE’s new global research centre in China wasformally opened in 2003, hiring 500 researchers;it is expected to employ 1,200 by 2005.5 Thereare similar figures in India, where, for instance,GE’s global research centre in Bangalore employs

around 2,400 people.6 High levels of educationare generally required for these jobs. For instance,in GE’s laboratory in China, more than 80% ofthe engineers hold PhD degrees, and in Bangalore60% of the employees have post-graduatequalifications in science. Also, foreign affiliatesoften offer better employment conditions: highersalaries, better working facili t ies and moresophisticated training (Zhang 2005).

Training. Many TNCs provide in-housetraining to their employees. Training undertakenby foreign affiliates conducting R&D can helpdevelop new and advanced skills among localengineers and researchers. The types of trainingmay range from on-the-job training to seminarsand overseas training, including at the parentcompany. For example, almost all the 250engineers and researchers recruited locally inThailand at the Toyota Technical Centre AsiaPacific (Thailand) (box IV.7) had been sent toJapan for training. In some host countries thegovernment has invited leading TNCs to help setup and run joint or cooperative training centres(chapter VII).

Supporting higher education. Some TNCsthat undertake R&D in developing countries to tappools of low-cost technical manpower supportlocal universities and engage in curriculumdevelopment and talent fostering. They may helpincrease or upgrade training in specific skills.Others provide internship and fellowshipprogrammes to high-performing students. Theirresearch collaboration with local universities canoffer a means of supporting higher education whilesimultaneously diffusing knowledge (sectionVI.B.3; box VI.1). However, host countries shouldensure that national school and university curriculado not become overly directed towards the needsof particular firms. The potential contributions byTNCs should be balanced against the risk ofbecoming too “asset-specific” in their R&D andeducation focus.

Human resource spillovers. Spillovers takeplace when trained employees move to other firmsor set up their own businesses. This is welldocumented from TNCs’ production activities,such as in the electronics industry in Malaysia(Hobday 1995). Spillovers from R&D activity havenot been analysed separately but are likely to besimilar. Research personnel trained in leading TNCaffiliates are bound to be highly prized by localfirms seeking to launch R&D. These effects onhuman resource development may be greater when

185CHAPTER VI

R&D by TNCs is linked to local production thanwhen it is conducted as a separate activity. Forexample, in India the main beneficiaries ofelectronics R&D by TNCs are probably theengineers directly engaged in research, whereasin China, where R&D is more often linked toproduction, larger spillovers may benefit localproducers and exporters.

Brain-drain effects. In some developingcountries the appearance of new careeropportunities in foreign affiliates (and domesticfirms that perform contractual R&D for TNCs)is contributing to a “reverse brain drain”. Manyscientists, engineers and entrepreneurs whomoved abroad to work in universities, R&Dinstitutions and TNC labs are returning home tosuch countries as China and India. The returningdiaspora often bring back knowledge of newresearch techniques and large-scale researchmanagement skills, in addition to their scientificknowledge. Some retain links with the firms orinstitutions abroad for which they worked: somebecome local managers of foreign affiliates orset up their own enterprises with contracts fromabroad. This has happened in Brazil, China, India,the Republic of Korea, Singapore and TaiwanProvince of China as well as in developedcountries such as Ireland. For example,Zhongguancun Science Park in Beijing hosts2,500 companies established by those returning

from abroad (box VI.2). In Taiwan Province ofChina, many companies were established bypeople who had worked abroad for TNCs (Linand Rasiah 2003).

The “reverse brain drain” may prove to beone of the most significant benefits of R&Dinternationalization. However, this benefit willaccrue only to developing countries that have theskills, infrastructure and other requirementsneeded to attract R&D. It may be more difficultfor other countries to encourage their besttechnical graduates to give up jobs in moreadvanced countries and return home (chapterVII).

3. Knowledge spillovers from R&Dby TNCs

Given the nature of knowledge as a publicgood, it can be expected that the R&D activitiesof a foreign affiliate will generate some spilloverbenefits to other firms and institutions in a hosteconomy. R&D activity builds upon the stock ofknowledge, both explicit and tacit , in anenterprise. Some of the knowledge that TNCR&D creates may only benefit the TNC itself (ifit is protected by patents or is so specialized thatit cannot be transferred). However, someknowledge can “leak” out to and benefit the

Box VI.2. Reverse brain drain: the case of the Zhongguancun Science Park in Beijing

Source: UNCTAD.

Zhongguancun Science Park, China’s first andlargest science park and home to 40 universitiesand 130 research institutes, has attracted foreignas well as domestic R&D centres. By 2004, 41foreign-invested R&D centres had been establishedby such leading TNCs as Hewlett-Packard, IBM,Intel, LG, Lucent, Microsoft, Motorola, Nokia,Nortel, Oracle, Samsung, Siemens, Sony, SunMicrosystems and Toshiba.

Returning members of the Chinese diasporaplay an important role in these R&D centres. SomeTNCs have appointed Chinese researchers whopreviously worked at their headquarters as headsor chief scientists of their R&D centres in Beijing.This has contributed to attracting back top Chinesescientists in specific areas, at least temporarily. Forexample, three consecutive directors of MicrosoftResearch Asia (box VI.1) were highly qualified

Chinese scientists in computer science working withMicrosoft in the United States. Although locallyrecruited researchers provide the main manpowerfor the activities of foreign-invested R&D centres,expatriate staff, particularly overseas Chinese, area valuable complement with their knowledge andexperience from working abroad. When somereturning diaspora leave foreign affiliates and joinlocal research entities or establish their owncompanies, they contribute further to theenhancement of local innovative capability.

For some returnees who decide to establishtheir own businesses, these foreign-invested R&Dcentres may also become important customers. Infact, out of the 14,000 high-technology firmslocated in the Park, 2,500 companies have beenestablished by graduates returning from abroad.

186 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

wider research community of a host country. Withthe establishment of foreign-invested R&Dcentres, tacit knowledge can be accessible locallyto domestic entit ies.7 Spillovers of tacitknowledge may be particularly valuable for a hostcountry. Tacit knowledge plays a critical role inR&D but is difficult and costly to create locally.

There may be some tension between theinterests of the host country and that of the TNCwith regard to knowledge spillovers. While the

former would seek to maximize the knowledgediffusion to other firms in the economy, the TNCoften may want to minimize “leakages”. Manycompromises are possible (after all, the situationis very similar in the home country of the TNC).IPR protection can limit the loss to the firm, ascan other strategies to limit the cost of spillovers(box VI.3).

Box VI.3. Protecting against the risks of technology leakage

Source: UNCTAD, based on Cannice et al. 2003, 2004.

While host countries see inward FDI as ameans of building technological capabilities,TNCs are often reluctant to transfer technologyor engage in local technological activity that mayhelp local firms to become competitors. TNCstherefore try to limit the ability of localcompetitors to appropriate their proprietarytechnology by various means (box table VI.3.1).

TNCs may insist on full ownership of theiraffiliates, thus limiting access to knowledge bylocal firms that could otherwise be joint-venturepartners. While local companies can still poachemployees from foreign affiliates, their accessto knowledge is likely to be more limited thanif they were able to share ownership and therebyhave their own people working on all activitiesin the foreign affiliates.

TNCs generally protect their corecompetencies (technologies) more than their non-core competencies, and are more willing totransfer the latter to foreign affiliates, outsourcethem or develop them in collaboration with local

partners. This need not mean that non-coretechnologies are obsolete or of low value to thehost country; they may be new and valuable butperipheral to the TNCs’ core activities.

TNCs may transfer some core technologyto foreign affiliates, which then work to improvetheir production through local process R&D. Theymay protect against its appropriation by makingthe outcome and production dependent on theparent firm, such as through local engagementin component production that has little valueexcept if combined with other components thatthe TNC produces elsewhere. TNCs may decideto develop new technologies using a system ofmultiple locations in which no foreign affiliatehas access to the full technological system.

TNCs may also transfer technology tacitlyrather than explicitly, thus slowing its absorptionby local employees and its re-transfer to a localpartner. This gives the TNC more time to developnew competencies while slowing the affiliates’development of their own R&D capabilities.

Box table VI.3.1. Actions by TNCs to limit risk of spillovers in a host country

Action Potential effect

Enter with wholly-owned Reduces monitoring costs and risk of loss because of difficulty for outsideoperations companies to become sufficiently knowledgeable about the technology in

order to appropriate it.

Transfer non-core technology Lower costs of loss to transferer from misappropriation, but transferees mayof low value to transferer be satisfied because of the asymmetrical value of the technology.

Transfer core (high value) If appropriated, the value is low for the transferee because the technologybut dependent (incomplete) can only be used in conjunction with complementary technologies held bytechnology the transferer.

Transfer technology in tacit Even if employees within the affiliate understand the technology, their transferrather than explicit form to another organization is slow because they must transfer it tacitly as well.

187CHAPTER VI

Knowledge spillovers can take placeprimarily through the mobility of labour,enterprise spin-offs and demonstration effects.If foreign affiliates are “embedded” in the hostcountry NIS, with close interaction betweenforeign affiliates, domestic firms, universitiesand research institutions, the scope for spilloversincreases (section VI.B.1).

As mentioned above, employee turnoveris one of the principal ways in which technologyand knowledge spill over to the domesticeconomy. This can be particularly valuable indeveloping countries, as this diffuses skills andexperience that are difficult to gain in other ways.It is particularly significant for R&D, since tacitknowledge is embedded in the knowledge andexperience of individuals rather than in hardwareor capital equipment. The extent of such diffusiondepends on whether domestic firms are asattractive employers as foreign ones. Indeveloping countries, foreign affiliates often offerbetter salaries than local competitors.8

Local firms or institutions can, of course,improve their attractiveness. For example, aresearch director of a TNC R&D centre in Chinarecruited a whole team of researchers back to theChinese Academy of Science, in part by offeringthem the opportunity of doing independentresearch. “Examples of individual seniorresearchers leaving TNC R&D centres to joinlocal companies are numerous, and it willcontinue to occur over the years” (Chen 2004,p. 37). In Malaysia, engineers who worked inlocal affiliates of TNCs like Motorola, TexasInstruments or Intel subsequently moved to R&Dmanagement jobs in local firms (Rasiah 1996).

Another channel for spillovers are “spin-off” firms or innovations from foreign affiliates.China Techfaith Wireless, China’s largestindependent R&D company for the design ofmobile phones, was formed by a 14-person teamthat left Motorola China in July 2002. The spin-off was later listed in NASDAQ in May 2005.9

Photonic Bridge, another R&D firm in China, wasfounded by a team of engineers and researchersfrom Lucent.

Knowledge spillovers are inherentlydifficult to measure. The few existing studies arebased mainly on data related to R&D by foreignaffiliates in developed countries. Studies basedon patents citation data suggest that R&Dspillover also takes place from foreign affiliatesto local firms in the United States (Almeida 1996,Branstetter 2000).10 Similarly, another study

found that foreign R&D had a significant positiveeffect on domestic innovation in 147 geographicsubregions of Europe, Canada and the UnitedStates (Peri 2004).

According to one study, R&D by foreignaffiliates in Singapore has acted as “a windowthrough which local Singaporean inventors tapinto a much larger knowledge pool” (Hu 2004,p. 798). Inventors in Singapore relied more onpatents from TNCs with a presence there thandid inventors in other countries. This differencewas particularly marked in computers andcommunications industries as well as in electricaland electronics industries – industries in whichforeign affil iates play an important role inSingapore (Hu 2004).

Studies conducted in the EU under theCommunity Innovation Survey programme,however, do not provide strong evidence ofspillovers from R&D by foreign affiliates. Asurvey of Belgian foreign and domestic R&Dfirms in manufacturing found no significanttechnology transfers from TNCs to the localeconomy (Veugelers and Cassiman 2004). Whileforeign affiliates in the survey were more likelythan domestic firms to describe themselves as“innovative”, acquire technology internationallyand cooperate in R&D with local firms, they wereless likely to be “locally networked” and totransfer technology to the local economy.11 Asimilar picture emerged in France, where foreignaffiliates used fewer local sources and cooperatedless with local partners than did domestic firms(Sachwald 2004b). In Italy, foreign affiliates withasset-seeking innovation strategies were foundto interact more with local firms and institutionsthan those with adaptive R&D strategies (Balcetand Evangelista 2005, box VI.4). A study of theproductivity effects of inward and outward FDIin Swedish manufacturing found no evidence ofR&D spillovers at the firm or industry level(Braconier et al. 2000).

Apart from paucity of data andmethodological problems that might explain theapparent lack of evidence of spillovers, it hasbeen suggested that spillovers between countriesthat are already technological leaders may in factbe limited (Braconier et al. 2000, p. 18). Indeed,a recent study confirms that the impact of inwardFDI in R&D on innovation and productivityvaries by the level of development of the hosteconomy (AlAzzawi 2004). In newlyindustrializing economies, inward-FDI-inducedR&D spillovers weighted by patent citations had

188 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

potentially a strong positive effect on localinnovation and productivity, especially if the FDIcame from technologically leading countries. Indeveloped countries on the other hand, inward-FDI-related R&D negatively affected localinnovation but sti l l had positive effects ondomestic productivity (AlAzzawi 2004, p. 28):

“FDI-induced R&D spillovers can bevery important for less advancedeconomies. This is true both if innovationor productivity is our variable of interest.It seems that the further apart the sourceand recipient are in terms of level oftechnological advancement, the larger the

Box VI.4. Asset-seeking foreign affiliates create more local R&D linkages:the case of Italy

Source: UNCTAD, based on Balcet and Evangelista 2005.

Foreign affiliates accounted for about 33%of all business enterprise R&D in Italy in 2001(annex table A.IV.1). Their levels of interactionwithin the local NIS differs considerably accordingto their strategies — notably whether they seekto penetrate the Italian market based on importedtechnologies or to exploit local technological andhuman resources. Drawing on data from the thirdCommunity Innovation Survey for the period1998-2000, a recent study assessed thetechnological contribution of foreign affiliates andtheir innovative activities (Balcet and Evangelista2005).

A simple comparison with domestic firmssuggests that foreign affiliates have a relativelyhigh propensity to innovate, that they devote moreresources to innovation and R&D activities,cooperate more with other firms and institutions,and establish formal technological linkages withother firms within the enterprise group to whichthey belong. However, much of this is explainedby the fact that foreign affiliates areoverrepresented in science-based and scale-intensive industries; it is also explained by theirgreater size. In fact when controlling for thesefactors, the propensity to innovate was lower inforeign affiliates. Affiliates did show a relativelyhigh propensity to introduce new productinnovations, to patent and to spend more on R&D.Meanwhile, external linkages with universitiesand R&D centres were less frequent and importantfor affiliates than for domestic firms.

Out of 535 manufacturing foreign affiliatescontained in the Italian data-set, low-technologyaffiliates (which basically import the technologythey need from abroad) and foreign affiliates withno innovative activities whatsoever accounted for42% of the sample. Among the remaining 312firms, most affiliates applied adaptive R&D andinnovation strategies, mainly targeting the

domestic market. There is thus a heavyconcentration of adaptive, low-technology andnon-innovative strategies among foreign affiliatesin Italy.

In general, “adaptive affiliates” displayedweak external linkages, often involving intra-group technology transfers from headquarters.Local sources of knowledge such as universitiesand R&D centres were generally not perceivedas important. Innovation (and also R&D) effortsof these affiliates were incremental and adaptivein nature. All types of industries were representedin this cluster in Italy.

About 50 affiliates were characterized as“asset-seeking”. They had a higher level and scopeof technological interactions with the externalenvironment. Innovation activities were mostlyundertaken in cooperation with other firms andinstitutions, such as universities and R&D centres.The most innovative asset-seeking affiliates hada strong internal commitment to innovation andR&D. The other asset-seeking affiliates dependedmore on knowledge, competencies and expertiseabsorbed from the external technological andscientific environment. The first type was stronglyrepresented in science-based industries, whereasthe industry composition of the other asset-seeking group was very mixed. Asset-seekingbehaviour was found not only in science-basedbut also in medium-technology industries as wellas in specific technological niches where Italianfirms hold a comparative advantage. Suchindustries include mechanical engineering, homeappliances and traditional industries like textilesand footwear.

The Italian case thus suggests that an “asset-seeking” pattern of internationalization can bepursued by different types of foreign affiliates,as long as the host country has accumulated asufficient stock of sharable knowledge.

189CHAPTER VI

potential positive spillover fromknowledge flows on the recipient.”

The experience of Italy (box VI.4) and theCzech Republic suggests that the situation maydiffer by industry. In the Czech automobileindustry, for instance, TNCs helped create asophisticated innovation system because of theirlong-term commitment to upgrading their R&Dcapabilities, patenting as well as cooperation withuniversities and R&D labs (Srholec 2005b).12

The R&D intensity of both foreign and domesticfirms in this industry was well above the nationalaverage, reaching levels similar to those of otherautomobile producing countries like France,Germany, Japan, Sweden and the United States.By contrast, TNCs in the Czech electronicsindustry largely undertook contractmanufacturing, and invested little in R&D. TheR&D intensity of foreign affil iates wassubstantially lower than that of domestic firmsand below the average for manufacturing. Forthe economy as a whole, foreign ownership wasfound to have a significant negative impact onthe propensity to conduct R&D (Srholec2005b).13 As in the other studies noted above,foreign affiliates were more likely to cooperatewith non-affiliated firms abroad but less likelyto cooperate with domestic firms and institutions.

4. Contributions to industrialupgrading

The internationalization of R&D may helphost countries move up the value chain andenhance competitiveness. Industrialcompetitiveness involves four interrelated typesof upgrading: process upgrading, productupgrading, functional upgrading (new mix ofactivities or different activities in the value chain)and chain upgrading (moving to a new valuechain in products of higher technology intensity)(Kaplinsky and Morris 2001). Industrialupgrading usually follows the sequence fromprocess upgrading through product upgrading andfunctional upgrading to chain upgrading (Gereffi1999, Lee and Chen 2000).14 R&D by TNCs cancontribute to all four. The extent to which itcontributes to process and product upgrading inhost-country industries depends on where theresults of the R&D are applied. Adaptive R&Dand some innovative R&D directed towards thedomestic market may contribute directly toprocess and product upgrading in domesticindustry, while the impact of innovative R&D

for global markets is likely to be more indirect.15

For developing countries with relatively lowlevels of innovative capabilities, product andprocess upgrading of industries may beparticularly important.

R&D by TNCs may lead to functionalupgrading in domestic industries: from assemblywork to R&D, design and other knowledge-basedactivities. Countries specializing in labour-intensive assembly are vulnerable to competitionfrom countries with lower wages.16 Economicrents in the value chain are increasingly to befound in areas outside production, such as R&D,branding and marketing. But developing countriesthat seek to move up along the value chain toR&D functions and other knowledge-basedactivities often encounter bottlenecks such as alack of resources and local demand for theseactivities. By transferring resources to a hostcountry, providing demand for R&D outcomesand stimulating the business innovation culture(sections VI.B.1 and VI.B.3), TNCs may helpdeveloping countries upgrade functionallytowards higher value-added activities.

R&D by TNCs may contribute to chainupgrading, from simple value chains to those forproducts involving more advanced technologies.Traditionally, low-income developing countrieswere considered to have a comparative advantageonly in low-technology industries. The emergenceof a developing country as a destination for theglobal or regional R&D centres of TNCs canchange the public perception of that country andhelp attract FDI in other knowledge-basedactivities as well. Indeed, countries that havebegun to attract innovative R&D by TNCs mayalready benefit from “reputation effects” as morecompanies start considering them for future R&Dexpansion. Some developing countries havesuccessfully built up more knowledge-intensiveindustries by leveraging R&D by TNCs. In China,for example, R&D by TNCs (box IV.8) and bydomestic companies (such as Huawei and ZTE)have contributed significantly to the rapidupgrading of the Chinese telecom equipmentindustry – from central office switches to mobiletelecommunications and other high-endequipment (Liang 2004). In Singapore, R&D byTNCs was a key factor in creating an innovationand industrial cluster around biomedical sciencessuch as pharmaceuticals and biotechnology (boxVI.5). Rather than remaining as exclusively low-cost manufacturing locations, these two countrieshave leveraged their relatively well-educated

190 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

populations and better innovation infrastructureto become centres of excellence for innovation.

R&D by TNCs can also contribute to theformation of industrial clusters at the regionallevel of a country. In the Pudong New Districtin Shanghai, for example, a complete value chainhas emerged since 2000, partly as a result of FDIinflows. By 2003 some 25 specialized chip designcompanies, four contract manufacturers, 14package and test companies, 22 equipmentsuppliers and some training and technical serviceproviders were present in the area.17 As of early2005, there were 129 chip design companies inShanghai employing 5,000 engineers andresearchers.18 Over time the cluster has madesignificant technological leaps in the area ofintegrated circuits and moved up the valuechain,19 and in 2004 sales of integrated circuitsincreased to above $2.4 billion, accounting forone-third of the national total. Government

policies at the local level significantly assistedthis process (section VII.D).

5. Potential concerns related toR&D internationalization

The potential costs of R&Dinternationalization for host countries depend onthe type of R&D and its motive, the mode of TNCentry to conduct R&D and the strength of the hostcountry’s innovation system. The main concernsrelate to the potential downsizing of R&Dfollowing cross-border M&As, unfair sharing ofintellectual property resulting from local R&D,crowding out of local firms from the market forresearchers, possible negative impacts of R&Dfragmentation, a race to the bottom in attractingR&D-related FDI and unethical behaviour byTNCs (table VI.1). These are taken in turn below.

Box VI.5. R&D by TNCs in the biomedical science industry in Singapore

In the “Industry 21 Vision”, a blueprint forSingapore’s economic strategy in the 21st century,the biomedical sciences industry was identifiedas a key growth engine for the country.a Sincethis initiative was launched in June 2000,Singapore has demonstrated rapid progress in theupgrading of this industry within a relatively shorttime span. Based on both domestic efforts andFDI by TNCs, Singapore has built world-classcapabilities across the entire value chain, fromR&D to manufacturing in biomedical sciencesand headquarters’ services in the biomedicalsciences industry. In manufacturing the overalloutput of the industry grew to $9.6 billion in2004. The total value added of manufacturing inbiomedical sciences was $6.1 billion (box figureVI.5.1), accounting for 21% of the country’s totalvalue added. Meanwhile, Singapore hassuccessfully obtained patents and developed newproducts in the biomedical sciences.

TNCs have contributed to the biomedicalsciences cluster in Singapore. They have playedan important role in industrial upgrading throughtheir R&D activities, ranging from basic research

Source: UNCTAD, based on ISPE 2003 and information from Economic Development Board, Singapore.

a This covers biomedical sciences, pharmaceuticals and medical technology.b For example, Eli Lilly invests $140 million in R&D and employs over 50 scientists and researchers.

to clinical development. Pharmaceutical companieslike Eli Lilly, Isis Pharmaceutical, VandaPharmaceuticals and Paradigm Therapeutics allconduct R&D in Singapore. Medical technologycompanies with an R&D presence include BD,Welch Allyn, Essilor, Siemens MedicalInstruments, Bracco, Applied Biosystems andFischer Scientific.

Box figure VI.5.1. Value-added of thebiomedical sciences industry in Singapore

(Millions of dollars)

Source: EDB Singapore.

0

1 000

2 000

3 000

4 000

5 000

6 000

7 000

2001 2002 2003 2004

191CHAPTER VI

Downsizing of existing R&D capacity andlosing control of technology. The internationa-lization of R&D is partly the result of TNCsacquiring companies that perform R&D.20 Suchacquisitions may lead to a reduction of R&Dactivity as part of rationalization programmes.Similarly, strategic R&D activities may berelocated as a result of a takeover; this is ofparticular concern to technology leaders but itmay also affect some developing countries ortransitional economies that have specialtechnological strengths.

A relevant factor here is whether acquiringand acquired firms are technologicallycomplementary or competitive. For instance, astudy of 62 firms in the EU found that there wasa reduction in R&D activity after a merger whenR&D activities were competitive (Cassiman etal. 2004). The remaining R&D became narrowerin scope (or more focused) and its time horizonbecame shorter. Key employees tended to leavemore often. These effects were stronger when thecompanies had been rivals before the merger.

In Latin America, R&D has rarely been themain reason for TNC entry, although manyacquired State-owned and private enterpriseswere R&D performers. In many cases, R&D wassubsequently downsized or closed entirely in amove to concentrate R&D activities atheadquarters or elsewhere within the TNCnetwork (Velho 2004, Cimoli 2001). In theautomotive and pharmaceuticals industries inBrazil and Argentina, some TNCs downsizedR&D but increased production (Velho 2004,Cimoli and Katz 2001). But not all takeovershave had the same outcomes. Two takeovers inthe auto parts industry in Brazil are illustrative.When the domestic producer of shock absorbers,Cofap, was acquired by Magnetti Marelli (Italy)in the 1990s, the R&D team was maintained,mainly because of their high level oftechnological competence. Conversely, in theacquisition by Lucas Varity (United Kingdom)of Freios Varga, a brakes producer, the R&D wasdismantled despite the competence that had beenaccumulated in the local firm. As an explanationfor these diverse results, it has been proposedthat brakes may require less local adaptation thanshock absorbers (Costa 2005). Some companies– including Ford (United States), Volkswagen(Germany) and Alcatel (France) — have reversedprevious decisions to close local R&D in orderto boost their competitive position in theBrazilian market (Queiroz et al., 2003; Costa

2005). In China also there are concerns relatingto the closure of R&D units in local firms thathave entered into joint ventures with foreignfirms.21

In Central and Eastern Europe, manycompanies were taken over by foreign TNCs aspart of privatization programmes. An UNCTADsurvey in 1999 covering 23 major privatizedcompanies found that the average annual growthof R&D expenditure fell from 23% to 14% afterprivatization, and R&D intensity (R&Dexpenditure as a percentage of sales) diminishedsignificantly (Kalotay and Hunya 2000).22 It ispossible that R&D expenditures were boostedbefore privatization to show better companyperformance before the sale, or that they werethe continuation of previous non-market-orientedand overstaffed programmes (ibid., p. 55). In oneprominent case, R&D activities were continuedand expanded: GE’s purchase of Tungsram inHungary initially involved layoffs but later ledto the company becoming GE’s centre for lightingactivities throughout the world, including R&D(ibid., p. 51).

The risks of R&D closure are likely to besmaller when FDI is undertaken to reap costadvantages from conducting R&D abroad or toaccess local technical skills and markets. Closuresdo not appear to have occurred to a high extentin R&D labs in developed countries such as theUnited Kingdom (Griffith et al. 2004). Similarly,a study of 35 companies privatized in eightEuropean countries found that while R&Dintensity decreased, R&D outputs (measured bythe number and quality of patents) increased(Munari and Sobrero 2005). There have beenseveral cases in the Canadian chemicals industryof TNCs reducing or closing local R&D afteracquisition; Shell closed its R&D capacity inOakville, and Diversey moved its R&D toChicago (Rugman and D’Cruz 2003). However,there are also examples of R&D expansion: theCanadian affiliate of Uniroyal Chemical (UnitedStates) retained a key role in the parentcompany’s global R&D, partly because of its hightechnical capacity.23

Unfair compensation for locally developedintellectual property. There may be concerns thatlocal firms, universities or research institutescollaborating with TNCs on R&D do not receivefair compensation for intellectual propertydeveloped locally, either before or afterpartnering with TNCs . Due to unbalanced

192 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

bargaining power, information asymmetry, marketfailures or institutional deficits, the contractualarrangements between TNCs and their localcounterparts may not reflect a fair allocation ofrights and responsibilities, to the disadvantageof local entities. This can lead either to unfairpricing of R&D inputs or to a biased allocationof ownership of the R&D outputs. Both issuesare closely related to IPRs.

The ownership of intellectual propertydetermines subsequent revenue flows in the formof patent fees or new product sales. On the onehand, host developing countries may fail to reapthe long-term financial benefits of FDI in R&Dwhen they do not have a fair share of ownershipof, and related economic rents from, the resultingintellectual property. Lack of ownership ofintellectual properties may also make adeveloping country dependent on TNCs for itstechnological progress. Moreover, a patent canbe framed to cover intellectual propertydeveloped by local research partners even priorto collaboration with a TNC. This could be fairif the local partner has given its consent and isappropriately compensated. However, the legalimplications of IPR protection may not be fullyappreciated by firms and universities indeveloping countries. If unaccustomed topatenting they may find it difficult to strike anappropriate deal with their foreign R&D partners,particularly in host countries that lack aneffective IPR system.24 The main approach toaddress these concerns is to strengthen relevantdomestic institutions (section VII.B) and theability of domestic firms and R&D institutionsto manage IPRs effectively.

Crowding out in the labour market andpotential harm to basic research. When foreignaffiliates enter a host location there may beconcerns about local research entities finding itmore difficult to attract or retain the best R&Dstaff, thus hampering their ability to innovate.25

In China, for example, some observers havenoticed a tendency for talented researchers toleave domestic companies and government labsto take up a career path in foreign affiliates’ R&Dunits (Simon 2005, p. 12). Even if the NIS as awhole would benefit , i t may represent anopportunity cost for individual local entities(research institutes, universities and enterprises).If the reallocation of human resources harms themanpower supply for basic research, the long-term efficiency of an NIS may also be negativelyaffected. Ultimately, what matters is the trade-

off between the contributions of TNCs to thestrengthening of the NIS on the one hand, andthe loss of skilled personnel to local R&D, whichmay or may not lead to a stronger NIS as a whole.The evidence on this is scanty, and it is not easyto assess the net impact.

Possible negative impact of fragmentationof R&D by TNCs. TNCs increasingly divide theirR&D activities into modules, allocating differenttasks to different countries. Some may confinetheir R&D activities in developing host countriesto low levels of skills or technology to protectvaluable proprietary technology. This can deprivehost countries of learning opportunities andreduce the spillover benefits. In Brazil , forexample, there is concern that the fragmentationof R&D is leading to a downgrading of humancapital in car production (Posthuma 2000). It hasalso been argued that fragmentation may bypassthe development sequence and limit the extentof real roots within the local NIS, making theR&D activity rather footloose (Pearce 2004). Onthe other hand, fragmentation may enable morecountries to participate in global R&D by TNCs.Moreover, economies of scale in researchspecialization could produce greater employmentin research and attract R&D by other TNCs ifthe country develops a good reputation forefficient research.

Race to the bottom and unethicalbehaviour. As competition for FDI intensifiesthere is a risk that governments will compete inoffering over-generous incentives to attract FDI.This could lead to losses in tax revenue or thelowering of regulatory standards (with associateddamage to the environment or workers’ welfare).One concern in this context is that TNCs tendto locate R&D in developing countries to takeadvantage of their relatively lax employment orsocial protection policies. In the pharmaceuticalsindustry, this could lead to the flouting of ethicalor medical standards found in developedcountries. TNCs may be tempted to conductclinical trials on new drugs in developingcountries where “the costs of conducting the trialsare lower and human subjects can be recruitedmore easily.”26 The issue here may be one of poorregulatory frameworks in host countries or it maybe chronic unemployment and poverty that makeclinical subjects willing to take health risks thatwould be unacceptable in developed countries.27

Meanwhile, there has been progress in theinternational harmonization of standards forclinical trials. TNCs, which depend mainly on

193CHAPTER VI

developed-country markets for profits,increasingly have to carry out multi-centre andmulti-ethnic clinical trials under theinternationally agreed standards (box VI.6).

C. Implications for homecountries

The home countries of TNCs also facebenefits and costs when their firms expand R&Dabroad. The benefits are that R&D abroad maylead to reverse technology transfers, lower costsand therefore increased R&D, leading toimproved competitiveness of the TNCs (whichcan also benefit other firms in the home country).The costs are that R&D internationalization maylead to a “hollowing-out” of domestic innovation,lost research jobs and leakage of valuableproprietary technologies. The net outcome is

difficult to predict. It depends on a range offactors: the motives for R&D internationalization,the degree to which the TNC is integrated in theNIS of home and host countries, and the levelsof development of home and host countries.

1. Improved overall R&D efficiency

As R&D grows more complex, it tends touse a more diverse set of information, skills andknowledge. This set may not be available withina single firm, or even a technology leader, orwithin a single country. Where this is so, R&Dinternationalization may be necessary in orderto conduct R&D efficiently by tapping a broaderrange of resources. The availability of researchmanpower or of a knowledge base abroad canaccelerate new product development. Lower costsin developing countries can make R&D moreeconomical. All these advantages to TNCs

Box VI.6. Clinical trials in India

Source: “Evidence regarding R&D investments in innovative and non-innovative medicines”, Financial Times, 14October 2003; Love 2003, “Eastern rebirth of the life sciences”, Financial Times, 10 June 2005.

Clinical trials – the approval process for newpharmaceutical products – are time-consuming,expensive and ethically difficult. They involverecruiting hundreds, often thousands, of peopleto volunteer for the testing of new medicines. Indiais an increasingly attractive destination for clinicalresearch for pharmaceuticals groups looking forfaster and more efficient ways to test drugs forwestern consumers.

India is well endowed with skilled R&Dpersonnel. It also has a relative abundance ofpeople with diseases that exist in developedcountries (including up to 30 million people withheart disease, 25 million with type-II diabetes and10 million with psychiatric disorders). Thisincludes a large pool of what are called “treatmentnaive” patients who have not yet been exposedto other drugs on the market. In addition, Indianrecruits are more likely to comply fully with thetrial process, unlike in developed countries wherea significant proportion of subjects drop out inorder to seek second opinions.

It has been estimated that firms can reducecosts by 20-30% by moving these R&D activitiesto India. Savings come from hiring clinical

researchers, nurses and IT staff at less than a thirdof wages in the West, in addition to differencesin the costs associated with the patients. Reflectingthis, it is estimated that the number of clinicalresearch organizations based in India increasedfourfold between 2001 and 2003. Indian firms,too, are participating in this new industrialactivity.

One factor apparently underpinning the shifthas been India’s newly adopted guidelines on“good clinical practices”, including the issue of“consent by the patients” in line with globalnorms. However, other commentators havequestioned what “consent” can mean in a drugtrial when patients are illiterate and might notadequately understand the experiment’s true risks;by definition, the drugs being tested haveunknown beneficial effects on the patient’s illnessor disease, and negative side effects are alsounknown.

There are some factors holding back thedevelopment of clinical research in India, suchas relatively slow approval processes. Anotherone is India’s reverence for animals, which makesit difficult to use certain animals (like monkeys).

194 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

potentially feed into the technologicalperformance of their home countries, and thusto their competitiveness and growth.

The efficiency gains for a TNC fromtapping into the competitively priced pools oftalent in Asia can be substantial. For example,a three-month, pre-clinical toxicology study onone compound might cost $850,000 in the UnitedStates but only $100,000 in India.28 Similarly,the collaboration between PalmOne (UnitedStates) and HTC (Taiwan Province of China) onthe Treo 650 smartphone helped reduce thedevelopment time of the product by severalmonths while decreasing the number of defectsby 50% (Engardio et al. 2005).

The internationalization of R&D can alsoallow home countries to retain and focus moreon higher value added activities, offshoring lesssophisticated or non-core innovative activitiesto developing countries (Reddy 2000). In thePalmOne case, resources in the United Stateswere focused on software while the hardwaredevelopment was shifted to HTC in TaiwanProvince of China.29

2. Reverse technology transferimplications

An important potential benefit to the homecountry from R&D internationalization is reversetransfer of technology, whereby knowledgeacquired by foreign affiliates through R&D (in-house, outsourced or collaborative) is channelledback to the home country. This knowledge helpsboth the TNC and the innovation system in whichit operates. However, such reverse transfers arelikely to be significant only if the host countryis technologically advanced (Kogut and Chang1991). Depending on the extent of diffusion athome, reverse transfers can improve theproductivity of the TNC, its vertically relatedenterprises (suppliers and buyers), its competitorsand the knowledge institutions with which itinteracts.

TNCs from the Republic of Korea andTaiwan Province of China have long located R&Dcentres in the United States, Europe and Asia togain access to new technologies (chapter V). Suchtechnologies have been applied in the homecountry to develop new products and processesfor global markets. More recently, companiesfrom China and India have set up R&D units inthe United States and Europe (chapter IV).

There are relatively few empirical studiesof the extent to which productivity growth inhome countries can be attributed to spilloversfrom overseas R&D, and most relate to developedcountries. The evidence suggests that the extentof reverse technology transfers hinges on thepurpose of the R&D. Studies of Japanese TNCssuggest that the scope for positive effects on theproductivity of firms in the home country is largewhen foreign affiliates undertake “innovative”R&D that tap into advanced knowledge centresabroad (Todo and Shimizutani 2005). AdaptiveR&D by foreign affiliates of Japanese TNCs,drawing on technology developed in Japan,served to improve productivity in the host countrybut did not contribute to enhanced productivityin the home country.30

TNCs from the United Kingdom that haveR&D investment in the United States havebenefited from reverse technology, and the effectswere particularly important in the case of R&Dunits set up to source technology (Griffith et al.2004). Meanwhile, foreign R&D by SwedishTNCs does not appear to have generatedsignificant spillovers in the home country, eitherat the firm level or the industry level (Braconieret al. 2000, Fors 1997), possibly because muchof this R&D is of the adaptive type drawing ontechnologies developed at home (Håkanson andZander 1986, Håkanson 1992).

A cross-country study of 152,000 firms in30 countries concluded that outward-FDI-inducedR&D had a positive impact on the home country’slevel of domestic innovation as measured bypatenting activity (AlAzzawi 2004). Suchbenefits were found in both developed countriesand in the newly industrializing economies.However, productivity benefits were found fornewly industrializing economies but not fordeveloped countries, suggesting that overseasR&D may be particularly beneficial for lessadvanced home countries.

3. Market expansion implications

Whereas adaptive R&D does not seem togenerate significant reverse knowledge transfersto the home economy, it may generate otherpositive effects such as promoting marketexpansion. Such R&D is typically performed toexpand sales in a foreign market by adapting aTNC’s products or processes to suit localpreferences and requirements. With the expansionof markets abroad, demand for material, inputs

195CHAPTER VI

and services procured in a home country forglobal operations is likely to increase.

In some cases products developed by localR&D cater exclusively to the local needs(Behrman and Fischer 1980, Bartlet and Ghoshal1991), while in others an expanded product lineas a result of local R&D may subsequentlybenefit sales in global markets as well (box VI.7).If local adaptive R&D evolves into innovativeR&D because a host market becomes a test bedfor product applications in the regional or globalmarket, or because it reaches a certain size, theoriginal adaptive R&D can open up opportunitiesfor expansion in other countries as well (box VI.8).

4. Home country concerns

The expansion of R&D by TNCs in theirforeign affiliates in the Triad, and, more recently,also in parts of the developing world, has givenrise to some concerns even among the mostadvanced home countries. The fact that TNCsnow consider a new set of locations as candidatesfor R&D activities has led some observers to callfor government intervention to mitigate possiblerisks associated with this development. Concernsare related to the possible consequences of R&Dabroad replacing domestic R&D, relating to ahollowing out of the home economy NIS and aloss of skills. A recent report from the AmericanElectronics Association is illustrative:

“As the United States takes its leadershipfor granted, countries around the worldhave caught on and are catching up.

While we begin to close our doors to thebest and the brightest minds, thesetalented individuals and the intellectualproperty and jobs they create here arelured elsewhere. As we cut funding forresearch and development (R&D) – acritical factor in the innovation that hasdriven our economy for a century – othercountries are investing in R&D, scientificeducation, and high-technologyinfrastructure.. . Americans may besurprised if the next revolutionarytechnology is produced abroad, but weshould not be” (American ElectronicsAssociation 2005, p. 5).

There may be cause for concern if TNCsreduce R&D at home due to perceivedweaknesses in the home-country NIS. Given therapid pace of technical change, such adjustmentsare often slower than the technological needs offirms, potentially resulting in “systemic inertia”(Narula and Zanfei 2004). Firms may thenacquire the technology they need from foreigncountries or invest in R&D abroad to draw onother countries’ NISs (Narula 2002). Theproblem, however, lies not in TNCs seeking toretain their competitive position, but in thestructural weaknesses of the domestic innovationsystem. The correct policy response would be toaddress structural weaknesses, not to preventlocal firms from competing effectively.

It is easy to overstate the risks of R&Dinternationalization. Innovating firms rarely shutdown their domestic R&D completely: this would

Source: Reddy 2000, pp. 138-143.

Box VI.7. Nestlé’s R&D centre in Singapore

Nestlé (Switzerland) established an R&Dcentre in Singapore in 1979 as part of its globalR&D network. Its main function was to developAsian-style convenience foods that werespecifically suited to the various cuisines,preparation techniques and eating habits withinthe Asia-Pacific region. The development ofculturally sensitive products such as food andbeverages requires local presence.

This R&D unit’s main activities focused oncreating new rice, cereal and noodle products formarkets in Asia and the world; developing newflavours through fermentation and enzymereactions; and bringing out new seasoning andcooking aids for the Asia-Pacific markets through

traditional food ingredients, spices and herbs. Itwas able to draw upon scientific knowledge heldwithin Nestlé’s global R&D network as well ason the specific knowledge related to productdevelopment.

The R&D also contributed to the expansionof the knowledge base of Nestlé’s global R&Dnetwork, relating to Asian cuisine and customerhabits. For instance, when Nestlé’s R&D unit inSweden developed the frozen vegetable product“Taste of Asia”, staff from Sweden went toSingapore to learn the cuisine. This product isnow marketed all over Europe. Similarly, stafffrom the Singapore unit assisted in introducingAsian noodle production in Europe.

196 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

risk losing valuable technological links at home,presumably the original base for the firms’competitive advantages.

Weaknesses of the home countryinnovation system may arise from the shortageof good researchers, the rising cost of conductingR&D or the lack of a manufacturing base withwhich researchers can interact. In science-basedindustries in particular, R&D may require acritical mass of researchers in differentdisciplines (De Meyer and Mizushima 1989). Ifthis critical mass is not available at home, TNCshave to locate R&D in countries that can offera suitable pool of talent. Even if it is available,bottom-line considerations may lead them to do

R&D abroad to lower costs or to interact withmanufacturing facilities. As manufacturing isoffshored, segments of innovative R&D have tomove with it. These factors have been importantin attracting chip design to East and South-EastAsia, for example (chapter V).

A growing global supply of skilled peopleat lower costs is a strong incentive for TNCs toexpand R&D abroad rather than at home. Forsome work categories this can lead to loss ofresearch jobs at home as well as downwardpressure on researchers’ wages. At the same time,given the growing need for R&D to respond toincreased competition in international marketsand to keep up with new technologies, the

Box VI.8. Mobile telecommunications R&D by TNCs in China

Source: UNCTAD.

a “Ten percent of Nokia handsets are designed by its Beijing centre, which is developing products for market fiveyears later”, West China Metropolitan News, 17 September 2004.

b “3G R&D distributed in nine cities”, Southern Metropolitan News, 16 November 2004.

Since the early 1990s, China’s mobiletelecommunications market has expanded rapidlyto become the world’s largest in terms of bothnetwork capacity and number of subscribers.Rapid infrastructure build-up has encouragedmany telecom equipment makers to invest in localproduction in the country. These enterprises alsoengage in local R&D in China (box table VI.8.1),which has come to play an increasingly importantrole in new product development.

Box table VI.8.1. R&D by selected TNCsin mobile telecoms technology in China,

2004

Number of R&D Number of R&DCompany centres in China employees in China

Motorola 15 1 300Nokia 5 800Ericsson 9 700Siemens 4 ..

Source: UNCTAD, based on Chinese newspaperaccounts and information from companies.

Initially the main function of these R&Dcentres was to adapt technology developed by theparent company to the specific marketrequirements in China. However, since mobiletelecommunications products are highlystandardized and the size and sophistication ofthe Chinese market has been rapidly increasing,

local adaptive R&D has evolved into globalinnovative R&D. For example, in the case ofmobile handsets, the Nokia 3610 model,introduced to the Asia-Pacific market in 2002,was the first product developed entirely by theNokia Product Development Centre in Beijing.Now every tenth mobile handset sold globallyby Nokia has been designed in Beijing.a Examplesof globally oriented R&D centres in China includeNokia China R&D Centre (1998), the MotorolaChina Research Institute (1999), Nortel ChinaR&D Centre (2001), Ericsson China Central R&DInstitute (2002) and Sony Ericsson’s global R&Dcentre in Beijing (2004).

Many of the R&D centres have capabilitiesin the area of 3G technologies and now developproducts for both the Chinese and global markets.Nine cities in China host 3G-related R&D centresowned by foreign TNCs or domestic companies(Huawei and ZTE), with emphasis on differentglobal standards recognized by the InternationalTelecommunication Union.b Although the ChineseGovernment has not granted 3G licences totelecom operators, 3G equipment developed andmanufactured locally by both foreign TNCs anddomestic firms has begun to supply the globalmarket. In this way the R&D activities in Chinahave helped the firms concerned expand theirbusiness in other locations as well, which in turnhas had positive effects on their respective homecountries.

197CHAPTER VI

increased internationalization of R&D may beparalleled by an increased demand for R&D skillsin the home countries as well.

Even if it is the less sophisticated or non-core R&D activities that are offshored todeveloping countries, some researchers at homewould have to be redeployed and some, at lowerlevels, might become redundant.31 A long-termworry is that this might lead to “the disruptionof the apprenticeship path”.32 New entrants toR&D will need more advanced skills to keepahead of competition from other countries. Thisprocess would entail adjustment costs andinstitutional changes to match education andtraining to needs for new skills.

The risk of technology leakage is anotherconcern. If R&D abroad results in the successfulimitation of TNCs’ technologies as well as ofother technologies developed in the home countryby foreign competitors, home countries may beworried that it may reduce the demand for theirproducts in the short term. In the longer term,a home country may fear losing control oversome key technologies, with an erosion of itsstrategic position in the global markets (OECDand Belgian Science Policy 2005).

It is important, however, to keep currentdevelopments in perspective. The volume of R&Dthat developing countries now attract is smallfrom a global perception. While there aresegments in which developing countries offerattractive conditions for R&D, this does not meanthat they have developed technologicalcapabilities to match those in developed countries(Reddy 2000). Although a larger share of highvalue-added, knowledge-intensive activities isbecoming subject to globalization, there is a longway to go before it can be considered a seriouscompetitive threat. It does however sharpen theneed for countries at all levels of developmentto ensure that their innovation systems have theskills needed to stay abreast of the technologyrace.

D. Conclusions

The internationalization of R&D by TNCsopens up new opportunities for developingcountries with strong skills and a technologicalbase to enhance the development of theirinnovative capabilit ies. It has importantimplications for developed countries as well asfor the world economy as a whole. It is still too

early to assess the full impact of thesedevelopments, but some implications are clear.

FDI in R&D can bring several benefits tohost countries. While the empirical evidence islimited, what exists suggests that such benefits– strengthening the NIS, promoting humanresource development, creating knowledgespillovers, upgrading industrial competitiveness– can be very important for developing countries.

Host countries attract innovative R&D byTNCs particularly in areas in which they haveestablished a competitive advantage. In Italy,TNCs are more likely to undertake innovativeR&D in medium-tech or low-tech industries. InIndia, strong domestic capabilit ies in thepharmaceuticals industry are now attracting TNCR&D in drug development. In China, similarly,the telecom equipment industry hosts some ofthe most innovative domestic firms as well assignificant R&D by TNCs.

At the same time, these benefits do notappear automatically. The most important factorfor realizing them is the absorptive capacity ofthe host country. Technological capabilities inthe domestic enterprise sector and technologyinstitutions are necessary not only to attract R&Dbut also to benefit from its spillovers. There maybe tensions between TNCs and host governmentsin that the former may seek to retain theirproprietary knowledge while the latter seek topromote as many spillovers as possible.

Although the benefits to developingcountries from R&D internationalization arelikely to outweigh the costs, the process can giverise to unwanted effects. Concerns may relate,for example, to the risk of foreign affiliatesattracting the best scientists and engineers frombasic research, or to unfair compensation of localcounterparts who collaborate with TNCs in R&D.These and other risks should be borne in mindby governments when designing andimplementing policies.

The nature of benefits to a host countrydepends on the type of R&D conducted, and onwhether the R&D is l inked to production.Generalizations are difficult, but a host countryis likely to benefit more when the results of R&Dare used in the host country and when the R&Dinvolves intense interaction between the TNC andlocal firms and institutions. R&D-relatedtechnology sourcing may give rise to someconcern among developed host countries oftechnology leakage. In developing host countries

198 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

the main potential costs are related to the riskof crowding out in the labour market, the closureof R&D units after acquisition, and insufficientcompensation for contributions to innovationwhen collaborating with TNCs.

The implications for home countries alsodepend on the type of R&D. It appears thattechnology sourcing and innovative R&D cangenerate significant knowledge spillovers to thehome economy, especially in developingcountries. The establishment of an R&D presencein leading technological centres abroad offersa potentially important way to link up with TNCR&D systems. Adaptive R&D abroad aimed atsupporting sales in foreign markets is also likelyto benefit home countries by improving thecompetitiveness of their TNCs and increasingindirect exports.

At the same time the expansion of R&Dto developing countries, motivated by weaknessesin the NIS of home countries or by lower R&Dcosts has given rise to concern in home countries,especially with regard to the risk of hollowingout and loss of jobs. Such offshoring is so newthat i ts assessment has to be tentative.Protectionist measures to limit the offshoring ofR&D by TNCs are unlikely to be effective inaddressing the root causes. In fact, restrictingthe ability of firms to raise their R&D efficiencywill have negative impacts on theircompetitiveness.

Instead, it will become more important toexplore new ways of collaborating with the newR&D locations, such as through joint researchprogrammes and outsourcing as well as throughinward and outward R&D-related FDI. Asdeveloping countries increase their number ofuniversity graduates, the historical near-monopoly of developed countries on scientistsand engineers and other highly educated workersis diminishing. Moreover, to the extent that alarger proportion of researchers and scientistsfrom developing countries decide to stay in theirown countries instead of migrating to Europe orthe United States, the latter economies may haveto rely more on developing their own domesticbase of human resources.

This makes it increasingly important fordeveloped countries to consider ways of makingtheir NISs more competitive, for example, byremoving bottlenecks and addressing "systemicinertia", and by identifying niches where theyare particularly strong. Similar to the case ofoffshoring of services in the broadest sense

(WIR04), R&D internationalization may requireappropriate policy responses to assist thoseworkers who are directly affected. Adjustmentto any change in employment patterns calls forgreater labour mobility and changes in the skillsprofile. In general, countries now face greaterpressure to make the necessary adjustments intheir institutional framework to enable theirworkers as well as their firms to move up thetechnology and skills ladder – also in the areaof R&D.

For the world economy as a whole, theinternationalization of R&D should help speedup the innovation process. By bringing morenational systems of innovation closer togetherit should also facilitate more cross-border flowsof knowledge and technology.

In the short to medium term, however, mostdeveloping countries are not in a position tobenefit from R&D internationalization. Manylack the skills and institutions to attract foreignR&D. Given the growing importance oftechnological and innovative capabilities forcompetitiveness, this may be a cause for concern.Countries that do not connect with these networksrisk falling further behind in terms oftechnological and innovative capabilities. Thereis no “quick fix” to this problem, but there arevital long-term policy issues that need to beaddressed now. The next chapter deals with someof these.

Notes1 For example, information may be exchanged between

foreign affiliates and TNC headquarters in the formof tacit knowledge or understandings that are notdescribed in a patent. Patent data may underestimatethe true degree of technology and knowledge transferthat has been possible. Similarly, patenting is arelatively new activity in many developing countries.Some countries may have been innovative but may nothave seen the importance of patenting their ideas.

2 For a discussion on the potential impacts of differenttypes of R&D by foreign affiliates on a host-countryNIS, see Pearce 2004.

3 See “A new transnational capitalist class? Capital flows,business networks and entrepreneurs in the Indiansoftware industry”, Economic and Political Weekly,27 November 2004.

4 TNCs tend to internalize their most valuabletechnologies rather than sell them to unrelated parties(WIR99).

5 Source: various news articles.6 This centre has filed 240 patents in the United States

and has already been granted 25 (see “Eastern rebirth

199CHAPTER VI

of the life sciences”, Financial Times, 10 June 2005).7 Tacit knowledge may include cognitive capacity,

experience and skills, or knowledge of routine,organizational structure, practices and norms.

8 See, for example, “Research labs power China’s nextboom”, International Herald Tribune, 13 September2004.

9 See “From the third type of fortune to the birth oftycoons”, New Fortune, 28 April 2005 (in Chinese),“Dexin lands successfully in NASDAQ, raising $142million”, www.tom.com, 7 May 2005 (in Chinese).

10 When a firm that is applying for a patent cites patentspreviously taken out by other firms, it indicates thatthere has been a path of learning and knowledge, fromthe first firm to those that followed its R&D trail.

11 Foreign affiliates made up the majority of the 445 firmsin the sample.

12 Foreign affiliates account for 47% of business R&Din the Czech Republic (chapter IV).

13 This result was sustained even after controlling forother explanatory factors relating to firms’ industrysector and location.

14 This accords with the upgrading process of enterprisesin some East Asian economies that have made thetransition from original equipment assembly (OEA)to original equipment manufacture (OEM), to owndesign manufacture (ODM) and own brand manufacture(OBM).

15 The impact of the innovative R&D on domesticinnovative capability and possible spillover effects may,however, be at least as important as for adaptive R&D.

16 Developing countries may even experience“immiserizing growth” if they become locked intostagnant incomes as producers face intense competitionand are engaged in a “race to the bottom” (Hubert 1995,Kaplinsky and Readman 2000, UNCTAD 2002a).

17 By early 2003 the Pudong New District had attracted66 FDI projects in microelectronics with investmentstotalling $8 billion. See “Shanghai Pudong New Districttries to establish a world-class industrial base inmicroelectronics”, China News Agency, 15 March 2003.

18 Shanghai Economic Commission “Shanghai’s ICindustry is leading the country”, 2 February 2005.

19 “Happiness and worries coexist in Pudong’smicroelectronics industry”, Shanghai Securities NewsCapital Weekly, 12 December 2003(www.stocknews.com.cn).

20 About 70% of all acquisitions are based on a market-

driven rationale (Kutschker 1989, p. 12, Granstrandet al. 1993, p. 416, Håkanson and Nobel 1993b, p. 402).

21 “Technology transfer from TNCs to China: new trendsand policy measures”, article posted on the websiteof MOFCOM 17 January 2005 (www.chinafdi.org.cn).

22 The companies were located in Croatia, the CzechRepublic, Hungary, Latvia, Poland and Slovenia.

23 In part this is attributed to the Canadian Government’ssupport of its research activities from 1962 to 1983(Rugman and D’Cruz 2003, p. 146.)

24 The experience of joint research with TNCs in theaerospace industry in the Russian Federation, forexample, suggests that local experience with thepatenting and marketing of innovative outputs, as wellas the legal and regulatory environment, are bothcritical in this regard (Ivanova 2004).

25 While an element of crowding out may also apply toinfrastructure, such physical capital can be expandedmore easily than human resources (Pearce 2004).

26 “Yet another sector embraces outsourcing to Asia: lifesciences”, International Herald Tribune, 25 February2005.

27 However, TNCs might be restrained from doing thisbecause if the drugs being tested are for consumptionin developed countries, clinical trials need to be carriedout on patients that have similar health and nutritionalstandards as those of the developed countries.

28 “Innovative Asia: how spending on research anddevelopment is opening the way to a new sphere ofinfluence”, Financial Times, 9 June 2005.

29 PalmOne’s designers provided the productspecifications, chose the key components and set theperformance needs of the product. HTC carried outmuch of the mechanical and electrical design (Engardioet al. 2005).

30 A study of Japanese TNCs’ R&D activities in theUnited States reached similar findings. A positiveimpact on the parent company’s R&D productivity interms of patents was noted for “research activities”by the foreign affiliates, but no such effect wasobserved in the case of “development-oriented R&D”(Iwasa and Odagiri 2004).

31 Such concerns have been voiced, for example, in thearea of software development (e.g. British ComputerSociety 2004).

32 “Innovative India”, The Economist, 3 April 2004.

A. Coherent policies andinstitutions make a

difference

The new trend towards the internationa-lization of R&D outside the Triad implies newopportunities for developing countries to connectwith the R&D networks of TNCs. However, todate most developing countries remain excludedfrom these networks. Thus the technological andinnovative capability gap between this lattergroup of countries and other economies continuesto widen. The challenge is to narrow this gap.

The experience of those developingcountries that have tapped into the TNCknowledge networks shows that policies andinstitutions are very important in TNCs’ decisionson where to locate their R&D. Investment inR&D is attracted more to “created assets” than“inherited endowments”, which means that it ispossible for governments to influence theoutcome of this decision-making process. Thischapter discusses how host countries can enhancetheir ability to benefit from R&D internationa-lization by TNCs. Chapter VIII considers theinternational framework for rule-making in thisarea.

The development of innovative capabilitieslies at the heart of economic growth anddevelopment (chapter III). While the preciseinterrelationship between technology andeconomic growth is open to debate, few, if any,countries have succeeded in achieving andsustaining high growth levels without investingin and exploiting technology. The promotion ofinnovation, with R&D being an integral part ofinnovative activity, is consequently becoming apolicy priority in countries at all levels ofdevelopment.

��������

THE ROLE OF NATIONAL POLICIES

The globalization process makes this evenmore important. A freer flow of goods, services,capital and labour adds competitive pressure onfirms — be they large or small, local ortransnational. Innovation is essential if firms areto use new technologies efficiently and staycompetitive in such an environment.

The ability of companies to innovate isintrinsically linked to the environment in whichthey operate. A useful framework for assessingthe role of policies in facilitating innovation isthe national innovation system (NIS) (chapterVI). An understanding of the NIS helpspolicymakers identify ways to enhance innovativeperformance and assist in pinpointing mismatcheswithin the system, both among institutions andin relation to government policies (OECD 1997b).Proper institutions – interpreted broadly to coverorganizations and the rules and incentivestructures governing innovation — are crucialto the effective functioning of an NIS (North1990, Metcalfe 1995, Edquist 1997).

Key policy objectives include providingan institutional setting that encourages andrewards innovation and strengthens innovativecapabilit ies in domestic enterprises andtechnology institutions. The ability to makecommercial use of results generated by R&D —by firms, universities or government agencies —depends on factors that can be influenced bygovernment action, such as the skills of the workforce, incentives for entrepreneurship and risk-taking, the quality of public institutions, accessto venture capital, trade and competition policiesand governance structures (Andersson 2005). Inaddition, governments can take measures to fosterinteraction among the various actors in the NIS.

As depicted in figure VII.1, various policyand institutional areas need to be addressed tomaximize the benefits that can be obtained from

202 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

R&D internationalization. The starting point isto build an institutional framework that fostersinnovation. Particular policy attention is neededin four areas: the availability, cost and qualityof human resources; the role of public research;intellectual property rights (IPRs); andcompetition policy. Efforts in these areas needto reflect the comparative advantage andtechnological specialization of each country aswell as the development trajectory along whicha country plans to move. FDI policy is also vitalto promote desired forms and impacts from FDI.Selective policies in this area include targetedinvestment promotion, performance requirementsand incentives, and science and technology parks.Finally, governments need to pay attention toboosting the capabilit ies of the domesticenterprise sector, notably through industry-specific policies and those relating to small andmedium-sized enterprises (SMEs). It is of coursealso important to ensure political and macro-economic stability and the proper functioning offinancial markets, but these aspects are beyondthe scope of this analysis.

While the long-term goals are similar,countries at different levels of development andwith different industrial structures have differentpolicy priorities. Throughout, this analysis seeksto draw lessons from countries — notably in Eastand South-East Asia — that have successfullymanaged to develop their innovation capabilities,sometimes, but by no means always, involvingTNCs in the process.

The chapter is structured as follows.Section B considers key policy areas that needto be addressed to strengthen the institutionalframework for fostering innovation with theinvolvement of TNCs, taking into account thedifferent comparative advantages anddevelopment strategies of countries. Section Caddresses the role of FDI policies, and sectionD discusses industry-specific policies and SMEpolicies for enhancing the benefits of R&Dinternationalization by TNCs. Section E considersthe role of home countries in enhancing theability of host countries to benefit from theinternationalization of R&D by TNCs. SectionF concludes.

Figure VII.1. National innovation systems and FDI in R&D: the policy dimension

Source: UNCTAD, adapted from Liang 2004, p. 171.

NISof the home country

International production system(global R&D network) of TNCs

NISof the host country

GovernmentNon-firm

institutions

Local firms

Foreignaffiliates

ForeignTNCs

Home-country policies (VII.E)• Promoting R&D internationalization• Strengthening the NIS of developing countries

Promoting FDI in R&D (VII.C)• Investment promotion• Performance requirements• R&D incentives• Science parks

Enhancing benefits from FDI in R&D (VII.D)• Industrial policy• SME policy• Incubation• Promoting linkages

Stengthening the institutional frameworkfor innovation (VII.B)• Human resource development• Research capacities of the public sector• Policies related to IPRs• Competition policy

203CHAPTER VII

B. Strengthening theinstitutional framework

for innovation

The policy agenda for promoting benefitsfrom R&D internationalization is wide. Thissection discusses four areas that are critical forstrengthening the institutional framework and thefunctioning of the NIS: human resources, publicR&D, the protection of IPRs and competitionpolicy.

1. Fostering human resources

The critical importance of human resourcesfor development is widely accepted. For example,a common denominator of the economic successof the various economies of East Asia is a strongemphasis on human capital at all levels (e.g.World Bank 1993). This applies directly topolicies concerning R&D internationalization.Company surveys show that access to skills isan overriding concern for most TNCs in decidingwhere to locate their R&D. As noted above(chapter V), the expansion of R&D in developingcountries – although still limited – is heavilyinfluenced by the availability of knowledgeworkers. The improved supply of highly skilledpeople is occurring as a result of deliberate andlong-term policies to raise educational standards,particularly at the tertiary level, as well as fromefforts to attract human resources from abroad.While education is important at all levels – fromprimary to tertiary – the discussion below focuseson higher education.

a. Development of skilled humanresources

Not all innovation requires people with auniversity education. Many important inventionshave been produced by people with limitedformal education. However, for R&D in largeprivate organizations such as TNCs, which seeka stream of incremental improvements in additionto new inventions, there is a clear need fortechnical and scientific skills developed throughhigher education (Baumol 2004). Moreover, thegrowing science base of many new industrialtechnologies makes it more difficult for the“gifted amateur” to innovate. To the extent thatcountries aspire to attract TNCs’ R&D, thedevelopment of relevant domestic skills andcapabilities is thus crucial. For countries that are

currently in a weak position to attract such R&D,skills development is even more relevant to boostdomestic capabilities.

In the past decade or so a few countriesin developing Asia, but also some othereconomies, have emerged as large sources ofworkers with tertiary education, and this trendis set to continue (chapter V). This is particularlyvisible in technical skills l ike science,engineering, mathematics and computing. China,India and the Russian Federation togetheraccounted for almost a third of all tertiary-leveltechnical students in the world in 2000/01.

While the number of qualified engineersand scientists clearly plays an important role inattracting R&D by TNCs, their quality andspecialization also matter. The skills required forapplied research in pharmaceuticals andbiotechnology are, for example, different fromthose required in automotive design. Similarly,the needs differ between different stages ofeconomic development. Policy-makers have toensure that the education system delivers the kindof skills that are the most in demand. Thus,efforts in the education area need to be closelycoordinated with policies in other fields. Forexample, the development of technicalcapabilities in the enterprise sector is importantto create local demand for university graduates.Without such demand, there is an increasing riskof people with higher education migrating toother countries in search of job opportunities.1

In this context, foreign affiliates can help byproviding new job opportunities (chapter VI).

One way to address this challenge is to usethe State as a “skills coordinator” (Green et al.1999). To accelerate skills formation in relevantareas, governments need an informed view of theskills that are in demand. Asia offers significantlessons. In Singapore, for example, the Ministryof Trade and Industry, the EconomicDevelopment Board and the Council forProfessional and Technical Education workclosely together to monitor future skills needs,drawing on inputs from foreign and localinvestors as well as from education and traininginstitutions. This information is matched againstnational policy objectives and used to buildtargets for various components of universities,polytechnics, schools and the Institute forTechnical Education (Green et al. 1999, p. 88).

In Latin America, the private sector hasexpressed concern that the skills generated byuniversities do not match its needs (Freeman

204 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

1995, de Ferranti et al. 2003, p. 228). Two outof three LAC researchers work in the publicsector, mostly in universities, and only one inten are employed in the business sector. Exceptfor Costa Rica — where around 25% ofresearchers are working in the business sector— that figure does not exceed 12% in any LACcountry. In terms of R&D spending, developmentwork (as opposed to basic or applied research)accounts for less than 30% in LAC as comparedwith more than 60% in countries l ike theRepublic of Korea or the United States (Velho2004, p. 17). Thus there appears to be amisalignment between the policies taken topromote skills and the demand from the privatesector, partly reflecting the current industrialspecialization towards natural resources andassembly operations based on low labour costs:

“Latin American and Caribbeanproduction patterns on the one handinduce private sector and enterprises toexpress a meagre demand for knowledgeand on the other hand lead domesticagents to mostly seek outward orientedlinkages and coordination, basicallyprivileging foreign companies andresearch laboratories that already havesound reputation and worldwide widelyrecognized experience in effective andefficient science and technology efforts.Thus a mismatch ensues between demandside needs and supply side offering,hampering policies’ impact ” (Cimoli etal. 2004, p. 11).

Education policies also need to evolve overtime as the demands from industry change andcountries develop. The case of the Republic ofKorea is illustrative. In the 1960s, a system oftechnical training was set up as part of broaderefforts to improve the infrastructure for scienceand technology. In the 1970s, the Governmentplaced emphasis on technical and engineeringeducation in the fields of heavy and chemicalindustries. In the 1980s focus shifted towards thetechnology-intensive industries and greaterefforts were made to bring back Korean scientistsworking overseas. Since 1990 more attention hasbeen given to promoting creativity, with thesetting up of the Creative Research Initiative in1997 to encourage a move from “imitation” to“innovation”. More recently, special incentiveshave been offered for universities to become lessteaching-oriented and more research-oriented.

It is important not just to educate peoplebut also to ensure that their skills are updatedcontinuously. This is especially true when thereis a mismatch between the supply and demandof specialized skills. Policies involving allstakeholders can help mitigate such problems ifall relevant actors recognize and accept the needfor actual implementation of specific policychanges (Vertzberger 2005, pp. 24-25). Policyintervention may be needed to re-skill and re-train production workers, technicians andengineers, expand the numbers of graduates withskills in special demand in industry,2 emphasizethe training of experienced managers, encourageentrepreneurs to upgrade their strategiccapabilities and align incentives for universitiesto interact with the private sector (e.g. throughinternships and sabbaticals).

Countries can involve foreign affiliates inthis process, for example by encouraging themto participate in joint projects with universitiesand other training institutions. This can be doneat different levels of education and training. CostaRica, for example, attracted a majorsemiconductor investment from Intel in 1996.Close links between Intel and the InstitutoTecnológico de Costa Rica helped securefinancial support from Intel to develop newprogrammes and increase enrolments ofengineering students (Mytelka and Barclay 2004).The auto parts maker Delphi collaborates withthe privately-run Tec de Monterrey in Mexicoto ensure adequate skills for its developmentwork in Ciudad Juarez.3 In India, Motorola workswith the Pune Institute of Advanced Technologiesto offer a postgraduate degree in advancedtelecommunications engineering with a softwarefocus (Reddy 2000). In Singapore, the efforts ofthe Economic Development Board to involveTNCs and foreign governments in trainingprogrammes helped ensure that they wererelevant and up to date (box VII.1). Without theseefforts the Board’s investment promotionactivities and subsequent upgrading into moreadvanced activities would have been crippled(Low et al. 1993, chapter 7).4

b. Importing human resources

Few countries can create all the skills theyneed; they therefore make use of a number ofexpatriate skills. In the OECD as a whole, some1.9 million students are enrolled in tertiaryeducation outside their country of origin (OECD

205CHAPTER VII

2004c, chart 3.6). The United States has been themain recipient of global knowledge migration inrecent decades. At the end of the 1990s over 50%of the post-doctoral students at MIT and Stanfordwere foreign citizens and more than 30% ofcomputer professionals in Silicon Valley had beenborn abroad.5 In Europe the growing importanceof the knowledge society and an aging populationhas made the attraction and retention of talenta key priority within the Lisbon agenda(European Commission 2004, p. 20). Also at thenational level, many European countries havetaken steps to attract foreign skills. For example,

the Government of France in 2004 launched aprogramme to attract the world’s leading expertsto growth sectors and to build teams around them(WIR04, p. 87); Belgium, Denmark, Finland,France, the Netherlands and Sweden haveintroduced special tax rules for foreign experts;and Germany and the United Kingdom haveestablished special programmes to facilitate easiermigration of foreign experts.6

Several developing countries are alsoseeking to attract foreign expertise. Singaporehas a liberal immigration policy to attract highly

Box VII.1. Engaging foreign affiliates in training: the Singapore case

In 1970 Singapore faced a serious andunexpected shortage of welders due to the rapidexpansion of its ship-repair industry.a TheGovernment addressed this problem partly byexpanding specially designed courses for thetraining of welders, and also by launching policiesto anticipate future needs for industrial skills (Lowet al. 1993). A number of joint industrial trainingprogrammes by the Economic Development Board(EDB) and leading TNCs were established: theTata-Government Training Centre (in 1972), theRollei-Government Training Centre (1973)b andthe Philips-Government Training Centre (1975).

The training programmes, with annualintakes of up to 100 people, were designed by theTNCs involved, which also managed theoperations through seconded directors and experts.All the programmes required in-plant training inthe TNC factories after completion of two yearsof in-centre training. Vocational institutessubsequently adopted many of the courses andcurricula. The EDB offered incentives (e.g. landand buildings and cost sharing) to induce TNCsto participate. It also launched a scheme thatrequired trainees to work in the TNC for a numberof years after the training, thereby assuring theTNCs a secure supply of skilled craftsmen. Whilethese institutes did not engage directly in

innovation or R&D, they contributed to thedevelopment of innovative capabilities inSingapore.

The model of joint training institutions wassubsequently refined, involving not only TNCsbut also foreign governments or governmentagencies. Between 1979 and 1982 the Japan-Singapore Technical Institute, the German-Singapore Institute and the French-SingaporeInstitute were started. In the mid-1980s a“transnational” approach was adopted, in whichresources and expertise were sourced from morethan one country.c The contributions of the TNCstook various forms (Low et al. 1993):

• Transfer of technology and know-how throughsecondment of experts;

• Training of EDB instructors and technical staffat the participating firms’ overseas locations;

• Assistance in curriculum and programmedevelopment;

• Donation and/or loan of equipment by theparticipating firms;

• Commitment by the participating firms toupgrade equipment and software; and

• Commitment to participate for a minimumduration of three years, subject to review andextension.

Source: UNCTAD.

a The shortage was a consequence of the closure of the Suez Canal, and the rapid growth of offshore oil explorationin South-East Asia. The demand for welders was further fuelled by the construction of new oil refineries and theexpansion of existing refineries.

b In 1982 this became the Brown Boveri-Government Training Centre following the failure of Rollei Werke.c For example, the German-Singapore Institute attracted the participation of several TNCs from the United States

(e.g. Hewlett-Packard), Europe (e.g. Siemens, Bull, Asea, Zeiss) and Japan (e.g. Seiko, Matsushita). The BrownBoveri-Government Training Centre was transformed into the Precision Engineering Institute in 1988, whichoversees a number of laboratories and manufacturing units (such as the Siemens-Nixdorf-EDB Centre for AdvancedTool and Die Making and the Japan-EDB Computer Numerical Control Laboratory).

206 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

skilled people to private firms and public researchinstitutes. By 2003, almost a third of doctoratelevel research and engineering scientists in thetertiary and public research institutions inSingapore were non-citizens.7 Such migrationcontributed to Singapore having the 7th highestratio of researchers per million inhabitants inthe world, just below that of the United Statesand ahead of countries such as France, Germanyand the United Kingdom. Singapore is spendingalmost $2 bill ion to recruit leading foreignscientists to develop research in the areas ofbiotechnology, genomics and nanotechnology.8

Many cities in China are actively seekingto attract highly skilled people in the largediaspora. For example, Shanghai is one of themost R&D-intensive areas of China. In 2002 theShanghai government announced a series ofmeasures, such as a preferential residential policyand a number of financial incentives, to attractuniversity graduates from elsewhere (Chen 2004,pp. 29-30). The Republic of Korea has not reliedmuch on skills immigration although variousefforts have been made to increase the return ofKorean scientists working abroad (box VII.2).9

Box VII.2. Policies in the Republic ofKorea to attract back scientists in the

diaspora

In the 1960s the Republic of Korea initiateda project to recruit Korean scientists workingabroad to meet the demand for human resourcesin science and technology. These efforts beganwith the establishment of the Korea Institute ofScience and Technology in 1966, and in 1968a specific project was launched to attract backqualified scientists in the diaspora. Asinducement measures they were offered modernlaboratories, competitive salaries and autonomyin their research. From 1968 to 1979, 238scientists returned to stay permanently in thecountry and another 255 scientists returnedtemporarily. These people played an importantrole in the 1970s and 1980s and contributed tocultivating new human resources in R&D. In1994 the work to attract qualified Koreanscientists from abroad was absorbed into a new“Invitation Program for Foreign Scientists &Engineers”.

Source: UNCTAD, based on Cho 2002.

What are the implications of the increasedmobility of highly skilled workers for the strengthof the NIS of a country? On the one hand, it mayaccentuate the brain drain from some developingeconomies, aggravating an already limited supplyof skilled human resources. Up to a third of R&Dprofessionals from developing countries residein the OECD area.10 On the other hand thediaspora is a potential source of skills,entrepreneurship, knowledge and capital for thehome countries. Bangalore in India has some35,000 “returned non-resident Indians”, manywith training and work experience in the UnitedStates.11 While some of these returnees joinforeign affil iates, others have set up newtechnology-intensive businesses in India (see alsochapter VI). To the extent that countries cancreate conditions that are conducive to suchreturn flows of human resources, the originalbrain drain can be turned into brain circulationwith positive implications for the NIS.

2. The role of research capabilitiesin the public sector

The public sector assumes an importantrole in every NIS, but notably in the area of basicresearch. In many developing countries, publicuniversities and research institutes even accountfor the bulk of R&D (chapter III), but such effortsare too often de-linked from the enterprise sector.For public R&D to provide spillovers and helpkick-start innovation by enterprises it is essentialthat enterprise R&D links with public R&Defforts, and that the public research institutespromote the spin-off of new companies.

Public research institutes can perform threeimportant functions within the NIS (Patel andPavitt 1994): undertake basic research andengineering/development work and produce newknowledge, some of which may be patentable;provide technical services (e.g. testing,consultancy) for firms as part of the infrastructurefor metrology, standards, testing and quality(MSTQ); and provide training to researchers. Ascountries develop, the nature of the workundertaken in public research institutes tends tobecome more sophisticated. In the mostdeveloped countries, universities and other publicresearch institutes assume key roles especiallyin the area of basic research. In general, publicR&D funding has played a more important rolein East Asia than in developed countries in

207CHAPTER VII

helping to develop innovative capabilities in keytechnological industries (Hu and Mathews 2003).The Industrial Technology Research Institute inTaiwan Province of China is a good illustrationof the role that public research institutes can play(box VII.3).

However, linkages between universities,public research institutes and enterprise R&D areoften weak (e.g. Ernst and Mowery 2004). Thisis a common situation in African countries. Astudy covering four African countries foundhardly any interaction between universities andthe private sector (Lall and Pietrobelli 2002).Moreover, the establishment of specialized R&Dinstitutes in Africa with the aim of supportingfirms in agriculture or manufacturing hasproduced meagre results (Adeboye 1997,Oyelaran-Oyeyinka 2004a). Public R&Dactivities tend to be insufficiently orientedtowards serving the needs of private-sectorclients, and industrial stakeholders are oftenunaware of the new technologies developed(Lewanika 2005). This failure has been explainedby the lack of an institutional base for innovation,a shortage of appropriate human capital, and theinability to tailor the activities of the institutesto the local context (Oyelaran-Oyeyinka 2004a).

In LAC, many public research instituteshave been in existence for many decades, mainlydealing with natural resources and health (Velho2004). There are also many industrial technologyinstitutes and some R&D institutes that focusspecifically on oil, telecoms, electricity and

space. However, in many instances their workhas not benefited the private sector directly.While the performance of the institutes varies,a common problem is that their researchers havelimited knowledge and understanding of thespecific needs of the private sector. To someextent this reflects the weak incentives providedto their researchers to interact with the privatesector (de Ferranti et al. 2003, p. 224; Cimoliet al. 2004). After the economic crisis in the1980s, however, public research institutes inmany LAC countries were required to increasetheir sourcing of funds from the private sector.As a consequence, where stronger links with theprivate sector emerged, institutes also began toconduct R&D that was more relevant to industry(Velho 2004).

It is possible to increase the relevance ofpublic research institutes to the private sector.India has a network of 38 laboratories and 45field/extension centres under the Council ofScientific and Industrial Research (CSIR),employing over 4,600 active scientists. In orderto revamp a system that had till then producedlittle that was of technological benefit to industry,the Government in the late 1980s launched amajor reform programme.12 It decided to limitthe level of public financing of the laboratories,and set a target for CSIR to earn 40% of itsexpenditures by selling research and otherservices to industry. The new annual budget ofeach laboratory was determined by its revenuesearning capability. As a result, the institutes’

A well-known public research institute thathas had a strong impact on innovation capabilitiesis the Industrial Technology Research Institute(ITRI) in Taiwan Province of China. Establishedin 1973 as a non-profit R&D organization, ITRIwas instrumental in establishing the integratedcircuit industry in Taiwan Province of China inthe 1970s by licensing fabrication technology fromRCA and transferring it to local companies. Itssubsidiary, the Electronics Research and ServiceOrganisation, was also instrumental in 1984 inhelping Acer develop what became the first 16-bit IBM-compatible personal computer fromTaiwan Province of China (Amsden and Chu2003).

Box VII.3. Spurring innovation in Taiwan Province of China

Source: UNCTAD.

According to ITRI’s president in aninterview in 1996, ITRI’s unique role was to trainprofessionals and then spin them off andencourage them to go into industry. Almost 10,000people have been trained at ITRI over the past20 years of which 73% joined industry. ITRI wasable to replace them by recruiting new graduatesfrom universities and expatriates from the UnitedStates. Personnel trained by ITRI made up thebackbone of the R&D and engineering workforcein the Taiwanese IT industry on an ongoing basis,together with the overseas Chinese returning toTaiwan Province of China with technical andmanagerial experience from companies anduniversities in the United States (Kim andTunzelmann 1998).

208 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

earnings almost tripled between 1992 and 1997to reach 2.1 billion rupees in 1996/97.13 By 2005,CSIR accounted for around 25% of all patentsfiled in India by Indians and a significant shareof all patents assigned by the USPTO to Indianinstitutions (chapter IV).14

Thus, while the building of innovativeskills in the public sector initially may be costly,it can provide vital resources for technologicaldevelopment if enterprise R&D grows andestablishes close links with public R&D.Institutes that have strong ties with the domesticprivate sector may also become partners withforeign affiliates. Foreign affiliates can interactwith the institutes in three main ways: by

subcontracting services to them; by undertakingjoint research projects or programmes; and byemploying skilled people from the institutes.Government-supported research institutes in theRepublic of Korea play an important role in thisregard, a role that has evolved over time (boxVII.4). In light of the internationalization ofR&D, there is a growing need to explore variousinternational dimensions of university-industrylinkages. Specifically, further analysis iswarranted of the role of TNCs as collaboratorswith national universities in developing countriesand of possible new avenues for the internationalexchange of scarce human resources (Ernst andMowery 2004).

Box VII.4. Government-supported research institutes in the Republic of Korea

Source: UNCTAD, based on Republic of Korea, Ministry of Science and Technology 2003a.

a These five research councils are: the Korea Research Council of Fundamental Science and Technology (S&T); theKorea Research Council of Industrial S&T; the Korea Research Council of Public S&T; the Korea Council ofEconomic and Social Research Institutes; and the Korea Council of Humanities and Social Research Institutes.

b In 2004, the three S&T-related councils were placed directly under the Deputy Prime Minister of Science andTechnology.

In the early stages of economicdevelopment, the Republic of Korea, lackingindigenous technological capability, had to relyon foreign sources for technologies required forindustrialization. With a view to developing theabsorptive capabilities of the country, in 1966the Government established the Korea Instituteof Science and Technology (KIST). KIST’s R&Dactivities were initially directed towards findingsolutions for simple and practical problemsarising from the application of the importedtechnology.

In the 1970s the Government createdspecific R&D organizations in strategic fieldssuch as electronics, telecommunications,machinery and metals, shipbuilding and chemicalsto support industrial development. These instituteshave been making important contributions tobuilding an indigenous R&D base.

As private R&D expanded, changes wereneeded in the role, operational efficiency andresearch performance of the institutes. Inresponse, the “Law for the Establishment,Operation and Development of GRIs”, enactedin January 1999, paved the way for the creationof five research councils to oversee the operationof the research institutes.a The councils were

placed directly under the Prime Minister’s Office,and individual institutes were given moreautonomy and responsibility. The changes wereexpected to improve research productivity,strengthen linkages between institutes, andincrease the transfer and commercialization ofresearch results. As of June 2005 there were 31government-sponsored research institutes in thecountry.b

The institutes actively interact with foreignresearch institutes and with TNCs. For example:

· The Paris-based Pasteur Institute set up abranch in KIST in April 2004. A joint projectcosting 146 million will initially focus onmalaria, tuberculosis and cancer research.

· Intel opened a research centre in Seoul in 2004to develop the next platform for state-of-the-art wireless communications technology andmultimedia compression technology. The centrewill also collaborate with the Electronics andTelecommunications Research Institute.

· The Korea Advanced Institute of Science andTechnology and the Cavendish Laboratory ofCambridge University opened a joint researchcentre in Daejeon City in November 2004. Itwill focus on nanoelectronics, fibre-opticelectronics and biophysics.

209CHAPTER VII

3. Policies related to intellectualproperty

A well-defined, balanced and enforceablesystem of IPRs is an important part of the NIS,especially in countries which have fairly welldeveloped innovative capabilities. By assigningownership to knowledge assets i t createsincentives for knowledge generation andfacilitates commercial exchange. It can also assistin protecting the interests of a host country’sfirms and institutions in making sure that theyare adequately rewarded in R&D collaborationswith TNCs (chapter VI). All members of theWTO are now required to meet minimumstandards of IPR protection as set out in theAgreement on Trade-related Aspects ofIntellectual Property Rights (TRIPS) (chapterVIII; UNCTAD and ICTSD 2005). Thus theprime issue is how to implement an IPR regimethat helps create an environment conducive toinnovative activities and maximize the benefitsof the country’s knowledge assets, including inthe context of the R&D internationalization byTNCs.

The main areas of intellectual propertyinclude copyright, geographical indications,patents, trademarks and undisclosed information(including trade secrets).15 For R&D — andinnovation in general — the most relevant typesare patents and trade secrets.16 Trade secrets mayin fact be even more important than patents fora country to be able to attract FDI in R&D. Tothe extent that the R&D process involvessensitive information, TNCs will always seek toprotect trade secrets against disclosure. A 1994survey of 1,478 R&D labs in the United Statesmanufacturing sector found that trade secrecywas effective for 51% of innovations, while thecorresponding figure for patents was only 35%(Cohen et al. 2000).

As noted in chapter V, the importance ofIPR protection for attracting R&D-related FDIis mixed and varies by industry (box V.3).Developing countries could increase theirattractiveness as locations for conducting R&Dby strengthening their protection of intellectualproperty, but it is not necessarily considered aprerequisite in the decision-making process ofTNCs. Other factors, such as the availability ofhuman resources, infrastructure and the domesticinnovative capacity in general, appear to be moreimportant. However, the development of domesticinnovative capacity, which does affect TNCs’

location decisions, is partly influenced by theIPR regime. Furthermore, to the extent that sucha regime facilitates sharing of knowledge andlearning, it can also help enhance the benefitsof FDI in R&D.

At the same time, IPR protection —especially a system of patents — may also entailcosts. It may, for example, place excessiveburdens on consumers. IPR protection assignsthe owner of intellectual property a degree ofmonopoly power. In order to balance the interestsof producers and consumers, countries thereforeneed to complement the introduction of IPRregimes with adequate competition policies(section B.4).

If well implemented, an IPR regime canhelp address the risk of negative effects fromR&D activities of TNCs (chapter VI). Whilecollaboration in R&D between TNCs and localR&D institutions can be beneficial to the hosteconomy by transferring tacit knowledge to thehost country, there are also potential pitfalls.Typical university–industry collaboration takesthe form of the outsourcing of a research projectto a university by a TNC. The latter may providefunding in exchange for the legal ownership ofthe research outcome, including the right topatent i t . If well designed and effectivelyimplemented, an IPR system may help protectthe local partners against unfair compensationfor their contributions (chapter VI).

Another example of the misappropriationof knowledge assets in developing countries isrelated to traditional knowledge.17 This broadlyrefers to the cumulative dynamic body ofknowledge, much of which is related to thenatural environment, held by an indigenous orlocal community that has been handed downthrough generations by oral transmission. Thereare two main issues of concern. First, indigenouscommunities that are holders of traditionalknowledge should be able to maintain their wayof life. Second, if commercialization based ontheir knowledge assets were to yield profits, theindigenous communities should be appropriatelycompensated. Governments or communities maytake measures to safeguard against the possibilityof others taking IPRs illegitimately.

One approach is to publish the details ofthe traditional knowledge before anyone tries topatent i t . This can be useful for traditionalknowledge that is clearly in the public domainand that entered the public domain with the freeand informed consent of the owners of this

210 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

knowledge (which is often not the case). Thisapproach has some limitations, however,including the fact that it puts the burden ofpublication fully on generally poor local andindigenous communities. Moreover, as i tincreases public access to the knowledge, withoutproper safeguards the possibili ty of i tsunauthorized commercial use increases.

Furthermore, governments may considerestablishing a legal framework that gives holdersof traditional knowledge the right to take actionagainst misuse or false claims in this area.Ascertaining whether the knowledge wasaccessed from the community with its free priorinformed consent and in accordance with itscustomary laws could be one component.

Given that traditional knowledge has longbeen used by indigenous communities, it may behard to claim that products based on traditionalknowledge are “novel” or involve “inventivesteps”, which opens the way for legal challengesagainst such patents. In March 2005 the world’sfirst legal challenge to a patent drawing ontraditional knowledge was concluded in favourof the challenger. In 1994, a patent on the methodfor controlling fungi using extracts from theNeem tree — a tree indigenous to the Indiansubcontinent — was granted to the United StatesGovernment and a United States-based TNC,W.R. Grace.18 However, a legal opposition to thispatent was subsequently filed, and after a processlasting ten years the European Patent Officeeventually revoked the patent.19 It is worthpointing out that actions to revoke inappropriatepatents are costly, no financial compensation isprovided to those opposing the patent to coverthese costs.

Apart from establishing a legal frameworkfor IPRs it is clear that many developingcountries need to build the capacity for i tsimplementation — including an efficient patentoffice and judicial system. In addition toknowledge of the legal system, a considerabledegree of expertise in science and technology isrequired for examining patent applications andclaims of infringement.20 In designing the IPRpolicy governments need to take into accounttheir countries’ economic needs as well as theircapacity for implementation.

In this area there is need for additionaltechnical assistance and capacity building.Although several initiatives exist to assistdeveloping countries in implementing the TRIPS

Agreement (chapter VIII), a significant gapremains between the development of legalsystems and their enforcement and management.Additional technical assistance may be requiredto help developing countries to:

• Manage and assess the value of theirknowledge resources;

• Integrate IPR systems in their nationaldevelopment strategies;

• Assess the performance and adequacy oftheir IPR systems; and

• Develop and implement IPR systems topromote R&D collaboration with TNCs.This involves an improved understandingof licensing agreements and the interfacebetween IPRs and competition law andpolicy.

Such assistance could also aim atstrengthening the capacities of entrepreneurs andgovernments to negotiate contracts and otherconditions or clauses for the transfer oftechnology and IPR protection, either asproviders or as receivers (UNCTAD 1996a, p. 4).

4. Competition policy andinnovation

Competition policy can play an importantrole in complementing the institutionalframework for ensuring that a country’s NIS isconducive to innovation, and that the benefitsfrom TNCs’ R&D are maximized while potentialcosts are minimized. Competition policy is nota proactive tool in encouraging FDI in R&D, butit can help boost innovation by maintaining andpromoting a competitive environment.Competition provides a general incentive forfirms — be they foreign or local — to innovate,for example, by encouraging them to invest inR&D and other innovatory activities (Nickell1996, Boone 2001).21 At the industry level, a keydeterminant of R&D intensity is the extent towhich the local institutional context rewardsinnovation (Furman et al. 2002). This dependson many factors, including the IPR regime at thenational level, as well as industry-specific factorssuch as government regulations, pressure fromlocal rivals and openness to internationalcompetition (Sakakibara and Porter 2000).22

The relationship between competition andinnovation is complex. Although the traditionalliterature on industrial organization predicts a

211CHAPTER VII

positive correlation between market concentrationand innovation,23 empirical work has shown apositive correlation between the level ofcompetition and innovative output (Geroski 1994,Blundell et al . 1999).24 This is particularlyevident in developing countries and transitioneconomies, where firms that face greater pressure,especially from TNCs, are more innovative thanfirms that feel less pressure (Carlin et al. 2001,World Bank 2004). Recent work has shown thatstricter competition laws and better enforcementof those laws have a positive impact oninnovation in low- and middle-income countries(Clarke 2005).

It is now commonly accepted thatcompetition policy needs to move beyond itstraditional focus on static efficiency (Ordoverand Baumol 1988).25 It should inter alia seekto evaluate the effects of business practices oninnovation and assess potential trade-offsbetween dynamic and static benefits. Firms donot innovate in isolation; close interaction withcustomers, competitors and suppliers is requiredfor the innovation process to take off.26 Findingthe right combination of competition andinteraction is therefore crucial (Wald andFeinstein 2004). Such considerations becomeeven more important when FDI enters the picture.

For promoting greater benefits from R&Dinternationalization by TNCs, some applicationsof competition policy are particularly relevantsuch as the licensing of IPRs, collaborationthrough joint ventures and alliances in high-technology industries, standard setting and patentpools, merger control, and policies to addressrestrictive business practices. These areas ofapplication all relate to business practices ofinternational scope, and represent the interfacebetween competition policy and R&D by TNCs.

One set of competition policy issues relatesto such IPR-related business practices asconditional licences and unconditional refusalsto license. Various jurisdictions, especially indeveloped countries, have introduced guidelinesregarding the licensing of IPRs.27 Specificguidelines have also been issued to tacklecompetition policy questions arising from variousforms of cooperation such as joint ventures,standard-setting and patent pools.28 In terms ofinternational competition policy enforcement,these regulations may facilitate cross-bordercollaborative business activities, some of whichare associated with R&D-related FDI.

Merger control is another relevantapplication area. Many firms haveinternationalized their R&D activities throughacquisitions of firms that also conduct R&D.From a host country perspective this may raiseconcerns that existing R&D activities may bedismantled after the takeover or that strategictechnology will be lost (chapter VI). From awider perspective, mergers between two majorplayers in an industry can have both positive andnegative effects on R&D and other innovativeactivities. On the one hand the combination oftwo firms’ sales and distribution networks maycreate better conditions for investing in R&D andinnovation. On the other hand the merger of twocompeting firms may result in a stronger (or evendominant) market position for the merged firms,and therefore weaker incentives to innovate. Suchconcerns may be particularly important in high-technology industries, in which technologychanges rapidly and the pressure for innovationis fierce. In the United States, for instance, manymore merger challenges were based on innovationconcerns during the 1995-1999 period than in the1990-1994 period (Gilbert and Tom 2001). Fordeveloping countries it may be important toimplement a more stringent competition policyfor dealing with TNCs entering through mergersand acquisitions,29 giving due consideration tothe enhancement of national innovativecapacities.

Finally, competition policies need toaddress possible restrictive business practicesby TNCs and their foreign affiliates. A prominentrole for foreign affiliates in an NIS implies thata competition authority may have to pay moreattention to possible obstacles to market entryfacing domestic firms. This is particularlyimportant if foreign companies engage in certainforms of restrictive business practices such asstrategic behaviour and vertical restrictions orinfluencing government policy-making.30 Thelatter might lead to regulatory capture, wherebythe public authorities involved lose sight of thepublic interest and protect the privileges ofestablished firms (Stigler 1971, Peltzman 1976).Unrestricted entry of domestic firms is crucialfor ensuring the existence of an active andinnovative domestic enterprise sector, and thusfor reaping benefits from spillovers from TNCR&D. In this regard, competition policy cancomplement other government efforts incountering TNC restrictions and influencing theformulation of relevant policies, and insafeguarding consumer interests (Liang 2004).

212 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

C. Promotion of R&D-related FDI

In the context of reaping benefits from theR&D internationalization by TNCs, FDI policiesassume an important role. FDI policies should inprinciple be derivatives of industrial, regional andscience and technology policies. Investmentpromotion agencies (IPAs) are important in thisprocess, especially if they act in close partnershipwith other government actors in an NIS. RelevantFDI policies also include the use of performancerequirements, incentives and science parks.

1. The role of investment promotionagencies

The appropriate role of an IPA in a country’soverall strategy to benefit from R&Dinternationalization by TNCs depends on severalfactors, including the country’s level ofdevelopment, comparative advantage, institutionalframework and development objectives. An IPAcan potentially serve two prime functions. The firstis to communicate and market existing investmentopportunities, e.g. through targeted promotion. Inthe specific case of R&D-related FDI, suchtargeting would have to be based on a carefulassessment of the location’s strengths andweaknesses, and a good understanding of thelocational determinants of potential R&D-relatedprojects.

If a location is unlikely to be able to offerthe conditions needed to attract R&D by TNCs,however, the main role of the IPA may not be toactively promote related FDI opportunities butrather to act in its policy advocacy role. IPAs maydraw the attention of relevant government bodiesto areas that are important for making a locationmore attractive for knowledge-intensive activitiesby TNCs. IPAs can potentially serve as a bridgebetween the private and public sector, helping toimprove the understanding of what is required tobenefit from R&D by TNCs.

For an IPA to play a constructive andeffective role in this regard it needs to be wellconnected with key government ministries and tohave a well-defined mandate to provide policyadvice on relevant issues (see box VII.5 for thecase of the Czech Republic).31In the Republic ofKorea, the Government in 2003 set up an IPA,Invest Korea, to promote FDI, including in R&D.

In addition, it also established in the same yearthe Korea Foundation for InternationalCooperation of Science and Technology (KICOS)to serve as a bridge between domestic and foreignnon-profit R&D centres. KICOS focuses onpromoting R&D centres involving prestigiousforeign research institutions and educationalorganizations. The two agencies both provideassistance to investors in R&D, as part of theGovernment's effort to make the Republic of Koreathe North-East Asian R&D Hub for theadvancement of science and technology.32

Preceding chapters have shown that asignificant presence of production activities canbe an asset when countries seek to develop R&Dactivities in an industry. The experiences of someAsian countries in the case of electronics andsemiconductors, and Brazil in automotive, areexamples from developing countries. From theperspective of investment promotion, this makesthe role of after-care services potentially important.Indeed, in many countries the greatest potentialfor R&D investment by TNCs is likely to be foundamong already existing foreign affiliates. Theexperiences of Singapore and Ireland (box VII.6),for example, suggest that close collaboration withexisting investors can pay off, if supported byother policies to make the host-countryenvironment more conducive to such investments.

The extent to which IPAs actively engagein the promotion of R&D-related FDI differs byregion and country. In an UNCTAD surveyconducted in February–April 2005, involving 84national IPAs,33 as many as 46 (or 55%) of theseIPAs reported that they actively promote FDI inR&D (table VII.1).34 A large number of IPAs indeveloped countries — including six of the newEU members — promote it (79%), as do 46% ofthe IPAs based in developing countries. Bysubregion, the highest percentage was noted forIPAs in Asia and Oceania. Conversely a minorityof IPAs in Africa actively promote R&D-relatedFDI, and only 11% of the LAC IPAs thatparticipated in the survey do so.

In terms of industry focus, computer andICT services are the most commonly targetedindustries by IPAs in both developed anddeveloping countries that promote R&D-relatedFDI. In developed countries (excluding the newEU members), many IPAs also target such FDI inchemicals and chemical products (includingpharmaceuticals) along with motor vehicles andother transport equipment; developing-country

213CHAPTER VII

IPAs pay relatively much attention to R&D byTNCs in agriculture. IPAs were also asked tospecify what tools they use to promote FDI inR&D. Most agencies mentioned “generalinvestment promotion” (such as missions,seminars and websites), followed by the settingup of science parks and the provision of taxincentives for R&D activities (table VII.2).35 Theuse and effectiveness of performancerequirements, incentives and science parks isdiscussed in more detail in subsequent sectionsof this chapter.

Table VII.1. Do IPAs actively target FDIin R&D?

(Number of responses)

Region Yes No

All economies 46 38Developed countries (excl. new EU members) 9 3New EU members 6 1South-East Europe and CIS 5 4Developing countries 26 30

Africa 9 13Latin America and the Caribbean 2 16Asia and Oceania 15 1

Source: UNCTAD survey of IPAs, February–April 2005.

Source: CzechInvest.

a The Department of Research, Development and Innovation has established steering committees for each of threepriority areas (life sciences, technical/engineering sciences and social sciences) while the Council for Research andDevelopment facilitates decisions on the efficient use of Government funding for research, which was about 550million euros for 2005.

IPAs frequently find themselves operatingin significant policy vacuums, partly due to a lackof coherence between FDI and science andtechnology policies. Only recently haveGovernment policies in the Czech Republic aimedat encouraging and fostering an innovation andtechnology culture, moved to centre stage.

Essentially, CzechInvest’s expanded role instimulating and securing R&D and innovativeactivities can be traced back to the year 2000, asa response to three main factors:

• Increased competitive pressure;• A shift in the agencies strategy from labour-

cost-sensitive manufacturing to businesssupport services and technology centres; and

• Positive results from a location audit thatbenchmarked the Czech Republic againstleading recipients of R&D-related FDI.

The results of the location audit alsosuggested a need for CzechInvest to help bridgethe gaps between different policy fields. Forexample, the incentive regime was exclusivelyaimed at manufacturing, the supply of suitableproperty options was limited and the link betweenuniversities and enterprises was not sufficientlystrong. CzechInvest had the expertise and strongsupport to initiate and oversee the administrationof a new incentive regime aimed at enhancingfactor conditions underpinning R&D andinnovation activities.

The number of business support services andtechnology centre projects in the Czech Republic

increased to 41 in 2004. This alone is insufficientto ensure the sustainable development of scienceand technology in the country. CzechInvestcontinues to fulfil a policy advocacy role, anddesigns and administers EU Structural Fundprogrammes aimed at enhancing innovation; ithas also fostered a deep-rooted partnership withkey constituents. It has joined with the Ministryof Industry and Trade to design and implementtwo programmes specifically aimed at supportinginnovation.

Also involved are the office of the DeputyPrime Minister, Economic Affairs; the Departmentof Research, Development and Innovation; andthe Council for Research and Development.a

The implementation of policies aimed atdeveloping the skills and capacities to sustain thegrowth of R&D activities and innovativeknowledge-based industries will take time. Allthe conditions needed to stimulate and sustaingrowth in knowledge-based industries cannot beprovided by domestic means and resources alone.Consequently, CzechInvest will continue to targetcompanies with mobile R&D and technologicallyadvanced innovative projects whilesimultaneously fulfilling a policy advocacy roleaimed at enhancing competitiveness. Such policyadvocacy could manifest itself in the creation ofa new technology agency modelled on bestinternational practices operating as an integralunit of, or running in close association with,CzechInvest.

Box VII.5. The IPA’s role in the Czech NIS

214 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

2. Performance requirements

Some countries have applied performancerequirements to induce TNCs to undertake moreR&D and other innovatory activities in theireconomies. In this context the most relevantinstruments are R&D requirements, buttechnology transfer and joint-venture/equity-ownership requirements may also play a role.

Both developed and developing countrieshave applied specific R&D requirements toforeign investors. For example, some developedcountries have imposed R&D requirements as acondition for entry to address the concern thatmost R&D activity of TNCs tends to remain inthe home country (UNCTAD 2003c, chapter VI).In India, R&D requirements have been imposedon both foreign and local investors to encourage

Source: UNCTAD, based on Ireland, Department of Enterprise, Trade and Employment 2004 and Barryforthcoming.

a Despite this increase in State spending on research, R&D expenditures in higher education and the public sectorremain below the EU average (Ireland, Department of Enterprise, Trade and Employment, 2004, p. 10).

b Science Foundation Ireland funds selected research programmes (153 by mid-2004, employing more than 750researchers) and five joint partnerships between tertiary level research institutions and industry.

Box VII.6. Enhancing the benefits from R&D-related FDI: the case of Ireland

In Ireland foreign affiliates account forabout two-thirds of business expenditures onR&D; this is mainly in ICT (75%) and another20% is in pharmaceuticals and medical devices.However, R&D expenditures per employee inforeign affiliates are still below levels prevailingin other European economies with a high-techindustrial structure such as Finland and Sweden.To boost innovation by both domestic and foreigncompanies, the development agencies — IDAIreland, Enterprise Ireland and Forfás (the nationalpolicy and advisory board for enterprise, trade,science, technology and innovation) — in the1990s jointly pushed for greater emphasis onscience, technology and innovation.

The release in 1996 of the first-ever IrishGovernment White Paper on Science, Technologyand Innovation emphasized the importance ofthese areas. As a result, under the NationalDevelopment Plan 2000-2006, there was a five-fold increase in investment in these areas, from0.5 bi l l ion in 1994 to 2.5 billion in 1999.a

Moreover, in 1998 the Programme for Researchin Third-Level Institutions was launched, whichestablished 24 major research centres as well asmajor programmes in human genomics andcomputational physics. A Technology Foresightexercise in 1999 identified biotechnology and ICTas priority areas for R&D support by ScienceFoundation Ireland.b Finally, a 20% tax creditfor incremental R&D was introduced in theFinance Act of 2004.

With a view partly to enhancing theinteraction between enterprises and academia inIreland, Science Foundation Ireland – in

collaboration with industrial partners — has setup six Centres for Science, Engineering andTechnology: three in the bio-medical field andthree in ICT. The development agencies alsoconduct various activities to promote business-academia linkages, including the promotion ofnetworks and clusters. In addition EnterpriseIreland, IDA Ireland and Science FoundationIreland are considering the introduction of pilotschemes to fund academic researchers to spendperiods working in industry and vice versa.

These policy efforts are expected to enhancethe benefits from R&D activities undertaken byforeign companies in Ireland. The country hasmanaged in recent years to attract severalsignificant R&D projects by TNCs. During theperiod 2002-2004 more than 40 such projects wererecorded (LOCOmonitor database). In severalcases the foreign companies have collaborated withlocal academic institutes. Examples include:

• Bell Labs’ R&D centre at Lucent Technologies’Dublin facility, linked with the establishmentof a collaborative academic centre at one ofthe city’s universities.

• Hewlett-Packard’s technology developmentcentre at its manufacturing facility outsideDublin, and the Digital Enterprise ResearchInstitute in collaboration with UniversityCollege Galway.

• Intel’s innovation centre outside Dublin andthe expansion of its R&D centre near Limerick.Intel has also partnered three Irish universitiesin the creation of an academic Centre forResearch on Adaptive Nanostructures andNanodevices.

215CHAPTER VII

them to set up in-house R&D facilities or to enterinto long-term consultancy agreements with localR&D institutions. However, requirements havetended to be minimal and are seldom closelymonitored (UNCTAD 2003c, chapter III).

In China, requirements to undertake R&Dare imposed as a condition for entry in selectedindustries where the inflows of FDI may beconsiderable but where TNCs have not undertakeR&D activities. A prominent example is thepassenger car industry. In an effort to tackle therelatively slow enhancement of domesticinnovation capability, the 2004 industrial policyrequired the establishment of an R&D centre withan investment of at least 500 million yuan (about$60 million) for any new automotive project tobe approved.36 Although the provision largelydeters the entry of domestic players into thatindustry, i t has contributed to changing theattitude of TNCs on R&D localization.37

The rationale for imposing a technology-transfer requirement may be to induce foreignaffil iates to adopt technologies that areappropriate to the factor endowments of thespecific host economy and to facilitate knowledgetransfer. However, TNCs are unlikely to channelproprietary information and knowledge unlessit is also in their interest. A review concludedthat explicit requirements for transferringtechnology have not been used very often(UNCTAD 2003c).38 In two studies of Japaneseand United States FDI, no positive impact wasfound of related performance requirements onthe extent to which technology was transferredto foreign affiliates (Urata and Kawai 2000,

Table VII.2. Policies and policy tools used by IPAs promoting FDI in R&D(Number of times the tool has been mentioned; multiple answers possible)

Developedcountries South-East

All (excl. new New EU Europe Developing Asia andPolicies and policy tools economies EU members) members and CIS economies Africa LACa Oceania

General investment promotion 36 7 6 5 18 7 1 10Setting up of science parks 26 5 5 2 14 4 - 10Tax incentives for R&D 26 3 3 3 17 7 1 9Promotion of l inkages between foreign

affi l iates and universities 24 4 4 2 14 6 1 7Strengthening of intellectual property rights 22 2 2 2 16 6 2 8Grants for R&D activities 20 4 6 2 8 2 - 6Reduced tariffs on imported R&D equipment 14 - - 1 13 8 - 5Special incentives to attract foreign researchers 9 3 - 2 4 2 - 2R&D requirements as a condition for entry 7 - 1 2 4 - 1 3Other policy tools 12 3 2 1 6 1 - 5

Source: UNCTAD survey of IPAs, February–April 2005.a Latin America and the Caribbean.

Note: Based on responses from the 46 IPAs that stated that they target FDI in R&D.

Blomström et al . 2000, pp. 216-217). Theimplementation of technology transferrequirements can be a challenge, mainly due tothe difficulties involved in measuring the extentto which transfers occur and in determining whattechnology is desirable.

Joint-venture and equity-ownershiprequirements have also been used to promotediffusion of knowledge and technology fromforeign affiliates to local counterparts, with mixedresults. Some researchers have found thattechnology employed in foreign affil iatesestablished in response to joint-venturerequirements tends to be three to ten years behindthe cutting edge for the industry concerned andthat the amount of technical training providedto local managers and workers is often a fractionof that received in wholly-owned affil iates(Moran 2002).39 Meanwhile, others argue thateven if the content and quality of technology aresuperior in the case of wholly-owned ventures,the presence of a local partner may increase theopportunities for local learning and diffusion ofwhatever knowledge is created locally ortransferred from abroad (Yun 2002).

There is always a risk that the use ofperformance requirements repels some FDI. Ingeneral, for countries in a stronger bargainingposition vis-à-vis the foreign investors (e.g.owing to a large domestic market), this risk islower. China and India have been able to attractconsiderable amounts of FDI in R&D whileimposing various requirements as mandatoryconditions for entry or as conditions for providingan incentive. The use of mandatory R&D or joint-

216 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

venture requirements in smaller economies wouldincrease the risk of losing FDI, unless the foreigninvestors are compensated (e.g. through variousfiscal or financial incentives). Even for such“voluntary” R&D requirements, without otherconditions in place – such as an adequate supplyof local capabilities and technical skills — fiscaland financial incentives linked to R&Drequirements are likely to have a limited impact.Conversely, if other factors are in place, theforeign investors may decide to locate R&Dactivities in the host country even without anextra inducement through incentives.

3. The use of R&D incentives isexpanding

Most developed countries and a growingnumber of developing countries use some formof incentives to attract R&D activities. In manycases government support is offered to bothdomestic and foreign firms on equal terms.Evidence suggests that R&D incentives couldhave a marginal impact (i.e. they might tilt thebalance in favour of a specific location) whencountries with similar factor endowments arecompeting for an FDI project. In general,however, other locational factors are moreimportant determinants. In considering the useof R&D incentives, governments should examinecarefully what incentives are the mostappropriate, taking into account budgetary andadministrative implications.

The rationale for government support forR&D is a presumption that, if left to the market,private firms will underinvest in R&D due to theproblems of appropriability and the high degreeof uncertainty associated with R&D investment.40

Incentives may aim to secure socially optimallevels of R&D.41

However, there are several pitfalls inproviding R&D incentives. As is the case withother types of incentives, internationalcompetition among countries in offeringincentives could result in the wasting of publicfunds as well as global economic distortions.Defining “R&D expenditures” is alsoproblematic. A broad and simple definition islikely to result in an unnecessarily generoussystem, while a more targeted system involvesmore complex administration. Whatever thedefinition, firms may resort to “re-labelling” sothat costs not related to R&D are counted as R&Dexpenditures in order to benefit from favourable

tax treatment. Another problem is related to theevaluation of R&D support programmes. It isalmost impossible to ascertain whether thebenefits (spillover effects) justify the costs ofsubsidies or foregone tax revenues. Finally, thereis a risk that a government might end upsupporting R&D projects that firms would haveundertaken even without its support.

Government support for R&D broadlytakes the form of financial and fiscal incentives(box VII.7). UNCTAD’s survey of IPAs (seeabove) indicates that more than half of theagencies that target FDI in R&D offer tax breaksfor such activities and in 43% of the cases R&Dgrants were provided (table VII.2). While thepicture is not uniform, the use of such R&Dincentives is on the increase, especially indeveloped countries. EU countries are makingthe greatest efforts to promote R&D activitiesby way of incentives42 and Austria, Denmark,Italy (for SMEs only), Portugal, Spain and theUnited Kingdom have some of the most generousR&D incentive systems (OECD 2003), whileFrance, Ireland and the United Kingdom all madetheir tax treatments of R&D more favourable in2004 (MacDougall 2004). Notable exceptionsamong the EU members are the Nordic countries.With regard to financial incentives, the EuropeanCommission in 2005 set out a seven-year planto increase R&D spending in the EU by way ofgrants wor th 70 b i l l ion .43 The plan was aresponse to the slow progress towards EUmembers’ pledge to increase R&D spending to3% of GDP by 2010.44 Outside Europe, the mostgenerous fiscal incentive schemes are offered byAustralia and Canada.45 In the United States, taxcredit for R&D is the most significant of theremaining domestic tax credits.46

Partly due to limited resources, developingcountries are more likely to apply fiscal thanfinancial incentive schemes. In the UNCTADsurvey of IPAs, more than twice as manydeveloping-country IPAs used tax incentives thanthose who used financial incentives (table VII.2).Many developing countries also charge lowertariffs on imported R&D equipment as a way ofpromoting technology transfer.

The two largest emerging-marketdestinations of FDI in R&D, China and India,have strengthened their systems of R&D support.In China, TNCs can set up R&D centres asindependent entities (under the rules applyingto Sino-foreign joint ventures), wholly foreign-owned enterprises or as independent departments

217CHAPTER VII

or branches of existing companies. Equipmentand parts imported by R&D centres meetingcertain requirements are exempt from customsduties and import value-added tax, and thetechnologies they develop and use are exemptfrom business tax.47 India offers inter alia a ten-year tax holiday for companies engagedexclusively in scientific R&D with commercialapplications (EIU 2004n).

Most other Asian countries that haveattracted significant FDI in R&D also provideextensive R&D support. In Malaysia, companies

can offset 100% of capital expenditure incurredwithin ten years against 70% of their income.48

Singapore allows a 100% deduction of R&Dexpenses (in certain cases 200%) and providesvarious grants and tax exemptions.49 Thailandrevamped its system of R&D incentives in 2004,after which firms can be entitled to a corporateincome tax holiday for up to eight years (EIU2004p).

In Latin America, the use of governmentsupport for R&D is less widespread. For exampleArgentina, Chile and Mexico50 do not havesignificant fiscal measures to promote R&D.Brazil, on the other hand, allows locally ownedIT firms51 to deduct some R&D expense fromtheir taxable income, and research financing isavailable for research projects in bioscience,physics, chemistry and environmental science(EIU 2004q). Colombia also offers investors inR&D centres certain fiscal tax exemptions (EIU2004r). While there is little information on theuse of incentives in Africa, South Africa allowsaccelerated depreciation of assets in certaintargeted areas, including R&D investment. Bothforeign and domestic firms are eligible for taxincentives. The Government also provides somedirect financial support for R&D (EIU 2004s).

Despite the proliferation of financialincentives for R&D, few studies have assessedtheir effectiveness. An analysis of the SmallBusiness Innovation Research Program in theUnited States found that firms awarded subsidiesunder this programme enjoyed greater sales andemployment growth and increased their chancesof receiving venture capital financing (Lerner1999). Another study, conversely, concluded thatthe subsidies granted under this programme didnot affect employment of R&D personnel.Furthermore, there is evidence that subsidies havecrowded out firm-financed R&D spending(Wallstein 2000).52

There are more studies on the effectivenessof fiscal incentives. They typically measure howmuch additional R&D expenditures are generatedby a 1% reduction in the costs of undertakingR&D. Various studies have noted that the long-term impact of R&D incentives may be moreimportant than the short-term ones (e.g. vanPottelsberghe et al. 2003, Bloom et al. 2002).However, it should be noted that these studiesdid not address the problems of re-labelling andinput price inflation.53

Box VII.7. Types of R&D incentives

Two main types of R&D incentives can bedistinguished: financial and fiscal. Financialincentives refer to direct funding of R&D projectsby the government through the granting ofpreferential loans or subsidies. Fiscal incentivesare tax based and can be further divided into sixtypes: accelerated depreciation, tax allowance,tax credit, tax holidays, income tax allowancesand import tariff exemption.

• Accelerated depreciation refers to a practicewhereby faster depreciation rates are appliedfor current and capital R&D expenditures.a

In most countries, non-capital R&Dexpenditures are treated as current expense,thus allowing the whole amount to be deductedfrom the taxable income during that year.

• Firms that are given tax allowances can deductR&D expenditures from taxable income at arate higher than 100%, resulting in a furtherreduction of corporate income tax liability.

• Tax credits also reduce a firm’s corporateincome tax, but the deductible amount iscalculated differently. In this case a certainpercentage of eligible R&D expenditures canbe deducted directly from corporate incometax.

• A tax holiday exempts firms investing in R&Dfrom paying taxes, or lowers the rates for agiven period of time.

• Tax allowances for personal income tax andimport tariff exemption can be targeted atpersonnel and products linked to the R&Dactivities of the firms.

Source: UNCTAD.

a This is an advantage for firms since R&Dinvestments would normally be treated as capitalexpenditures, in which case only the amount thatcorresponds to the depreciation of such assetscould be deducted from taxable income each year.

218 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Regarding the factors that most influenceTNCs’ decisions on where to locate their R&D,a recent survey found that incentives, whileimportant, are not a major determinant (EIU2004a). Infrastructure and a tradition ofinnovation have the greatest impact.Nevertheless, government support can tip thebalance in favour of a certain location when otherfactors are equally attractive (Cantwell andMudambi 2000, chapter V).

As incentives are only one of many factorsthat influence the location of R&D, countries thatcontinue to compete by offering incentives toattract such FDI need to be aware of the risk thatthe costs involved may eventually outweigh thebenefits. In designing an incentive policy,governments also need to decide whether a moretargeted approach or a more universal approachis the most appropriate. A targeted approach ismore complicated and is likely to involve higheradministration costs. Complicated incentiveschemes also tend to be less effective.54 A moreuniversal approach (primarily fiscal incentives)requires larger resources, some of which willinevitably be used to support R&D projects thatdo not require any support.

One way to enhance the potential benefitsfrom incentives is to promote R&D collaborationamong local firms and/or institutions. Such ameasure may help build domestic R&Dcapabilities by providing local R&D entities moreopportunities for learning and funding. In Brazil,for example, some R&D incentives are providedonly on the condition that the R&D is donejointly with research institutes and schools ofhigh academic standing (EIU 2004q). Amongdeveloped countries, in Denmark firms canreceive extra tax deductions on research projectsco-financed by enterprises and public researchinstitutions; in the United Kingdom companiesare able to claim credit for R&D work which theysubcontract to certain institutions includinguniversities, charities and scientific researchorganizations (United Kingdom, Inland Revenue2002).

4. Using science parks asattractors

Science parks are used to create a moreconducive environment for innovation and R&Din enterprises, often in close proximity touniversities and other public technical institutes.

In UNCTAD’s survey of IPAs, the setting up ofscience parks was the second most commonlymentioned policy tool used by those that targetFDI in R&D. While their precise goals differ,such parks offer various kinds of support andnetworking activities, help newly started venturesand enhance cooperation between existingcompanies in the park. Many of them provide thespecialized infrastructure needed to undertakeR&D work. As locations for R&D-related FDI,science parks may offer attractive features byfacilitating clustering and networking, offeringaccess to skilled people, providing the necessaryinfrastructure and administrative support and, lastbut not least, offering a pleasant living andworking environment. According to theInternational Association of Science Parks therewere about 600 science parks in 2004 worldwide,hosting some 65,000 companies.55 Two-thirds ofall parks are located in the United States andEurope, and East Asia accounts for the bulk ofall science parks in developing economies(Andersson et al. 2004, p. 152).56

A well-known case in Asia is the HsinchuScience Park set up in 1980 in Taiwan Provinceof China. While it was originally established witha view to serving local companies, non-Taiwanesecompanies have also been attracted. In 2004, 52out of 384 companies in the park were non-Taiwanese.57 In Singapore, the first science parkwas also established in 1980 and now hosts 300local and foreign companies (Zhang 2004). TheZhongguancun Science Park (Beijing) is China’sfirst and largest science park with more than14,000 high-technology firms, including 1,600foreign affil iates (see also box VI.2). Theoffshoring of software development to India hasoften benefited from the presence of dedicatedtechnology parks for IT services (WIR04). As of2003 there were 39 such parks, accounting for80% of all India’s software exports in 2002/03.

A few science and technology parks havebeen established in different parts of Africa,especially in North Africa. Algeria, Egypt,Morocco and Tunisia all have at least one suchpark in place.58 Madagascar and Senegalsimilarly host technology parks;59 and inUNCTAD’s survey of IPAs Ghana, Kenya, Maliand Nigeria stated that they use science parksto attract FDI in R&D. In South Africa a newpark — “The Innovation Hub” — will becomethe first African science park that is accreditedinternationally. Its main objective will be toattract a variety of enterprises active in, inter

219CHAPTER VII

alia, ICT, electronics, l ife sciences andaerospace.60

There has been limited use of science andtechnology parks in Latin America (IADB 2001).The first attempt at promoting innovative clusterswas Brazil’s creation of 13 “technologicalinnovation nuclei” in selected universities andresearch centres in 1982 (Quandt 1999). Mexicostarted to create business incubators in 1990 withthe support of the National Council for Scienceand Technology and the Association of Incubatorsand Technological Parks.

There is little evidence concerning theeffectiveness of science parks. There appears tobe some consensus that they can contribute tocommercializing university-based knowledge andtechnology and can act as an important node ininnovative clusters and in the NIS more broadly.As such they can also be useful tools in attractingFDI and embedding foreign affiliates in an NIS.However, establishing a park does not guaranteesuccess. One issue concerns the financing of thepark and the role of government support. It hasbeen argued that governments should ensurestrong private sector interest in any project beforeextending financial support, and that governmentsupport should be reduced as the park develops(IADB 2001). Another issue concerns theassigning of IPRs, especially if a science parkfacilitates the commercialization of university-based knowledge. Thirdly, in developingcountries it is important to find a balance betweenproviding employment opportunities foruniversity students and avoiding the risk ofdraining skills away from universities (Anderssonet al. 2004, p. 154). Fourthly, as science parkscan constitute a key tool for the regionaldevelopment of innovative clusters, the role ofsub-national and local governments is decisive.

D. Industry-specificpolicies to enhance thebenefits of FDI in R&D

In addition to specific policies geared toattract R&D by TNCs, various “flanking policies”are important to enhance the benefits from suchactivities. In this context industry-specificpolicies deserve particular attention as they haveplayed an important role in encouragingindigenous production and innovation capabilitiesin developing countries. Such capabilities arecentral to sustaining technological and economic

development and to reaping the benefits fromR&D by TNCs. Policy formulation needs toreflect the fact that the nature of differentindustries varies considerably.

Industry-specific policies need to bedefined in light of a country’s overalldevelopment strategy. Within such a strategy, anindustry-based vision can form the basis fordeciding what R&D by TNCs to target and howto benefit from it, highlighting the need for closeinteraction between industrial and FDI policies.For developing countries it is important to takeaccount of their development level andcomparative advantage so that policy objectivesare realistically set. For many low-incomecountries it may be appropriate to give priorityto the development of less technology-intensiveindustries and services rather than high-technology ones.

What policy tools should be used dependsin part on the industries a country seeks topromote. Appropriate policy formulation anddesign therefore requires in-depth knowledge ofan industry, its production and technologicalcapabilit ies and the kind of R&D that isundertaken locally. In countries that lack theknowledge base necessary for new productdevelopment in an industry, the enhancement ofmanufacturing rather than R&D capacity is likelyto be the first priority for industrial development.Before moving towards R&D-based activities,a country first has to develop basic productioncapabilit ies (chapter III). For developingcountries that already have significantmanufacturing capabilities in some technology-intensive industries, policymakers may firstconsider promoting experimental development(by foreign affil iates as well as by localenterprises) in these industries. For moreadvanced developing countries with strongmanufacturing capabilit ies in some high-technology industries, going from developmentto (applied) research is the major challenge.61

The ultimate test for most countries is to fosternational innovative capabilities in technology-intensive industries.

As many policy measures target R&D inspecific industries, the boundary betweenindustrial policy and science and technologypolicy becomes blurred, requiring closecoordination between the two. In some countriesthe policy focus is shifting from “industries” to“clusters”, which reflects the growing emphasison inter-organizational relationships and networks

220 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

in R&D and production (Freeman and Barley1989, Olk and Young 1997). By fosteringknowledge-based cluster formation,62 industrialpolicy can encourage joint R&D efforts andknowledge spillovers involving both domesticfirms and foreign affiliates.63

To enhance benefits from R&D interna-tionalization, industry-specific policies also needto support entrepreneurship and foster theemergence of technology start-up SMEs. Thereis growing recognition of the role of SMEs inan NIS.64 Small-sized technology start-up firmsare often responsible for important innovations.While the relatively high concentration of R&Din large firms is a natural consequence of theirability to manage fixed costs and risk, SMEs tendto be more flexible and can therefore drivetechnological change at a faster pace than largefirms. Thus SMEs can be especially importantin high-technology industries.

However, small firms face severaldifficulties that can prevent them from fullyrealizing their technological and commercialpotential. By making resources accessible andaffordable to them, active SME policies cancontribute to the emergence of an innovativedomestic enterprise sector in new areas.

In high-technology industries governmentscan foster technology enterprise developmentthrough business incubation systems fortechnology start-ups. Such systems can provideyoung start-ups with the necessary resources andservices (e.g. access to financing, networking,technical assistance and business consulting),help reduce non-commercial risks, supportentrepreneurship and, thereby, thecommercialization of R&D by these firms.

As part of efforts to build domesticenterprise capabilities in an industry there canbe a need to strengthen the environment fortechnology start-ups by upgrading existingfinancial intermediaries and by introducing suchfinancial instruments as seed and venture capitalfunds. Venture capital has been perceived mainlyas a private sector activity, and in mostdeveloping countries governments have playeda limited role in this area. However, there canbe a role for public venture capital funds tocompensate for the lack of private sources ofventure capital needed to encourage R&Dinvestment (Andersson and Napier 2005).

When carefully designed, businessincubator and venture capital programmesfunction as complementary approaches. Whilebusiness incubators help prepare the ground forgrowing firms and may compensate for some ofthe market failures that hinder the growth of newfirms, venture capital provides both financialcapital and expertise. Despite the obviousbenefits and synergies deriving from closecollaboration between incubators and venturecapitalists, in reality such collaboration is farfrom automatic. Active policies are often neededto catalyse it.

In many instances industry-specificpolicies and SME policies directed towardstechnology start-ups need to be implemented atthe local rather than national level. This isparticularly important in large countries wherecomparative advantages and resourceendowments of various locations may differconsiderably. In Shanghai, for example, policiesat the central level were complemented by stronglocal government support to attract FDI in thesemiconductor industry and to build up aninternationally competitive industrial base (boxVII.8).

E. The role of homecountries

Developed countries can help securebenefits from the internationalization of R&Dto developing countries in different ways,including through the promotion of R&Dinternationalization and measures aimed atstrengthening the NISs of developing countries.

The limited information available on home-country policies related to R&Dinternationalization suggests, however, thatrelatively few countries have specific measuresin this area. A recent review of home-countrymeasures in developed countries concluded thatfew governments support firms financially thatwant to conduct R&D abroad (OECD and BelgianScience Society 2005). Some financialmechanisms encourage joint collaboration inR&D activities, such as the EU FrameworkProgrammes. In a few countries indirect fundingof R&D (e.g. tax credits) is also granted if R&Dexpenditure is incurred outside the country (e.g.purchase of R&D services from foreign researchinstitutes). Most jurisdictions among developed

221CHAPTER VII

countries that grant R&D incentives do soirrespectively of whether the R&D supported isundertaken inside or outside the country.However, Belgium, France, Japan and Spainrequire (at least for some incentives) that theR&D is conducted in the respective country(IBFD 2004, pp. 222-230).

Some developed countries provide supportto domestic public institutions to undertake R&Dactivities abroad, including in developingcountries. For example the Australian Centre forInternational Agricultural Research promotespartnerships between Australian and developing-country institutions. It supported more than 50R&D projects in Viet Nam between 1993 and2003 (UNCTAD 2004, p. 10). The French Centrefor International Cooperation on AgriculturalResearch for Development provides new andemerging technologies related to sustainableagricultural development and conservation of theenvironment in Africa, Asia and Oceania, LatinAmerica and Europe. Its researchers, posted in50 countries, work with national research

organizations or provide technical support indevelopment projects (ibid., p. 27).

A growing number of developing-countryTNCs — mainly from Asia — also conduct R&Dabroad to access technologies, skilled humanresources, unique innovative networks andattractive innovation environments (chapter IV).Some Asian governments, such as those of China,India, Malaysia and Singapore, actively facilitateand encourage outward FDI,65 but fewspecifically encourage FDI in R&D. The onlyknown example in this regard is China. In thecontext of i ts “go global” strategy66 theGovernment of China has promulgated a seriesof regulations and circulars in recent years tomanage and encourage overseas investment byChinese enterprises.67 The country adopts aselective support strategy.

In October 2004 the National Developmentand Reform Commission (NDRC) and the Export-Import Bank of China (EIBC) jointly issued acircular encouraging overseas investment projects

Source: UNCTAD.

a This took place right after the Central Government had introduced “Several Policies to Encourage the Developmentof the Industries of Software and Integrated Circuit” (File No. 18).

Box VII.8. The role of local governments in building domestic capabilities: the case ofShanghai

Following decisions taken by the CentralGovernment in China in June 2000, the municipalGovernment of Shanghai took a series of stepsto develop the local semiconductor industry.a

• For projects on integrated circuit (IC)manufacturing it granted exemptions andreductions of local taxes and fees, facilitatedthe import, export and international travel ofcompany employees and provided a 1% interestdeduction of commercial loans denominatedin renminbi.

• For IC design, it provided preferential treatmentto firms and set up specific funds for theestablishment of a technical platform, includinga semiconductor intellectual property bank.

• Various agencies of the municipal governmentworked together to accelerate the upgradingof the semiconductor industry. Specific fundingprogrammes (e.g. the Product-Design-ChipProject) were introduced and existing ones (e.g.the Technology-oriented SME InnovationFund) were leveraged to enhance local

technological levels and innovativecapabilities.

• In terms of manpower development, educationand research centres in relevant areas at localuniversities were encouraged and specificpolicies were adopted to attract highly skilledhuman resources from within China andabroad. The municipal government alsoestablished a programme to attract Chinesereturnees to form start-ups for conducting R&Din Shanghai.

• In 2003, a semiconductor intellectual propertyexchange centre was set up to serve as aplatform for IPR protection and trading, anda specialized guarantee fund was launched toaddress the financing problem facing small ICdesign companies.

• To encourage linking together downstream andupstream firms in the value chain, the localgovernment also introduced the SpecializedProject to Encourage the Collaborationbetween Final Product Industry and IC DesignIndustry.

222 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

in four areas, including “overseas R&D centresthat can util ize internationally advancedtechnologies, managerial skills and professionals”(see also chapter II). Preferential credit is grantedfor investments in these four areas and the NDRCand the EIBC have established a joint supportivemechanism to promote such outward FDI. TheEIBC specifically arranges “special loans foroverseas investments” within its export creditplan in order to support the identified investmentprojects. It accelerates the process of projectscreening, and the NDRC also facilitates contactswith other agencies to improve the risk-controlmechanism for overseas investment. Theencouragement of R&D abroad reflects the effortsof the Government to enhance China’s innovativecapability by leveraging foreign resources.

An indirect way for home countries to helpdeveloping countries derive greater benefits fromR&D internationalization is to assist them instrengthening their NISs. However, bilateral aidorganizations rarely focus on science andtechnology, and when they do the aid tends notto be effective:

“A few bilateral development agencieshave a strong focus on science andtechnology. But even where suchprograms exist, they lack strong linkswith domestic scientific institutions indonor countries…Aid programs need toreflect the view that the best way toaddress poverty is to stimulate economicgrowth. This will require a focus onscience, technology, and innovation” (UNMillennium Project 2005, p. 165).

There is scope for more bilateralcooperation to foster policy formulation andstronger innovation systems in developingcountries (UNCTAD 2005d). A key area in thisregard is human resource development. Thedomestic educational systems in many poorcountries, especially in Africa, are not sufficientlyflexible or well-funded to achieve the neededincrease in the number of tertiary students. Theinternational community could play a more activerole in this area, for example, by strengtheningthe local educational infrastructure and bymaking education opportunities to developingcountries available in developed countries. Manydeveloped countries already provide developing-country students with scholarships for highereducation in their countries.68 Some also providedeveloping countries’ academic, research andprofessional institutions with research and

equipment support (UNCTAD 2004, p. 11). It hasbeen proposed that developed countries establisha second “Colombo Plan” for sub-Saharan Africaunder which students from African countriescould study abroad.69 To address the risk of braindrain, special provisions would, however, haveto be made to ensure that students return to theirhome countries upon completion of their studies(UN Millennium Project 2005).

Efforts by home countries to improve theinstitutional framework for innovation indeveloping countries could help establishtechnical standards and certification systemsthrough access to and provision of testingequipment for standard setting and qualityassessment (UNCTAD 2004, p. 15). Similar stepscould be taken in the area of IPRs and throughR&D collaboration between institutions indeveloped and developing countries. In the healthsector some developed-country governments havefunded R&D public and private institutions indeveloping countries to develop drugs andvaccines. Such support has mainly involvedfinancing research, conducting trials andproviding mechanisms for delivery of servicesto end-users (UNCTAD 2004, p. 9). Moreover,the EU has contributed to the NISs of developingcountries by encouraging an exchange ofscientists and closer interaction betweenuniversities in developing countries and EUmember countries (UNCTAD 2005d, para 27).

F. Concluding remarks

Today, no country can rely entirely onknowledge created within its borders. Thechallenge facing countries is therefore to ensurethat they connect in the most effective way withglobal R&D networks of TNCs and theinnovation systems of other countries. Inwardas well as outward R&D-related FDI can hereplay a role. In order to derive benefits from thecurrent trend of R&D internationalization, thischapter has underlined the need for activegovernment policies in a number of areas, andthat such intervention is done in a coherent way.For the many developing countries that arecurrently not taking part in the process of R&Dinternationalization, important lessons could bedrawn from the experience of other countries.

In all the developing economies that havebeen successful in improving their innovationcapabilities and in attracting R&D by foreigncompanies, the government has played a key role.

223CHAPTER VII

In particular, while their strategies differ theyhave all sought to strengthen their innovationsystems by enhancing their "created assets",notably their human resources, and theirinstitutional frameworks affecting the incentivesand conditions for firms to innovate. But in orderto be effective, such policies demand politicalcommitment and a clear, long-term vision. Acountry that simply opens up to trade andinvestment and passively waits for newknowledge and technology to flow in from abroadis likely to be at a competitive disadvantage vis-à-vis those that actively adapt and strengthentheir policies and institutions.

The ability of a country to benefit fromR&D internationalization depends first andforemost on the strength of its NIS. The strongerthe NIS, the greater is the likelihood not onlyof attracting R&D by TNCs but also of spilloverbenefits arising from inward as well as outwardFDI in R&D. Policies on human resourcedevelopment, promotion of linkages betweenR&D activities in the public and enterprisesectors, strategic use of IPR systems andcompetition policies are key in this respect.Efforts in these areas need to be closelycoordinated with investment policies. Indeed, acoherent approach is required to ensure thatgovernment interventions are effective insecuring benefits from R&D internationalization.In essence, policies in the areas of innovation,education, competition, FDI as well as thosetargeting the needs of specific industries andSMEs need to be seen as part of a vision aimedat enhancing competitiveness and development.

Active and coherent policies are among themost striking features of those developingcountries that are now emerging as nodes in theR&D networks of TNCs. The success of someAsian economies is no coincidence. In most ofthem the starting point has been a long-termvision of how to move the economy towardshigher value-added and knowledge-basedactivities. In many instances, targeted governmentpolicies aimed at strengthening the NIS andfacilitating knowledge inflows. Such policieshave included:

• Active promotion of imports of technology,know-how, people and capital from abroad;some have relied on inward FDI while othershave linked up with the TNCs throughcontractual arrangements.

• Strategic investment in human resources tosupport technological upgrading in the

private sector - typically with a strong focuson science and engineering.

• Continuous improvement of educationalsystems.

• Promoting immigration or the return ofskilled workers in the diaspora.

• Development of infrastructure (such asscience parks, public R&D labs, incubators)that helps promoting innovation in the NISby both foreign and local firms.

• Use of performance requirements and/orincentives as part of an overall strategy toattract FDI in targeted activities.

• Strategic implementation of IPR protection.

Effective implementation of policies inthese areas requires collaboration betweenrelevant government bodies and coordination atthe highest level. There is also a need to delineatethe responsibilities of individual ministries andagencies at both the national and sub-nationallevels. Because R&D activities have a strongtendency towards geographic clustering,government agencies at the local level can playan important role in attracting FDI in R&D tospecific localities by establishing science parks,providing specific incentives and facilitating themobility and availability of technically qualifiedpeople. Moreover, in designing and implementingtheir policies, governments need to understandwhat determines the location of R&D, how R&Dby TNCs interacts with other actors within theNIS of a country and how to connect effectivelywith other systems of innovation.

For many developing countries at the lowerranks of the UNCTAD Innovation CapabilityIndex, any expectation of a major influx of R&Dby TNCs would be unrealistic in the short-term.That is not a reason for inaction, however. Rather,countries should consider how to begin a processthrough which economic and technologicalupgrading can be fostered. As argued by oneexpert (Lewanika 2005, p. 12):

“An important starting point fordeveloping countries is to increase thepercentage of GNP devoted to education,science teaching as well as research anddevelopment. The notion that investingin science and technology is a t ime-consuming, wasteful and costly activitywill condemn developing countries toperpetual economic illness. Initiatives toassist Third World countries to develop

224 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

must include science, technology andinnovation as one of its main themes.”

Successful efforts at the interface ofinvestment, technology and industrial policiesare essential in order for more countries to benefitfrom the current trend towards greaterinternationalization of R&D. Recentdevelopments have shown that developingcountries can play a role even in highlysophisticated R&D by TNCs. Currently thephenomenon is confined to relatively fewdeveloping and transition economies, but R&Dinternationalization is expected to deepen andpotentially involve more countries. This processis still in its infancy. The fostering of innovativecapabilities is a long-term task for governments.For latecomer countries, ensuring that a processaimed at strengthening the NIS gains momentumcan be seen as a first necessary step.

For developed home countries, currenttrends accentuate the need for relying more onthe creation, diffusion and exploitation ofscientific and technological knowledge as ameans of promoting growth and productivity.Rather than regarding R&D internationalizationas a threat, these countries may seek to seize theopportunities it offers. Reverting to protectionismmost likely would harm the ability of their firmsto compete. Instead, i t will be important toexplore new ways of collaborating with the newR&D locations, such as through joint researchprogrammes and outsourcing as well as throughinward and outward R&D-related FDI. Tofacilitate such collaboration, and to help morecountries build the necessary capabilities toparticipate in the process, developed countriesmay decide to offer additional support aimed atstrengthening various aspects of the innovationsystems of countries which currently have weakinnovative capabilities. Such contributions couldeffectively help in the overall efforts to narrowthe technology and innovative capability gap thatmay otherwise continue to widen.

Notes1 In many parts of Africa, for example, academic

education has often produced skills demanded in publicadministration rather than in industry. Africaninstitutions of higher education enrol 60% of studentsin the arts and humanities and only 40% in scienceand engineering (Oyelaran-Oyeyinka 2004a, p. 20).

2 In Malaysia, for example, such needs were identifiedfor electrical and electronics engineering, informationtechnology, communications technology and circuit

design personnel who are able to combine hardware,software and application knowledge (Ernst 2004).

3 UNCTAD interviews.4 See also WIR01 , chapter V, for examples on how

countries have involved foreign affiliates in trainingand technological upgrading of suppliers.

5 Among those working in science and engineeringoccupations in 2000, 17% with bachelors’ degrees, 29%with masters’ degrees and 38% with doctorates wereforeign-born (Ernst 2005a). In 2001, 133,000 foreigncitizens were enrolled at the graduate level in scienceand engineering in the United States. This correspondedto more than 30% of the total number of science andengineering graduates that year, an increase from 20%in 1983 (United States, NSF 2004, appendix table 2-12).

6 See ISA (2004) and “The brain drain: old myths, newrealities”, OECD Observer, May 2002.

7 Tertiary technical institutions in Singapore also attracta large number of foreign students. According to figuresfrom the Singaporean Agency for Science, Technologyand Research (A*STAR), 79% of full-time postgraduateresearch students in science and engineering wereforeigners in 2003.

8 “Singapore aims to be a biotechnology hub”, FinancialTimes, 10 June 2005.

9 In 2003 only 2.7% of all professors were non-Korean,foreign students at universities accounted for only 0.2%of all enrolled students and there were few foreignworkers in the private and public sectors (Kwon 2003).

10 Some estimates suggest that around 300,000professionals from the African continent live and workin Europe and North America (see “The brain drain:old myths, new realities”, OECD Observer, May 2002).Many countries in Latin America have also beenexporting human resources (ECLAC 2002, chapter 8).At the beginning of the 1990s, some 300,000 LatinAmerican and Caribbean professionals and technicianswere living outside their home country; over two-thirdsof that total were concentrated in the United States(Villa and Martínez 2001).

11 “In a ‘brain gain’, India’s Westernized émigrés returnhome”, International Herald Tribune, 26 July 2004.

12 This programme coincided with a World BankIndustrial Technology Development project, onecomponent of which was to upgrade the technologyinstitutions and strengthen their linkages with industry.The project helped shape the direction of reform andprovided technical assistance to help reorient thelaboratories and train their managers and staff.

13 In 1996, outside earnings contributed to 16.4% of totalexpenditures, and some laboratories, such as theNational Chemical Laboratory in Pune, were even moresuccessful, earning over 50% from industry, the bulkof which came from foreign contracts.

14 Communication from CSIR, May 2005.15 The TRIPS Agreement also covers industrial designs

and layout designs of integrated circuits.16 Not all pieces of knowledge can be patented. For

example, information obtained in the process of anR&D project before its completion may be valuable,but it does not constitute an invention as such andtherefore cannot be patented. Even when an inventioncan potentially be patented firms may prefer not to

225CHAPTER VII

disclose the details of their intellectual property throughpatenting. In these circumstances such information canbe kept as a trade secret.

17 For more information, see UNCTAD (2004), and relevantpapers at www.unctad.org/trade_env/TK2.htm.

18 While the legal process was taking place the patentwas sold on to other firms although the Governmentof the United States remained the co-proprietor of thepatent.

19 ICTSD Bridges Weekly News Trade Digest, 23 May2005 (www.ictsd.org).

20 In countries where the IPR regime allows patentingin new areas such as software, business models andfinancial formulae, the required range of expertise iseven wider.

21 See also Cimoli and Primi forthcoming and ECLAC2004c.

22 According to the 1995 United States antitrust guidelinethere is a clear division of powers between IPR policyand competition policy in this area.

23 Schumpeter’s economic analysis is a commonly usedstarting point. It argues that firms must make high levelsof profits and enjoy some monopoly power in orderto be able to invest in innovation.

24 Several attempts have been made to provide atheoretical explanation for this positive impact (e.g.,Aghion and Howitt 1998, and Sutton 1998).

25 In the short run, competition is necessary to enhancethe allocative efficiency of an economy and maximizeconsumer welfare. Beyond this static function,competition is a key driving force behind technologicalprogress and long-term economic growth, thusinfluencing dynamic economic efficiency.

26 Innovations are strongly influenced by horizontalrivalry between competitors, but vertical relations arealso important (Edquist 1997).

27 In the United States, “Guidelines for the Licensing ofIntellectual Property” were issued in 1995 by theDepartment of Justice and the Foreign TradeCommission; the European Commission published itsRegulation No 240/96 concerning Technology TransferAgreements in 1996; and in Japan, “Guidelines forPatent and Know-How Licensing Agreements” wereissued in July 1999.

28 In 2000, the “Antitrust Guidelines for CollaborationAmong Competitors” were published in the UnitedStates, and the European Commission issuedregulations regarding specialization agreements andR&D agreements. In 2001 the European Commissionfurther introduced “Guidelines on the Applicabilityof Article 81 of the EC Treaty to Horizontal Co-operation Agreements”.

29 For a discussion of cross-border M&As and their effectson competition, including limitations in this respect,see WIR2000, and Singh 2003.

30 Strategic behaviour aims to deter potential entrantsrather than to destroy actual competitors. Verticalrestrictions occur when a firm at one stage of the lineof value-added activity imposes restraints on the termsof trading by firms at another stage. Vertical restraintsinclude resale price maintenance, selective distributionsystems, tying arrangements (tie-in sales), exclusivedealing and refusal to sell.

31 In Mauritius, for example, the Board of Investmentis actively participating in the work of the National

Productivity and Competitiveness Council to boost thecountry’s innovation performance.

32 Republic of Korea, Ministry of Science and Technology2003b; KICOS 2004.

33 The response rate was 55%.34 This number is higher than that reported in the 2004

UNCTAD survey on FDI in services, in which onlyone-third of the IPAs responded that they promotedFDI in R&D activities (WIR04). This may imply thatan increasing number of IPAs are quickly respondingto new opportunities created by the internationalizationof R&D networks.

35 Frequent use of general promotion was also reportedby IPAs that do not actively target R&D-related FDI.

36 This applies to both foreign and local companies.37 For example, as part of the joint venture agreement

between Nissan Motor and Dongfeng MotorCorporation, the building of an R&D centre began inDecember 2004. Meanwhile DaimlerChrysler, HondaMotor and Hyundai Motor, together with theirrespective local joint-venture partners, have announcedplans to establish R&D centres in China. In additionShanghai GM and Shanghai Volkswagen have bothdecided to expand their existing R&D centres. Anestimated four billion yuan (about $0.5 billion) willbe invested in R&D in the industry. See “Can fourbillion Yuan in R&D heal the pain of China’sautomotive technology?” Jiefang Daily, 26 January2005. [In Chinese].

38 In South Africa for example, the Foreign InvestmentGrant, which was established in 2000, intended amongother things to induce investors to bring in newmachinery and equipment. As of 2003, the system hadnot been able to induce the level of technology transferthat the Government had hoped for (UNCTAD 2003c,pp. 196-197).

39 Similar findings were reported by Ernst 1999.40 It is argued that Governments are better placed to take

on risk than firms or financial institutions for tworeasons. First, they may be able to spread risks overa larger number of projects. Secondly, they may assessthe risks differently; even a commercially unsuccessfulR&D project could be worth pursuing if it generatesenough knowledge from which the society can benefit.

41 Although the estimates of the social return from R&Dinvestment vary, empirical studies indicate that thereare important spillover effects (Jones and Willams1998).

42 For a recent comprehensive report on the tax treatmentof the 25 EU member countries as well as the UnitedStates and Japan, see IBFD 2004. For more informationon various schemes in use in developed countries, seeGregory and Botha, 2003. For country-specificinformation on R&D support programmes, see also EIU(2004b) for Australia, EIU (2001) for Austria, EIU(2004c) for Belgium, EIU (2004d) for Canada, EIU(2004e) for France, EIU (2004f) for Germany, EIU(2004g) for Greece, EIU (2004h) for Italy, EIU (2004i)for Japan, EIU (2004j) for Spain, EIU (2004k) forSwitzerland, EIU (2004l) for the United Kingdom andEIU (2004m) for the United States.

43 “Brussels hopes research money will aid innovation”,Financial Times, 3 April 2005.

44 A high-level group chaired by the former PrimeMinister of the Netherlands, Wim Kok, had already

226 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

proposed that incentives be used to strengthen thescience base and make the EU more competitive(European Commission 2004, p. 21).

45 See EIU (2004b) for Australia and www.irap-pari.nrc-cnrc.gc.ca for Canada.

46 Network World Fusion, www.nwfusion.com.47 Circular of the Ministry of Foreign Trade and Economic

Cooperation concerning the establishment of foreign-invested research and development centres, file No.218, 18 April 2000

48 The rules differ somewhat between firms performingR&D on a contractual basis and those undertaking R&Din-house R&D (EIU 2004o).

49 Singaporean Economic Development Board,www.sedb.com.

50 In 2003 Mexico revoked its R&D credit, which wasaimed at encouraging United States TNCs to shift moreR&D activities to their Mexican affiliates, as thecomplexity of the system rendered it ineffective (EIU2004t).

51 A company is regarded as “locally owned” if 51% ofits voting shares are owned by Brazilian institutionsor individuals.

52 In Israel an extra dollar of R&D subsides granted tothe manufacturing sector was found to increase thelong-run company-financed R&D expenditures by 41cents on average, which was lower than expected giventhe dollar-for-dollar matching requirements upon whichmany subsidized projects are based (Lach 2000).

53 The problem of input price inflation refers to thedifficulty of distinguishing the volume effect from theprice effect. If incentives do stimulate extra R&D, theywill result in an increased demand for input into R&D.If the supply of R&D inputs such as highly skilledprofessionals is limited, increased demand will resultin raising the price of R&D inputs. Thus an increasein R&D expenditures as a result of public support maypartly be accounted for by the inflated price of theseinputs.

54 In Thailand, for example, few firms took advantageof an R&D tax break because of complicatedregulations and the cost of the investment involvedin the R&D schemes (EIU 2004p). The Governmentof Mexico in 2003 decided to revoke a 30% R&D taxcredit that had provided incentives for United Statescompanies to shift more R&D activities to their foreignaffiliates in Mexico. The original measure had beenineffective due to too many exceptions and clauses(EIU 2004t).

55 See Sanz 2004. Other estimates suggest that there wereover 1,000 science parks worldwide in 1990 (Anderssonet al. 2004, p. 152).

56 In the Russian Federation, the State Duma approvedthe first draft legislation in June 2005 to set up researchzones of up to 2 square kilometres and offered taxincentives. See “Duma bill aims for tech parks”, TheMoscow Times, 10 June 2005.

57 Communication from Hsinchu Science-Based IndustrialPark, April 2005.

58 See, for example, UNIDO, Technology Parks: Tunisia,at: www.unido.org.

59 See e.g., United Nations Educational Scientific andCultural Organization (UNESCO), www.unesco.org.

60 See e.g., www.theinnovationhub.com.61 In Singapore for instance, the transition from

manufacturing-related R&D to applied and even basicresearch began to take place as a consequence ofproactive government policies targeted at TNCsinvolved in such manufactures as hard disk drives andtelecom equipment in the 1990s (Amsden and Tschang2003), and the biotechnology industry in recent years.In economies such as the Republic of Korea and TaiwanProvince of China, the upgrading from manufacturingto development and then to research was mainlythrough domestic efforts rather than foreign TNCs.

62 See for instance, Porter 1997 and Dunning 1997.63 See Roelandt and Den Hertog (1999) for cluster-based

policy measures in various countries. In Thailand, forexample, the Board of Investment (BOI) in 2004initiated new investment packages for specific industrialclusters concerned with the manufacture of hard diskdrives and semiconductors. Eligible firms in theseclusters are not only final producers but also suppliers.

64 The relationship between firm size and innovativeactivity has been found to be ambiguous (e.g. Vossen1996).

65 For instance the “Regionalisation Finance Scheme”in Singapore is a fixed-cost financing programmedesigned to assist Singapore-based enterprises set upoperations abroad. It is part of the Government’s effortto assist Singapore-based enterprises to internationalizetheir operations, sell in the global market place andleverage global resources in order to grow.

66 The “go global” strategy of the Government of Chinawas formulated in the mid-1990s and formallyannounced in the Suggestion from the CentralCommission of the CCP on the Tenth Five-Year Planon National Economy and Social Development passedin October 2000 (www.people.com.cn).

67 These regulations include the 2004 InterimAdministrative Measures on the Approval of OverseasInvestment Projects (NDRC), the 2004 Provisions onIssues Concerning the Approval of Overseas Investmentand Establishment of Enterprises (MOC), the 2004Circular on the Supportive Credit Policy on KeyOverseas Investment Projects Encouraged by the State(NDRC and EIBC) and various other regulations andcirculars on foreign currency management, statistics,performance assessment and State-owned assetsmanagement (People’s Republic of China, MOFCOM,www.fec.mofcom.gov.cn).

68 An example of mutually beneficial cooperation existsbetween France and universities in China. Thiscooperation has resulted in the training of highlyqualified researchers who could find employment bothin local institutes and firms, and in foreign affiliatesof French TNCs (UNCTAD 2005d).

69 Under the Colombo Plan for Cooperative EconomicDevelopment in Asia and Oceania, donor countries haveoffered scholarships and fellowships to developingcountries in the region since 1951. The Plan supportedthe development of scientific and technologicalexpertise in a number of countries (UN MillenniumProject 2005, p. 92).

As shown in Chapter VII, national policiesare critical for strengthening the NIS, as well asfor encouraging and facili tating foreigninvestment in R&D and maximizing the benefitsfrom it. However, in an increasingly integratedglobal economic system, national policies cannotbe pursued in isolation. Their reach and impactare often influenced by legal and regulatoryarrangements at the international level. In fact,most of the recent international investmentagreements (IIAs) contain specific provisionsgoverning FDI in R&D.1 Other internationalregulatory frameworks that have a direct bearingon FDI in R&D include those that addressintellectual property right (IPR) regimes and thegeneration, transfer and diffusion of science,technology and innovation, agreements thatencourage home-country measures and corporatesocial responsibili ty, and internationalcooperation agreements in science andtechnology.

This chapter examines these variousagreements in turn, and identifies issues ofspecial relevance to FDI in R&D, both in termsof facilitating national policies to encourage FDIin R&D or restricting the policy space availableto countries to design and implement suchpolicies. These issues include entry andestablishment of investment in R&D,performance requirements, use of incentives toencourage FDI in R&D, free movement of keypersonnel, protection of investment in R&D,home-country measures and corporate socialresponsibili ty, protection of IPRs andinternational cooperation in R&D.

��������

THE INTERNATIONAL FRAMEWORK

A. Internationalinvestment agreements

1. Entry and establishment

In general, IIAs do not impose restrictionson the entry and establishment of R&D-relatedinvestment, unless, for example, reasons ofnational security are involved.

Of special significance in relation to theentry and establishment of FDI in R&D is theWTO General Agreement on Trade in Services(GATS). The GATS addresses market access forR&D services through commercial presence (akinto FDI) if scheduled in a member country’s listof commitments. It applies to any measureaffecting trade in services (if supplied on acommercial basis), and R&D is defined andconsidered as one of the many services.2 Underits positive list approach, countries indicate theindustries they want to l iberalize – with orwithout conditions. A number of countries haveundertaken liberalization commitments in R&Dservices, some of them with partial limitationsattached.

As of March 2005, 49 out of 136 members’schedules included commitments on R&Dservices (i.e. about 36% of WTO members haveundertaken commitments in this area).3 Themajority of these (27 schedules) includedcommitments in all three categories of R&D:natural sciences, social sciences and humanities,and interdisciplinary R&D (figure VIII.1).

228 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Twenty-six developing-country members,including two LDCs (Gambia, Nepal), 11developed countries and 12 transition economies(including 6 new EU members) had undertakencommitments on FDI in R&D.4

In terms of the number of commitments indifferent fields,5 more countries undertookcommitments on R&D in the social sciences andhumanities than in the natural sciences andinterdisciplinary R&D (figure VIII.2). A littlemore than half of the commitments made bydeveloped countries were related to R&D in thesocial sciences and humanities; the rest weredistributed equally between the natural sciencesand interdisciplinary R&D. Commitments bydeveloping countries, on the other hand, wereevenly distributed across all three fields.Countries formerly classified as transitioneconomies scheduled two-fifths of theircommitments in relation to R&D in the socialsciences and humanities, with the remainderbeing distributed equally between the naturalsciences and interdisciplinary R&D.

Most commitments have no limitationsattached. However, developing countriesundertook more partial commitments with respectto market access, while developed countriesundertook more partial commitments regardingnational treatment. Countries formerly classifiedas transition economies made no partialcommitments (figure VIII.3).

Partial limitations relate mainly to theownership and control of enterprises involvedin R&D services. Typical restrictions on FDIin R&D listed in the GATS schedules includerequirements to have a local partner in jointventures, a limit on the shares of foreign capital,

Source: UNCTAD, based on GATS schedules of specificcommitments (as of March 2005).

Figure VIII.1. Schedules with commitmentson commercial presence in R&D services

Figure VIII.2. Level of commitments undercommercial presence for R&D activities

Source: UNCTAD based on GATS schedules of specificcommitments (as of March 2005).

Note: MA = market access (Article XVI); NT = national treatment(Article XVII). Figures on “full commitments” relateto schedules’ entries where WTO members havecommitted to apply no MA or NT limitations (i.e. “none”entries, in terms of GATS). “Partial limitations” countthose services in which only particular l istedrestrictions apply, as listed in members’ schedulesof specific commitments.

nationality requirements for members of theboard of directors and key personnel, and variouslicensing and registration requirements. Mostlimitations reflect the desire to maintain somedegree of national control, while at the same timecreating an enabling framework for the inflowof investments into R&D. This is the main effectof limitations on the participation of foreigncapital, particularly in those cases in which thelimit is set at 49% of the equity share or below.The combination of l imits on foreignparticipation with the requirement to conductR&D through a joint venture with local partnersmay be intended to ensure that spill-over oftechnological innovation to local partners takesplace. A similar objective may also be sought bythe requirements to employ nationals as keypersonnel and as members of the board ofdirectors.

Some WTO members have includedlicensing and registration requirements aslimitations in their schedules. In principle, priorlicensing and registration requirements are notnecessarily contrary to the GATS, and it is notmandatory to list them as limitations in theschedules of commitments, unless a countrywishes to use them as instruments to discriminateagainst the establishment of a foreign commercialpresence. This may be done to ensure that onlysuch R&D that meets national policy

Full commitments Partial commitments

Nu

mb

er

of

co

mm

itm

en

ts

45

40

35

30

25

20

15

10

5

0MA MA MANT NT NT

R&D naturalsciences

R&D socialsciences

InterdisciplinaryR&D

87 scheduleswithout any

commitment onR&D services

(64%)

22 scheduleswith commitments

only on someR&D services

(16%)

27 scheduleswith commitments

on all threeR&D services

(20%)

229CHAPTER VIII

requirements is permitted, and to protect nationalR&D development against external competition.

In sum, R&D is generally not a restrictedactivity in IIAs. Rather, international agreementsconfirm the predominance of policies seeking toencourage and facilitate FDI in R&D. However,as the experience of GATS suggests, countriesmay restrict liberalization in this area in orderto increase the likelihood of reaping the fullbenefits from FDI in R&D.

2. Performance requirements

BITs generally do not address performancerequirements with regard to the entry of FDI. Asto the post-entry treatment of FDI, nationaltreatment and other standards of treatment andprotection apply across the board.

A small number of IIAs contain specificprovisions prohibiting the use of performancerequirements that mandate investment in R&Dactivities as a condition for entry and operation,unless they are attached to the receipt orcontinued receipt of an advantage. For example,the 1998 BIT between Bolivia and the UnitedStates (as well as 12 other BITs concluded bythe United States) prohibits countries to “mandateand enforce, as a condition for the establishment,

acquisition, expansion, management,conduct or operation of a coveredinvestment, any requirement (including anycommitment or undertaking in connectionwith the receipt of a governmentalpermission or authorization)” to “[…] (f)carry out a particular type, level orpercentage of research and developmentin the Party’s territory” (Article IV).6

Similar prohibitions can be found in the2002 BIT concluded between Japan and theRepublic of Korea7 and in the 2002 NewAge Economic Partnership Agreementbetween Japan and Singapore.8

This approach limits the possibilityfor countries to devise policies to mandateR&D activities by foreign investors as acondition for their entry and operation, andtherefore narrows their policy space, or atthe least the mandatory character of suchpolicies. They will have to be used onlyin connection with an encouragement toforeign investors (i.e. an incentive) but notas a self-standing obligation.

A different approach has been takenby NAFTA, where there is no prohibition

of performance requirements attached to the entryand operation of FDI that mandate R&D activitiesin the territory of the host country (Article1106(1)). Moreover, NAFTA explicitly allowstheir use as a condition for the receipt orcontinued receipt of an advantage (Article1106(4)).9 This approach implies that countriesare free to attach conditions to the entry andoperation of investments in the form ofmandatory involvement in R&D activities,provided other core disciplines of the applicableagreements (such as national treatment, MFN,protection against expropriation) are adhered to.It also implies that countries are specificallyallowed to apply such conditions by attachingthem to an incentive.

Following the NAFTA approach, the 2004version of the United States model BIT10 and allinvestment chapters in subsequent FTAsconcluded by the United States, the 2004Canadian Foreign Investment Protection andPromotion Agreement model (hereinafter theCanada BIT model), the 2004 Japan-MexicoNew-Age Economic Partnership Agreement, andthe 2004 BIT between the United States andUruguay do not prohibit the use of performancerequirements relating to the establishment andoperation of FDI in R&D.

Figure VIII.3. Level of commitments of R&Dservices, by group of countriesa

Source: UNCTAD, based on the GATS schedules of specificcommitments (as of March 2005).

a Based on former United Nations geographical classification; seefootnote 4.

b Excluding LDCs.

Note: MA = market access (Article XVI); NT = national treatment(Article XVII). Total commitments of developing countries(excluding LDCs) amount to 51, those of developed countriesto 19, those of countries formerly classified as transitioneconomies to 29 and those of LDCs to 4. All of these coverboth market access and national treatment.

Full commitments Partial commitments

Num

ber

ofcom

mitm

ents

0

10

20

30

40

50

60

MA NT MA NT MA NT MA NT

Developingcountriesb

Developedcountries

Economiesin transition LDCs

230 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

3. Incentives

As stated above, R&D performancerequirements relating to the entry and operationof FDI may be expressly and specifically allowedwhen they are a quid pro quo for investmentincentives (i.e. when they are a condition for thereceipt or continued receipt of an advantage).This further illustrates the importance countriesaccord to R&D policies and encouragements atthe international level.

In a number of countries, R&D hastraditionally been undertaken or encouraged andsupported by the government. Of key importancehere is the protection or denial of access byforeign investors to government-funded R&Dprogrammes. This practice has been identifiedas a barrier to investment by the 2005 UnitedStates National Trade Estimate Report on ForeignTrade Barriers .11 It is a sensitive issue forcountries that grant substantial support to publicand private research, but also for developingcountries wishing to foster indigenous R&Dcapacity. Some countries have seen the need tosafeguard flexibility for targeted encouragementand support policies at the international level byintroducing reservations and exceptions to theircore commitment on non-discrimination.

One approach for achieving this objectiveis to list a specific reservation relating to R&Dsubsidies. This approach has been favoured, forexample, in the 2004 Agreement between Japan andthe United Mexican States for the Strengtheningof the Economic Partnership, in which the scheduleof Japan under Annex 7 (Reservations for FutureMeasures) provides that “National Treatment maynot be accorded to investors of Mexico and theirinvestment with respect to subsidies for researchand development”.12

In other cases, countries do not single outR&D incentives, grants or governmentprogrammes, but have adopted generalreservations and exceptions to national treatment,MFN, and provisions on entry of personnelrelating to “subsidies or grants provided by aParty or a state enterprise, including government-supported loans, guarantees and insurance”(Canada BIT model, Article 9.5 (b)), with a viewto denying foreign investors access to suchsubsidies. This approach also applies to anysubsidies, grants or government programmes inthe area of R&D. It may be added that a preferredavenue for dealing with incentives for aninvestment in R&D activities is the conclusion

of individual investment or State contracts,whereby a government enters into an agreementwith an investor that can also include provisionson subsidies for FDI in R&D (UNCTAD2004d).13 General standards of treatment stillapply across the board, however, including toState contracts.

At the multilateral level, the WTOAgreement on Subsidies and CountervailingMeasures (SCM) deals specifically withsubsidies, including R&D subsidies. It aims atreducing and eventually eliminating subsidies thatdistort international trade in goods. Although itregulates subsidies related to trade in goods only,R&D subsidies could be challenged under theSCM Agreement if they are provided for servicesthat are used in the production of exported goods,and hence can be considered a cross-subsidy ongoods. This is also relevant in the case ofsubsidies for R&D services that form an inputinto traded goods.

The GATS is directly relevant to subsidiesfor FDI in R&D. Fifteen WTO members havelodged horizontal limitations (i.e. measures thataffect all services listed in the schedule) tonational treatment as far as access to R&Dprogrammes is concerned,14 thus ensuring againstaccess of foreign investors to such subsidies. Ifsuch limitations are not scheduled, it may be thatnational treatment and MFN treatment apply tosubsidies in industries that have been liberalized.On the other hand, unlike the General Agreementon Tariffs and Trade (GATT), which issupplemented by the SCM in the field of tradein goods, the GATS does not have specificdisciplines on subsidies in relation to services.However, Article XV of the GATS envisagesfuture negotiations to develop disciplines in thisarea.

Some WTO members have also introducedbroader limitations to national treatment withregard to government subsidies throughhorizontal restrictions in their schedules, whichapply across the board to all sectors and typesof subsidies, including those in R&D. Specifichorizontal limitations on subsidies may followfrom national policies that reserve governmentassistance only to national research institutionsand/or firms. Finally, a few limitationsconcerning subsidies may also be found in WTOmembers’ schedules dealing with particularindustries. In these, restrictions on subsidies mayapply also to public assistance to R&D.

231CHAPTER VIII

4. Key personnel

To ensure the effective operation of aninvestment, TNCs may wish to employ key foreignpersonnel with relevant technical skills, includingR&D personnel, while host countries may wish toensure that their nationals have the advantage ofworking in foreign affiliates so as to facilitate thetransfer of knowledge and skills. To this end, a hostcountry may impose restrictions on the employmentof key foreign personnel. While a majority of IIAshave no specific provisions dealing with themovement of key personnel,15 some treaties includeprovisions related to the admission of individualsor employees of an investor in connection with aninvestment so as to facilitate the employment ofkey personnel, including R&D personnel. Theseprovisions apply to investors of the othercontracting party, and, specifically, to personnelemployed by an investor, for the purpose ofestablishing, administering or advising on theoperation of an investment (see, for example, theUnited States-Romania BIT of 1992, Article II.3).The Canadian BIT model (Article 6) specificallyseeks to facilitate the entry of foreign nationalsemployed in a capacity that requires specializedknowledge. The Australia-Thailand FTA, like manyrecent FTAs, has a separate chapter on the“Movement of natural persons” that covers naturalpersons employed by an investor in respect of aninvestment, with a separate entry for “specialists”.16

The same approach is taken by the GATS mode 4(“Presence of natural persons”) where countrieshave specifically scheduled commitments on marketaccess concerning “intra-corporate transferees”,including specialists.

Two approaches prevail in IIAs. One consistsof an obligation by the host country to permit entryand sojourn subject to its laws and regulations onthe entry of aliens (e.g. the 1992 BIT between theUnited States and Kazakhstan and the 2004Canadian BIT model). Another approach providesfor an obligation by the contracting parties to “givesympathetic consideration to applications” for theentry and sojourn of persons (e.g. the 2003 China-Germany BIT).

5. General protection of FDI in R&D

In terms of protection, most IIAs do notaddress the issue of FDI in R&D specifically, butrefer to the protection of investment in general.Three issues are particularly relevant to theprotection of FDI in R&D: the protection of IPRs

by including them in the definition of investment;provisions on the free transfer of returns arisingfrom R&D activities; and the application of thenational treatment/MFN standard to foreigninvestors investing in R&D activities.

By using a broad definition of the term“investment”, IIAs provide protection to bothtangible property (e.g. research and testlaboratories) and intangible assets such as IPRsthat form part of the assets of an investor (e.g.patents or test data on R&D results). Theinclusion of IPRs in the definition of assets takesinto account their economic value. This has cometo be of crit ical importance and central toinvestment protection (UNCTAD 1998, p. 35).The vast majority of IIAs define IPRs broadly.For example, the 1999 BIT between Croatia andFinland states in Article 1 that:

‘’The term ‘investment’ means every kindof asset established or acquired byinvestors of one Contracting Party in theterritory of the other Contracting Party inaccordance with its laws and regulationsand shall include in particular, though notexclusively:

…

d) intellectual property rights including,but not limited to, copyrights andneighbouring rights, industrial propertyrights, trademarks, patents, industrialdesigns and technical processes, rights inplants [sic] varieties, know-how, tradesecrets, trade names and goodwill;’’

This emphasis on IPRs forming part of theprotected assets signifies that their economicvalue will be taken into account in case ofcompensation. This may provide additionalcomfort for investment in R&D where IPRs arecrucial – both the ones that form part of the assetscontributed by the investor when making theinvestment, and the ones that derive from theoperation of the investment (i.e. the carrying outof R&D activities by TNCs in a host country).

Furthermore, by including IPRs in thedefinition of protected investment, the protectionagainst direct and indirect expropriation offeredby an agreement could potentially also encompassprotection against compulsory licensing, whereit can be shown that this has an expropriatorypurpose and that it is carried out in breach of theprotective standards of treatment contained inthe applicable IIA and in disregard of the relevantprovisions of IPR agreements (UNCTAD 2001).

232 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

To avoid such a far-reaching interpretation of theexpropriation provision, several agreements, suchas the recent United States and Canada modelBITs, explicitly carve out from the scope ofexpropriation “the issuance of compulsory licensesgranted in relation to IPRs in accordance with theTRIPS Agreement” (United States model BIT,Article 6.5; see also the Canada model BIT, Article13.5).

When it comes to the free transfer of funds– licence fees, “royalties, technical assistance andtechnical fees … accruing from any investmentof the investors” (Article 6 of the 1997 Malaysia-Ghana BIT) – relevant provisions in IIAs alsoapply to FDI in R&D, as the proceeds of suchinvestment generally take the form of licence feesand other royalties.

As indicated above, countries have to becareful when designing and implementing theirnational policies if they want to reserve somespecial treatment to local R&D companies, andif they do not want to give to foreign investorsaccess to all their available incentives or supportpackages.

Finally, investors can also benefit from thegeneral protection provided by national treatmentand MFN standards. A direct implication forinvestment projects in R&D is that (unlessexceptions apply) any subsidies, grants andgovernment funds are available to foreign investorson the same conditions as they are for nationalcompanies performing R&D. But, as indicatedearlier, several treaties seek to “carve out” accessto such programmes from the scope of theprovision on national treatment. A more generalcarve-out, as far as taxation issues are concerned(e.g. as in the Canada model BIT of 2004), mayalso provide for the possibility to give specialtreatment to domestic firms when governmentpolicy takes the form of tax incentives.

6. Home-country measures andcorporate social responsibility

Some international agreements encouragethe use of home-country measures in the area ofR&D. For example, Article 66 (2) of the WTOAgreement on Trade-related Aspects of IntellectualProperty Rights (TRIPS) states that “[d]evelopedcountry Members shall provide incentives toenterprises and institutions in their territories forthe purpose of promoting and encouragingtechnology transfer to least developed country

Members in order to enable them to create a soundand viable technological base”.

In addition, home countries of TNCs canalso encourage their firms to participate activelyin international cooperation on R&D by investingin R&D activities and establishing linkages withlocal and regional firms. Such encouragement isalso sought by international agreements (boxVIII.1).

Even when provisions are hortatory (i.e.non-binding), some of these instruments providean enabling framework within which TNCs areencouraged to operate and invest in R&D-relatedactivities in developing countries.

Source: OECD.

Box VIII.1. The OECD Guidelines forMultinational Enterprises

Chapter VIII (“Science and Technology”) ofthe 2000 OECD Guidelines for MultinationalEnterprises provides that corporations should:

“1. Endeavour to ensure that their activities arecompatible with the science and technology (S&T)policies and plans of the countries in which theyoperate and as appropriate contribute to thedevelopment of local and national innovativecapacity.

2. Adopt, where practicable in the course of theirbusiness activities, practices that permit the transferand rapid diffusion of technologies and know-how,with due regard to the protection of intellectualproperty rights.

3. When appropriate, perform science andtechnology development work in host countries toaddress local market needs, as well as employ hostcountry personnel in an S&T capacity andencourage their training, taking into accountcommercial needs.

4. When granting licenses for the use of intellectualproperty rights or when otherwise transferringtechnology, do so on reasonable terms andconditions and in a manner that contributes to thelong term development prospects of the hostcountry.

5. Where relevant to commercial objectives, developties with local universities, public researchinstitutions, and participate in co-operative researchprojects with local industry or industryassociations.”

233CHAPTER VIII

B. International rulesrelating to IPRs

International rules on IPR protection areincreasingly setting parameters for nationalpolicies in the area of the generation, transferand diffusion of technology.17 Such rules mayprovide incentives for TNCs to undertake FDIin R&D,18 but at the same time they may alsorestrict a country’s freedom to implement nationalpolicies concerning IPRs and R&D development.

Most relevant here is the TRIPSAgreement. That Agreement recognizes in itspreamble “the underlying public policy objectivesof national systems of intellectual property,including developmental and technologicalobjectives”. Furthermore, in Article 7 (entitled“Objectives”), it states as objectives of IPR

protection and enforcement to “contribute to thepromotion of technological innovation and to thetransfer and dissemination of technology, […]in a manner conducive to social and economicwelfare, and to a balance of rights andobligations”. It also recognizes the authorizationof WTO members to control the abuse of IPRs.19

The TRIPS Agreement establishes internationalminimum standards of protection andenforcement for R&D-relevant IPRs such aspatents and undisclosed information (tradesecrets).20 These standards may contribute tomaking host countries safer destinations for FDIin R&D by obliging the provision of effectiveprotection of IPRs (box VIII.2; see also chaptersV and VII). A number of recent IIAs haveextended the TRIPS minimum standards, thussetting further disciplines on the nationalregulation of IPRs (“TRIPS-plus”) (box VIII.3),21

Box VIII.2. TRIPS minimum IPR standards of relevance to FDI in R&D and TRIPSflexibilities of relevance to host-country R&D

Minimum standards

• The TRIPS Agreement contains provisions onnational treatment and MFN. Both apply tonatural and juridical persons with regard to theprotection of intellectual property. Theseprovisions remove any discrimination betweendomestic and foreign firms in the protectionof intellectual property.

• The Agreement extends protection to bothproduct patents and process patents in all fieldsof technology, including, with certainqualifications, pharmaceutical andbiotechnological products.a

• It also obliges members to make patentsavailable, without discrimination as to the placeof invention, the field of technology or whetherthe products are imported or produced locally.b

The latter may be interpreted as prohibitingthe imposition on foreign investors of “localworking” requirements for patents (providingcompulsory licensing or revocation of thepatent if the protected product is not producedlocally but imported).c The protection offoreign investors’ R&D assets is thereby madeless dependent on a particular performance.

Flexibilities

• The TRIPS Agreement leaves members thefreedom to define criteria of patentability,namely novelty, inventive step and industrialapplicability (Article 27.1).

• It appears not to contain obligations to makepatents available for new uses of knownproducts (“second uses”) (although there is noWTO practice on this matter).

• It contains no obligation to provide patentson “plants and animals other than micro-organisms, and essentially biological processesfor the production of plants or animals otherthan non-biological and microbiologicalprocesses” (Article 27.3(b)).

• The TRIPS Agreement gives WTO membersthe discretion to include in their patent lawsthe obligation for a patent applicant to indicatethe best mode for carrying out the inventionknown to the inventor at the filing date or,where priority is claimed, at the priority dateof the application (Article 29). Thiscomplements members’ obligations under thesame provision to require a patent applicantto disclose his/her invention in return forobtaining a patent. This information is usuallypublished 18 months after the filing of theapplication. The repository of patentinformation is perhaps the single largestexisting source of technological informationavailable for developing countries.

• The Agreement also allows limited exceptionsto exclusive patent rights “provided that suchexceptions do not unreasonably conflict witha normal exploitation of the patent and do notunreasonably prejudice the legitimate interests

/...

234 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

assuming that this may provide an additionalincentive for FDI in R&D.

At the same time, these internationalobligations restrict national policy space withinwhich IPRs and R&D development policies canbe implemented. One example is the TRIPSprovision that denies WTO members the right toexempt certain fields of technology from patentprotection or to limit the latter to processes only(which would leave all new products in the publicdomain).22 Also, the potential prohibition of localworking requirements (box VIII.2) could reducea host country’s possibilities of promoting access

by local researchers to foreign technologies.Moreover, TRIPS-plus agreements make it moredifficult for local R&D actors to access first-generation inventions (due to some TRIPS-plusobligations to extend the patent term for unknownuses of already patented products). As a result,some countries have resisted the inclusion of thefull range of provisions noted in box VIII.3.

The effects that limitations of nationalpolicy space may have on technologicaldevelopment often depend on a country’s levelof domestic technological capacity (chapter VII).In the past, some countries have used lax IPR

Source: UNCTAD, based on UNCTAD-ICTSD 2005.

a Article 27.3(b), TRIPS Agreement, contains optional exemptions from patentability in the area of biotechnologyproducts. Article 65.4, TRIPS Agreement, authorized for developing countries a transition period until 1 January2005 for products not protectable under national patent law on 1 January 2000. This applies mainly to pharmaceuticalproducts.

b Article 27.1, TRIPS Agreement.c According to some views in the literature, the non-discrimination obligation under Article 27 TRIPS does not

apply to bona fide distinctions between local and foreign production, in particular in the area of public health andthe promotion of affordable access to essential medicines. For details, see UNCTAD-ICTSD 2005, chapter 25(“Patents: Non-voluntary Uses (Compulsory Licenses)”).

d Except for some basic obligations such as national treatment and MFN, see Article 66.1. The TRIPS Agreementalso provides that this transition period has to be extended by the WTO Council for TRIPS upon duly motivatedrequest by an LDC. The Maldives is the first LDC to have been granted such an extension (Decision of the TRIPSCouncil on 15 June 2005, IP/C/35).

e See paragraph 7 of the Doha Declaration on the TRIPS Agreement and Public Health, WT/MIN(01)/DEC/W/2,and the Decision of the Council for TRIPS of 22 June 2002, IP/C/25.

of the patent owner, taking account of thelegitimate interests of third parties” (Article30). This provision has been used in mostjurisdictions to establish exceptions to patentrights for some forms of experimental orresearch uses. The scope of such exceptionsvaries by country.

• The Agreement leaves members the freedomto issue compulsory licences to third parties,provided a number of conditions are met(Article 31), as confirmed in the DohaDeclaration on TRIPS and Public Health. Inorder to facilitate the use of compulsorylicensing by members lacking sufficientdomestic pharmaceutical manufacturingcapacities, the General Council Decision of 30August 2003 (WT/L/540) on “Implementationof Paragraph 6 of the Doha Declaration on theTRIPS Agreement and Public Health” waivedcertain requirements under Article 31 on atemporary basis. Members of the WTO have

Box VIII.2. TRIPS minimum IPR standards of relevance to FDI in R&D and TRIPSflexibilities of relevance to host-country R&D (concluded)

so far failed to replace this transitional waiverwith a permanent amendment to theAgreement.

• The TRIPS Agreement also gives membersdiscretion to choose their own regime ofexhaustion of IPRs (Article 6), equallyconfirmed in the Doha Declaration on TRIPSand Public Health.

• The Agreement authorizes the control of IPRabuses through competition laws and policies,in particular in licensing agreements (e.g.between local researchers and TNCs) (Article8.2 and Article 40).

• LDC members have been allowed extendabletransition periods for the implementation ofthe TRIPS minimum standards (1 January2006 in general;d 1 January 2016 for theapplication of patent rights and rules on theprotection of undisclosed information topharmaceutical products).e

235CHAPTER VIII

protection to encourage development in someindustries, and strengthened their IPR protectionpolicies once these industries had prospered. TheIndian pharmaceutical industry and its interactionwith patent regimes is one example. This industryhas attained its high level of development partlybecause the Indian Patent Act of 1971 deniedpatent protection to pharmaceutical products. Thisgave the domestic industry an opportunity tobuild up capabilit ies in imitative productinnovation. Some Indian companies developedtheir own expertise and technological capacity,reflected in sharply increased R&D expendituresin the 1990s, from $36.5 million in 1990/91 to$73.6 million in 1999/00 (UNCTAD 2003d, p.109). The introduction of patent protection from1 January 2005, in fulfilment of India’s TRIPSobligations, corresponds with calls from Indianpharmaceutical companies for enhancedprotection of their new assets. Existing firms cannow enjoy patent protection for their earliertechnological innovations.

For these and other reasons, it is essentialfor countries, in particular developing countriesand LDCs, to understand and make use of theflexibilities contained in the TRIPS Agreement(box VIII.2). There is also a need for additionaltechnical assistance and capacity building, asprovided for in Article 67 of the TRIPSAgreement, with a view to facili tating thedevelopment-oriented implementation of IPRsfor the promotion of local R&D capacities.

C. Internationalcooperation in R&D

Several international agreements also aimto encourage international cooperation in the areaof R&D. They do so by establishing cooperationamong the State parties to the agreements,thereby providing an enabling framework forprivate-sector R&D projects and FDI in R&D.Such cooperation can either take place in abroader context, for example at the regional level,or be encouraged through specific science andtechnology cooperation agreements. Given thewill of the parties to do so, both approaches canhelp build domestic innovatory capacity andprovide a framework in which national policiesaimed at encouraging FDI in R&D can bedeveloped to benefit from the greater impact and/or stronger support of the international community.

As far as the broader cooperation contextis concerned, some IIAs, particularly some recentFTAs, contain provisions promoting R&Dcollaboration in scientific and industrialendeavours. This may involve joint researchprojects in fields of common interest, theexchange of scientists and researchers and thefostering of relations between research centres.23

Several agreements extend this to the promotionof FDI in R&D (box VIII.4). Internationalcooperation in the area of R&D is particularlypertinent at the regional level within the contextof regional economic integration. Here, it could

Box VIII.3. TRIPS-plus provisions of potential relevance to FDI in R&D and local R&D

Source: UNCTAD.

a There is no WTO jurisprudence or authoritative interpretation on this matter.

A number of recent IIAs require their partiesto:

• Extend the patent term in cases of delays inthe granting of the patent caused by theregulatory approval process (mainly in the fieldof medicines).

• Provide patents for new uses of knownproducts (“second uses”), as opposed to theTRIPS Agreement.

• Extend patent protection to plants and animals.• Provide for exclusive rights in pharmaceutical

test data (Article 39.3 of the TRIPS Agreementmay be interpreted as leaving Members thefreedom to protect such data through non-exclusive rights only).a

These provisions make R&D activities moreexpensive and complicated than before forcompetitors. For example, a system of dataexclusivity prevents regulatory authoritiesresponsible for granting marketing approval fromrelying on test data first submitted by the dataoriginator. In order to receive marketing approval,the competitor has to carry out the same tests asalready undertaken by the data originator. Thecompetitors are thus obliged to focus their R&Dactivities on the reproduction of expensive testing,instead of concentrating efforts on follow-on R&Dthat could improve the existing products or adaptthem to particular local needs.

236 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

lead to the development of competitive industries,in which members can pool their resources, sharethe costs and risks and enhance opportunities forregional or local enterprises.24 However, wherethis approach ignores foreign investors fromoutside the region, i t may risk excluding asignificant source of technology and cooperation(UNCTAD 2001).

Science and technology cooperationagreements are another avenue of internationalR&D cooperation that has a direct bearing on FDIin R&D. These agreements focus specifically oninternational R&D cooperation, and offer aframework within which countries can developpolicies encouraging local and foreign investorsto participate in specific R&D projects.Frameworks established by such agreements canfacilitate the flow of information, the formation

of alliances, the pooling of financial resources,the joining of technological expertise andendowments, the financing of technologymatchmaking and the creation of public-privatesector partnerships. These global approaches areimportant for promoting FDI in R&D and, morebroadly, the internationalization of R&D.

* * *Within a globalized world economy,

national policies aimed at the development ofR&D capabilit ies are increasingly beingcomplemented by rule-making at the internationallevel. As this overview of internationalagreements has shown, R&D activities are givenspecial attention in a number of internationaltreaties, ranging from IIAs to international IPRregimes to international cooperation agreementsin the field of science and technology. This

Box VIII.4. The promotion of R&D investment in regional agreements

Source: UNCTAD.

a Similar provisions can be found in other agreements entered into by the European Community; see for examplethe Association Agreements with Algeria (art. 51), Chile (art. 36), Jordan (art. 64, 73), and the Palestinian LiberationOrganization (art. 38, 49), the Cooperation Agreements with Armenia (art. 51, 56), Brazil (art. 10, 16), the LaoPeople’s Democratic Republic (art. 10), Sri Lanka (art.4, 9), and South Africa (art. 55); and the PartnershipAgreements with Georgia (art.53) and the Russian Federation (art. 62, 77).

Several regional IIAs address investment inR&D. For example, Article 46 of the AssociationAgreement between the European Community andEgypt (which came into effect in 2004) – entitled“Investments and promotion of investments” –states that: “Cooperation shall aim at increasingthe flow of capital, expertise and technology toEgypt through, inter alia: appropriate means ofidentifying investment opportunities andinformation channels on investment regulations.[...] Cooperation may extend to the planning andimplementation of projects demonstrating theeffective acquisition and use of basic technologies”.In addition, the agreement includes specific clauseson cooperation in science and technology (Article43) and industrial cooperation, including in R&D(Article 45). In this case, the agreement seeks topromote R&D and facilitate FDI and technologytransfer to the developing partner, therebystrengthening its R&D capacity.a

Some regional groupings of developingcountries place emphasis on cooperation amongtheir members in the area of science and technology,identifying R&D as a specific area for cooperation.This approach has been taken in Protocol III onIndustrial Policy Amending the Treaty Establishingthe Caribbean Community: “Mindful of the

imperatives of research and development andtechnology transfer and adaptation for thecompetitiveness of Community enterprises on asustainable basis” (Preamble), the contractingparties seek to adopt measures to promote market-led research, technological development, encouragepublic/private sector cooperation in research andtechnological development activities and facilitatecooperation between private sector enterprises tointegrate the results of research and technologicaldevelopment (Article VIII, replacing Article 44with Article 43.1 and 2(a) and (c)).

Within the broader ASEAN cooperationframework, the Agreement on ASEAN EnergyCooperation of 1986 (as amended in 1995)specifically identifies private sector participationin the cooperation among ASEAN membercountries in the area of technological research,development and demonstration (Article 1.2(iii)).

Similarly, the Treaty establishing theCommon Market for Eastern and Southern Africa(COMESA) includes specific clauses that seek topromote industrial R&D, the transfer, adaptationand development of technology, and linkagesthrough the provision of investment incentives toindustries (Treaty Establishing COMESA, Article100d).

237CHAPTER VIII

multifaceted framework imposes legal andregulatory measures and standards that affect theability of countries to devise their own policiesin this regard and to develop their innovativecapabilit ies, including through theinternationalization of R&D. The implicationsfor R&D development vary by the level ofdevelopment of countries and by the types ofinternational agreements involved.

Notes1 Unless referenced, all IIAs (for a definition, see chapter

I, footnote 45) referred to in this chapter can be foundin UNCTAD 2005e, 2004c, 2002b, 2000 and 1996band at www.unctad.org/iia.

2 According to the Provisional Common ProductClassification (CPC), used by most WTO members inthe GATS context, the definition of R&D coversservices relating to scientific progress achieved in thevarious fields of the natural sciences (CPC 851), socialsciences and humanities (CPC 852) andinterdisciplinary R&D (CPC 853) in three areas: basicresearch, applied research and experimentaldevelopment. FDI in any of these fields is then coveredby the concept of commercial presence.

3 Compared to commitments in other sectors and areasof activity, this number is quite modest.

4 One schedule represents one WTO member country,except for the case of the European Communities andits 12 members (at the time of the signature of theGATS), which are counted as one single WTO member.All other current EU members are counted separately,as they all have presented individual lists ofcommitments to the GATS. From that perspective, thetotal number of WTO members is therefore consideredto be 136, instead of the official number of 148. The12 transition economies include six countries of South-East Europe and the CIS (Bulgaria, Croatia, The formerYugoslav Republic of Macedonia, Georgia, Kyrgyzstan,the Republic of Moldova) and six new EU members(the Czech Republic, Estonia, Hungary, Latvia,Slovakia, Slovenia).

5 WTO members were mostly free to choose thescheduling technique of their preference; not allcountries followed the CPC structure, or they modifiedit in certain aspects. Therefore, an interpretative effortwas required in some cases to make the content ofdifferent schedules compatible. As a result, these datashould be considered indicative in nature and may varyfrom those in other studies in this area.

6 The 12 other BITs concluded by the United States thatinclude similar provisions are with: Georgia (1994),Trinidad and Tobago (1994), Uzbekistan (1994),Albania (1995), Honduras (1995), Nicaragua (1995),Croatia (1996), Azerbaijan (1997), Jordan (1997),Mozambique (1998), Bahrain (1999), El Salvador(1999). The texts of these BITs can be found at(www.unctad.org/iia).

7 See Article 9.1.h, expressly prohibiting the use ofrequirements to “(h) achieve a given level or valueof research and development in its territory as acondition for investment and business activities in itsterritory”.

8 See Article 75, stating similarly in connection with“the establishment, acquisition, expansion,management, operation, maintenance, use or possession

of investments”.9 Some countries have chosen to go beyond this

permissive approach by making specific reservationsin this regard. For example, Canada has listed areservation in Annex 1 of NAFTA stating thatprohibiting the use of performance requirements(Article 1106(1)) does not apply to any requirement,commitment or undertaking imposed or enforced inconnection with a review under the Investment CanadaAct, to “carry out research and development”.

10 The 1994 model BIT of the United States prohibitedthe use of R&D performance requirements (ArticleVI(f)), not including conditions for the receipt orcontinued receipt of an advantage

11 See www.ustr.gov.12 See also Article 10, paragraph 8, of the Energy Charter

Treaty, available at: www.unctad.org/iia.13 Individual State contracts can provide more favourable

conditions for the investor than a treaty. This is usuallyconfirmed in BITs through a provision undertaking torespect all commitments made in specific agreements,including State contracts.

14 A typical horizontal measure concerning subsidies onR&D reads: “3) Unbound for subsidies for researchand development”. Australia, Brazil, Cambodia,Canada, Croatia, the EU (12), Finland, Iceland, Japan,the Republic of Korea, Mexico, Norway and Sloveniahave recorded such a restriction. Kuwait and Qatar havesimilar horizontal measures listed in their limitationson market access.

15 BITs, for example, typically submit this issue tonational laws and regulations.

16 Article 1002 of the FTA between Australia and Thailandstates that: “… i. “specialist” means a natural personwithin an organisation who possesses knowledge atan advanced level of technical expertise, and whopossesses proprietary knowledge of the organisation’sservice, research equipment, techniques, ormanagement; or a natural person with high-leveltechnical or professional qualifications and skills andexperience.’’

17 International IPR standards (as contained, for example,in the Paris Convention for the Protection of IndustrialProperty) lay out the main principles for the interactionof national IPR laws with foreign investors: nationaltreatment, right of priority and the independence ofpatents obtained for the same invention in differentcountries. Another core principle of the internationalintellectual property architecture, the MFN treatmentobligation, was only introduced with the WTO-TRIPSAgreement.

18 IPR protection may also help to build domestic R&Dcapacity and encourage domestic innovation – a matternot further explored here.

19 See Article 8.2: “Appropriate measures, provided thatthey are consistent with the provisions of thisAgreement, may be needed to prevent the abuse ofintellectual property rights by right holders or the resortto practices which unreasonably restrain trade oradversely affect the international transfer oftechnology.” Note that this provision does not allowmembers to deviate from their obligations, requiringthat national measures be “consistent with theprovisions of this agreement”. Unlike other WTOAgreements such as the GATT, the TRIPS Agreementdoes not contain a general exception clause.

20 Other categories of IPRs covered under the TRIPSAgreement are copyright and related rights, trademarks,geographical indications, industrial designs, and layoutdesigns of integrated circuits.

21 For an overview, see Fink and Reichenmiller 2005.

238 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

22 Article 27.1, TRIPS Agreement; see also box VIII.1.23 For example, the Australia-Thailand Free Trade

Agreement of 2004. In chapter 8 on “Trade inServices”, “Part III: Cooperation”, it spells out severalareas of cooperation. Article 808 states that “1. TheParties shall strengthen and enhance existingcooperation efforts in service sectors and developcooperation in sectors that are not covered by existing

cooperation arrangements, through inter alia: (a)research and development...”

24 R&D cooperation within the European Union illustratesthe benefits of this regional approach (see Article 163of the Treaty Establishing the European Community).Such cooperation among European countries in areasthat are very sensitive to security and economiccompetitiveness was facilitated at an early stage bythe Treaty Establishing the European Community.

REFERENCES

A.T. Kearney (2004). FDI Confidence Index, Vol. 7(Alexandria, VA: A.T. Kearney).

Abdel Musik, Guillermo (2004). “Trade innovationperformance of Mexico after NAFTA”, Working Paper,2004-01 (Vancouver: Centre for Policy Research onScience and Technology, Simon Fraser University)(www.sfu.ca/cprost/docs/guillermo1.doc).

Adeboye, Titus (1997). “Models of innovation and sub-SaharanAfrica’s development tragedy”, Technology Analysisand Strategic Management, 9, 2, pp. 213-237.

Aghion, Philippe and Peter Howitt (1998). Endogenous GrowthTheory (Cambridge, MA: MIT Press).

AlAzzawi, Shireen (2004). “Innovation, productivity andforeign direct investment-induced R&D spillovers”(Davis, CA: University of California), mimeo.

Alfaro, Carlos (2004). “Argentina: ICSID arbitration and BITschallenged by the Argentine Government”, Alfaro -Abogados, 21 December, www.mondaq.com.

Almeida, Paul (1996). “Knowledge sourcing by foreign MNEs:patent citation analysis in the US semiconductorindustry”, Strategic Management Journal, 17, WinterSpecial Issue, pp. 155-165.

Altshuler, Rosanne and Harry Grubert (2003). “Taxes,repatriation strategies and multinational financialpolicy”, Journal of Public Economics, 87, 1, pp. 73-107.

Ambos, Björn (2005). “Foreign direct investment in industrialresearch and development: a study of German MNCs”,Research Policy, 34, 4, pp. 395-410.

American Electronics Association (2005). Losing theCompetitive Advantage? The Challenge for Science andTechnology in the United States (Santa Clara, CA:American Electronics Association).

Amsden, Alice and Ted Tschang (2003). “A new approach toassess the technological complexity of differentcategories of R&D (with examples from Singapore)”,Research Policy, 32, 4, pp. 553-572.

Amsden, Alice and Wan Wen Chu (2003). Beyond LateDevelopment: Taiwan’s Upgrading Policies (Cambridge,MA: MIT Press).

Andersson, Thomas (2005). “Linking national science,technology and innovation policies with FDI policies”.Paper presented at UNCTAD Expert Meeting on FDIand Development, Geneva, 24-26 January.

Andersson, Thomas and Glenda Napier (2005). The VentureCapital Market: Global Trends and Issues for NordicCountries (Malmö, Sweden: IKED).

Andersson, Thomas, Sylvia Schwaag Serger, Jens Sörvik andEmily Wise Hansson (2004). The Cluster PoliciesWhitebook (Malmö, Sweden: Holmbergs i Malmö).

Archibugi, Daniele and Alberto Coco (2004). “A new indicatorof technological capabilities for developed anddeveloping countries”, World Development, 32, 4, pp.629-654.

________ (2005). “Measuring technological capabilities atthe country level: a survey and a menu for choice”,Research Policy, 34, 2, pp. 175-194.

Archibugi, Daniele and Jonathan Michie (1995). “Theglobalisation of technology: a new taxonomy”,Cambridge Journal of Economics, 19, 1, pp. 121-140.

Archibugi, Daniele and Simona Iammarino (2002). “Theglobalisation of technological innovation: definitionand evidence”, Review of International PoliticalEconomy, 9, 1, pp. 98-122.

Armbrecht, Ross F.M. (2003). “Siting industrial R&D in China:notes from pioneers”, slide presentation (Arlington, VA:Industrial Research Institute).

Arndt, Sven W. and Henryk Kierzkowski (eds.) (2001).Fragmentation: New Production Patterns in the WorldEconomy (Oxford: Oxford University Press).

Arora, Ashish and Alfonso Gambardella (2004). “Theglobalization of the software industry: perspectives andopportunities for developed and developing countries”,NBER Working Paper, No. 10538 (Cambridge, MA:NBER).

Arora, Ashish, V.S. Arunachalam, Jai Asundi and RonaldFernandes (2000). “The Indian software servicesindustry”, Research Policy, 30, 8, pp. 1267-1287.

Azpiazú, Daniel (2004). “Les privatisations en Argentine:concentration du capital et ‘capture’ des institutionsrégulatrices”, Problèmes d’Amérique Latine, 53, pp.105-131

Balcet, Giovanni and Rinaldo Evangelista (2005). “Globaltechnology: innovation strategies of foreign affiliatesin Italy”. Paper prepared for UNCTAD, mimeo.

Baldwin, Carliss W. and Kim B. Clark (2000). Design Rules:The Power of Modularity (Cambridge, MA: MIT Press).

Barry, Frank (forthcoming). “FDI, transfer pricing and themeasurement of R&D-intensity”, Research Policy.

Bartlet, Christopher A. and Sumatra Ghoshal (1991). ManagingAcross Borders: The Transnational Solution (Boston,MA: Harvard Business School Press).

Baumol, William J. (2004). “Education for innovation:entrepreneurial breakthroughs vs. corporate incrementalimprovements”, NBER Working Paper, No. 10578(Cambridge, MA: NBER).

Behrman, Jack N. and William A. Fischer (1980). OverseasR&D Activities of Transnational Companies (Cambridge,MA: Oelgeschlager, Gunn and Hain).

Belderbos, Rene (2003). “Entry mode, organisational learningand R&D in foreign affiliates: evidence from Japanesefirms”, Strategic Management Journal, 24, 3, pp. 235-259.

Belitz, Heike (2004). “Forschung und Entwicklung inmultinationalen Unternehmen”, Studien zum deutschenInnovationssystem, 8-2004 (Berlin: German Instituteof Economic Research).

Bell, Martin and Keith Pavitt (1993). “Technologicalaccumulation and industrial growth: contrasts betweendeveloped and developing countries”, Industrial andCorporate Change, 2, 2, pp. 157-210.

Bevan, Alan A. and Julian Fennema (2003). “Finance,restructuring and performance in privatised Russianenterprises”, Comparative Economic Studies, 45, 2, pp.117-147.

240 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

BIAC (2005). “Internationalisation of research anddevelopment”, Discussion Paper, March (Paris: Businessand Industry Advisory Committee to the OECD).

BIS (2004). “Foreign direct investment in the financial sectorof emerging market economies”, March (Basel: Bankof International Settlements Committee on the GlobalFinancial System).

Blanke, Jennifer and Augusto Lopez-Carlos (2005).“Measuring competitiveness in the Arab World”, inWorld Economic Forum, Arab World CompetitivenessReport 2005 (Geneva: World Economic Forum), pp. 155-170

Blomström, Magnus, Ari Kokko and Mario Zejan (2000).Foreign Direct Investment: Firm and Host CountryStrategies (London: Macmillan Press, and New York:St. Martin’s Press).

Blomström, Magnus, Robert E. Lipsey, and Ksenia Kulchycky(1988). “US and Swedish direct investment andexports”, in Robert Baldwin, ed., Trade Policy Issuesand Empirical Analysis (Chicago: University of ChicagoPress), pp. 259-297.

Bloom, Nicholas, Rachel Griffith and Alexander Klemm(2001). “Issues in the design and implementation of anR&D tax credit for UK firms”, Institute of Fiscal StudiesBriefing Note, No. 15 (London: IFS).

Bloom, Nick, Rachel Griffith and John Van Reenen (2002).“Do R&D tax credits work? Evidence from a panel ofcountries 1979-1997”, Journal of Public Economics,85, 1, pp. 1-31.

Blundell, Richard, Rachel Griffith and John Van Reenen(1999). “Market share, market value and innovation ina panel of British manufacturing firms”, Review ofEconomic Studies, 66, pp. 529-554.

von Boehmer, Alexander (1995). Internationalisierungindustrieller Forschung und Entwicklung – Typen,Bestimmungsgründe und Erfolgsbedingungen(Wiesbaden: Deutscher Universitätsverlag)

Boone, Jan (2001). “Intensity of competition and the incentiveto innovate”, International Journal of IndustrialOrganization, 19, pp. 705-726.

Bouzas, Roberto and Chudnovsky Daniel (2004). “Foreigndirect investment and sustainable development: therecent argentine experience”, Universidad de San AndrésVictoria (Argentina), mimeo.

Braconier, Henrik, Karolina Ekholm and Karen-HeleneMidelfart Knarvik (2000). “Does FDI work as a channelfor R&D spillovers? Evidence based on Swedish data”,Discussion Paper, No. 2469 (London: Centre forEconomic Policy Research).

Branstetter, Lee (2000). “Is FDI a channel for knowledgespillovers? Evidence from Japanese FDI in US”, NBERWorking Paper, No. 8015 (Cambridge, MA: NBER).

Branstetter, Lee, Raymond Fisman and C. Fritz Foley (2004).“Do stronger intellectual property rights increaseinternational technology transfer? Empirical evidencefrom U.S. firm-level panel data”, World Bank PolicyResearch Working Paper, No. 3305 (Washington, D.C.:World Bank).

Brash, Donald T. (1966). American Investment in AustralianIndustry (Cambridge, MA: Harvard University Press).

British Computer Society (2004). Offshoring: A Challengeor Opportunity for British IT Professionals? (York: YorkPublishing Services).

Brockhoff, Klaus (1998). Internationalization of Researchand Development (Heidelberg: Springer).

Brook, Anne-Marie, Robert Price, Douglas Sutherland, NielsWesterlund and Christophe André (2004). “Oil pricedevelopments: drivers, economic consequences andpolicy responses”, OECD, Economic DepartmentWorking Papers, No. 412 (Paris: OECD).

Campbell-Kelly, Martin and William Aspray (1996).Computer: A History of the Information Machine (NewYork: Basic Books).

Cannice, Mark V., Roger Chen and John D. Daniels (2003).“Managing international technology transfer risk: a caseanalysis of U.S. high technology firms in Asia”, Journalof High Technology Management Research, 14, pp. 171-187.

________ (2004). “Managing international technology transferrisk: alternatives and complements to ownershipstructure”, Management International Review, 44, pp.129-152.

Cantwell, John A. and Odile E.M. Janne (1999).“Technological globalization and innovative centres:the role of corporate technological leadership andlocational hierarchy”, Research Policy, 28, 2-3, pp. 119-144.

Cantwell, John and Massimo G. Colombo (2000).“Technological and output complementarities: inter-firm cooperation in information technology ventures”,Journal of Management and Governance, 4, 1-2, pp.117-147.

Cantwell, John and Ram Mudambi (2000). “The location ofMNE R&D activity: the role of investment incentives”,Management International Review, 40, 1, pp. 127- 148.

Carlin, Wendy, Steven Fries, Mark Schaffer and Paul Seabright(2001). “Competition and enterprise performance intransition economies: evidence from a cross-countrysurvey”, Working Paper, #63 (London: European Bankfor Reconstruction and Development).

Carrincazeaux, Christope, Yannick Lung and Alain Rallet(2001). “Proximity and localisation of corporate R&Dactivities”, Research Policy, 30, 5, pp. 777-789.

Cassiman, Bruno, Massimo Colombo, Paola Garrone andReinhilde Veugelers (2004). “The impact of M&A onthe R&D process: an empirical analysis of the role oftechnological and market relatedness”, mimeo.

Chang, Ha-Joon and Robert Rowthorn (1995). “Role of thestate in economic change: entrepreneurship and conflictmanagement”, in Ha-Joon Chang and Robert Rowthorn,eds., Role of the State in Economic Change (Oxford:Oxford University Press), pp. 31-50.

Chantasasawat, Busakorn, K.C. Fung, Hitomi Iizaka and AlanSiu (2003). “The giant sucking sound: is China divertingforeign direct investments from other Asianeconomies?”. Paper presented at the 6th Asian EconomicPanel Meeting, Seoul, 9-10 October.

Chaparro, Eduardo (2005). “Condiciones y características deoperación de la industria minera en América Latinadurante el bienio 2004-2005”. Informe preparado parala VII edición de la Conferencia de Ministros de Mineríade las Américas (CAMMA), Buenos Aires, May.

Chen, Yun-Chung (2004). “Restructuring the Shanghaiinnovation system: the role of multinationalcorporations’ R&D centres in Shanghai”. Paperpresented at the ASIANLICS International conferenceon “Innovation Systems and Clusters in Asia –Challenges and Regional Integration”, Bangkok, 1-2April.

241REFERENCES

Cheong, Y-R. (2000). “The impact of China’s entrance to theWTO on neighbouring East Asian economies”, ChinaEconomic Review, 11, pp. 419-422.

Chesbrough, Henry W. (2003). Open Innovation: The NewImperative for Creating and Profiting from Technology(Boston, MA: Harvard Business School Press).

Chiesa, Vittorio (1996). “Managing the internationalizationof R&D activities”, IEEE Transactions on EngineeringManagement, 43, 1, pp. 7-23.

China, Ministry of Education (2004). Statistical Report onthe National Education Development 2003 (Beijing:China, Ministry of Education) (www.moe.edu.cn/edoas/website18/info5515.HTM).

Cho, Hwang Hee (2002). Review of S&T Human ResourcePolicies in Korea (Seoul: Science & Technology PolicyInstitute)

Cimoli, Mario (2001). “Networks, market structures andeconomic shocks: the structural changes of innovationsystems in Latin America”. Paper presented at theseminar on “Redes peroductivas e institucionales enAmerica Latina”, Buenos Aires, 9-12 April.

Cimoli, Mario and Annalisa Primi (forthcoming). “El diseñoy la implementación de las políticas tecnológicas enAmérica Latina: un (lento) proceso de aprendizaje”,in Dante Avaro, eds., América Latina y la Sociedad delConocimiento (Plaza y Valdes, Mexico: Cátedra UEALCSociedad del Conocimiento de FLACSO).

Cimoli, Mario and Jorges Katz (2001). “Structural reforms,technological gaps and economic development: a LatinAmerican perspective”. Paper presented at the DRUID-Nelson and Winter conference, Aalborg, 12-15 of June.

Cimoli, Mario, João Carlos Ferraz and Annalisa Primi (2004).“Science and technology policies in open economies:the case of Latin America and the Caribbean”. Paperpresented at the first meeting of ministers and highauthorities on science and technology, Lima, Peru, 11-12 November.

Clarke, George R.G. (2005). “Do Government policies thatpromote competition encourage or discourage newproduct and process development in low and middle-income countries?”, World Bank Policy Research Paper,No. 3471 (Washington, DC: World Bank).

Cohen, Wesley M. and, Daniel A. Levinthal (1989).“Innovation and learning: the two faces of R&D”,Economic Journal, 99, 4, pp. 569-96.

Cohen, Wesley M., Richard R. Nelson and John P. Walsh(2000). “Protecting their intellectual assets:appropriability conditions and why U.S. manufacturingfirms patent (or not)”, NBER Working Paper, No. 7552(Cambridge, MA: NBER)

Consoni, Flavia (2004). “Da tropicalização ao projeto deveículos: um estudo das competências emdesenvolvimento de produtos nas montadoras deautomóveis no Brasil”, Ph.D. thesis (Campinas: IG-Unicamp), mimeo.

Consoni, Flavia and Ruy Quadros (2003). “Betweencentralisation and decentralisation of productdevelopment competencies: recent trajectory changesin Brazilian subsidiaries of car assemblers”. Paperpresented at the XI International Colloquium ofGERPISA on “Company Actors on the Look out for NewCompromises: Developing GERPISA’s New AnalyticalSchema”, Paris, 11-13 June.

D’Costa, Anthony (2003). “Uneven and combineddevelopment: understanding India’s software exports”,World Development, 31, 1, pp. 211-226.

Costa, Ionara (2005). “Notes on R&D and TNCs affiliatesin Brazil”. Paper presented at the UNCTAD ExpertMeeting on the Impact of FDI on Development, Geneva,24-26 January.

Dahlman, Carl. J., Bruce Ross-Larson and Larry E. Westphal(1987). “Managing technological development: lessonsfrom newly industrializing countries”, WorldDevelopment, 15, 6, pp. 759-775.

Dalton, Donald H. and Manuel G. Serapio (1995). GlobalizingIndustrial R&D (Washington, D.C.: Office ofTechnology Policy).

Dashevsky, Steven and Dmitry Loukashov (2004). “Caspianoils: who wants to be a millionaire?” (Moscow: AtonCapital Equity Research), mimeo.

Dernis, Hélène, Dominique Guellec and Bruno vanPottelsberghe de la Potterie (2001). “Using patent countsfor cross-country comparisons of technology output”,STI Review, 27, pp. 129-146.

Desai, Mihir A., C. Fritz Foley and James R. Hines, Jr. (2003).“A multinational perspective on capital structure choiceand internal capital markets”, Harvard NOM ResearchPaper, No. 03-27 (Cambridge, MA: Harvard BusinessSchool).

Deutsche Bundesbank (2005). “The German balance ofpayments in 2004”, Monthly Report, May.

DIHK (2005a). Investitionen im Ausland: Ergebnisse einerDIHK-Umfrage bei den Industrie- und Handelskammern,Frühjahr 2005 (Berlin: Deutsche Industrie- undHandelskammertag).

DIHK (2005b). R&D Offshoring: Examination of Germany’sAttractiveness as a Place to Do Research (Berlin:Deutsche Industrie- und Handelskammertag).

Dumbaugh, Kerry and Mark Sullivan (2005). “China’s growinginterest in Latin America”. Congressional ResearchService (CRS) report for Congress (Washington, D.C.:United States Department of State).

Dunning, John H. (1958). American Investment in BritishManufacturing Industry (London: Allen and Unwin).

________ (1989). Transnational Corporations and the Growthof Services: Some Conceptual and Theoretical Issues(New York: United Nations)

________ (1997). Alliance Capitalism and Global Business(London: Routledge).

________ and Rajneesh Narula (1995). “The R&D activitiesof foreign firms in the United States”, InternationalStudies of Management & Organization, 25, 1-2, pp.39-73.

________ (2005). Multinationals and Industrial Competitiveness:A New Agenda (Cheltenham and Northampton: EdwardElgar).

ECB (2004). “Economic integration in selected regions outsidethe European Union”, ECB Monthly Bulletin, October(Frankfurt: European Central Bank).

________ (2005). ECB Monthly Bulletin, June (Frankfurt:European Central Bank).

ECLAC (2002). Globalization and Development, LC/G.2157(SES.29/3)/I (Santiago, Chile: United Nations).

________ (2003). Foreign Investment in Latin America andthe Caribbean 2002 (Santiago, Chile: United Nations).

________ (2004a). Balance Preliminar de las Economías deAmérica Latina y el Caribe 2004 (Santiago, Chile:United Nations).

242 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

________ (2004b). Foreign Investment in Latin America andthe Caribbean 2003 (Santiago, Chile: United Nations).

________ (2004c). Productive development in open economies,LC/G.2234/SES.30/3 (Santiago, Chile: ECLAC).

________ (2005). Foreign Investment in Latin America andthe Caribbean 2004 (Santiago, Chile: United Nations).

EDA Consortium (2004). Market Statistics Service Survey(San José: Electronic Design Automation Consortium),August.

Edler, Jakob, Frieder Meyer-Krahmer and Guido Reger (2002).“Changes in the strategic management of technology:results of a global benchmark survey”, R&DManagement, 32, 2, pp. 149-164.

Edquist, Charles (1997). System of Innovation: Technologies,Institutions and Organisations (London: Pinter).

Edquist, Charles and Maureen McKelvey (eds.) (2001).Systems of Innovation: Growth, Competitiveness andEmployment (Cheltenham: Edward Elgar).

Eedes, James (2005). “Afghanistan sets out its stall”, FDIMagazine, 7 June (www.fdimagazine.com).

Egger, Peter (2001). “European exports and outward foreigndirect investment: a dynamic panel data approach”,Weltwirtschaftliches Archiv, 137, pp. 427-449.

EIU (2001). Country Commerce: Austria (New York: EIU).

________ (2004a). Scattering the Seeds of Innovation: theGlobalization of Research and Development (London:EIU).

________ (2004b). Country Commerce: Australia (New York:EIU).

________ (2004c). Country Commerce: Belgium (New York:EIU).

________ (2004d). Country Commerce: Canada (New York:EIU).

________ (2004e). Country Commerce: France (New York:EIU).

________ (2004f). Country Commerce: Germany (New York:EIU).

________ (2004g). Country Commerce: Greece (New York:EIU).

________ (2004h). Country Commerce: Italy (New York: EIU).

________ (2004i). Country Commerce: Japan (New York:EIU).

________ (2004j). Country Commerce: Spain (New York:EIU).

________ (2004k). Country Commerce: Switzerland (NewYork: EIU).

________ (2004l). Country Commerce: United Kingdom (NewYork: EIU).

________ (2004m). Country Commerce: United States (NewYork: EIU).

________ (2004n). Country Commerce: India (New York:EIU).

________ (2004o). Country Commerce: Malaysia (New York:EIU).

________ (2004p). Country Commerce: Thailand (New York:EIU).

________ (2004q). Country Commerce: Brazil (New York:EIU).

________ (2004r). Country Commerce: Colombia (New York:EIU).

________ (2004s). Country Commerce: South Africa (NewYork: EIU).

________ (2004t). Country Commerce: Mexico (New York:EIU).

________ (2005a). CEO Briefing: Corporate Priorities for2005 (London: EIU).

________ (2005b). One Year on: the Impact of EUEnlargement (London: EIU).

Engardio, Pete and Bruce Einhorn (2005). “Outsourcinginnovation”, Business Week, 21 March, pp. 51-57.

Ernst & Young (2005). European Attractiveness Survey:Emerging Economies Stake Their Claim (www.ey.com/GLOBAL/content.nsf/International/RECH_-_Press_Release_-_European_Attractiveness_Survey_2005).

Ernst, Dieter (1994). What are the Limits to the Korean Model?The Korean Electronics Industry Under Pressure(Berkeley, CA: University of California at Berkeley).

________ (1999). “Globalization, convergence, and diversity:the Asian production networks of Japanese electronicsfirms”, in Michael Borrus, Dieter Ernst, and StephenHaggard, eds., Rivalry or Riches: InternationalProduction Networks in Asia (Cornell, NY: CornellUniversity Press).

________ (2000). “Inter-organizational knowledgeoutsourcing: what permits small Taiwanese firms tocompete in the computer industry?”, Asia PacificJournal of Management, 17, 2, pp. 223-255.

________ (2002). “Global production networks and thechanging geography of innovation systems: implicationsfor developing countries”, Economics of Innovation andNew Technologies, 11, 6, pp. 497-523.

________ (2003). “Internationalisation of innovation: whyis chip design moving to Asia?”, Working Paper, 67(Honolulu, HI: East-West Center).

________ (2004). “Global production networks in East Asia’selectronics industry and upgrading perspectives inMalaysia”, in Shahid Yusuf, M. Anjum Altaf, KaoruNabeshima, eds., Global Production Networking andTechnological Change in East Asia (Washington, D.C.:World Bank and Oxford University Press).

________ (2005a). “Complexity and internationalisation ofinnovation: why is chip design moving to Asia?”,International Journal of Innovation Management, 9,1, pp. 47-73.

________ (2005b). “Has high-tech, high value-addedoutsourcing worked?”. Paper presented at the Councilon Foreign Relations conference on “The EvolvingGlobal Talent Pool: Issues, Challenges and StrategicImplications”, New York, 16-17 June.

Ernst, Dieter and Barry Naughton (2005). “Buildingcapabilities within global networks: China’s upgradingand innovation in the IT sector” (Honolulu, HI: East-West Center), mimeo.

Ernst, Dieter and David Mowery (2004). “University-industrylinkages in the Pacific Rim: public policy issues”(Honolulu, HI: East-West Center), mimeo.

Ernst, Dieter and David O’Connor (1992). Competing in theElectronics Industry. The Experience of NewlyIndustrialising Economies, Development Centre Studies(Paris: OECD).

243REFERENCES

Ernst, Dieter, Tom Ganiatsos and Lynn Mytelka (eds.) (1998).Technological Capabilities and Export Performance:Lessons from East Asia (Cambridge: CambridgeUniversity Press).

European Commission (2004). Facing the Challenge: TheLisbon Strategy for Growth and Employment. Reportfrom the High Level Group chaired by Wim Kok,November (Brussels: European Commission).

Ferguson, Charles H. and Charles R. Morris (1993). ComputerWars: How the West can win in a post-IBM World (NewYork: Times Books).

de Ferranti, David, Guillermo Perry, Indermit Gill, WilliamMaloney, Jose Luís Guasch, Carolina Sanchez-Paramoand Norbert Schady (2003). Closing the Gap inEducation and Technology (Washington, D.C: WorldBank).

Figueiredo, Paulo. N. (2001). Technological Learning andCompetitive Performance (Cheltenham: Edward Elgar).

Fink, Carsten and Patrick Reichenmiller (2005). “TighteningTRIPS: the intellectual property provisions of recentUS free trade agreements”. Trade note (Washington,D.C.: World Bank), mimeo.

Fisch, Jan Hendrik (2003). “Optimal dispersion of R&Dactivities in multinational corporations with a geneticalgorithm”, Research Policy, 32, 8, pp. 1381-1396.

Flamm, Kenneth (1988). Creating the Computer. Government,Industry and High Technology (Washington, D.C: TheBrookings Institution).

Fors, Gunnar (1997). “Utilization of R&D results in the homeand foreign plants of MNEs”, Journal of IndustrialEconomics, 45, pp. 341-358.

Freeman, Christopher (1987). Technology Policy and EconomicPerformance: Lessons from Japan (London: Pinter).

________ (1995). “The national system of innovation inhistorical perspective”, Cambridge Journal ofEconomics, 19, pp. 5-24.

Freeman, John and Stephen Barley (1989). “The strategicanalysis of interorganizational relations inbiotechnology”, in Ray Loveridge and Mart Pitt, eds.,Strategic Management of Technological Innovation(New York: John Wiley and Sons), pp. 127-155.

Froot, Kenneth A. and Jeremy C. Stein (1991). “Exchangerates and foreign direct investment: an imperfect capitalmarkets approach”, Quarterly Journal of Economics,106, pp. 1191-1217.

Frost and Sullivan (2004). Outsourcing Technology in Asia:Analysis of the Changing Face of R&D (San Antonio,TX: Frost and Sullivan Company).

Furman, Jeffrey L., Michael E. Porter and Scott Stern (2002).“The determinants of national innovation capacity”,Research Policy, 31, 6, pp. 899-933.

Furtado João, Mariana Zanatta, Ionara Costa, EduardoStrachman and Sérgio Queiroz (2003). “FDI andtechnology policies in Brazil”, European Associationof Evolutionary Political Economy (EAEPE)Conference, Maastricht, 7-10 November.

Galina, Simone V.R. (2003). “Desenvolvimento global deprodutos: o papel das subsidiárias brasileiras defornecedores de equipamentos do setor detelecomunicações”. PhD Thesis (São Paulo: Universityof São Paulo), mimeo.

Gassman, Oliver and Maximilian von Zedtwitz (1999). “Newconcepts and trends in international R&D Organization”,Research Policy, 28, 3, pp. 231-250.

Gassmann, Oliver and Zheng Han (2004). “Motivations andbarriers of foreign R&D activities in China”, R&DManagement, 34, 4, pp. 423-438.

Gerber, Jim and Jorge Carrillo (2002). “Are Baja California’smaquiladora plants competitive?”. Paper prepared forthe San Diego Dialogue Border Forum on “The Futureof Manufacturing in Baja California: The Electronicsand Auto Parts Sectors”, San Diego, 18 June.

Gereffi, Gary (1999). “The organization of buyer-driven globalcommodity chains: how U.S. retailers shape overseasproduction networks”, in Gary Gereffi and MiguelKorzeniewicz, eds., Commodity Chains and GlobalCapitalism (London: Praeger), pp. 95-122.

Geroski, Paul A. (1994). Market Structure, CorporatePerformance and Innovative Activity (Oxford and NewYork: Oxford University Press and Clarendon Press).

Gertler, Meric S. (2003). “Tacit knowledge and the economicgeography of context”, Journal of Economic Geography,3, 1, pp. 75-99.

Gerybadze, Alexander and Guido Reger (1999). “Globalizationof R&D: recent changes in the management ofinnovation in transnational corporations”, ResearchPolicy, 28, 2-3, pp. 251-274.

Gilbert, Richard J. and Willard K. Tom (2001). “Is innovationking at the antitrust agencies? The intellectual propertyguidelines five years later”, Antitrust Law Journal, 69,pp. 43-86.

Goldman Sachs (2005). Global Healthcare: IndianPharmaceuticals, Goldman Sachs Global InvestmentResearch, 7 April.

Goldstein, Harry and Ronil Hira (2004). “Spectrum R&D 100:The world’s biggest R&D spenders”, IEEE Spectrum,November (www.spectrum.ieee.org/WEBONLY/publicfeature/nov04).

Gorter, Joeri and Ashok Parikh (2003). “How sensitive is FDIto differences in corporate income taxation within theEU?”, De Economist, 151, 2, pp. 193-204.

Granstrand, Ove (1999). “Internationalization of corporateR&D: a study of Japanese and Swedish corporations”,Research Policy, 28, 2-3, pp. 275-302.

Granstrand, Ove, Lars Håkanson and Sören Sjölander (1993).“Internationalization of R&D – a survey of some recentresearch”, Research Policy, 22, pp. 413-430.

Green, Francis, David Ashton, Donna James and Johny Sung(1999). “The role of the State in skill formation:evidence from the Republic of Korea, Singapore, andTaiwan”, Oxford Review of Economic Policy, 15, 1, pp.82-96.

Gregory, Adrian and Anne-Maré Botha (2003). Tax Incentives– A Way to Stimulate R&D and Innovation (London:PricewaterhouseCoopers) (www.pwcglobal.com/gx/eng/ins-sol/publ/ipvalue/pwc_12.pdf).

Griffith, Rachel, Stephen Redding and Helen Simpson (2004).“Foreign ownership and productivity: new evidencefrom the service sector and the R&D lab”, CEPRDiscussion Paper Series, No. 4691 (London: Centrefor Economic Policy Research).

Guellec, Dominique and Bruno van Pottelsberghe de la Potterie(2001). “The internationalisation of technology analysedwith patent data”, Research Policy, 30, 8, pp. 1253-1266.

________ (2004a). “From R&D to productivity growth: Dothe sources of funds and institutional settings matter?”,Oxford Bulletin of Economics and Statistics, 66, 3, pp.353-376.

244 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

________ (2004b). “Measuring the internationalisation of thegeneration of knowledge: an approach based on patentdata”, in Henk F. Moed, Wolfgang Glänzel and UlrichSchmoch, eds., Handbook of Quantitative Science andTechnology Research: the Use of Publication and PatentStatistics in Studies of Science and Technology Systems(Dordrecht: Kluwer Academic Publishers), pp. 645-662.

Hagedoorn, John (2002). “Inter-firm R&D partnerships: anoverview of patterns and trends since 1960”, ResearchPolicy, 31, 4, pp. 477–492.

Håkanson, Lars (1992). “Locational determinants of foreignR&D in Swedish multinationals”, in Ove Granstrand, LarsHåkanson and Sören Sjölander, eds., TechnologyManagement and International Business (Chichester:Wiley & Sons), pp. 97-116.

Håkanson, Lars and Robert Nobel (1993a). “Foreign researchand development in Swedish multinationals”, ResearchPolicy, 22, 5-6, pp. 373-396.

________ (1993b). “Determinants of foreign R&D in Swedishmultinationals”, Research Policy 22, 5-6, pp. 397-411.

Håkanson, Lars and Udo Zander (1986). Managing InternationalResearch & Development (Stockholm: Mekanförbundet).

Hatzichronoglou, Thomas (1997). “Revision of the hightechnology sector and product classification”, STI WorkingPapers, 1997/2 (Paris: OECD).

Hobday, Michael (1995). Innovation in East Asia: The Challengeto Japan (Aldershot: Edward Elgar).

________ (2001). “The electronics industries of Pacific Asia:exploiting international production networks for economicdevelopment”, Asia Pacific Economic Literature, 15, 1,pp. 13-29.

Hoff, Karl and Joseph. E. Stiglitz (2001). “Modern economictheory and development”, in Gerald M. Meier and J. E.Stiglitz, eds., Frontiers of Development Economics(Oxford: Oxford University Press), pp. 389-459.

Howells, Jeremy (2000). “Innovation and services: newconceptual frameworks”, CRIC Discussion Paper, No.38 (Manchester: CRIC) (les1.man.ac.uk/cric/Pdfs/DP38.pdf).

Howells, John (1997). “Research and technology outsourcing”,CRIC Discussion Paper, No. 6 (Manchester: Centre forResearch on Innovation and Competition) (les1.man.ac.uk/cric/Pdfs/DP6.pdf).

Hu, Albert Guangzhou (2004). “Multinational corporations,patenting and knowledge flow: the case of Singapore”,Economic Development and Cultural Change, 52, 4, pp.781-800.

Hu, Mei-Chih and John A. Mathews (2003). “Nationalinnovative capacity in East Asia”, MGSM Working Paper,2003-17 (Sydney: Macquarie Graduate School ofManagement).

Hubert, Schmitz (1995). “Small shoemakers and Fordist giants:tales of a supercluster”, World Development, 23, 1, pp.9-28.

Hunya, Gábor (2004). “Manufacturing FDI in new EU MemberStates: foreign penetration and location shifts between1998 and 2002”, WIIW Research Reports , No. 311(Vienna: Vienna Institute for International EconomicStudies).

________ (2005). WIIW Database on Foreign Direct Investmentin Central, East and Southeast Europe 2005:Opportunities for Acquisition and Outsourcing (Vienna:Vienna Institute for International Economic Studies).

IADB (2001). Competitiveness: the Business of Growth:Economic and Social Progress in Latin America(Washington, D.C: Inter-American Development Bank).

IADB (2004). “The emergence of China: opportunities andchallenges for Latin America and the Caribbean”.Preliminary draft for discussion at the Conference on“Emergence of China: Opportunities and Challenges forLatin America and the Caribbean”, Washington, D.C.,1 October.

IBFD (2004). Tax Treatment of Research and DevelopmentExpenses (Brussels: European Commission)(europa.eu.int/comm/taxation_customs/resources/documents/eu_rd_final_rep_dec_2004.pdf).

ICSID (2005). “Laudo en el procedimiento entre CMS GasTransmisión Company (demandante) y la RepúblicaArgentina (demandado)”, Caso No. ARB/01/8.

IIF (2004). “Corporate governance in Russia – an investorperspective”. Report of the task force established by theIIF’s Equity Advisory Group (Washington, D.C.: IIF)(www.iif.com/data/public/ Russia_corpgov_1004.pdf).

________ (2005). “Capital flows to emerging marketeconomies”, 19 January (Washington, D.C.: Institute ofInternational Finance).

IISD (2005). “First domino falls on Argentina as tribunal rulesin financial crisis arbitration”, Investment Law and PolicyNews Bulletin, 27 May.

IMD (various years). World Competitiveness Yearbook(Lausanne: IMD).

IMF (2004). Islamic Republic of Iran: 2004 Article IVConsultation—Staff Report; Staff Supplement; StaffStatement, and Public Information Notice on the ExecutiveBoard Discussion, IMF Country Report, No. 04/306(Washington, D.C.: International Monetary Fund).

________ (2005). World Economic Outlook April 2005:Globalization and External Imbalances (Washington,D.C.: International Monetary Fund).

IMF/OECD (2004). Surveys of Implementation ofMethodological Standards for Direct Investment(Washington, D.C. and Paris: IMF and OECD).

Intarkumnerd, Patarapong and Peerapol Sittivijan (2005).“Foreign direct investment in R&D in Thailand: evolutionand challenges”. Paper presented at ASEAN-UNCTADSeminar on “Key Issues of FDI: Attracting Quality FDI”,Bangkok, 5-6 May.

Ireland, Department of Enterprise, Trade and Employment(2004). Building Ireland’s Knowledge Economy: The IrishAction Plan for Promoting Investment in R&D to 2010(Dublin: Irish Government Publications).

ISA (2004). Klimatet för utländska investeringar i Sverige –Rapport till regeringen 2004 (Stockholm: Invest inSweden Agency).

Islamic Republic of Iran, Chamber of Commerce, Industriesand Mines (undated). Guide to Iranian Market(www.iccim.org/english/Law/guide_to_iranian_market/buyback_transactions.htm).

Islamic Republic of Iran, OIETAI (2004). Guide to Invest inIran (Tehran: Ministry of Economic Affairs & FinanceOrganization for Investment, Economic and TechnicalAssistance of Iran).

ISPE (2003). “Singapore – the premier biomedical hub of Asia”,Pharmaceutical Engineering, July/August 2003, pp. 1-13.

iSuppli (2003). China’s Fabless Firms Race Beyond FoundationStage (El Segundo, CA: iSuppli).

245REFERENCES

ITPS (2003). Forskning och Utveckling i InternationellaFöretag, S2003:007 (Stockholm: Elanders Gotab).

________ (2005). Forskning och Utveckling i InternationellaFöretag, S2005:005 (Stockholm: Elanders Gotab).

ITRS (2002). International Technology Roadmap forSemiconductors 2001 Edition (Austin, TX:Semiconductor Industry Association).

________ (2004). International Technology Roadmap forSemiconductors 2003 Edition (Austin, TX:Semiconductor Industry Association).

Ivanova, Natalya (2004). “Cooperate to survive: strategicalliances in the Russian aerospace industry”, in SlavoRadosevic and Bert M. Sadowski, eds., InternationalIndustrial Networks and Industrial Restructuring inCentral and Eastern Europe (Boston, Dordrecht andLondon: Kluwer Academic Publishers), pp. 131-154.

Iwasa, Tomoko and Hiroyuki Odagiri (2004). “Overseas R&D,knowledge sourcing, and patenting: an empirical studyof Japanese R&D investment in the US”, ResearchPolicy, 33, 5, pp. 807–828.

Jaffe, Adam B. (1998). “Measurement issues”, in Lewis M.Branscomb and James Keller, eds., Investing inInnovation (Cambridge, MA: MIT Press), pp. 64-84.

Jankowski, John E. (2001). “Measurement and growth of R&Dwithin the service economy”, Journal of TechnologyTransfer, 26, 4, pp. 323–336.

Japan, METI (2004). Dai 32-kai Wagakuni Kigyo no KaigaiJigyo Katsudo (Tokyo: National Printing Bureau).

JBIC (2005). “Survey report on overseas business operationsby Japanese manufacturing companies – results of JBICFY2004 survey: outlook for Japanese foreign directinvestment (16th annual survey)”, Journal of JBICInstitute, 22 (February), pp. 6-106.

________ (various years). Survey Report on Overseas BusinessOperations by Japanese Manufacturing Companies(Tokyo: Japan Bank For International Cooperation ).

Jones, Charles I. and John C. Williams (1998). “Measuringthe social return to R&D”, Quarterly Journal ofEconomics, 113, 4, pp. 1119-1135.

Jones, Gary K. and Hildy J. Teegen (2001). “Global R&Dactivity of U.S. MNCs: does national culture affectinvestment decisions?”, Multinational Business Review,9, 2, pp. 1-7.

________ (2003). “Factors affecting foreign R&D locationdecisions: management and host policy implications”,International Journal of Technology Management, 25,8, pp. 791-813.

Kalotay, Kálmán (2000). “Book review on Mihály Simai, PeterFarkas, Andrea Éltetö and Peter Gál, A müködötökekivitele és a technikai fejlQ´dés a 21. század küszöbén(Budapest: Ministry of Education, 2000)”, TransnationalCorporations, 9, 2, pp. 162-167.

Kalotay, Kálmán and Gábor Hunya (2000). “Privatization andforeign direct investment in Central and EasternEurope”, Transnational Corporations, 9, 1, pp. 39-66.

Kaplinsky, Raphael and Jeff Readman (2000). “Globalisationand upgrading: what can (and cannot) be learnt frominternational trade statistics in the wood furnituresector?” (Brighton: Centre for Research in InnovationManagement, University of Brighton and Institute ofDevelopment Studies, University of Sussex), mimeo.

Kaplinsky, Raphael and Mike Morris (2001). A Handbook forValue Chain Research (Brighton: Institute ofDevelopment Studies).

Katz, Jorge M. (eds.) (1987). Technology Generation in LatinAmerican Manufacturing Industries (London:Macmillan).

KICOS (2004). The Best Partner for Global R&D Cooperation(Seongnam-Si: Korea Foundation for InternationalCooperation of Science & Technology).

Kim, Linsu (2003). “Technology transfer & intellectualproperty rights”, Issue Paper, No. 2 (Geneva: ICTSDand UNCTAD).

Kim, Seok-Ran and Nick von Tunzelmann (1998). “Aligninginternal and external networks: Taiwan’s specializationin IT” (Sussex: SPRU, University of Sussex), mimeo.

Kogut, Bruce and Sea-Jin Chang (1991). “Technologicalcapabilities and Japanese foreign direct investment inthe US”, Review of Economics and Statistics, 73, pp.401-413.

KPMG (2005). Global Tax Rates Survey 2005 (Amsterdam:KPMG International).

Kuemmerle, Walter (1997). “Foreign direct investment inindustrial research in the pharmaceutical and electronicsindustries – results from a survey of multinationalfirms”, Research Policy, 28, 2-3, pp. 179-193.

________ (1999). “The drivers of foreign direct investmentinto research and development: an empiricalinvestigation”, Journal of International Business Studies,30, 1, pp. 1-24.

Kumar, Nagesh (1996). “Intellectual property protection,market orientation and location of overseas R&Dactivities by multinational enterprises”, WorldDevelopment, 24, 4, pp. 673-688.

________ (2001). “Determinants of location of overseas R&Dactivity of multinational enterprises: the case of US andJapanese corporations”, Research Policy, 30, 1, pp. 159-174.

Kuntz, Lucia I. (1999). “Brazil: a bridgehead to the tropics”,The UNESCO Courier, 52, 5(May), p. 34.

Kutschker, M. (1989). “Akquisition, internationale”, in KlausMacharzina and Martin K. Welge, eds.,Handwörterbuch: Export und internationaleUnternehmung (Stuttgart: Schäffer-Poeschel), pp. 2-22.

Kwon, Yong Soo (2003). Korea’s S&T globalization: PastDevelopment and Issues Ahead (Seoul: Science &Technology Policy Institute).

Lach, Saul (2000). “Do R&D subsidies stimulate or displaceprivate R&D? Evidence from Israel” (ssrn.com/abstract=240228), mimeo.

Lall, Sanjaya (1979). “The international allocation of R&Dactivity by US multinationals”, Oxford Bulletin ofEconomics and Statistics, 41, 4, pp. 313-331.

________ (1992). “Technological capabilities andindustrialization”, World Development, 20, 2, pp. 165-86.

________ (2001a). Competitiveness, Technology and Skills(Cheltenham: Edward Elgar).

________ (2001b). “Competitiveness indices and developingcountries: an economic evaluation of the GlobalCompetitiveness Report”, World Development, 29, 9,pp. 1501-1525.

________ (2003). “Indicators of the relative importance ofIPRs in developing countries”, Research Policy, 32, 9,pp. 1657-1680.

246 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Lall, Sanjaya and Carlo Pietrobelli (2002). Failing to Compete:Technology Development and Technology Systems inAfrica (Cheltenham: Edward Elgar).

Lall, Sanjaya and Manuel Albaladejo (2002). “Indicators ofthe relative importance of IPRs in developing countries”,Working Paper, 85 (Oxford: Oxford University, QueenElizabeth House).

Lall, Sanjaya and Manuel Albaladejo and Jinkang Zhang(2004). “Mapping fragmentation: electronics andautomobiles in East Asia and Latin America”, OxfordDevelopment Studies, 32, 3, pp. 407-432.

Lazonick, William (2005). “Evolution of the New Economybusiness model”, in Eric Brousseau and Nicholas Curien,eds., The Economics of the Internet (Cambridge:Cambridge University Press).

Le Bas, Christian and Christophe Sierra (2002). “Locationversus home country advantages in R&D activities:some further results on multinationals’ locationstrategies”, Research Policy, 31, 4, pp. 589-609.

Lee, Ji-Ren and Jen-Shyang Chen (2000). “Dynamic synergycreation with multiple business activities: toward acompetence-based growth model for contractmanufacturers”, in Ron Sanchez and Aime Heene, eds.,Research in Competence-Based Management, Vol. 6 inAdvances in Applied Business Strategy Series (Stamford,CT: JAI Press), pp. 209-228.

Lerner, Josh (1999). “The government as venture capitalist:the long-run impact of the SBIR program”, The Journalof Business, 72, 3, pp. 285- 318.

Lewanika, Mwananyanda M. (2005). “Zambia’s experienceswith FDI and R&D”. Paper presented at UNCTADExpert Meeting on FDI and Development, Geneva, 24-26 January.

Li, Jiatao and Jing Zhong (2003). “Explaining the growth ofinternational R&D alliances in China”, Managerial andDecision Economics, 24, 2-3, pp. 101-115

Liang, Guoyong (2004). New Competition: Foreign DirectInvestment and Industrial Development in China, ERIMPh.D. Series, Research in Management, 47 (Rotterdam:RSM, Erasmus University.)

Lin, Yeo and Rajah Rasiah (2003). “Structure, technical changegovernment intervention: the development ofinformation hardware industry in Taiwan”, InternationalJournal of Business and Society, 4, 2, pp. 135-171.

Lopez-Claros, Augusto (2004). “Executive summary”, in WorldEconomic Forum, World Competitiveness Report 2004(Geneva: World Economic Forum), pp. xi-xxiii.

Love, James (2003). “Evidence regarding research anddevelopment investments in innovative and non-innovative medicines” (Washington, D.C: ConsumerProject on Technology), mimeo.

Low, Linda, Toh Mun Heng, Soon Teck Wong, Tan Kong Yamand Helen Hughes (1993). Challenge and Response:Thirty Years of the Economic Development Board(Singapore: Times Academic Press).

Lu, Qiwen (2000). China’s Leap into the Information Age:Innovation and Organization in the Computer Industry(Oxford: Oxford University Press).

Lundvall, Bengt-Åke (1992a). Innovation, Growth and SocialCohesion: The Danish Model (Cheltenham: EdwardElgar).

________ (1992b). National Systems of Innovation: TowardsA Theory of Innovation and Interactive Learning(London: Pinter).

MacDougall, Diarmuid (2004). “Why are EU countriesfocusing on R&D credits?”, Journal of InternationalTaxation,15, 7, pp. 38-43, 64.

Malek, Jan (2000). “R&D outsourcing that works”,Pharmaceutical Executive, March, pp. 70-78.

Manea, Julia (2002). Multinational Enterprises’ Strategiesand their Implications for the Processes ofTransformation and Development in Central and EasternEuropean Transition Economies, PhD thesis (Reading:University of Reading).

Manea, Julia and Robert Pearce (2001). “Multinationalstrategies and sustainable industrial transformation inCEE transition economies: the role of technology”, inJames H. Taggart, Maureen Berry and MichaelMcDermott, eds., Multinationals in a New Era(Basingstoke: Palgrave), pp. 118-140.

Mansfield, Edwin (1994). “Intellectual property protection,foreign direct investment, and technology transfer”,International Finance Corporation Discussion Paper,No. 19 (Washington, D.C.: World Bank).

________ (1995). “Intellectual property protection, directinvestment, and technology transfer: Germany, Japan,and the United States”, International FinanceCorporation Discussion Paper, No. 27 (Washington,D.C.: World Bank).

Mariani, Myriam (2002). “Next to production or totechnological clusters? The economics and managementof R&D location”, Journal of Management andGovernance, 6, pp. 131-152.

Marseille Innovation and ANIMA (2005). Innovation, pôlestechnologiques et attraction de l’investissement: LaMéditerranée intelligente (Marseille and Paris: MarseilleInnovation and Réseau Euro-Méditerranéen des Agencesde Promotion des Investissements).

Martin, Grant and Henry Chang (eds.) (2003). Winning theSoC Revolution: Experiences in Real Design (Boston,MA: Kluwer Academic Publishers).

Maskus, Keith (2005). “The role of intellectual property rightsin encouraging foreign direct investment and technologytransfer”, in Carsten Fink and Keith Maskus, eds.,Intellectual Property and Development: Lessons fromRecent Economic Research (New York: World Bank andOxford University Press).

Mathews, John A. and Dong-Sung Cho (2000). TigerTechnology: The Creation of a Semiconductor Industryin East Asia (Cambridge: Cambridge University Press).

McKinsey (2005). Global Survey of Business ExecutivesConfidence Index, www.mckinseyquarterly.com.

McKinsey Global Institute (2005). The Emerging Global LaborMarket: Part II – The Supply of Offshore Talent inServices (San Francisco, CA: McKinsey & Company).

Medcof, John W. (1997). “A taxonomy of internationallydispersed technology units and its application tomanagement issues”, R&D Management, 27, 4, pp. 301-318.

Metcalfe, J. Stan (1995). “Technology systems and technologypolicy in an evolutionary framework”, CambridgeJournal of Economics, 19, 1, 25-46.

De Meyer, Arnold and Atsuo Mizushima (1989). “Global R&Dmanagement”, R&D Management, 19, pp. 135-46.

Meyer-Krahmer, Frieder and Guido Reger (1999). “Newperspectives on the innovation strategies of multinationalenterprises: lessons for technology policy in Europe”,Research Policy, 28, 7, pp. 751-776.

247REFERENCES

Molero, José (1998). “Patterns of internationalization ofSpanish innovatory firms”, Research Policy, 27, 5, pp.541-558.

Moosa, Imad A. (2002). Foreign Direct Investment, Theory,Evidence and Practice (Basingstoke, Hampshire:Houndmills).

Moran, Theodore H. (2002). “The relationship between trade,foreign direct investment, and development: newevidence, strategy, and tactics under the DohaDevelopment Agenda negotiations”. Paper prepared forADB’s Study on “Regional Integration and Trade:Emerging Policy Issues for Selected DevelopingMember Countries”, November.

Morgan, Jane (2002). Research and Experimental DevelopmentStatistics 2000 (Newport: United Kingdom, Office ofNational Statistics) (www.statistics.gov.uk/articles/economic_trends/ET_Aug02_Morgan.pdf)

Moris, Francisco (2005a). “Foreign direct investment, R&D,and innovation: concepts and data”. Background noteprepared for UNCTAD (Arlington, VA: U.S. NationalScience Foundation), mimeo.

________ (2005b). “R&D investments by U.S. TNCs inemerging and developing markets in the 1990s”.Background paper prepared for UNCTAD (Arlington,VA: U.S. National Science Foundation), mimeo.

Mowery, David, Joanne Oxley and Brian Silverman (1998).“Technological overlap and interfirm cooperation:implication for the resource-based view of the firm”,Research Policy, 27, 5, pp. 507-524.

Mukherjee, Shoibal (2005). “Pharmaceutical research anddevelopment in India: looking up?”, Business Briefing:Pharma Outsourcing 2005 (London: BusinessBriefings), pp. 98-103.

Munari, Federico and Maurizio Sobrero (2005). “The effectsof privatisation on R&D investments and productivity:an empirical analysis of European firms”, mimeo.

Mytelka, Lynn and Lou A. Barclay (2004). “Using foreigninvestment strategically for innovation”, EuropeanJournal of Development Research, 16, 3, pp. 527-555.

Narula, Rajneesh (1999). “In-house R&D, outsourcing oralliances? Some strategic and economic considerations”,Working Paper (Oslo: ESST, University of Oslo).

________ (2001). “Choosing between internal and non-internalR&D activities: some technological and economicfactors”, Technology Analysis & Strategic Management,13, pp. 365-387.

________ (2002). “Innovation systems and ‘inertia’ in R&Dlocation: Norwegian firms and the role of systemic lock-in”, Research Policy, 31, pp. 795-816.

________ (2003). Globalization & Technology:Interdependence, Innovation Systems and IndustrialPolicy (Oxford: Polity Press).

Narula, Rajneesh and Antonello Zanfei (2004). “Globalizationof innovation: the role of multinational enterprises”,in Jan Fagerberg, David C. Mowery and Richard R.Nelson, eds., The Oxford Handbook of Innovation(Oxford: Oxford University Press), pp. 318-345.

Naughton, Barry and Adam Segal (2002). “Technologydevelopment in the new millennium: China in searchof a workable model”, in William Keller and RichardSamuels, eds., Crisis and Innovation: Asian Technologyafter the Millennium (New York, NY: CambridgeUniversity Press), pp. 137-160.

Nelson, Richard R. (1990). “On technological capabilitiesand their acquisition”, in Robert E. Evenson and GustavRanis, eds., Science and Technology: Lessons forDevelopment Policy (Boulder, CO: Westview Press),pp. 71-80.

________ (ed.) (1993). National Innovation Systems: AComparative Analysis (Oxford and New York: OxfordUniversity Press).

Nelson, Richard R. and David C. Mowery (eds.) (1999).Sources of Industrial Leadership: Studies of SevenIndustries (Cambridge: Cambridge University Press).

Nelson, Richard R. and Sidney J. Winter (1982). AnEvolutionary Theory of Economic Change (Cambridge,MA: Harvard University Press).

Nickell, Stephen J. (1996). “Competition and corporateperformance”, Journal of Political Economy, 104, pp.724-746.

Nobel, Robert and Julian Birkinshaw (1998). “Innovation inmultinational corporations: control and communicationpatterns in international R&D operations”, StrategicManagement Journal, 19, 5, pp. 479-496.

North, Douglas C. (1990). Institutions, Institutional Changeand Economic Performance (Cambridge: CambridgeUniversity Press).

Nureev, Rustem and Anton Runov (2003). “Back to privateownership or on to private ownership?”, Social Sciences,34, 2, pp. 29-48.

OECD (1997a). Proposed Guidelines For Collecting andInterpreting Technological Innovation Data (OsloManual) (Paris: OECD).

________ (1997b). National Innovation Systems (Paris:OECD).

________ (2000). “Corporate tax incentives for foreign directinvestment”. Background Paper for Special Session IIIon FDI and the restructuring of transition and emergingeconomies, presented at the UN/ECE RegionalConference, Geneva, 6-7 December.

________ (2002a). “Corporate taxation and foreign directinvestment: India’s tax competitiveness”. Paperpresented at the International Conference on “TaxationPolicy for Accelerating Investment: Domestic andForeign”, New Delhi, 13-14 November.

________ (2002b). Proposed Standard Practice for Surveyson Research and Experimental Development (FrascatiManual) (Paris: OECD).

________ (2003). Tax Incentives for Research andDevelopment: Trends and Issues (Paris: OECD).

________ (2004a). OECD Investment Policy Reviews: RussianFederation – Progress and Reform Challenges (Paris:OECD).

________ (2004b). Compendium of Patent Statistics 2004(Paris: OECD).

________ (2004c). Education at a Glance (Paris: OECD).

________ and Belgian Science Policy (2005).“Internationalisation of R&D: trends, issues andimplications for S&T policies: a review of theliterature”. Background report for the Forum on theInternationalisation of R&D, Brussels, 29-30 March.

Olk, Paul and Candace Young (1997). “Why members stayin or leave an R&D consortium? Performance andconditions of membership as determinants ofcontinuity”, Strategic Management Journal, 18, pp. 855-877.

248 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Ordover, Jansuz-Aleksander and William Baumol (1988).“Antitrust policy and high-technology industries”,Oxford Review of Economic Policy, 4, pp. 13-34.

Oyelaran-Oyeyinka, Banji (2004a). “Learning and localknowledge institutions in African Industry”, DiscussionPaper Series, #2004-2 (Maastricht: United NationsUniversity).

________ (2004b). “How can Africa benefit fromglobalization?”, ATPS Special Paper Series, No. 17(Nairobi: African Technology Policy Studies Network).

Panagariya, Arvind (2003). “TRIPS and the WTO: an uneasymarriage”, Economics Working Paper Archive,“International Trade” series, No. 0309002 (St. Louis,MO: WUSTL).

Patel, Pari and Keith Pavitt (1991). “Large firms in theproduction of the world’s technology: an important caseof non-globalization”, Journal of International BusinessStudies, 22, 1, pp. 1-21.

________ (1994). “National innovation systems: why theyare important and how they might be measured andcompared”, Economics of Innovation and NewTechnology, 3, pp. 77-95.

________ (2000). “National systems of innovation understrain: the Internationalisation of Corporate R&D”, inRay Barrell, Geoffrey Mason and Mary O’Mahoney,eds., Productivity, Innovation and EconomicPerformance (Cambridge: Cambridge University Press).

Patel, Pari and Modesto Vega (1999). “Patterns ofinternationalisation of corporate technology: locationvs. home country advantages”, Research Policy, 28, 2-3, pp. 145-155.

Paul, Lauren Gibbons (1998). “Outside inspiration”, CIOEnterprise Magazine, April 15 (www.cio.com/enterprise/041598).

Pavitt, Keith (1999). Technology, Management and Systemsof Innovation (Cheltenham: Edward Elgar).

Pavlínek, Petr (2004). “Regional development implicationsof foreign direct investment in Central Europe”,European Urban and Regional Studies, 11, 1, pp. 47-70.

Pearce, Robert D. (1989). The Internationalisation of Researchand Development by Multinational Enterprises (London:Macmillan).

________ (1999). “Decentralised R&D and strategiccompetitiveness: Globalised approaches to generationand use of technology in multinational enterprises(MNEs)” Research Policy, 28, 2-3, pp.157–178.

________ (2004). “National systems of innovation and theinternational technology strategy of multinationals”,Discussion Paper, 006-04 (Reading: University ofReading).

Pelto, Elina, Peeter Vahtra and Kari Liuhto (2003). “Cyp-Rusinvestment flows to Central and Eastern Europe –Russia’s direct and indirect investments via Cyprus toCEE”, Electronic Publications, 2/2003 (Turku: Pan-European Institute), www.tukkk.fi/pei.

Peltzman, Sam (1976). “Toward a more general theory ofregulation”, Journal of Law and Economics, 19, pp. 211-240.

Peri, Giovanni (2004). “Catching-up to foreign technology?Evidence on the ‘Veblen-Gerschenkron’ effect of foreigninvestments”, NBER Working Paper, No. 10893(Cambridge, MA: NBER).

Pfaffermayr, Michael (1996). “Foreign outward directinvestment and exports in Austrian manufacturing:substitutes or complements?”, WeltwirtschaftlichesArchiv, 132, pp. 501-522.

Porter, Michael E. (1997) “Knowledge-based clusters andnational competitive advantage”. Paper presented atTechnopolis 97, Ottawa, 9-12 September.

Porter, Michael E. and Scott Stern (2001). “Innovation:location matters”, MIT Sloan Management Review, 42,4, pp. 28-36.

Posthuma, Anna (2000). “Trickle down or draining out? Theimpacts of liberalisation and globalisation on in-housetechnological activities in the Brazilian automotiveproduction chain” (Geneva: International LabourOffice), mimeo.

van Pottelsberghe de la Potterie, Bruno, Steve Nysten andEsmeralda Megally (2003). “Evaluation of current fiscalincentives for business R&D in Belgium”, SSTC andCEB Working Paper (Brussels: Solvay Business School,ULB).

Prencipe, Andrea, Andrew Davies and Michael Hobday (2003).The Business of Systems Integration (Oxford: OxfordUniversity Press).

PricewaterhouseCoopers (2005). Eighth Annual Global CEOSurvey (www.pwcglobal.com).

Puffer, Sheila M. and Daniel J. McCarthy (2003). “Theemergence of corporate governance in Russia”, Journalof World Business, 38, 4, pp. 284-298.

Quandt, Carlos (1999). “The concept of virtual technopolesand the feasibility of incubating technology-intensiveclusters in Latin America and the Caribbean”. Paperpresented at Technopolis 97, Ottawa, 9-12 September.

Queiroz, Sérgio, Mariana Zanatta and Carolina Andrade(2003). “Internationalization of MNCs’ technologicalactivities: what role for Brazilian subsidiaries?”. Paperpresented at SPRU Conference in honour of Keith Pavitt,Brighton, 13-15, November.

Quinteros, Carolina (2004). “Puede el CAFTA sacar adelantea la maquila centroamericana?”; Revista “Centroaméricaen la Economía Mundial del Siglo XXI”, ASIES, mimeo.

Radosevic, Slavo (1999). International Technology Transferand Catch-Up in Economic Development (Cheltenham:Edward Elgar).

Ramb, Fred and Alfons J. Weichenrieder (2005). “Taxes andthe financial structure of German inward FDI”, DeutscheBundesbank Discussion Paper, Series 1, No. 05/2005.

Ramirez, Paulina (2003). “The globalization of research amongleading European and US pharmaceutical multinationals:The shift of European investment to the USA”, CISCWorking Paper, No. 09 (Galway: National Universityof Ireland, www.nuigalway.ie/cisc.

Rasiah, Rajah (1996). “Institutions and innovations: movingup the technology ladder in the electronics industry inMalaysia”, Industry and Innovation, 3, 2.

Reddy, Prasada (2000). Globalization of Corporate R&D:Implications for Innovation Systems in Host Countries(London and New York: Routledge).

________ (2005). “Global R&D activities in India”. Paperpresented at UNCTAD Expert Meeting on FDI andDevelopment, Geneva, 24-26 January.

Reinert, Erik (1995). “Competitiveness and its predecessors:a 500 year cross-national perspective”, StructuralChange and Economics Dynamics, 6, pp. 23-42.

249REFERENCES

Republic of Korea, Ministry of Commerce, Industry andEnergy (2005). “The summary of Korean experiencesrelated to government policies on building the nationalR&D system and innovation capability, attracting FDIin R&D” (Seoul: Ministry of Commerce, Industry andEnergy), mimeo.

Republic of Korea, Ministry of Science and Technology(2003a). “MOST: Roles and Activities” (Seoul: Ministryof Science and Technology).

________ (2003b). “Policy to support the advancement offoreign R&D centers into Korea” (Seoul: Ministry ofScience and Technology).

Roberts, Edward B. (2001). “Benchmarking global strategicmanagement of technology”, Research TechnologyManagement, 44, 2, pp. 25-36.

Roelandt, Theo and Pim Den Hertog (eds.) (1999). BoostingInnovation: The Cluster Approach (Paris: OECD).

Ronstadt, Robert C. (1977). Research and Development Abroadby US Multinationals (New York, NY: Praeger).

Rugman, Alan and Joseph D’Cruz (2003). Multinationals asFlagship Firms: Regional Business Networks (Oxford:Oxford University Press.)

Sachs, Jeffrey D. and John W. McArthur (2005). “TheMillennium Project: a plan for meeting the MillenniumDevelopment Goals”, Lancet, 365, pp. 347-353,www.thelancet.com.

Sachwald, Frédérique (2004a). “Globalization of innovationnetworks—the case of French firms”. Paper presentedat the 29th Annual Conference of the EuropeanInternational Business Academy, Copenhagen, 11-13December.

Sachwald, Frédérique (2004b). “Internationalisation de la R&Ddes enterprises et attractivité de la France”, IFRI,Opération Futuris Recherche Innovation Société, mimeo.

Safarian, A. Edward (1966). Foreign Ownership of CanadianIndustry (Toronto: McGraw Hill).

Sakakibara, Mariko and Michael E. Porter (2000). “Competingat home to win abroad”, Review of Economics andStatistics, 53, pp. 310-322.

Salacuse, Jeswald and Nicholas Sullivan (2005). “Do BITsreally work: an evaluation of bilateral investment treatiesand their grand bargain”, Harvard Journal ofInternational Law, 46, pp. 67- 130.

Santangelo, Grazia D. (2000). “Inter-European regionaldispersion of corporate research activity in informationand communications technology: the case of German,Italian and UK regions”, International Journal of theEconomics of Business, 7, 3, pp. 275-295.

Sanyal, Prabuddha (2004). “Intellectual property rightsprotection and location of R&D by multinationalenterprises”, Journal of Intellectual Capital, 5, 1, pp.59-76.

Sanz, Luis (2004). “From science parks to metropolitaninnovation environments”. PowerPoint presentation,Stockholm 25 August.

Schmaul, Bernd (1995). Organisation und Erfolginternationaler Forschungs- und Entwicklungseinheiten(Wiesbaden: Gabler).

Shelley, Toby (2005). Oil: Politics, Poverty and the Planet(London and New York: Zed Books).

Shen, Xiaobai (1999). The Chinese Road to High Technology:a Study of Telecommunications Switching Technologyin the Economic Transition (New York: St. Martin’sPress).

Shlapentokh, Vladimir (2004). “Wealth versus political power:the Russian case”, Communist and Post-CommunistStudies, 37, 2, pp. 135-160.

SIA (1999). International Technology Roadmap forSemiconductors (San Jose, CA: Semiconductor IndustryAssociation ).

Sigurdson, Jon (2004).”People’s Republic of China’stechnological capability”. Paper presented at the 2004Laeba Annual Conference, Beijing, 3-4 December.

________ (2005a). “Foreign R&D investment in China”. Paperpresented at the Workshop on Global R&D in China,organized by the Nanjing Municipal Development andReform Commission, Nanjing, 27-29 May.

________ (2005b). Technological Superpower China(Cheltenham: Edward Elgar).

Simon, Denis Fred (2005). Presentation for the hearing onChina’s high-technology development by the US-ChinaEconomic & Security Review Commission, 21-22 April(Stanford: Stanford University), mimeo.

Singh, Ajit (2003). “Multilateral competition policy andeconomic development: a developing countryperspective on the European Community proposals”,UNCTAD Series on Issues in Competition Law andPolicy (New York and Geneva: United Nations).

Sinn, Hans Werner and Wolfgarg Ochel (2003). “Social union,convergence and migration”, Journal of Common MarketStudies, 41, pp. 869-896.

Sorg, Armin (2005). “Globalization of R&D and TNCstrategies: Siemens”. Paper presented at the UNCTADExpert Meeting on the Impact of FDI on Development,Geneva, 24-26 January.

Srholec, Martin (2005a). “Research and development activityof foreign affiliates: firm-level evidence from the CzechRepublic” (Oslo: University of Oslo), mimeo.

________ (2005b). “Innovation activities of foreign affiliatesin the Czech Republic”. Paper presented at the UNCTADExpert Meeting on the Impact of FDI on Development,Geneva, 24-26 January.

Stigler, George J. (1971). “The theory of economic regulation”,Bell Journal of Economics and Management Science,2, pp. 3-21.

Sutton, John (1998). Technology and Market Structure(Cambridge, MA: MIT press)

Teece, David (1992). “Competition, cooperation, andinnovation: organizational arrangements for regimesof rapid technological progress”, Journal of EconomicBehavior & Organization, 18, 1, pp. 1-25.

________ (2000). Managing Intellectual Capital (Oxford:Oxford University Press).

Tether, Bruce (2002). “The sources and aims of innovationin services: variety between and within sectors”, CRICDiscussion Paper, No. 55 (Manchester: CRIC)(les1.man.ac.uk/cric/Pdfs/DP55.pdf ).

________ (2004). “Do services innovate (differently)?”, CRICDiscussion Paper, No. 66 (Manchester: CRIC).

Todo, Yasuyuki and Satoshi Shimizutani (2005). “OverseasR&D activities by Japanese multinational enterprises:causes, impacts, and interactions with parent firms”.

250 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Paper presented at the ASEAN- UNCTAD seminar inBangkok, 5-6 May.

Toh, Mun Heng (2005). “R&D activities of TNCs in Singapore:its role in building the national innovation system andeconomic development”. Paper presented at the ASEAN-UNCTAD Seminar on “Key Issues of FDI: AttractingQuality FDI”, Bangkok, 5-6 May.

UN Millennium Project (2005). Innovation: ApplyingKnowledge to Development (London and Sterling:Earthscan Publishing).

UNCTAD (1999 – WIR99). World Investment Report 1999:Foreign Direct Investment and the Challenge forDevelopment (New York and Geneva: United Nations).

________ (2000 – WIR00). World Investment Report 2000:Cross-border Mergers and Acquisitions andDevelopment (New York and Geneva: United Nations).

________ (2002 – WIR02). World Investment Report 2002:Transnational Corporations and Export Competitiveness(New York and Geneva: United Nations).

________ (2003 – WIR03). World Investment Report 2003:FDI Policies for Development: National andInternational Perspectives (New York and Geneva:United Nations).

________ (2004 – WIR04). World Investment Report 2004:The Shift Towards Services (New York and Geneva:United Nations).

________ (1996a). The TRIPS Agreement and DevelopingCountries (New York and Geneva: United Nations).

________ (1996b). International Investment Instruments: ACompendium, Vols. I, II and III (Geneva: UnitedNations).

________ (2000). International Investment Instruments: ACompendium, Vols. IV and V (Geneva: United Nations).

________ (2001). Transfer of Technology. UNCTAD Serieson Issues in International Investment Agreements (NewYork and Geneva: United Nations).

________ (2002a). The Least Developed Countries Report2002: Escaping the Poverty Trap (New York andGeneva: United Nations)

________ (2002b). International Investment Instruments: ACompendium, Vols. VII, VIII, IX and X (Geneva: UnitedNations).

________ (2003a). World Investment Directory: Vol. VIIICentral and Eastern Europe 2003 (New York andGeneva: United Nations).

________ (2003b). Transfer of Technology for SuccessfulIntegration into the Global Economy (New York andGeneva: United Nations).

________ (2003c). Foreign Direct Investment andPerformance Requirements: New Evidence from SelectedCountries (New York and Geneva: United Nations).

________ (2003d). “Transfer of technology for successfulintegration into the global economy: a case study ofthe pharmaceutical industry in India”, United Nationsdocument, No. UNCTAD/ITE/IPC/2003/6 (New Yorkand Geneva: United Nations).

________ (2004a). The Least Developed Countries Report2004: Linking International Trade with PovertyReduction (Geneva and New York: United Nations).

________ (2004b). Facilitating Transfer of Technology toDeveloping Countries: A Survey of Home-CountryMeasures (New York and Geneva: United Nations).

________ (2004c). International Investment Instruments: ACompendium, Vols. XI and XII (Geneva: UnitedNations).

________ (2004d). State Contracts. UNCTAD Series on Issuesin International Investment Agreements (New York andGeneva: United Nations).

________ (2005a). “Global players from the BRICs: the riseof outward FDI from Brazil, Russia, India and China”,mimeo.

________ (2005b). TNCs and the Removal of Textiles andClothing Quotas (New York and Geneva: UnitedNations).

________ (2005c). Trade and Development Report 2005 (NewYork and Geneva: United Nations).

________ (2005d). “Report of the expert meeting on the impactof FDI on development”, UNCTAD document, TD/B/COM.2/EM.16/3 (Geneva: UNCTAD).

________ (2005e). International Investment Instruments: ACompendium – Vols. XIII and XIV (Geneva: UnitedNations).

________ (forthcoming a). South-South Cooperation inInternational Investment Arrangements. UNCTAD Serieson International Investment Policies for Development(Geneva and New York: United Nations).

________ (forthcoming b). “Many BITs have still not enteredinto force”, Occasional Note (Geneva and New York:United Nations).

________ (forthcoming c). Investment-State Disputes ArisingFrom Investment Treaties: A Review, UNCTAD Serieson International Investment Policies for Development(Geneva: United Nations).

________ (forthcoming d). Investment Policy Review: Kenya(New York and Geneva: United Nations).

________ (forthcoming e). Investment Policy Review Brazil(New York and Geneva: United Nations).

UNCTAD and ICSTD (2005). Resource Book on TRIPS andDevelopment (New York: Cambridge University Press).

UNDESA (2001). “Financial flow statistics”. BackgroundPaper No.19 prepared by Department of Economic andSocial Affairs for Sustainable Development,Commission on Sustainable Development, NinthSession, 16-27 April 2001.

UNDESA and UNCTAD (2005). World Economic Situationand Prospects 2005 (New York and Geneva: UnitedNations).

UNDP (2001). Human Development Report 2001: Making NewTechnologies Work for Human Development (New Yorkand Oxford: Oxford University Press).

________ (2002). Arab Human Development Report 2002:Building a Knowledge Society – Creating Opportunitiesfor Future Generations (New York: United NationsDevelopment Programme).

________ (2005). UN Millennium Project, 2005. Investingin Development: A Practical Plan to Achieve theMillennium Development Goals (London: Earthscan).

UNIDO (2003). Industrial Development Report 2002/2003(Vienna: UNIDO).

United Kingdom, Commission for Africa (2005). Report ofthe Commission for Africa: Our Common Interest(London: Commission for Africa).

251REFERENCES

United Kingdom, DTI (2004). The 2004 R&D Scoreboard:The top 700 UK and 700 International Companies byR&D Investment (London: DTI)(www.innovation.gov.uk/projects/rd_scoreboard/home.asp.)

United Kingdom, Inland Revenue (2002). Budget 2002:Research and Development (R&D) Tax Credit(www.inlandrevenue.gov.uk/budget2002/revbn16.pdf).

United States, NSB (2004). Science and EngineeringIndicators 2004 (Arlington, VA: National ScienceBoard), www.nsf.gov/statistics.

United States, NSF (2004). “Industrial R&D employment inthe United States and in U.S. multinationalcorporations”, InfoBrief, NSF 05-302 (Arlington, VA:United States, National Science Foundation)(www.nsf.gov/sbe/srs/infbrief/nsf05302/start.htm).

________ (forthcoming). Research and Development inIndustry: 2002 (Arlington, VA: United States, NationalScience Foundation).

Urata, Shujiro and Hiroki Kawai (2000). “Intrafirm technologytransfer by Japanese manufacturing firms in Asia”, inTakatoshi Ito and Anne O. Krueger, eds., The Role ofForeign Direct Investment in East Asian EconomicDevelopment (Chicago: University of Chicago Press),pp.49-77.

Velho, Lea (2004). “Science and technology in Latin Americaand the Caribbean: An Overview”, Discussion Paper,2004-4 (Maastricht: UNU-INTECH).

Vertzberger, Yaacov Y.I. (2005). “Strategic policy reform:policy resilience, learning and change”. Paper presentedat the 46th annual convention of the InternationalStudies Association, Honolulu, 1-5 March.

Veugelers, Reinhilde and Bruno Cassiman (2004). “Importanceof international linkages for local know-how flows:some econometric evidence from Belgium”, EuropeanEconomic Review, 48, 2, pp. 455-476.

Villa, Miguel and Jorge Martínez (2001). “El mapa migratoriointernacional de América Latina y el Caribe: patrones,perfiles, repercusiones e incertidumbres”. Paperpresented at the Seventh International Symposium on“Latin American Emigration: Interregional Comparisonamong North America, Europe and Japan”, Osaka, 11-13 December.

Voelker, Rainer and Richard Stead (1999). “New technologiesand international location choice for research anddevelopment units: evidence from Europe”, TechnologyAnalysis and Strategic Management, 11, 2, pp. 199-209.

Vossen, Rob (1996). R&D Decisions, Firm Size and MarketStructure (Capelle: Labyrint Publication).

Wagner, Caroline, Irene Brahmakulam, Brian Jackson, AnnyWong and Tatsuro Yoda (2001). Science and TechnologyCollaboration: Building Capacity in DevelopingCountries? (Santa Monica, CA: Rand Corporation).

Wald, Douglas L. and Deborah L. Feinstein (2004). “Mergerenforcement in innovation markets: the latest chapter– Genzyme/Novazyme”, The Antitrust Source, July2004.

Wallstein, Scott J. (2000). “The effects of government-industryprograms on private R&D: the case of the SmallBusiness Innovation Research program”, Rand Journalof Economics, 31, 1, pp. 82-100.

Walsh, Kathleen (2003). Foreign High-Tech R&D in China:Risks, Rewards and Implications for US-China Relations(Washington, D.C.: Henry L. Stimson Centre)(www.stimson.org/techtransfer/pdf/FinalReport.pdf).

WEF (various years). Global Competitiveness Report (Geneva:World Economic Forum).

van Wersch, Mark P.F.M. (2003). Balance of payments in theNetherlands: the Road to a Modern Survey System,Statistical Bulletin, Special Issue (Amsterdam: DeNederlandsche Bank) (www.dnb.nl/dnb/bin/doc/se2003m05_tcm13-36503.pdf)

World Bank (1993). The East Asian Miracle: Economic Growthand Public Policy (New York: Oxford University Press).

________ (2004). World Development Report 2005(Washington, D.C.: World Bank).

________ (2005a). Global Development Finance 2005(Washington, D.C.: World Bank).

________ (2005b). Global Economic Prospects 2005(Washington, D.C.: World Bank).

________ (2005c). “Private telecom projects – private activitydown 15 percent in 2003”, Public Policy for the PrivateSector Journal, Issue 288, April, pp. 1-4(rru.worldbank.org/Documents/PublicPolicyJournal/288izaguirre.pdf).

________ (2005d). Doing Business in 2005 (Washington, D.C.:World Bank).

________ (2005e). World Development Report 2005: A BetterInvestment Climate for Everyone (Washington, D.C.:World Bank and New York: Oxford University Press).

Yegorov, Igor (2004). “Delayed restructuring through gradualglobal integration in aviation and space industry in theUkraine”, in Slavo Radosevic and Bert M. Sadowski,eds., International Industrial Networks and IndustrialRestructuring in Central and Eastern Europe (Boston,Dordrecht and London: Kluwer Academic Publishers),pp. 155-166.

Yousef, Tarik M. (2005). “Structural reforms, the investmentclimate, and private sector development in the ArabWorld”, in World Economic Forum, Arab WorldCompetitiveness Report 2005 (Geneva: World EconomicForum), pp. 21-32.

Yun, Mikyung (2002). “Comments on ‘The relationshipbetween trade, foreign direct investment, anddevelopment: new evidence, strategy, and tactics underthe Doha Development Agenda begotiations’”. Paperpresented at a conference organized by the AsianDevelopment Bank in November 2002, mimeo.

Zander, Ivo (1994). “The tortoise evolution of the multinationalcorporations’ foreign technological activity in Swedishmultinational firms 1890-1990” (Stockholm: Instituteof International Business), unpublished dissertation.

Zanfei, Antonello (2000). “Transnational firms and thechanging organisation of innovative activities”,Cambridge Journal of Economics, 24, 5, pp. 515-542.

von Zedtwitz, Maximilian (2004). “International R&D incompanies from developing countries” (Beijing: Schoolof Economics and Management, Tsinghua University),mimeo.

________ (2005). “International R&D strategies in companiesfrom developing countries – the case of China”. Paperpresented at the UNCTAD Expert Meeting on the Impactof FDI on Development, Geneva, 24-26 January.

von Zedtwitz, Maximilian and Olivier Gassmann (2002).“Market versus technology drive in R&Dinternationalization: four different patterns of managingresearch and development”, Research Policy, 31, 4, pp.569-588.

252 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Zhang, Yanzhong (2005). “Globalization of R&D: new trendof economy globalization”. Paper presented at UNCTADExpert Meeting on FDI and Development, Geneva, 24-26 January.

Zhang, Yuehua (2004). “Effecting science park managementfor development: tracing the 20-year history ofSingapore Science Park for experience and lesson”(Waikato, New Zealand: University of Waikato), mimeo.

Zhao, Minyuan (2004). “Doing R&D in countries with weakIPR protection: can corporate management substitutefor legal institutions?”, mimeo.

Zhou, Yuping and Sanjaya Lall (2005). “The impact of China’sFDI surge on FDI in South-East Asia: panel data analysisfor 1986-2001", Transnational Corporations, 14, 1, pp.41-66.

ANNEXES

An

ne

x t

ab

le A

.I.1

. C

ros

s-b

ord

er

M&

A d

ea

ls w

ith

va

lue

s o

f o

ve

r $

1 b

illi

on

co

mp

lete

d i

n 2

00

4

Valu

eIn

dust

ry o

f th

eR

ank

($ b

illio

n)A

cqui

red

com

pany

Hos

t ec

onom

yIn

dust

ry o

f th

e ac

quir

ed c

ompa

nyA

cqui

ring

com

pany

Hom

e ec

onom

yac

quir

ing

com

pany

1

15.8

Abb

ey N

atio

nal P

lcU

nite

d K

ingd

omB

anks

San

tand

er C

entr

al H

ispa

no S

AS

pain

Ban

ks2

11

.1Jo

hn H

anco

ck F

inan

cial

Uni

ted

Sta

tes

Life

insu

ranc

eM

anul

ife F

inan

cial

Cor

pC

anad

aLi

fe in

sura

nce

3

10.5

Cha

rter

One

Fin

lU

nite

d S

tate

sS

avin

gs in

stitu

tions

, fe

dera

lly c

hart

ered

Citi

zens

Fin

anci

al G

roup

Uni

ted

Sta

tes

Nat

iona

l com

mer

cial

ban

ks4

9.

6A

mer

sham

Plc

Uni

ted

Kin

gdom

Bio

logi

cal p

rodu

cts,

exc

ept

diag

nost

icG

ener

al E

lect

ric

Uni

ted

Sta

tes

Pow

er,

dist

ribu

tion,

and

subs

tanc

essp

ecia

lty t

rans

form

ers

5

7.8

John

Lab

att

Ltd

Can

ada

Mal

t be

vera

ges

Am

bev

Bra

zil

Mal

t be

vera

ges

6

4.5

GA

GFA

H-H

ousi

ng P

ortf

olio

Ger

man

yD

wel

ling

oper

ator

s, e

xcep

t ap

artm

ents

Fort

ress

Deu

tsch

land

Gm

bHG

erm

any

Inve

stor

s, n

ec7

4.

0B

raco

SA

Bra

zil

Mal

t be

vera

ges

Inte

rbre

w S

AB

elgi

umM

alt

beve

rage

s8

3.

9E

diso

n M

issi

on E

nerg

y C

o.A

ustr

alia

Cog

ener

atio

n, a

ltern

ativ

e en

ergy

sou

rces

Inve

stor

Gro

upU

nite

d K

ingd

omIn

vest

ors,

nec

9

3.9

Gru

po F

inan

cier

o B

BVA

Ban

com

erM

exic

oB

anks

BB

VAS

pain

Ban

ks10

3.

7TX

U A

ustr

alia

Ltd

Aus

tral

iaE

lect

ric

serv

ices

Sin

gapo

re P

ower

Pte

Ltd

Sin

gapo

reE

lect

ric

serv

ices

11

3.4

SIC

OR

Inc

Uni

ted

Sta

tes

Pha

rmac

eutic

al p

repa

ratio

nsTe

va P

harm

a In

ds L

tdIs

rael

Pha

rmac

eutic

al p

repa

ratio

ns12

3.

4M

esse

r G

ries

heim

-Ind

Gas

Ops

Ger

man

yIn

dust

rial

gas

esA

ir L

iqui

de S

AFr

ance

Indu

stri

al g

ases

13

3.1

Can

ary

Wha

rf G

roup

Plc

Uni

ted

Kin

gdom

Land

sub

divi

ders

and

dev

elop

ers,

exc

l. ce

met

erie

sS

ongb

ird

Acq

uisi

tion

Ltd

Uni

ted

Sta

tes

Inve

stor

s, n

ec14

2.

9D

ial C

orp

Uni

ted

Sta

tes

Soa

p &

oth

er d

eter

gent

s, e

xcep

tH

enke

l KG

aAG

erm

any

Per

fum

es,

cosm

etic

s, a

ndsp

ecia

lty c

lean

ers

othe

r to

ilet

prep

arat

ions

15

2.8

Moo

re W

alla

ce I

ncC

anad

aM

anifo

ld b

usin

ess

form

sR

R D

onne

lley

& S

ons

Co

Uni

ted

Sta

tes

Com

mer

cial

pri

ntin

g,lit

hogr

aphi

c16

2.

7C

ellte

ch G

roup

Plc

Uni

ted

Kin

gdom

Com

mer

cial

phy

sica

l and

bio

logi

cal r

esea

rch

UC

B S

AB

elgi

umM

edic

inal

che

mic

als

and

bota

nica

l pro

duct

s17

2.

7D

ynam

it N

obel

AG

Ger

man

yE

xplo

sive

sR

ockw

ood

Spe

cial

ties

Gro

upU

nite

d S

tate

sC

hem

ical

s an

d ch

emic

alpr

epar

atio

ns,

nec

18

2.6

Hid

roca

ntab

rico

Spa

inE

lect

ric

serv

ices

ED

PP

ortu

gal

Ele

ctri

c se

rvic

es19

2.

6V

NU

Wor

ld D

irec

tori

esN

ethe

rlan

dsM

isce

llane

ous

publ

ishi

ngIn

vest

or G

roup

Uni

ted

Kin

gdom

Inve

stor

s, n

ec20

2.

4JC

Pen

ney-

Eck

erd

Sto

res

Uni

ted

Sta

tes

Dru

g st

ores

and

pro

prie

tary

sto

res

Jean

Cou

tu G

roup

(PJC

)Inc

Can

ada

Dru

g st

ores

and

pro

prie

tary

stor

es21

2.

4Le

aseP

lan

Cor

p N

VN

ethe

rlan

dsP

asse

nger

car

leas

ing

Inve

stor

Gro

upG

erm

any

Inve

stor

s, n

ec22

2.

4G

SW

Ger

man

yS

ingl

e-fa

mily

hou

sing

con

stru

ctio

nIn

vest

or G

roup

Uni

ted

Sta

tes

Inve

stor

s, n

ec23

2.

3To

m B

row

n In

cU

nite

d S

tate

sC

rude

pet

role

um a

nd n

atur

al g

asE

nCan

a C

orp

Can

ada

Cru

de p

etro

leum

& n

atur

al g

as24

2.

2C

elan

ese

AG

Ger

man

yIn

dust

rial

org

anic

che

mic

als,

nec

Bla

ckst

one

Gro

up L

PU

nite

d S

tate

sIn

vest

ors,

nec

25

2.1

EnC

ana(

UK

)Hol

ding

s Lt

dU

nite

d K

ingd

omC

rude

pet

role

um a

nd n

atur

al g

asN

exen

Inc

Can

ada

Cru

de p

etro

leum

& n

atur

al g

as26

2.

0D

DI

Poc

ket

Inc

Japa

nR

adio

tele

phon

e co

mm

unic

atio

nsIn

vest

or G

roup

Uni

ted

Sta

tes

Inve

stor

s, n

ec27

2.

0P

roct

er &

Gam

ble-

Hut

chis

on L

tdC

hina

Soa

p &

oth

er d

eter

gent

s, e

xcep

tP

roct

er &

Gam

ble

Co

Uni

ted

Sta

tes

Soa

p &

oth

er d

eter

gent

s,sp

ecia

lty c

lean

ers

exce

pt s

peci

alty

cle

aner

s28

1.

9Fo

ndo

Imm

obili

Pub

blic

iIt

aly

Rea

l est

ate

inve

stm

ent

trus

tsIn

vest

or G

roup

Uni

ted

Kin

gdom

Inve

stor

s, n

ec29

1.

9A

quila

Ste

rlin

g Lt

dU

nite

d K

ingd

omE

lect

ric

serv

ices

Pow

erG

en P

LcU

nite

d K

ingd

omE

lect

ric

serv

ices

30

1.9

STM

icro

elec

tron

ics

NV

Sw

itzer

land

Sem

icon

duct

ors

and

rela

ted

devi

ces

Cas

sa D

epos

iti e

Pre

stiti

SpA

Ital

yP

erso

nal c

redi

t in

stitu

tions

31

1.9

US

Pre

miu

m O

ffice

Pro

pert

ies

Uni

ted

Sta

tes

Ope

rato

rs o

f no

n-re

side

ntia

l bui

ldin

gsIn

vest

or G

roup

Sin

gapo

reIn

vest

ors,

nec

32

1.8

WC

M-R

esid

entia

l Ppt

yG

erm

any

Rea

l est

ate

agen

ts a

nd m

anag

ers

Bla

ckst

one

Gro

up L

PU

nite

d S

tate

sIn

vest

ors,

nec

33

1.8

ATU

Aut

o-Te

ile-U

nger

Gm

bHG

erm

any

Aut

o an

d ho

me

supp

ly s

tore

sK

ohlb

erg

Kra

vis

Rob

erts

& C

oU

nite

d S

tate

sIn

vest

ors,

nec

34

1.7

BoC

OM

MC

hina

Ban

ksH

SB

C H

oldi

ngs

Plc

Uni

ted

Kin

gdom

Ban

ks35

1.

7If

Ska

defo

rsak

ring

Hol

ding

AB

Sw

eden

Life

insu

ranc

eS

ampo

Oyj

Finl

and

Inve

stm

ent

advi

ce36

1.

7Vo

dafo

ne K

KJa

pan

Rad

iote

leph

one

com

mun

icat

ions

Voda

fone

Int

l Hol

ding

sN

ethe

rlan

dsR

adio

tele

phon

eco

mm

unic

atio

ns37

1.

7S

ecur

icor

Plc

Uni

ted

Kin

gdom

Sec

urity

sys

tem

s se

rvic

esG

roup

4 F

alck

A/S

Den

mar

kIn

vest

ors,

nec

38

1.7

New

Rea

l SpA

Ital

yR

eal e

stat

e in

vest

men

t tr

usts

Exc

elsi

a O

tto

Ger

man

yS

peci

al p

urpo

se f

inan

ce39

1.

7B

renn

tag

AG

Ger

man

yC

hem

ical

s an

d al

lied

prod

ucts

, ne

cB

ain

Cap

ital I

ncU

nite

d S

tate

sIn

vest

ors,

nec

/...

255ANNEX A

256 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.I.1

. C

ros

s-b

ord

er

M&

A d

ea

ls w

ith

va

lue

s o

f o

ve

r $

1 b

illi

on

co

mp

lete

d i

n 2

00

4 (

co

nc

lud

ed

)

Valu

eIn

dust

ry o

f th

eR

ank

($ b

illio

n)A

cqui

red

com

pany

Hos

t ec

onom

yIn

dust

ry o

f th

e ac

quir

ed c

ompa

nyA

cqui

ring

com

pany

Hom

e ec

onom

yac

quir

ing

com

pany

40

1.7

Gas

Tra

nsm

issi

on N

W C

orp

Uni

ted

Sta

tes

Nat

ural

gas

tra

nsm

issi

onTr

ansC

anad

a C

orp

Can

ada

Ele

ctri

c se

rvic

es41

1.

7K

onin

klijk

e Ve

ndex

KB

B N

VN

ethe

rland

sD

epar

tmen

t sto

res

VD

XK

Acq

uisi

tion

BV

Uni

ted

Sta

tes

Spe

cial

pur

pose

fina

nce

42

1.7

Pic

ard

Sur

gele

s S

AFr

ance

Gro

cery

sto

res

BC

Par

tner

s Lt

dU

nite

d K

ingd

omIn

vest

ors,

nec

43

1.6

Kor

am B

ank

Kor

ea,

Rep

ublic

of

Ban

ksC

itigr

oup

Inc

Uni

ted

Sta

tes

Nat

iona

l com

mer

cial

ban

ks44

1.

6S

ophi

a S

AFr

ance

Less

ors

of r

eal p

rope

rty,

nec

GE

Cap

ital R

eal E

stat

eU

nite

d S

tate

sLe

ssor

s of

rea

l pro

pert

y, n

ec45

1.

5Ve

rizo

n-C

anad

ian

Dir

ecto

ry B

usC

anad

aM

isce

llane

ous

publ

ishi

ngB

ain

Cap

ital I

ncU

nite

d S

tate

sIn

vest

ors,

nec

46

1.5

Ash

anti

Gol

dfie

lds

Co

Ltd

Gha

naG

old

ores

Ang

logo

ld L

tdS

outh

Afr

ica

Gol

d or

es47

1.

5C

onta

ct E

nerg

y Lt

dN

ew Z

eala

ndE

lect

ric

serv

ices

Ori

gin

Ene

rgy

Ltd

Aus

tral

iaE

lect

ric

serv

ices

48

1.5

MO

NY

Gro

up I

ncU

nite

d S

tate

sLi

fe in

sura

nce

AX

A F

inan

cial

Inc

Uni

ted

Sta

tes

Life

insu

ranc

e49

1.

5B

SN

Gla

sspa

ck S

AFr

ance

Gla

ss c

onta

iner

sO

wen

s-Ill

inoi

s In

cU

nite

d S

tate

sG

lass

con

tain

ers

50

1.5

Chi

pPA

C I

ncU

nite

d S

tate

sS

emic

ondu

ctor

s an

d re

late

d de

vice

sS

T A

ssem

bly

Test

Ser

vice

s Lt

dS

inga

pore

Inst

rum

ents

to

mea

sure

elec

tric

ity51

1.

5M

obilT

el A

GB

ulga

riaR

adio

tele

phon

e co

mm

unic

atio

nsB

idC

o A

DA

ustr

iaS

peci

al p

urpo

se fi

nanc

e52

1.

4V

impe

lcom

OJS

CR

ussi

an F

eder

atio

nR

adio

tele

phon

e co

mm

unic

atio

nsA

lfa T

elec

om L

tdB

ritis

h V

irgin

Tele

phon

e co

mm

unic

atio

ns,

Isla

nds

exce

pt r

adio

tele

phon

e53

1.

4Fo

ur S

easo

ns H

ealth

Car

e Lt

dU

nite

d K

ingd

omN

ursi

ng&

pers

onal

car

e fa

cils

Inve

stor

Gro

upG

erm

any

Inve

stor

s, n

ec54

1.

4Vo

dafo

ne H

oldi

ngs

KK

Japa

nTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Voda

fone

Int

l Hol

ding

sN

ethe

rlan

dsR

adio

tele

phon

eco

mm

unic

atio

ns55

1.

4La

Rin

asce

nte

SpA

Ital

yD

epar

tmen

t st

ores

Auc

han

SA

Fran

ceG

roce

ry s

tore

s56

1.

4S

avoy

Gro

up L

tdU

nite

d K

ingd

omH

otel

s an

d m

otel

sIn

vest

or G

roup

Irel

and

Inve

stor

s, n

ec57

1.

4N

K S

lavn

eft

Rus

sian

Fed

erat

ion

Oil

and

gas

field

exp

lora

tion

serv

ices

Bri

tish

Pet

role

um P

lcU

nite

d K

ingd

omS

oap

& o

ther

det

erge

nts,

exce

pt s

peci

alty

cle

aner

s58

1.

4B

urns

Phi

lp-H

erbs

& S

pice

Div

Uni

ted

Sta

tes

Food

pre

para

tions

, ne

cA

ssoc

iate

d B

ritis

h Fo

ods

Plc

Uni

ted

Kin

gdom

Bre

ad a

nd o

ther

bak

ery

prod

ucts

, ex

cept

coo

kies

59

1.3

Tele

foni

ca M

ovil

Chi

le S

AC

hile

Rad

iote

leph

one

com

mun

icat

ions

Tele

foni

ca M

ovile

s S

AS

pain

Rad

iote

leph

one

com

mun

icat

ions

60

1.2

IGE

N I

nter

natio

nal I

ncU

nite

d S

tate

sP

harm

aceu

tical

pre

para

tions

Roc

he H

oldi

ng A

GS

witz

erla

ndP

harm

aceu

tical

pre

para

tions

61

1.2

Com

mun

ity F

irst

Ban

ksha

res

Uni

ted

Sta

tes

Sta

te b

anks

,mem

ber

fed

rese

rve

Ban

cWes

t C

orp,

Hon

olul

u,H

IU

nite

d S

tate

sS

tate

ban

ks62

1.

2G

old

Fiel

ds L

tdS

outh

Afr

ica

Gol

d or

esN

orim

et L

tdU

nite

d K

ingd

omM

etal

s se

rvic

e ce

ntre

s an

dof

fices

63

1.2

Ban

k of

Ber

mud

a Lt

dB

erm

uda

Ban

ksH

SB

C H

oldi

ngs

Plc

Uni

ted

Kin

gdom

Ban

ks64

1.

2Te

lcel

Cel

ular

SA

Vene

zuel

aR

adio

tele

phon

e co

mm

unic

atio

nsTe

lefo

nica

Mov

iles

SA

Spa

inR

adio

tele

phon

eco

mm

unic

atio

ns65

1.

1FI

H E

rhve

rvsb

ank

A/S

Den

mar

kB

anks

Kau

pthi

ng B

unad

arba

nki h

fIc

elan

dB

anks

66

1.1

Voda

fone

Hol

ding

s K

KJa

pan

Tele

phon

e co

mm

unic

atio

ns,

exce

pt r

adio

tele

phon

eVo

dafo

ne I

ntl H

oldi

ngs

Net

herl

ands

Rad

iote

leph

one

com

mun

icat

ions

67

1.1

Lora

l Spa

ce-S

atel

lites

(6)

Uni

ted

Sta

tes

Com

mun

icat

ions

ser

vice

s, n

ecIn

tels

at L

tdB

erm

uda

Com

mun

icat

ions

ser

vice

s, n

ec68

1.

1W

eeta

bix

Ltd

Uni

ted

Kin

gdom

Cer

eal b

reak

fast

foo

dsLa

timer

Acq

uisi

tions

Ltd

Uni

ted

Kin

gdom

Spe

cial

pur

pose

fin

ance

com

pany

69

1.1

Bel

lsou

th-C

olom

bian

Ope

ratio

nsC

olom

bia

Rad

iote

leph

one

com

mun

icat

ions

Tele

foni

ca M

ovile

s S

AS

pain

Rad

iote

leph

one

com

mun

icat

ions

70

1.0

Firs

t Act

ive

Plc

Irel

and

Mor

tgag

e ba

nker

s an

d lo

an c

orre

spon

dent

sR

oyal

Ban

k of

Sco

tland

Gro

upU

nite

d K

ingd

omB

anks

71

1.0

ICA

AB

Sw

eden

Gro

cery

sto

res

Kon

inkl

ijke

Aho

ld N

VN

ethe

rlan

dsG

roce

ry s

tore

s72

1.

0D

olla

ram

aC

anad

aD

epar

tmen

t st

ores

Bai

n C

apita

l Inc

Uni

ted

Sta

tes

Inve

stor

s, n

ec73

1.

0C

alW

est

LLC

-Ind

l Ppt

y P

ortf

oli

Uni

ted

Sta

tes

Ope

rato

rs o

f no

nres

iden

tial b

uild

ings

Inve

stor

Gro

upA

ustr

alia

Inve

stor

s, n

ec74

1.

0Fr

iedr

ich

Gro

he A

G &

Co

KG

Ger

man

yP

lum

bing

& h

eatin

g eq

uipm

ent &

sup

plie

s(hy

dron

ics)

Inve

stor

Gro

upU

nite

d S

tate

sIn

vest

ors,

nec

75

1.0

Pet

rom

SA

Rom

ania

Cru

de p

etro

leum

and

nat

ural

gas

OM

V A

GA

ustr

iaC

rude

pet

role

um &

nat

ural

gas

So

urc

e:

UN

CT

AD

, cr

oss

-bo

rde

r M

&A

da

tab

ase

.

No

te:

Wh

ere

th

e u

ltim

ate

pa

ren

t co

mp

an

y is

dif

fere

nt,

M&

A d

ea

ls w

ith

in t

he

sa

me

eco

no

my

are

sti

ll c

on

sid

ere

d c

ross

-bo

rde

r M

&A

s.

257ANNEX A

Annex table A.I.2. Number of greenfield FDI projects, by destination, 2002-2004

Total Total

Destination economy/region 2002 2003 2004 Destination economy/region 2002 2003 2004

World 5 656 9 303 9 796 Other Africa 95 181 158Developed countries 2 721 3 843 4 070 West Africa 27 53 30

Europe 1 814 2 651 2 941 Benin - 1 -European Union 1 770 2 565 2 851 Burkina Faso - 1 1

Austria 13 80 96 Cape Verde 1 - -Belgium 62 62 99 Côte d’Ivoire - 1 -Cyprus 9 8 6 Ghana 2 13 4Czech Republic 94 141 136 Guinea 4 2 3Denmark 25 73 86 Guinea-Bissau - 1 -Estonia 32 29 40 Mauritania 1 2 1Finland 17 31 30 Niger - 1 -France 126 155 201 Nigeria 17 25 18Germany 130 264 247 Senegal 2 2 3Greece 26 42 56 Sierra Leone - 4 -Hungary 210 213 211 Central Africa 11 21 19Ireland 93 137 128 Angola 6 14 16Italy 72 110 123 Cameroon 2 1 1Latvia 38 42 27 Chad 1 - -Lithuania 36 43 22 Congo - 1 1Luxembourg 4 12 12 Congo, Dem. Rep. of 1 3 1Malta 4 3 3 Equatorial Guinea 1 2 -Netherlands 42 100 82 East Africa 22 44 50Poland 91 154 230 Burundi 1 - -Portugal 42 58 69 Eritrea - 1 1Slovakia 44 63 85 Ethiopia - 2 1Slovenia 13 23 16 Kenya 4 12 15Spain 153 215 241 Madagascar - 3 3Sweden 68 93 123 Mauritius 6 3 7United Kingdom 326 414 482 Mozambique 2 5 4

Other developed Europe 44 86 90 Seychelles 1 - 2Iceland 1 4 1 Somalia - - 1Liechtenstein 2 - - Tanzania, United Rep. of 2 6 6Norway 7 26 23 Uganda 2 5 5Switzerland 34 56 66 Zambia 4 4 4

North America 632 829 801 Zimbabwe - 3 1Canada 218 241 223 Southern Africa 35 63 59United States 414 588 578 Botswana 3 4 5

Other developed countries 275 363 328 Lesotho - 1 -Australia 137 180 139 Namibia 1 3 3Greenland 1 2 1 South Africa 31 55 49Israel 7 16 16 Swaziland - - 2Japan 106 132 152 Latin America and the Caribbean 562 794 794New Zealand 24 33 20 South and Central America 525 742 743

Developing countries and territories 2 355 4 446 4 758 South America 367 530 556Africa 169 306 262 Argentina 44 64 73

North Africa 74 125 104 Bolivia 10 9 14Algeria 15 21 19 Brazil 175 287 258Egypt 23 40 32 Chile 38 61 55Libyan Arab Jamahiriya 2 4 7 Colombia 26 43 47Morocco 23 35 32 Ecuador 11 9 21Sudan 3 10 5 Guyana - - 1Tunisia 8 15 9 Paraguay 1 3 2

/...

258 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.I.2. Number of greenfield FDI projects, by destination, 2002-2004 (concluded)

Total Total

Destination economy/region 2002 2003 2004 Destination economy/region 2002 2003 2004

Peru 26 30 32 Macao, China 2 3 6Suriname 1 2 - Mongolia 2 6 2Uruguay 12 4 10 Taiwan Province of China 41 113 83Venezuela 23 18 43 South Asia 284 514 725

Central America 158 212 187 Afghanistan 2 6 4Costa Rica 7 13 7 Bangladesh 9 17 7El Salvador 6 4 7 India 250 457 685Guatemala 3 5 3 Maldives 1 - -Honduras 4 7 5 Nepal 1 1 1Mexico 129 168 154 Pakistan 13 23 17Nicaragua 3 8 1 Sri Lanka 8 10 11Panama 6 7 10 South-East Asia 361 784 726

Caribbean and other America 37 52 51 Brunei Darussalam 1 2 2Antigua and Barbuda 1 - - Cambodia 1 5 7Aruba - 1 - Indonesia 31 60 59Bahamas 2 3 1 Lao People’s Dem. Rep. - 5 3Barbados 2 - 1 Malaysia 79 183 125Bermuda 1 1 - Myanmar - 5 1Cuba 4 6 5 Philippines 28 74 74Dominican Rep. 7 11 9 Singapore 107 156 173Guadeloupe - 1 - Thailand 59 161 121Haiti 1 - - Timor-Leste - 1 -Jamaica 3 5 4 Viet Nam 55 132 161Martinique - 1 - Oceania 3 6 6Puerto Rico 12 19 29 Fiji - 2 -Saint Lucia - 1 - New Caledonia 2 - 2Trinidad and Tobago 4 3 2 Papua New Guinea 1 4 4

Asia and Oceania 1 624 3 346 3 702 Transition economies 580 1 014 968Asia 1 621 3 340 3 696 South-East Europe 299 352 398

West Asia 233 421 403 Albania 12 10 7Bahrain 24 24 15 Bosnia and Herzegovina 15 28 18Iran, Islamic Rep. of 10 29 23 Bulgaria 77 96 109Iraq - 32 5 Croatia 33 45 39Jordan 4 15 11 Macedonia, TFYR 8 10 7Kuwait 4 7 19 Romania 112 116 168Lebanon 8 20 23 Serbia and Montenegro 42 47 50Oman 10 11 14 CIS 281 662 570Qatar 14 22 26 Armenia 2 16 6Saudi Arabia 21 32 37 Azerbaijan 9 25 25Syrian Arab Rep. 2 8 6 Belarus 1 15 10Turkey 46 69 64 Georgia 4 3 6United Arab Emirates 88 146 156 Kazakhstan 6 36 30Yemen 2 6 4 Kyrgyzstan - 6 1

South, East and South-East Asia 1 388 2 919 3 293 Moldova, Republic of 5 8 14East Asia 743 1 621 1 842 Russian Federation 202 432 377

China 581 1 299 1 529 Tajikistan - 6 4Hong Kong, China 57 90 122 Turkmenistan 5 13 3Korea, Dem. People’s Rep. of - 1 - Ukraine 28 72 79Korea, Republic of 60 109 100 Uzbekistan 19 30 15

Source: UNCTAD, based on information from OCO Consulting, LOCOmonitor website (www.locomonitor.com).

Note: Includes new (greenfield) and expansion FDI projects, both announced and realized.

259ANNEX A

Annex table A.I.3. Number of greenfield FDI projects, by industry, 2002-2004

Sector/industry 2002 2003 2004 2002 2003 2004 2002 2003 2004 2002 2003 2004

Latin America and World Developed countries Africa the Caribbean

Total 5 656 9 303 9 796 2 721 3 843 4 070 169 306 262 562 794 794

Primary 217 473 281 31 83 45 31 76 45 78 128 83Extraction 217 473 281 31 83 45 31 76 45 78 128 83

Manufacturing 1 929 3 176 3 108 809 1 134 1 119 66 103 88 218 318 304

Services 3 510 5 654 6 407 1 881 2 626 2 906 72 127 129 266 348 407Electricity 26 76 58 12 33 24 4 2 2 5 17 8Construction 263 403 369 113 130 140 14 12 15 34 47 62Retail 741 1 217 1 463 420 632 774 16 16 21 73 87 95Internet or ICT infrastructure 72 148 137 45 59 55 4 3 2 8 12 31Business services 592 1 034 940 254 446 388 14 45 22 41 61 40Research and development 331 516 642 168 188 194 1 8 3 3 17 21Sales, marketing and support 581 1 041 1 462 336 525 682 8 18 31 42 53 76Headquarters 292 319 358 195 185 210 1 3 5 6 6 10Testing 58 47 39 31 18 11 - - 1 3 4 2Customer support centre 125 176 146 77 104 65 2 3 6 12 11 11Logistics and distribution 275 425 507 157 203 269 6 8 11 29 26 39Maintenance/service 41 107 74 16 43 26 - 5 4 3 2 3Shared services centre 42 73 77 18 34 25 - 1 2 2 2 1Technical support centre 18 17 62 13 7 26 - - 3 - 1 2Training 53 55 73 26 19 17 2 3 1 5 2 6

of which: South, East and South-East Europe

Asia and Oceania West Asia South-East Asia and CIS

Total 1 624 3 346 3 702 233 421 403 1 388 2 919 3 293 580 1 014 968

Primary 56 122 66 15 31 21 39 90 44 21 64 42Extraction 56 122 66 15 31 21 39 90 44 21 64 42

Manufacturing 580 1 246 1 241 47 85 81 533 1 158 1 158 256 375 356

Services 988 1 978 2 395 171 305 301 816 1 671 2 091 303 575 570Electricity 3 17 15 1 4 - 2 13 15 2 7 9Construction 62 155 101 19 48 23 43 107 78 40 59 51Retail 130 299 404 40 58 74 90 241 329 102 183 169Internet or ICT infrastructure 10 51 33 1 7 2 9 44 31 5 23 16Business services 198 377 376 42 74 66 156 301 309 85 105 114Research and development 155 291 405 - 1 5 155 290 400 4 12 19Sales, marketing and support 159 341 564 25 52 80 134 289 483 36 104 109Headquarters 86 116 127 18 21 19 68 95 108 4 9 6Testing 19 24 23 3 1 1 15 23 22 5 1 2Customer support centre 34 57 61 2 5 2 32 52 59 - 1 3Logistics and distribution 69 137 137 14 14 19 55 123 118 14 51 51Maintenance/service 19 43 30 2 12 5 17 31 25 3 14 11Shared services centre 21 35 49 1 - - 20 35 49 1 1 -Technical support centre 5 8 27 1 - - 4 8 27 - 1 4Training 18 27 43 2 8 5 16 19 38 2 4 6

Source: UNCTAD, based on information from OCO Consulting, LOCOmonitor website (www.locomonitor.com).

Note: The items under the main sectors refer to the key business function or the primary activity of each project.

260 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.I.4. Estimated world inward FDI stock, by sector and industry, 1990, 2003(Millions of dollars)

1990 2003

South-EastSector/industry Developed Developing Developed Developing Europe and

countries economies World countries economies CIS World

Primary 145 404 24 727 170 131 428 831 143 993 21 498 594 321Agriculture, hunting, forestry and fishing 3 326 4 253 7 579 6 854 22 579 205 29 637Mining, quarrying and petroleum 142 078 18 337 160 415 417 878 121 414 21 294 560 585Unspecified primary - 2 137 2 137 4 099 - - 4 099

Manufacturing 595 142 150 410 745 552 2 081 645 779 112 15 345 2 876 102Food, beverages and tobacco 66 744 10 010 76 754 211 181 25 983 6 278 243 442Textiles, clothing and leather 22 277 5 224 27 501 49 055 8 545 41 57 641Wood and wood products 19 280 4 563 23 843 78 160 15 990 1 114 95 264Publishing, printing and reproduction

of recorded media 14 444 568 15 013 59 310 4 51 59 365Coke, petroleum products and nuclear fuel 51 526 3 147 54 672 59 309 17 702 450 77 460Chemicals and chemical products 115 342 45 481 160 823 437 022 76 201 2 475 515 699Rubber and plastic products 12 225 1 838 14 064 36 427 2 440 37 38 904Non-metallic mineral products 16 079 2 835 18 914 59 928 5 526 832 66 286Metals and metal products 47 540 15 104 62 643 153 850 23 415 1 463 178 728Machinery and equipment 44 776 10 015 54 792 130 369 28 990 1 765 161 124Electrical and electronic equipment 68 291 17 311 85 603 222 348 57 125 347 279 821Precision instruments 11 312 478 11 789 32 062 1 811 92 33 964Motor vehicles and other transport

equipment 45 085 8 124 53 208 277 764 13 123 52 290 940Other manufacturing 17 896 3 173 21 069 91 257 15 348 35 106 640Unspecified secondary 42 324 22 539 64 863 183 604 486 910 313 670 827

Services 717 147 157 950 875 097 4 015 555 1 110 757 27 514 5 153 826Electricity, gas and water 6 804 2 784 9 588 143 734 41 016 1 130 185 880Construction 15 919 5 267 21 185 56 441 38 216 776 95 433Trade 191 244 24 399 215 644 762 879 152 371 5 548 920 798Hotels and restaurants 20 269 4 004 24 274 66 177 24 029 378 90 585Transport, storage and communications 15 367 11 988 27 355 343 859 90 296 10 843 444 998Finance 264 677 88 920 353 597 1 299 225 251 082 4 832 1 555 138Business activities 107 279 14 341 121 620 831 063 448 307 a 3 700 1 283 070 a

Public administration and defence - 57 57 1 831 383 65 2 279Education 90 - 90 490 2 18 510Health and social services 952 - 952 9 382 4 378 1 13 761Community, social and personal

service activities 12 795 20 12 815 68 089 5 186 192 73 468Other services 68 585 4 517 73 102 369 438 36 711 2 406 152Unspecified tertiary 13 165 1 654 14 819 62 947 18 778 28 81 753

Private buying and selling of property .. .. .. 1 000 - - 1 000Unspecified 8 822 4 229 13 051 57 055 74 859 3 245 135 159

Source: UNCTAD.a A considerable share of investment in this industry is in Hong Kong (China), which accounted for 60% of developing

economies stock and 21% of the world total stock. Hong Kong (China) data include investment holding companies.

Note: Data should be interpreted with caution. The world total was extrapolated on the basis of data covering50 countries in 1990 and 63 countries in 2003, or latest year available. They accounted for over four-fifthsof world inward FDI stock in 1990 and 2003. Only countries for which data for the three main sectors wereavailable, were included. The distribution share of each industry of these countries was applied to estimatethe world total in each sector and industry. As a result, the sum of the sectors for each economic groupis different from the totals shown in annex table B.2. Approval data were used for Mongolia in 1992. Howeverin the case of Cambodia, China, Indonesia, the Lao People’s Democratic Republic, Mongolia (2002), Myanmar,Nepal, Sri Lanka, Taiwan Province of China and Viet Nam, the actual data were estimated by applying theimplementation ratio of realized FDI to approved FDI to the latter (33% in 1994 for Cambodia, 54% in 2002for China, 30% in 1997 for Indonesia, 10% in 1990 and 7% in 1999 for Lao People's Democratic Republic,44% in 2002 for Mongolia, 39% in 1990 and 45% in 2002 for Myanmar, 41% in 1990 and 47% in 1999 forNepal, 62% in 1995 for Sri Lanka, 74% in 1990 and 63% in 2002 for Taiwan Province of China and 15%in 1990 for Viet Nam). The world total in 1990 includes South-East Europe and the CIS, although data bysector and industry are not available for that region.

261ANNEX A

Annex table A.I.5. Estimated world outward FDI stock, by sector and industry, 1990, 2003(Millions of dollars)

1990 2003

South-EastSector/industry Developed Developing Developed Developing Europe and

countries economies World countries economies CIS World

Primary 158 187 867 159 054 400 733 3 178 554 404 465Agriculture, hunting, forestry and fishing 5 135 285 5 420 3 470 697 1 4 168Mining, quarrying and petroleum 153 052 582 153 634 394 607 2 481 553 397 641Unspecified primary .. .. .. 2 657 - - 2 657

Manufacturing 773 322 6 109 779 432 2 117 367 103 414 392 2 221 174Food, beverages and tobacco 74 023 420 74 443 233 185 2 060 59 235 304Textiles, clothing and leather 19 142 187 19 329 108 596 2 712 7 111 315Wood and wood products 21 041 80 21 121 53 803 1 462 - 55 265Publishing, printing and reproduction of recorded media 2 218 - 2 218 11 799 - - 11 800Coke, petroleum products and nuclear fuel 38 500 - 38 500 28 069 274 6 28 349Chemicals and chemical products 147 763 762 148 525 436 793 4 351 230 441 373Rubber and plastic products 14 240 101 14 341 27 166 969 - 28 135Non-metallic mineral products 12 845 183 13 028 19 830 864 6 20 701Metals and metal products 64 963 85 65 048 219 894 2 618 - 222 512Machinery and equipment 41 162 22 41 183 81 679 406 - 82 085Electrical and electronic equipment 95 467 1 018 96 485 169 149 15 854 - 185 002Precision instruments 13 246 - 13 246 31 040 405 - 31 445Motor vehicles and other transport equipment 58 996 10 59 006 351 904 1 512 52 353 468Other manufacturing 34 096 10 34 106 53 994 186 33 54 213Unspecified secondary 135 621 3 231 138 852 290 466 69 742 - 360 209

Services 815 717 11 350 827 067 5 058 640 562 409 795 5 621 844Electricity, gas and water 9 417 - 9 417 108 142 27 - 108 169Construction 17 861 178 18 038 37 526 6 805 6 44 337Trade 136 983 1 836 138 819 498 761 65 342 43 564 146Hotels and restaurants 6 978 - 6 978 82 072 8 486 - 90 558Transport, storage and communications 38 930 501 39 431 457 599 41 093 112 498 804Finance 389 831 7 027 396 858 1 731 335 153 304 74 1 884 714Business activities 53 959 1 283 55 242 1 700 643 271 469 526 1 972 639Public administration and defence .. .. .. 5 677 - - 5 676Education 422 - 422 877 1 - 878Health and social services 838 - 838 839 - - 839Community, social and personal service activit ies 3 354 - 3 354 16 133 202 - 16 336Other services 108 148 526 108 674 371 099 13 258 34 384 391Unspecified tertiary 48 995 - 48 995 47 936 2 421 - 50 358

Private buying and selling of property .. .. .. 1 155 - - 1 155Unspecified 3 413 240 3 653 143 616 51 870 21 195 507

Source: UNCTAD.

Notes: Data should be interpreted with caution. The world total was extrapolated on the basis of data covering24 countries in 1990 and 37 countries in 2003, or latest year available. They accounted for around four-fifths of world outward FDI stock in 1990 and 2003. The distribution share of each industry of these countrieswas applied to estimate the world total in each sector and industry. As a result, the sum of the sectors foreach economic group is different from the totals shown in annex table B.2. Approval data were used forTaiwan Province of China. For 1990, the world total includes South-East Europe and the CIS although databy sector and industry were not available for that region. Moreover, as major home developing economieswere not covered due to lack of data, the respective shares for developing economies were underestimatedfor that year.

262 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.I.6. Estimated world inward FDI flows, by sector and industry, 1989-1991 and2001-2003

(Millions of dollars)

1989-1991 2001-2003

South-EastSector/industry Developed Developing Developed Developing Europe and

countries economies World countries economies CIS World

Primary 9 552 3 725 13 276 34 260 14 597 2 780 51 637Agriculture, hunting, forestry and fishing - 11 610 599 81 1 699 44 1 823Mining, quarrying and petroleum 9 523 3 115 12 638 34 215 12 899 2 736 49 850Unspecified primary 39 - 39 - 36 - - - 36

Manufacturing 50 915 16 880 67 795 96 424 68 997 1 760 167 181Food, beverages and tobacco 5 100 2 581 7 681 11 249 3 110 731 15 089Textiles, clothing and leather 2 148 263 2 411 2 580 1 114 2 3 697Wood and wood products 2 032 254 2 286 1 871 260 155 2 286Publishing, printing and reproduction of recorded media 915 - 915 1 537 124 - 1 661Coke, petroleum products and nuclear fuel - 1 013 325 - 688 7 032 178 68 7 278Chemicals and chemical products 11 270 2 265 13 535 13 789 5 363 103 19 255Rubber and plastic products 954 32 987 1 260 231 2 1 493Non-metallic mineral products 1 372 233 1 604 2 112 243 16 2 371Metasl and metal products 4 115 1 343 5 457 8 571 1 355 48 9 973Machinery and equipment 5 158 3 077 8 235 7 536 5 268 320 13 124Electrical and electronic equipment 3 877 1 011 4 888 6 639 5 206 9 11 854Precision instruments 880 - 880 18 83 22 123Motor vehicles and other transport equipment 3 728 317 4 045 9 952 1 939 1 11 891Other manufacturing 2 410 1 032 3 442 10 067 1 573 6 11 646Unspecified secondary 7 970 4 148 12 118 12 211 42 952 277 55 440

Services 82 694 12 027 94 721 353 428 103 402 4 198 461 028Electricity, gas and water 872 1 247 2 118 18 621 5 543 94 24 258Construction 527 700 1 227 3 047 2 028 118 5 192Trade 16 426 2 599 19 025 32 914 14 787 1 287 48 988Hotels and restaurants 3 782 945 4 727 1 433 1 251 18 2 703Transport, storage and communications 1 702 1 290 2 993 60 339 14 090 1 384 75 813Finance 33 841 2 553 36 393 92 600 20 923 583 114 105Business activities 11 591 1 565 13 155 98 293 34 072 a 673 133 038 a

Public administration and defence 2 435 - 2 436 2 590 - 5 2 595Education 7 5 12 - 4 38 7 41Health and social services 71 24 94 - 241 149 - 1 - 93Community, social and personal service activities 2 391 9 2 400 5 113 3 549 23 8 685Other services 8 191 672 8 863 32 697 3 919 2 36 618Unspecified tertiary 859 419 1 277 6 026 3 054 6 9 085

Private buying and selling of property 120 - 120 552 - - 552Unspecified 7 614 4 018 11 632 11 583 6 024 664 18 272

Source: UNCTAD.a A considerable share of investment in this industry is in Hong Kong (China), which accounted for 67% of inward

flows to developing economies and 17% of total inward flows. Hong Kong (China) data include investment holdingcompanies.

Note: Data should be interpreted with caution. The world total was extrapolated on the basis of data covering67 countries in 1989-1991 and 84 countries in 2001-2003, or the latest three-year period average available.They accounted for 89 and 78 per cent of of world inward FDI flows respectively in the periods 1989-1991and 2001-2003. Only countries for which data for the three main sectors were available were included. Thedistribution share of each industry of these countries was applied to estimate the world total in each sectorand industry. As a result, the sum of the sectors for each economic group is different from the totals shownin annex table B.1. Approval data was used for Israel (1994 instead of 1989-1991), Mongolia (1991-1993instead of 1989-1991) and Sri Lanka (2000-2002 instead of 2001-2003). In the case of Cambodia, China,Indonesia, Japan, Kenya, the Lao People’s Democratic Repbulic, Mongolia, Myanmar, Nepal, Papua NewGuinea, Solomon Islands, Sri Lanka, Taiwan Province of China, Turkey and Zimbabwe, the actual data wasestimated by applying the implementation ratio of realized FDI to approved FDI to the latter: Cambodia (9%in 1994-1995 and 92% in 2000-2002), China (47% in 1989-1991), Indonesia (15% in 1989-1991), Japan (21%in 1989-1991 and 41% in 2001-2003), Kenya (7% in 1992-1994), the Lao People’s Democratic Republic (1%in 1989-1991), Mongolia (45% in 2000-2002), Myanmar (21% in 1992-1994), Nepal (30% in 1989-199153%in 1996-1998), Papua New Guinea (20% in 1993-1995 and 36% in 1996-1998), Solomon Islands (1% in 1994-1995), Sri Lanka (47% in 1995), Taiwan Province of China (65% in 1989-1991 and in 2000-2002), Turkey(40% in 1989-1991) and Zimbabwe (23% in 1993-1995). The world total for 1989-1991 includes South-East Europe and the CIS, although data by sector and industry are not available for that region.

263ANNEX A

Annex table A.I.7. Estimated world outward FDI flows, by sector and industry, 1989-1991 and2001-2003

(Millions of dollars)

1989-1991 2001-2003

South-EastSector/industry Developed Developing Developed Developing Europe and

countries economies World countries economies CIS World

Primary 10 821 79 10 900 35 174 117 120 35 411Agriculture, hunting, forestry and fishing 530 42 572 268 4 - 272Mining, quarrying and petroleum 10 140 37 10 177 35 021 113 120 35 254Unspecified primary 151 - 151 - 116 - - - 116

Manufacturing 82 351 1 498 83 849 102 851 4 444 3 107 298Food, beverages and tobacco 13 326 136 13 461 15 598 39 4 15 641Textiles, clothing and leather 2 044 61 2 104 - 1 488 46 - - 1 442Wood and wood products 5 343 40 5 383 6 576 5 - 6 581Publishing, printing and reproduction of recorded media 156 - 156 789 - - 789Coke, petroleum products and nuclear fuel 122 - 122 3 055 - - 3 056Chemicals and chemical products 13 150 212 13 362 16 389 108 - 16 498Rubber and plastic products 587 35 621 1 653 14 - 1 667Non-metallic mineral products 1 195 70 1 265 725 5 - 730Metals and metal products 6 328 168 6 495 12 330 21 - 12 351Machinery and equipment 7 475 7 7 483 3 337 8 - 3 345Electrical and electronic equipment 10 419 305 10 725 8 078 765 - 8 843Precision instruments 655 - 655 2 430 20 - 2 449Motor vehicles and other transport equipment 5 712 - 5 712 12 617 65 - 1 12 681Other manufacturing 8 602 5 8 607 1 478 14 - 1 492Unspecified secondary 7 236 460 7 696 19 283 3 334 - 22 618

Services 117 209 1 020 118 229 463 975 26 778 14 490 767Electricity, gas and water 1 015 - 1 015 14 388 51 - 14 439Construction 2 445 31 2 476 2 096 169 - 2 265Trade 15 594 270 15 864 48 681 3 660 2 52 343Hotels and restaurants 416 4 420 4 636 - 322 - 4 315Transport, storage and communications 7 689 33 7 722 83 378 632 2 84 011Finance 49 567 446 50 013 151 620 2 893 4 154 517Business activities 26 642 19 26 661 143 497 18 128 6 161 630Public administration and defence - - - 475 - - 475Education 20 - 20 142 - - 142Health and social services - 124 - - 124 74 - - 74Community, social and personalservice activit ies 568 - 568 1 510 1 - 1 511Other services 8 873 217 9 090 8 722 1 354 - 10 077Unspecified tertiary 4 505 - 4 505 4 756 212 - 4 968

Private buying and selling of property 576 - 576 2 067 - - 2 067Unspecified 9 673 90 9 763 27 916 2 338 2 30 256

Source: UNCTAD.

Note: Data should be interpreted with caution. The world total was extrapolated on the basis of data covering27 countries in 1989-1991 and 38 countries in 2001-2003, or the latest three-year period average available.They accounted for 94 and 79 per cent of of world outward FDI flows respectively in the periods 1989-1991and 2001-2003. Only countries for which data for the three main sectors were available were included. Thedistribution share of each industry of these countries was applied to estimate the world total in each sectorand industry. As a result, the sum of the sectors for each economic group is different from the totals shownin annex table B.1. Approval data was used for Taiwan Province of China. In the case of India and Japan,the actual data was estimated by applying the implementation ratio of realized FDI to approved FDI to thelatter: India (57% in 2001-2003) and Japan (75% in 1989-1991 and 95% in 2001-2003). The world totalin 1989-1991 includes South-East Europe and the CIS, although data by sector and industry are not availablefor that region.

264 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.I.8. Number of parent corporations and foreign affiliates, by region andeconomy, latest available year

(Number)

Parent Foreign Parent Foreigncorporations affiliates corporations affiliates

based in located in Region/economy based in located inRegion/economy Year economya economya Year economya economya

Developed economies 50 520b 247 241 b Guinea 2004 .. 31Guinea-Bissau 2004 .. 5

Europe 41 461b 209 788 b Liberia 2004 .. 18Mali 2004 1 21

European Union 36 003b 199 303 b Mauritania 2004 2 p 4Austria 2003 969 2 679 c Niger 2004 1 p 4Belgium 2003 991d 2 341 d Nigeria 2004 .. 124Cyprus 2004 4 500 2 000 Senegal 2004 6 p 63Czech Republic 1999 660e 71 385 f Sierra Leone 2004 1 p 7Denmark 1998 9 356 2 305 g Togo 2004 3 p 14Estonia 2003 351 2 858Finland 2001 900h 2 030 c, g Central Africa 4 b 274 b

France 2002 1 267 10 713 Angola 2004 .. 68Germany 2003 6 010 9 314 Cameroon 2004 .. 92Greece 2003 170 750 Central African Republic 2004 .. 1Hungary 2003 .. 26 793 i Chad 2004 .. 9Ireland 2001 39 j 1 225 k Congo 2004 .. 45Italy 1999 1 017 l 1 843 l Congo, Democratic Republic of 2004 4 p 1Luxembourg 2002 41m 773 m Equatorial Guinea 2004 .. 7Latvia 2004 7 450 Gabon 2004 .. 51Lithuania 2003 150 2 652Malta 2004 23 132Netherlands 2004 1 608n 13 365 East and Southern Africa 148 b 1 711 b

Poland 2001 58 j 14 469 o East Africa 51 b 792 b

Portugal 2002 600p 3 000 Burundi 2004 .. ..Slovakia 2004 .. 2 128 Comoros 2004 .. 1Slovenia 2000 .. 1 617 q Djibouti 2004 1 p 4Spain 2004 857 r 6 340 Ethiopia 2004 4 p 21Sweden 2002 4 260s 4 656 c Kenya 2004 8 170United Kingdom 2004 2 169 13 485 Madagascar 2004 .. 49

Malawi 2004 .. 16Other developed Europe 5 458b 10 485 b Mauritius 2004 4 71

Gibraltar 2004 34 98 Mozambique 2004 5 p 68Iceland 2000 18 55 Rwanda 2004 2 13Norway 1998 900 5 105 t Seychelles 2004 .. 24Switzerland 2004 4 506u 5 227 Somalia 2001 1 ..

Uganda 2003 .. 255 North America 3 857b 28 332 b United Republic of Tanzania 2004 15 p 61

Canada 1999 1 439 3 725 c Zambia 2004 11 13United States 2002 2 418 24 607 Zimbabwe 2004 .. 26

Other developed countries 5 202b 9 121 b Southern Africa 97 b 919 b

Botswana 2004 .. 6Australia 2001 682 2 352 Lesotho 2004 .. 1Israel 2004 154 37 Namibia 2004 .. 6Japan 2003 4 149v 4 710 w South Africa 2004 85 845New Zealand 2004 217e 2 022 Swaziland 2002 12 61

Latin America and the Caribbean 2 914 b 35 617 b

Developing economies 18 029b 335 338 b

South and Central America 2 301 b 33 811 b

Africa 324b 5 846 b

South America 2 077 b 6 654 b

North Africa 157b 3 286 b Argentina 2004 42 1 383Algeria 2004 .. 54 Bolivia 2003 .. 364Egypt 2004 10 271 Brazil 2004 1 225 2 820Morocco 2004 3 295 Chile 2004 478 x 575Sudan 2004 2p 7 Colombia 2004 302 u 427Tunisia 2004 142h 2 659 Ecuador 2004 2 218

Guyana 2002 4 h 56Other Africa 167b 2 560 b Paraguay 2004 .. 38West Africa 15b 575 b Peru 2004 10 e,y 329

Benin 2004 .. 19 Suriname 2004 1 11Burkina Faso 2004 .. .. Uruguay 2002 .. 164 z

Côte d’Ivoire 2004 .. 174 Venezuela 2004 13 545Gambia 2004 .. 13Ghana 2004 1 78 Central America 224 b 26 881 b

/...

265ANNEX A

Annex table A.I.8. Number of parent corporations and foreign affiliates, by region andeconomy, latest available year (continued)

(Number)

Parent Foreign Parent Foreigncorporations affiliates corporations affiliates

based in located in Region/economy based in located inRegion/economy Year economya economya Year economya economya

Belize 2004 4 13 South Asia 1 764 b 3 237 b

Costa Rica 2004 8 173 Afghanistan 2004 1 2El Salvador 2003 .. 304 Bangladesh 2004 2 28Guatemala 2004 .. 151 Bhutan 1997 .. 2Honduras 2004 1 69 India 2004 1 700 ag 1 181Mexico 2002 .. 25 708 Maldives 2004 1 3Nicaragua 2004 .. 51 Nepal 2004 1 p 9Panama 2003 211 412 Pakistan 2001 59 ah 582

Sri Lanka 2004 .. 1 430The Caribbean and other America 613 b 1 806 b

Antigua and Barbuda 2004 .. 13 South-East Asia 314 b 33 892 b

Aruba 2004 .. 32 Brunei Darussalam 2004 1 34Bahamas 2004 44 158 Cambodia 2002 .. 23 ai

Barbados 2004 11 146 Indonesia 2004 313 aj 721Bermuda 2004 362 348 Lao People’s Democratic Republic 2004 .. 161 ak

British Virgin Islands 2002 .. 129 Malaysia 1999 .. 15 567 al

Cayman Islands 2004 85 470 Myanmar 2004 .. 15Dominica 2004 .. 2 Philippines 2004 .. 411Dominican Republic 2004 2 147 Singapore 2002 .. 14 052 am

Grenada 2004 .. 13 Thailand 1998 .. 2 721Haiti 2004 1 13 Viet Nam 2004 .. 187Jamaica 2004 .. 74Netherlands Antilles 2004 101 159 Oceania 22 b 448 b

Saint Kitts and Nevis 2004 5 10 Fiji 2002 2 151 e

Saint Lucia 2004 1 20 Kiribati 2004 .. 1Saint Vincent and the Grenadines 2004 1 11 New Caledonia 2004 .. 3Trinidad and Tobago 2004 .. 61 Papua New Guinea 2004 .. 208

Samoa 2004 7 p 43Asia and Oceania 14 791 b 293 875 b Solomon Islands 2004 7 p 18

Tonga 2004 .. 5Asia 14 769 b 293 427 b Vanuatu 2002 6 19 an

West Asia 1 642 b 10 988 b South-East Europe and the CIS 1 178 b 107 812 b

Bahrain 2004 13 79 South-East Europe 124 b 97 407 b

Iran, Islamic Republic of 2004 13 42 Albania 2004 .. 14Jordan 2004 2 26 Bosnia and Herzegovina 2004 1 51Kuwait 2004 15 35 Bulgaria 2000 26 j 7 153 ao

Lebanon 2004 9 64 Croatia 2004 70 e 191Oman 2004 92 aa 49 Macedonia, TFYR 2002 .. 6Qatar 2003 5 30 Romania 2002 20 j 89 911 ap

Saudi Arabia 2004 .. 167 Serbia and Montenegro 2004 7 81Syrian Arab Republic 2004 .. 11Turkey 2004 1 474 9 616 CIS 1 054 b 10 405 b

United Arab Emirates 2004 13 865 Armenia 2004 .. 347Yemen 2002 6 p 4 Azerbaijan 2004 1 30

Belarus 2004 .. 25South, East and South-East Asia 13 127 b 282 439 b Georgia 1998 .. 190 aq

East Asia 11 049 b 245 310 b Kazakhstan 2004 101 1 575China 2003 2 000 ab 215 000 ac Kyrgyzstan 1998 .. 4 004 ar

Hong Kong, China 2003 948 ad 9 072 Moldova, Republic of 2002 951 2 670Korea, Republic of 2004 7 460 ae 16 181 Russian Federation 2004 .. 1 176Macao, China 2003 35 723 Ukraine 2004 1 367Mongolia 1998 .. 1 400 Uzbekistan 2004 .. 21Taiwan Province of China 2004 606 af 2 934

World 69 727 690 391

Source: UNCTAD, based on national sources.a The number of parent companies/foreign affi l iates in the economy shown, as defined by that economy. Deviations from the definition adopted in

the World Investment Report (see section on “definitions and sources” in annex B) are noted below. The data for Afghanistan, Albania, Algeria,Angola, Antigua and Barbuda, Argentina, Aruba, Azerbaijan, Bahamas, Bahrain, Bangladesh, Barbados, Belarus, Belize, Benin, Bermuda, Bosniaand Herzegovina, Botswana, Brazil, British Virgin Islands, Brunei Darussalam, Burkina Faso, Burundi, Cameroon, Cayman Islands, Central AfricanRepublic, Chad, Chile, Colombia, Congo, Costa Rica, Cô te d’Ivoire, Democratic Republic of the Congo, Djibouti, Dominica, Dominican Republic,Ecuador, Equatorial Guinea, Ethiopia, Gabon, Gambia, Ghana, Gibraltar, Grenada, Guatemala, Guinea-Bissau, Haiti, Honduras, India, Islamic Republic

266 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

of Iran, Israel, Jamaica, Jordan, Kenya, Kiribati, Kuwait, Latvia, Lebanon, Lesotho, Liberia, Madagascar, Malawi, Maldives, Mali, Malta, Mauritania,Mauritius, Morocco, Mozambique, Myanmar, Namibia, Nepal, the Netherlands, Netherlands Antilles, New Caledonia, New Zealand, Nicaragua, Niger,Nigeria, Panama, Qatar, Paraguay, the Philippines, Saint Lucia, Saint Kitts and Nevis, Saint Vincent and the Grenadines, Samoa, Saudi Arabia,Serbia and Montenegro, Senegal, Seychelles, Sierra Leone, Slovakia, Solomon Islands, Somalia, South Africa, Spain, Sudan, Suriname, Switzerland,Syrian Arab Republic, Togo, Tonga, Uganda, United Arab Emirates, United Republic of Tanzania, Uzbekistan, Vanuatu, Viet Nam, Western Samoaand Zimbabwe are from Who Owns Whom 2004 (London, Dun & Bradstreet). For Argentina, Bermuda, Israel and South Africa, the number of parentcorporations based in the economy refers to only those that have affi l iates abroad and affi l iates in the home economy. Therefore, the number ofparent corporations is underestimated in those four countries.

b Includes data only for the countries shown below.c Majority-owned foreign affi l iates.d Provisional figures by Banque Nationale de Belgique (2003).e As of 1997.f Of this number, 53,775 are fully-owned foreign affi l iates; includes joint ventures.g Directly and indirectly owned foreign affi l iates (subsidiaries and associates), excluding branches.h As of 1999.i Source: Hungary Statistics Office.j As of 1994.k Refers to the number of foreign-owned affi l iates in Ireland in manufacturing and services activit ies that receive assistance from the Investment

and Development Authority (IDA).l Relates to parent companies’ and foreign affiliates’ industrial activities (based on Consiglio Nazionale dell’Economia e del Lavoro, “Italia Multinazionale,

2000, inward and outward FDI in Italian industry in 1998 and 1999” April 2002.m Excludes special purpose entit ies (SPEs), i.e. holding companies.n Data refers to October 1993.o Cumulative number of companies with a foreign capital share that participated in the statistical survey.p As of 2001.q Source: Bank of Slovenia.r Data refer to 1998: includes those Spanish parent enterprises which are controlled, at the same time, by a direct investor.s Data provided by Sveriges Riksbank; includes those Swedish parent companies that are controlled, at the same time, by a direct investor.t Data refers to Norwegian non-financial joint-stock companies with foreign shareholders owning more than 10 per cent of the total shares in 1998.u As of 1995.v Data refer to Japanese companies that had overseas affiliates as of fiscal year ending in March 2004, except for financial, insurance and real estates

industries (source: Ministry of Economy, Trade and Industry, Survey of Overseas Business Activities) where Japanese firms had at least two foreignaffi l iates with a more than 20% equity share as of November 2003 (source: Toyokeizai, Kaigai Shinshutsu Kigyo Soran 2004, Tokyo: ToyokeizaiShimposha, 2004).

w Data refer to the number of foreign affiliates in which foreign investors hold more than one-third of the stocks or shares, except for financial, insuranceand real estate industries as at the end of March 2003 (source: Ministry of Economy, Trade and Industry, Survey of Trends in Business Activitiesof Foreign Affiliates) and the number of foreign affiliates in financial, insurance and real estate industries as of December 2002 (source: Toyokeizai,Gaishikei Kigyo Soran 2003, Tokyo: Toyokeizai Shimposha, 2003).

x Estimated by Comité de Inversiones Extranjeras 1998.y Less than 10.z Number of enterprises included in the Central Bank survey (all sectors).aa As of May 1995.ab Estimated by UNCTAD.ac Currently existing registered foreign-invested enterprises, which include: (i) equity joint ventures (foreign equity>25%), (ii) contractual joint ventures

(no equity arrangements), and (ii i) wholly foreign-owned enterprises (100% foreign ownership).ad Number of regional headquarters as at 1 June 2002.ae As of 1999. Data refer to the number of investment projects abroad.af Number of approved new investment projects abroad in 1998.ag Data refer to the number of approved FDI projects as of 2003.ah As of 1998.ai Data refer to the number of approved foreign investment projects, including joint-venture projects with local investors. Wholly owned Cambodian

projects are excluded.aj As of 1996.ak Number of projects l icensed since 1988 up to end 2004.al May 1999. Refers to companies with foreign equity stakes of 51 per cent and above. Of this, 3,787 are fully owned foreign affi l iates.am Number of wholly owned foreign companies.an Data refer to the number of projects implemented.ao The number refers to registered investment projects between 1992 and 2000, data from Bulgarian Foreign Investment Agency.ap Data refer to the cumulative number of companies with FDI as at end December 2002.aq Number of cases of approved investments of more than 100,000 dollars registered during the period January 1996 up to March 1998.ar Joint-venture companies established in the economy.

Note: The data can vary significantly from preceding years, as data become available for countries that were notbeen covered before, as definitions change, or as older data are updated.

267ANNEX A

An

ne

x t

ab

le A

.I.9

. T

he

wo

rld

’s t

op

10

0 n

on

-fin

an

cia

l T

NC

s,

ran

ke

d b

y f

ore

ign

as

se

ts,

20

03

a

(Mill

ion

s o

f d

olla

rs a

nd

nu

mb

er

of

em

plo

yee

s)

R

anki

ng b

y:

Ass

ets

S

ales

E

mpl

oym

ent

N

o. o

f af

filia

tes

Fore

ign

TNIb

asse

tsTN

IbIIc

Cor

pora

tion

Hom

e ec

onom

yIn

dust

ryd

Fore

ign

Tota

lFo

reig

neTo

tal

Fore

ign

Tota

l(P

er c

ent)

Fo

reig

n

Tota

lIIc

177

37G

ener

al E

lect

ricU

nite

d S

tate

sE

lect

rical

& e

lect

roni

c eq

uipm

ent

258

900

647

483

54 0

8613

4 18

715

0 00

030

5 00

043

.21

068

1 39

876

.39

27

95Vo

dafo

ne G

roup

Plc

Uni

ted

Kin

gdom

Tele

com

mun

icat

ions

243

839

f26

2 58

150

070

59 8

9347

473

60 1

0985

.171

201

35.3

23

7212

Ford

Mot

or C

ompa

nyU

nite

d S

tate

sM

otor

veh

icle

s17

3 88

2f

304

594

60 7

6116

4 19

613

8 66

3g

327

531

45.5

524

623

84.1

14

9065

Gen

eral

Mot

ors

Uni

ted

Sta

tes

Mot

or v

ehic

les

15 4

466

f44

8 50

751

627

185

524

104

000

294

000

32.5

177

297

59.6

05

1078

Brit

ish

Pet

role

um C

ompa

ny P

lcU

nite

d K

ingd

omP

etro

leum

exp

l./re

f./di

str.

141

551

177

572

192

875

232

571

86 6

5010

3 70

082

.160

117

51.2

86

3141

Exx

onM

obil

Cor

pora

tion

Uni

ted

Sta

tes

Pet

role

um e

xpl./

ref./

dist

r.11

6 85

3f

174

278

166

926

237

054

53 7

48g

88 3

0066

.121

829

474

.15

722

80R

oyal

Dut

ch/S

hell

Gro

upU

nite

d K

ingd

om/

Net

herla

nds

Pet

role

um e

xpl./

ref./

dist

r.11

2 58

7h

168

091

129

864

201

728

100

000

g11

9 00

071

.845

492

948

.87

868

94To

yota

Mot

or C

orpo

ratio

nJa

pan

Mot

or v

ehic

les

94 1

64f

189

503

87 3

5314

9 17

989

314

264

410

47.3

124

330

37.5

89

1648

Tota

lFr

ance

Pet

role

um e

xpl./

ref./

dist

r.87

840

f10

0 98

994

710

118

117

60 9

3111

0 78

374

.141

960

269

.60

1062

69Fr

ance

Tel

ecom

Fran

ceTe

leco

mm

unic

atio

ns81

370

f12

6 08

321

574

52 2

0288

626

218

523

48.8

118

211

55.9

211

1458

Sue

zFr

ance

Ele

ctric

ity, g

as a

nd w

ater

74 1

47f

88 3

4333

715

44 7

2011

1 44

517

2 29

174

.760

594

763

.89

1289

34E

lect

ricite

De

Fran

ceFr

ance

Ele

ctric

ity, g

as a

nd w

ater

67 0

6918

5 52

716

062

50 6

9951

847

i16

7 30

932

.920

426

477

.27

1380

63E

.On

Ger

man

yE

lect

ricity

, gas

and

wat

er64

033

f14

1 26

018

659

52 3

3029

651

69 3

8341

.247

879

060

.51

1485

74D

euts

che

Tele

kom

AG

Ger

man

yTe

leco

mm

unic

atio

ns62

624

146

601

23 8

6863

023

75 2

41g

248

519

37.0

9717

854

.49

1559

67R

WE

Gro

upG

erm

any

Ele

ctric

ity, g

as a

nd w

ater

60 3

4598

592

23 7

2949

061

53 5

54g

127

028

50.6

377

650

58.0

016

2323

Hut

chis

on W

ham

poa

Lim

ited

Hon

g K

ong,

Chi

naD

iver

sifie

d59

141

80 3

4010

800

18 6

9910

4 52

912

6 25

071

.41

900

2 35

080

.85

1732

40S

iem

ens

AG

Ger

man

yE

lect

rical

& e

lect

roni

c eq

uipm

ent

58 4

63f

98 0

1164

484

83 7

8424

7 00

041

7 00

065

.375

310

1174

.48

1853

46Vo

lksw

agen

Gro

upG

erm

any

Mot

or v

ehic

les

57 8

53f

150

462

71 1

9098

367

160

299

334

873

52.9

203

283

71.7

319

2135

Hon

da M

otor

Co

Ltd

Japa

nM

otor

veh

icle

s53

113

f77

766

54 1

9970

408

93 0

06g

131

600

72.0

102

133

76.6

920

3489

Viv

endi

Uni

vers

alFr

ance

Div

ersi

fied

52 4

21f

69 3

6015

764

28 7

6132

348

j49

617

65.2

106

238

44.5

421

4283

Che

vron

Texa

co C

orp.

Uni

ted

Sta

tes

Pet

role

um e

xpl./

ref./

dist

r.50

806

81 4

7072

227

120

032

33 8

4361

533

59.2

9320

146

.27

223

30N

ews

Cor

pora

tion

Aus

tral

iaM

edia

50 8

0355

317

17 7

7219

086

35 6

04j

38 5

0092

.521

326

979

.18

2365

29P

fizer

Inc

Uni

ted

Sta

tes

Pha

rmac

eutic

als

48 9

60f

116

775

18 3

4445

188

73 2

00g

122

000

47.5

7392

79.3

524

9385

Tele

com

Ital

ia S

paIta

lyTe

leco

mm

unic

atio

ns46

047

101

172

6 81

634

819

14 9

1093

187

27.0

3373

45.2

125

5018

BM

W A

GG

erm

any

Mot

or v

ehic

les

44 9

4871

958

35 0

1447

000

26 0

8610

4 34

254

.012

915

782

.17

2660

53E

ni G

roup

Italy

Pet

role

um e

xpl./

ref./

dist

r.43

967

f85

042

29 3

4158

112

36 6

5876

521

50.0

154

226

68.1

427

49

Roc

he G

roup

(U

moe

AS

)S

witz

erla

ndP

harm

aceu

tical

s42

926

48 0

8922

790

23 1

8357

317

g65

357

91.8

139

159

87.4

228

9579

Dai

mle

rChr

ysle

r AG

Ger

man

y/U

nite

d S

tate

sM

otor

veh

icle

s41

696

f22

5 14

355

195

153

992

76 9

9336

2 06

325

.228

155

850

.36

2944

32Fi

at S

paIta

lyM

otor

veh

icle

s41

552

79 1

6036

078

53 3

5388

684

162

237

58.3

339

436

77.7

530

153

Nes

tlé S

AS

witz

erla

ndFo

od &

bev

erag

es41

078

f72

402

44 3

0865

329

247

506

253

000

74.1

471

501

94.0

131

555

IBM

Uni

ted

Sta

tes

Ele

ctric

al &

ele

ctro

nic

equi

pmen

t40

987

f10

445

755

369

89 1

3118

0 51

5g

319

273

52.6

315

342

92.1

132

8347

Con

ocoP

hilli

psU

nite

d S

tate

sP

etro

leum

exp

l./re

f./di

str.

36 5

10f

82 4

0229

428

90 4

9114

982

j39

000

38.4

103

147

70.0

733

4631

Son

y C

orpo

ratio

nJa

pan

Ele

ctric

al &

ele

ctro

nic

equi

pmen

t35

257

f84

880

44 3

6664

661

96 4

0016

2 00

056

.623

629

978

.93

3458

71C

arre

four

SA

Fran

ceR

etai

l34

323

f49

335

39 3

6879

780

138

283

419

040

50.6

128

231

55.4

135

9624

Wal

-Mar

t Sto

res

Uni

ted

Sta

tes

Ret

ail

34 0

1810

4 91

247

572

256

329

361

765

1500

000

25.0

6783

80.7

236

6954

Tele

foni

ca S

AS

pain

Tele

com

mun

icat

ions

33 4

66k

66 8

2510

508

32 0

5485

765

g14

8 28

846

.913

319

966

.83

3736

70Ve

olia

Env

ironn

emen

t SA

Fran

ceW

ater

Sup

ply

33 3

99f

49 1

5417

578

32 2

8320

5 69

430

9 56

362

.942

476

055

.79

3843

21P

roct

er &

Gam

ble

Uni

ted

Sta

tes

Div

ersi

fied

33 3

6157

048

27 7

1951

407

68 6

94g

110

000

58.3

174

215

80.9

39 /...

268 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.I.9

. T

he

wo

rld

’s t

op

10

0 n

on

-fin

an

cia

l T

NC

s,

ran

ke

d b

y f

ore

ign

as

se

ts,

20

03

a (

co

nti

nu

ed

)(M

illio

ns

of

do

llars

an

d n

um

be

r o

f e

mp

loye

es)

R

anki

ng b

y:

Ass

ets

S

ales

E

mpl

oym

ent

N

o. o

f af

filia

tes

Fore

ign

TNIb

asse

tsTN

IbIIc

Cor

pora

tion

Hom

e ec

onom

yIn

dust

ryd

Fore

ign

Tota

lFo

reig

neTo

tal

Fore

ign

Tota

l

(Per

cen

t)

Fore

ign

T

otal

IIc

3941

10S

anof

i-Ave

ntis

Fran

ceP

harm

aceu

tical

s33

024

f44

484

12 2

9122

247

36 5

7675

567

59.3

335

385

87.0

140

5716

Hew

lett-

Pac

kard

Uni

ted

Sta

tes

Ele

ctric

al &

ele

ctro

nic

equi

pmen

t32

144

f74

708

43 8

4373

061

73 1

58j

142

000

51.5

179

215

83.2

641

9277

Mits

ubis

hi C

orpo

ratio

nJa

pan

Who

lesa

le tr

ade

31 2

58f

78 3

4220

054

130

912

14 7

65j

49 2

1928

.417

031

454

.14

4287

13D

euts

che

Pos

t AG

Ger

man

yTr

ansp

ort a

nd s

tora

ge29

524

195

748

19 7

1445

166

175

775

383

173

34.9

341

406

83.9

4335

62U

nile

ver

Uni

ted

Kin

gdom

/ N

ethe

rland

sD

iver

sifie

d28

654

h47

952

27 6

3548

186

179

000

234

000

64.5

316

522

60.5

444

649

Phi

lips

Ele

ctro

nics

Net

herla

nds

Ele

ctric

al &

ele

ctro

nic

equi

pmen

t28

524

36 6

2631

594

32 7

7313

6 75

016

4 43

885

.826

337

869

.58

4563

91N

issa

n M

otor

Co

Ltd

Japa

nM

otor

veh

icle

s28

517

73 3

8842

002

64 0

8250

836

l12

3 74

848

.558

146

39.7

346

1127

Lafa

rge

SA

Fran

ceN

on-m

etal

lic p

rodu

cts

28 1

2731

365

13 1

1715

415

50 5

2475

338

80.6

389

489

79.5

547

6656

Rep

sol Y

PF

SA

Spa

inP

etro

leum

exp

l./re

f./di

str.

27 9

3348

034

14 5

1540

710

14 9

2430

644

47.5

8112

465

.32

4848

28B

AS

F A

GG

erm

any

Che

mic

als

27 0

9942

437

21 9

9937

653

37 0

5487

159

54.9

206

259

79.5

449

2533

Com

pagn

ie D

e S

aint

-Gob

ain

SA

Fran

ceN

on-m

etal

lic m

iner

al p

rodu

cts

27 0

5638

008

23 8

3433

967

122

696

172

811

70.8

612

791

77.3

750

456

Nov

artis

Sw

itzer

land

Pha

rmac

eutic

als

26 7

48f

49 3

1716

076

24 8

6441

031

78 5

4157

.023

225

690

.63

5184

75M

itsui

& C

o Lt

dJa

pan

Who

lesa

le tr

ade

26 2

62f

62 7

0947

508

105

936

10 8

26j

39 7

3538

.019

836

454

.40

5286

14A

ltria

Gro

up In

cU

nite

d S

tate

sTo

bacc

o25

711

f96

175

34 3

7160

704

40 5

57m

165

000

36.0

196

234

83.7

653

7881

End

esa

Spa

inE

lect

ric s

ervi

ces

25 4

8858

155

6 22

818

328

12 9

3926

777

42.0

8517

748

.02

548

4A

lcan

Inc.

Can

ada

Met

al a

nd m

etal

pro

duct

s25

275

f31

957

13 1

7213

640

38 0

0049

000

84.4

306

328

93.2

955

2690

BH

P B

illito

n G

roup

Aus

tral

iaM

inin

g &

qua

rryi

ng24

254

36 6

7517

673

24 9

4325

294

35 0

7069

.742

9842

.86

5628

50G

laxo

smith

klin

e P

lcU

nite

d K

ingd

omP

harm

aceu

tical

s23

893

42 8

1332

296

35 0

0656

360

h10

0 91

968

.015

822

969

.00

5782

55R

enau

lt S

AFr

ance

Mot

or v

ehic

les

22 3

42f

71 2

8327

330

42 3

5334

921

g13

0 74

040

.913

620

765

.70

5837

93A

nglo

Am

eric

anU

nite

d K

ingd

omM

inin

g &

qua

rryi

ng21

623

43 1

0510

872

18 5

6215

1 00

019

3 00

062

.319

752

437

.60

5919

87K

onin

klijk

e A

hold

Net

herla

nds

Ret

ail

20 8

8429

552

47 7

4463

282

189

945

h25

7 14

073

.374

164

45.1

260

2064

AE

S C

orpo

ratio

nU

nite

d S

tate

sE

lect

ricity

, gas

and

wat

er20

871

f29

904

6 25

78

415

216

22j

30 0

0072

.156

9360

.22

6154

22D

ow C

hem

ical

Com

pany

Uni

ted

Sta

tes

Che

mic

als

20 0

3941

891

19 8

1032

623

22 9

64g

46 4

0052

.721

626

780

.90

6218

43Vo

lvo

AB

Sw

eden

Mot

or v

ehic

les

19 4

5131

787

23 1

6024

023

47 6

0375

740

73.5

233

319

73.0

463

4757

Pin

ault-

Prin

tem

ps R

edou

te S

AFr

ance

Who

lesa

le tr

ade

19 2

54f

30 6

4916

828

30 7

6751

847

100

779

56.3

287

449

63.9

264

7442

Bay

er A

GG

erm

any

Pha

rmac

eutic

als/

chem

ical

s18

892

47 0

2017

033

32 3

3448

700

115

400

45.0

236

320

73.7

565

11

Thom

son

Cor

pora

tion

Can

ada

Med

ia18

418

18 7

327

943

8 15

938

350

39 0

0098

.030

031

096

.77

6633

36S

ingt

el L

td.

Sin

gapo

reTe

leco

mm

unic

atio

ns17

911

21 6

684

672

6 88

48

642

j19

081

65.3

2330

76.6

767

3052

Brit

ish

Am

eric

an T

obac

co P

lcU

nite

d K

ingd

omTo

bacc

o17

871

h33

891

27 9

72n

41 8

3268

702

86 9

4166

.224

836

368

.32

6881

99N

atio

nal G

rid T

rans

coU

nite

d K

ingd

omE

nerg

y17

563

41 7

807

673

15 8

489

029

27 3

0841

.242

244

17.2

169

242

Nok

iaFi

nlan

dTe

leco

mm

unic

atio

ns17

050

f29

273

36 7

6337

202

28 9

7951

359

71.2

9810

395

.15

7099

84H

itach

i Ltd

Japa

nE

lect

rical

& e

lect

roni

c eq

uipm

ent

16 2

9689

545

21 1

7780

602

80 2

2632

6 34

423

.030

967

046

.12

7149

20U

nite

d Te

chno

logi

es C

orpo

ratio

nU

nite

d S

tate

sTr

ansp

ort e

quip

men

t16

212

34 6

4814

257

31 0

3414

3 00

020

3 30

054

.434

542

281

.75

7294

68P

etro

nas

- P

etro

liam

Nas

iona

l Bhd

Mal

aysi

aP

etro

leum

exp

l./re

f./di

str.

16 1

14i

53 4

578

981

25 6

613

625

30 6

3425

.723

441

156

.93

7338

92M

cDon

ald’

s C

orpo

ratio

nU

nite

d S

tate

sR

etai

l15

913

25 5

2511

101

17 1

4024

0 14

241

8 00

061

.526

6738

.81

7427

25S

tora

Ens

o O

YJ

Finl

and

Pap

er15

910

22 6

4610

382

15 3

7329

156

42 8

1468

.622

928

580

.35

7561

44D

u P

ont (

E.I.

) de

Nem

ours

Uni

ted

Sta

tes

Che

mic

als

15 8

40f

37 0

3914

888

o26

996

39 6

57j

81 0

0049

.011

515

872

.78

7612

82R

io T

into

Plc

Uni

ted

Kin

gdom

Min

ing

& q

uarr

ying

15 4

1924

015

9 77

310

009

26 0

0036

016

78.0

6814

546

.90

7798

86D

uke

Ene

rgy

Cor

pora

tion

Uni

ted

Sta

tes

Ele

ctric

ity, g

as a

nd w

ater

15 4

14f

56 2

035

537

22 5

294

652

j23

800

23.8

3373

45.2

1 /...

269ANNEX A

An

ne

x t

ab

le A

.I.9

. T

he

wo

rld

’s t

op

10

0 n

on

-fin

an

cia

l T

NC

s,

ran

ke

d b

y f

ore

ign

as

se

ts,

20

03

a (

co

nc

lud

ed

)(M

illio

ns

of

do

llars

an

d n

um

be

r o

f e

mp

loye

es)

Ran

king

by:

A

sset

s

S

ales

Em

ploy

men

t

No.

of

affil

iate

sFo

reig

nTN

Ib

asse

tsTN

IbIIc

Cor

pora

tion

Hom

e ec

onom

yIn

dust

ryd

Fore

ign

Tota

lFo

reig

neTo

tal

Fore

ign

Tota

l (P

er c

ent)

F

orei

gn

Tot

al

II

c

7840

38Lv

mh

Moë

t-H

enne

ssy

Loui

s Vu

itton

SA

Fran

ceTe

xtile

s an

d le

athe

r15

386

p24

356

8 28

515

063

35 3

6056

241

60.4

296

390

75.9

079

7366

Thys

senk

rupp

AG

Ger

man

yM

etal

s an

d m

etal

pro

duct

s15

237

f36

641

20 0

7445

641

92 1

79j

184

358

45.2

389

668

58.2

380

6715

Abb

ott L

abor

ator

ies

Uni

ted

Sta

tes

Pha

rmac

eutic

als

15 2

1426

715

7 70

319

681

33 1

66j

72 1

8147

.310

312

383

.74

8170

76M

atsu

shita

Ele

ctric

Indu

stria

l Co.

, Ltd

.Ja

pan

Ele

ctric

al &

ele

ctro

nic

equi

pmen

t14

739

f69

449

42 0

2569

839

170

965

290

493

46.8

274

504

54.3

782

100

100

Veriz

on C

omm

unic

atio

nsU

nite

d S

tate

sTe

leco

mm

unic

atio

ns13

831

f16

5 96

82

449

67 7

5217

269

j20

3 10

06.

813

219

5.94

8376

88M

etro

AG

Ger

man

yR

etai

l13

600

f33

571

32 1

0467

690

107

210

242

010

44.1

245

549

44.6

384

2939

Nor

sk H

ydro

Asa

Nor

way

Div

ersi

fied

13 4

2932

729

23 1

5825

716

30 8

6644

602

66.8

254

335

75.8

285

5297

Chr

istia

n D

ior

SA

Fran

ceTe

xtile

s13

388

31 8

958

461

15 7

4536

391

56 8

1553

.310

3528

.57

862

8C

RH

Plc

Irel

and

Lum

ber &

oth

er b

uild

ing

mat

eria

l dea

lers

13 1

84f

13 9

7613

070

13 6

0851

694

54 2

3995

.242

147

688

.45

8764

61S

cotti

sh P

ower

Uni

ted

Kin

gdom

Ele

ctric

util

ities

12 9

9124

665

4 75

310

352

6 66

314

339

48.4

7111

561

.74

8871

72A

lcoa

Uni

ted

Sta

tes

Met

als

and

met

al p

rodu

cts

12 9

31f

31 7

118

319

21 5

0470

700

120

000

46.1

104

188

55.3

289

911

Pub

licis

Gro

upe

SA

Fran

ceB

usin

ess

serv

ices

12 9

19f

13 4

004

367

4 87

921

451

35 1

6682

.329

534

286

.26

9097

73M

arub

eni C

orpo

ratio

nJa

pan

Who

lesa

le tr

ade

12 8

14f

39 7

2225

175

73 8

151

723

j24

417

24.5

161

293

54.9

591

1360

Hol

cim

AG

Sw

itzer

land

Non

-met

allic

min

eral

pro

duct

s12

808

h20

091

6 59

610

187

46 9

4648

200

75.3

105

170

61.7

692

517

Cad

bury

Sch

wep

pes

Plc

Uni

ted

Kin

gdom

Food

& b

ever

ages

12 8

0414

209

8 86

210

525

48 3

9055

799

87.0

9411

383

.19

9379

26W

yeth

Uni

ted

Sta

tes

Pha

rmac

eutic

als

12 7

7629

727

6 26

915

851

21 6

17j

52 3

8541

.367

8479

.76

9488

96S

tato

il A

saN

orw

ayP

etro

leum

exp

l./re

f./di

str.

12 7

2133

174

9 68

437

239

7 49

119

326

34.4

3511

231

.25

9517

98B

AE

Sys

tem

s P

lcU

nite

d K

ingd

omTr

ansp

ort e

quip

men

t12

695

16 8

0217

530

22 4

5048

900

72 3

0073

.857

248

22.9

896

5619

Rob

ert B

osch

Gm

bHG

erm

any

Mac

hine

ry a

nd e

quip

men

t12

683

q40

410

32 7

61q

45 9

1912

3 00

023

2 00

051

.921

025

682

.03

9751

45M

otor

ola

Inc

Uni

ted

Sta

tes

Tele

com

mun

icat

ions

12 6

1832

098

17 9

8327

058

48 4

00j

88 0

0053

.679

109

72.4

898

3951

Ber

tels

man

nG

erm

any

Med

ia12

498

f25

466

14 6

9421

219

46 1

5773

221

60.5

320

468

68.3

899

757

Sam

sung

Ele

ctro

nics

Co

Ltd

Rep

ublic

of K

orea

Ele

ctric

al &

ele

ctro

nic

equi

pmen

t12

387

p56

524

41 3

6254

349

19 0

26j

55 3

9744

.180

8989

.89

100

9159

Inte

rnat

iona

l Pap

er C

ompa

nyU

nite

d S

tate

sP

aper

12 0

8835

525

6 99

225

179

28 9

8082

800

32.3

9314

862

.84

So

urc

e:

UN

CT

AD

/Era

smu

s U

niv

ers

ity

da

tab

ase

.a

All

data

are

bas

ed o

n th

e co

mpa

nies

' ann

ual r

epor

ts u

nles

s ot

herw

ise

stat

ed.

b TN

I, t

he T

rans

natio

nlity

Ind

ex,

is c

alcu

late

d as

the

ave

rage

of

the

follo

win

g th

ree

ratio

s: f

orei

gn a

sset

s to

tot

al a

sset

s, f

orei

gn s

ales

to

tota

l sal

es a

nd f

orei

gn e

mpl

oym

ent

to t

otal

em

ploy

men

t.c

II,

the"

Inte

rnai

onla

izat

ion

Inde

x",

is c

alcu

late

d as

the

num

ber

of f

orei

gn a

ffilia

tes

divi

ded

the

num

ber

of a

ll af

filia

tes

(not

e: a

ffilia

tes

coun

ted

in t

his

tabl

e re

fer

to o

nly

maj

ority

-ow

ned

affil

iate

s).

dIn

dust

ry c

lass

ifica

tion

for

com

pani

es f

ollo

ws

the

Uni

ted

Sta

tes

Sta

ndar

d In

dust

rial

Cla

ssifi

catio

n as

use

d by

the

Uni

ted

Sta

tes

Sec

uriti

es a

nd E

xcha

nge

Com

mis

sion

(S

EC

).e

Fore

ign

sale

s ar

e ba

sed

on t

he o

rigi

n of

the

sal

es,

unle

ss o

ther

wis

e st

ated

.f

In a

num

ber

of c

ases

, co

mpa

nies

rep

orte

d on

ly p

artia

l for

eign

ass

ets.

In

thes

e ca

ses,

the

rat

io o

f th

e pa

rtia

l for

eign

ass

ets

to t

he p

artia

l (to

tal)

ass

ets

was

app

lied

to t

otal

ass

ets

to c

alcu

late

the

tot

al f

orei

gn a

sset

s. I

nal

l cas

es,

the

resu

lting

fig

ures

wer

e se

nt f

or c

onfir

mat

ion

to t

he c

ompa

nies

.g

Fore

ign

empl

oym

ent

data

are

cal

cula

ted

by a

pply

ing

the

shar

e of

for

eign

em

ploy

men

t in

tot

al e

mpl

oym

ent

of t

he p

revi

ous

year

to

tota

l em

ploy

men

t of

200

3.h

Dat

a fo

r ou

tsid

e E

urop

e.i

Dat

a w

ere

obta

ined

fro

m t

he c

ompa

ny in

res

pons

e to

an

UN

CTA

D s

urve

y.j

Fore

ign

empl

oym

ent

data

are

cal

cula

ted

by a

pply

ing

the

shar

e of

bot

h fo

reig

n as

sets

in t

otal

ass

ets

and

fore

ign

sale

s in

tot

al s

ales

to

tota

l em

ploy

men

t.k

Fore

ign

asse

ts d

ata

are

calc

ulat

ed b

y ap

plyi

ng t

he s

hare

of

fore

ign

asse

ts in

tot

al a

sset

s of

the

pre

viou

s ye

ar t

o to

tal a

sset

s of

200

3.l

Fore

ign

empl

oym

ent

data

are

cal

cula

ted

by a

pply

ing

the

aver

age

of t

he s

hare

s of

for

eign

em

ploy

men

t in

tot

al e

mpl

oym

ent

of a

ll co

mpa

nies

in t

he s

ame

indu

stry

(om

ittin

g th

e ex

trem

es)

to t

otal

em

ploy

men

t.m

Fore

ign

empl

oym

ent

data

are

cal

cula

ted

by a

pply

ing

the

shar

e of

for

eign

em

ploy

men

t in

tot

al e

mpl

oym

ent

of P

hilip

Mor

ris

in t

he p

revi

ous

year

to

tota

l em

ploy

men

t of

Altr

ia G

roup

thi

s ye

ar.

nIn

a n

umbe

r of

cas

es c

ompa

nies

rep

orte

d on

ly p

artia

l reg

ion-

spec

ified

sal

es.

In t

hese

cas

es,

the

ratio

of

the

part

ial f

orei

gn s

ales

to

the

part

ial (

tota

l) s

ales

was

app

lied

to t

otal

sal

es t

o ca

lcul

ate

the

tota

l for

eign

sal

es.

In a

ll ca

ses,

the

res

ultin

g fig

ures

hav

e be

en s

ent

for

conf

irm

atio

n to

the

com

pani

es.

oFo

reig

n sa

les

are

base

d on

cus

tom

er lo

catio

n.p

Fore

ign

asse

ts d

ata

are

calc

ulat

ed b

y ap

plyi

ng t

he s

hare

of

both

for

eign

sal

es in

tot

al s

ales

and

for

eign

em

ploy

men

t in

tot

al e

mpl

oym

ent

to t

otal

ass

ets.

qD

ata

for

outs

ide

Wes

tern

Eur

ope.

Not

e: T

he li

st c

over

s no

n-fin

anci

al T

NC

s on

ly.

In s

ome

com

pani

es,

fore

ign

inve

stor

s m

ay h

old

a m

inor

ity s

hare

of

mor

e th

an 1

0 pe

r ce

nt.

270 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.I.1

0.

Th

e t

op

50

no

n-f

ina

nc

ial

TN

Cs

fro

m d

ev

elo

pin

g e

co

no

mie

s,

ran

ke

d b

y f

ore

ign

as

se

ts,

20

03

a

(Mill

ion

s o

f d

olla

rs,

nu

mb

er

of

em

plo

yee

s)

Ran

king

by:

Ass

ets

S

ales

Em

ploy

men

t

No.

of

affil

iate

sFo

reig

nTN

Ib

asse

tsTN

IbIIc

Cor

pora

tion

Hom

e ec

onom

yIn

dust

ryd

Fore

ign

Tota

lFo

reig

neTo

tal

Fore

ign

Tota

l(P

er c

ent)

Fore

ign

Tota

lIIc

17

41H

utch

ison

Wha

mpo

a Li

mite

dH

ong

Kon

g, C

hina

Div

ersi

fied

59 1

4180

340

10 8

0018

699

104

529

126

250

71.4

1900

2350

80.8

52

2739

Sin

gtel

Ltd

.S

inga

pore

Tele

com

mun

icat

ions

17 9

11f

21 6

684

672

68 8

488

642

g21

716

43.1

2330

76.6

73

4235

Pet

rona

s -

Pet

rolia

m N

asio

nal B

hd h

Mal

aysi

aP

etro

leum

exp

l./re

f./di

str.

16 1

1453

457

8 98

125

661

3 62

530

634

25.7

167

234

71.3

74

2648

Sam

sung

Ele

ctro

nics

Co.

, Ltd

.R

epub

lic o

f Kor

eaE

lect

rical

& e

lect

roni

c eq

uipm

ent

12 3

8756

524

41 3

6254

349

19 0

26g

55 3

9744

.180

8989

.89

512

36C

emex

S.A

.M

exic

oC

onst

ruct

ion

Mat

eria

ls11

054

16 0

215

189

7 16

717

051

g25

965

69.0

3548

72.9

26

2337

Am

éric

a M

óvil

Mex

ico

Tele

com

mun

icat

ions

8 67

613

348

3 10

77

649

8 40

3g

18 4

7150

.412

1675

.00

731

24C

hina

Oce

an S

hipp

ing

(Gro

up)

Com

pany

Chi

naTr

ansp

ort a

nd s

tora

ge8

457

f18

007

6 07

6i

9 16

34

600

64 5

8640

.122

5639

.29

846

7P

etro

leo

Bra

sile

iro S

.A. -

Pet

robr

asB

razi

lP

etro

leum

exp

l./re

f./di

str.

7 82

753

612

8 66

542

690

5 81

048

798

15.6

1379

16.4

69

2547

LG E

lect

roni

cs In

c.R

epub

lic o

f Kor

eaE

lect

rical

& e

lect

roni

c eq

uipm

ent

7 11

8j

20 1

7314

443

i29

846

36 2

6863

951

46.8

134

151

88.7

410

1634

Jard

ine

Mat

heso

n H

oldi

ngs

Ltd

Hon

g K

ong,

Chi

naD

iver

sifie

d6

159

k8

949

5 54

0k

8 47

757

895

g11

0 00

062

.316

2369

.57

1110

14S

appi

Lim

ited

Sou

th A

fric

aP

aper

4 88

7f

6 20

33

287

4 29

99

454

g16

939

70.4

115

456

25.2

212

3345

Sas

ol L

imite

dS

outh

Afr

ica

Indu

stria

l che

mic

als

4 22

610

536

5 03

39

722

5 64

331

150

36.7

2125

84.0

013

5030

Chi

na N

atio

nal P

etro

leum

Cor

pora

tion

Chi

naP

etro

leum

exp

l./re

f./di

str.

4 06

0f

97 6

535

218

57 4

2322

000

1 16

7 12

95.

011

920

458

.33

1422

2C

apita

land

Lim

ited

Sin

gapo

reR

eal e

stat

e3

936

10 3

161

449

2 25

25

033

g10

175

50.7

261

3.28

158

43C

ity D

evel

opm

ents

Lim

ited

Sin

gapo

reH

otel

s3

879

l7

329

703

l93

011

549

13 7

0370

.922

827

582

.91

164

49S

hang

ri-La

Asi

a Li

mite

dH

ong

Kon

g, C

hina

Hot

els

and

mot

els

3 67

24

743

436

542

12 6

19m

16 3

0078

.429

3193

.55

1715

33C

itic

Pac

ific

Ltd.

Hon

g K

ong,

Chi

naD

iver

sifie

d3

574

f7

167

2 40

93

372

8 04

512

174

62.5

23

66.6

718

4516

CLP

Hol

ding

sH

ong

Kon

g, C

hina

Ele

ctric

ity, g

as a

nd w

ater

3 56

49

780

298

3 63

948

84

705

18.3

311

27.2

719

4121

Chi

na S

tate

Con

stru

ctio

n E

ngin

eerin

g C

orp.

Chi

naC

onst

ruct

ion

3 41

79

677

2 71

69

134

17 0

5112

1 54

926

.428

7537

.33

2024

22M

TN G

roup

Lim

ited

Sou

th A

fric

aTe

leco

mm

unic

atio

ns3

374

4 81

91

308

3 59

52

601

6 06

349

.86

1637

.50

212

26A

sia

Food

& P

rope

rtie

sS

inga

pore

Food

& b

ever

ages

3 33

13

537

1 23

21

273

32 2

95n

41 8

0089

.42

450

.00

2211

46Fl

extr

onic

s In

tern

atio

nal L

td.

Sin

gapo

reE

lect

rical

& e

lect

roni

c eq

uipm

ent

3 20

6l

5 63

44

674

8 34

080

091

g82

000

70.2

9210

686

.79

2330

17C

ompa

nhia

Val

e do

Rio

Doc

eB

razi

lM

inin

g &

qua

rryi

ng3

155

f11

434

6 51

37

001

224

29 6

3240

.516

5529

.09

2429

10Y

TL C

orp.

Ber

had

Mal

aysi

aU

tiliti

es2

878

6 24

848

91

060

1 51

8g

4 89

541

.124

115

20.8

725

2038

Hon

Hai

Pre

cisi

on In

dust

ries

Taiw

an P

rovi

nce

of C

hina

Ele

ctric

al &

ele

ctro

nic

equi

pmen

t2

597

6 03

24

038

10 7

9378

575

n93

109

54.9

2533

75.7

626

911

Chi

na R

esou

rces

Ent

erpr

ises

hH

ong

Kon

g, C

hina

Pet

role

um e

xpl./

ref./

dist

r.2

364

4 03

42

542

4 45

076

364

g80

000

70.4

419

21.0

527

491

Oil

& N

atur

al G

as C

orp.

Indi

aP

etro

leum

exp

l./re

f./di

str.

2 32

815

249

648

9 37

04

515

n38

033

11.4

140

2.50

285

31N

eptu

ne O

rient

Lin

es L

td.

Sin

gapo

reTr

ansp

ort a

nd s

tora

ge2

266

4 06

44

705

5 52

310

367

g11

322

77.5

610

60.0

029

374

Uni

ted

Mic

roel

ectr

onic

s C

orpo

ratio

nTa

iwan

Pro

vinc

e o

f Chi

naE

lect

rical

& e

lect

roni

c eq

uipm

ent

2 25

110

302

1 89

12

781

1 04

5g

10 5

7633

.27

759.

3330

4420

Sin

gapo

re A

irlin

es L

imite

dS

inga

pore

Tran

spor

t and

sto

rage

2 11

811

278

2 93

26

182

2 46

5g

29 7

3424

.84

1136

.36

3132

19M

etal

urgi

ca G

erda

u S

.A. h

Bra

zil

Met

al a

nd m

etal

pro

duct

s2

056

f4

770

2 09

64

531

5 33

419

597

38.9

1953

35.8

532

2127

Bar

low

orld

Ltd

Sou

th A

fric

aD

iver

sifie

d1

967

3 48

22

380

4 57

410

514

22 7

4951

.62

450

.00

3340

50Q

uant

a C

ompu

ter

Inc

Taiw

an P

rovi

nce

of C

hina

Com

pute

r an

d re

late

d ac

tiviti

es1

934

4 59

373

78

657

5 77

2n

17 3

1828

.04

410

0.00

341

42Fi

rst P

acifi

c C

ompa

ny L

imite

d h

Hon

g K

ong,

Chi

naE

lect

rical

& e

lect

roni

c eq

uipm

ent

1 91

02

074

2 16

22

162

46 9

2646

951

97.3

2834

82.3

535

4340

Hyu

ndai

Mot

or C

ompa

nyR

epub

lic o

f Kor

eaM

otor

veh

icle

s1

780

o20

334

12 0

0820

935

4 82

551

837

25.1

810

80.0

036

3618

Taiw

an S

emic

ondu

ctor

Taiw

an P

rovi

nce

Man

ufac

turin

g C

o Lt

d o

f Chi

naC

ompu

ters

and

rel

ated

act

iviti

es1

539

f11

957

3 78

76

067

4 44

216

997

33.8

412

33.3

3 /..

271ANNEX A

An

ne

x t

ab

le A

.I.1

0.

Th

e t

op

50

no

n-f

ina

nc

ial

TN

Cs

fro

m d

ev

elo

pin

g e

co

no

mie

s,

ran

ke

d b

y f

ore

ign

as

se

ts,

20

03

a (

co

nc

lud

ed

)(M

illio

ns

of

do

llars

, n

um

be

r o

f e

mp

loye

es)

Ran

king

by:

A

sset

s

S

ales

Em

ploy

men

t

N

o. o

f af

filia

tes

Fore

ign

TN

Ib

asse

tsTN

IbIIc

Cor

pora

tion

Hom

e ec

onom

yIn

dust

ryd

Fore

ign

Tota

lFo

reig

neTo

tal

Fore

ign

Tota

l(P

er c

ent)

Fore

ign

Tota

lIIc

3717

3B

enq

Cor

p.Ta

iwan

Pro

vinc

e o

f Chi

naC

ompu

ters

and

rel

ated

act

iviti

es1

497

2 77

81

838

3 77

411

100

14 9

1159

.01

147.

1438

4823

Chi

na N

atio

nal O

ffsho

re O

il C

orp.

hC

hina

Pet

role

um a

nd n

atur

al g

as1

467

14 4

791

877

6 50

71

000

24 0

0014

.441

105

39.0

539

1928

Fras

er &

Nea

ve L

imite

dS

inga

pore

Food

& b

ever

ages

1 39

54

536

1 23

22

140

9 95

1g

12 8

7855

.280

148

54.0

540

3925

Sw

ire P

acifi

c Li

mite

d h

Hon

g K

ong,

Chi

naB

usin

ess

serv

ices

1 38

7f

12 0

601

042

2 26

318

791

56 7

0030

.223

147

049

.15

4135

9K

eppe

l Cor

pora

tion

Lim

ited

Sin

gapo

reD

iver

sifie

d1

361

5 92

889

83

496

11 3

6420

402

34.8

1053

18.8

742

329

Yue

Yuen

Indu

stria

l Hol

ding

s Li

mite

d h

Hon

g K

ong,

Chi

naTe

xtile

s an

d le

athe

r1

317

2 56

92

470

2 52

024

1 80

024

2 00

083

.118

3354

.55

4313

32A

cer

Inc.

Taiw

an P

rovi

nce

of C

hina

Ele

ctric

al &

ele

ctro

nic

equi

pmen

t1

244

3 45

13

637

4 64

05

374

n6

368

66.3

914

64.2

944

288

Del

ta E

lect

roni

cs In

c.Ta

iwan

Pro

vinc

e o

f Chi

naE

lect

roni

cs1

219

1 86

126

61

458

1 35

6l

3 23

841

.913

7816

.67

4538

12G

rupo

Bim

bo S

A D

e C

vM

exic

oFo

od1

156

2 71

61

417

4 15

315

525

70 6

4432

.950

232

21.5

546

145

Chi

na M

inm

etal

s C

orpo

ratio

n h

Chi

naM

etal

s m

inin

g an

d pr

oces

sing

1 15

05

352

1 93

31

168

973

33 0

0063

.322

162

13.5

847

644

The

MU

I Gro

upM

alay

sia

Hot

els

1 13

51

840

499

621

9 02

2n

11 0

0074

.716

619

983

.42

4818

13G

rum

a S

.A. D

e C

.V.

Mex

ico

Food

& b

ever

ages

1 08

62

081

1 39

62

051

8 51

915

104

58.9

1252

23.0

849

476

Hon

gkon

g E

lect

ric H

oldi

ngs

Lim

ited

Hon

g K

ong,

Chi

naE

lect

ricity

, gas

and

wat

er1

074

7 30

525

81

449

295

m2

092

15.5

17

14.2

950

3415

Nan

Ya

Pla

stic

s C

orpo

ratio

nTa

iwan

Pro

vinc

e o

f Chi

naR

ubbe

r an

d pl

astic

s1

030

8 57

693

65

052

16 4

78n

22 0

2735

.13

1127

.27

So

urc

e:

UN

CT

AD

/ E

rasm

us

Un

ive

rsit

y d

ata

ba

se.

aA

ll da

ta a

re b

ased

on

the

com

pani

es’ a

nnua

l rep

orts

unl

ess

othe

rwis

e st

ated

.b

TNI

is c

alcu

late

d as

the

ave

rage

of

the

follo

win

g th

ree

ratio

s: f

orei

gn a

sset

s to

tot

al a

sset

s, f

orei

gn s

ales

to

tota

l sal

es a

nd f

orei

gn e

mpl

oym

ent

to t

otal

em

ploy

men

t.c

II is

cal

cula

ted

as t

he n

umbe

r of

for

eign

affi

liate

s di

vide

d by

num

ber

of a

ll af

filia

tes

(not

e: a

ffilia

tes

coun

ted

in t

his

tabl

e re

fer

to o

nly

maj

ority

-ow

ned

affil

iate

s).

dIn

dust

ry c

lass

ifica

tion

for

com

pani

es f

ollo

ws

the

Uni

ted

Sta

tes

Sta

ndar

d In

dust

rial

Cla

ssifi

catio

n as

use

d by

the

Uni

ted

Sta

tes

Sec

uriti

es a

nd E

xcha

nge

Com

mis

sion

(S

EC

).e

Fore

ign

sale

s ar

e ba

sed

on t

he o

rigi

n of

the

sal

es.

In a

num

ber

of c

ases

com

pani

es r

epor

ted

sale

s on

ly b

y de

stin

atio

n.f

In a

num

ber

of c

ases

, com

pani

es r

epor

ted

only

par

tial f

orei

gn a

sset

s. In

thes

e ca

ses,

the

ratio

of t

he p

artia

l for

eign

ass

ets

to th

e pa

rtia

l (to

tal)

asse

ts w

as a

pplie

d to

tota

l ass

ets

to c

alcu

late

the

tota

l for

eign

asse

ts.

gFo

reig

n em

ploy

men

t da

ta a

re c

alcu

late

d by

app

lyin

g th

e sh

are

of f

orei

gn e

mpl

oym

ent

in t

otal

em

ploy

men

t of

the

pre

viou

s ye

ar t

o to

tal e

mpl

oym

ent

of 2

003.

hD

ata

wer

e ob

tain

ed f

rom

the

com

pany

in r

espo

nse

to a

n U

NC

TAD

sur

vey.

iFo

reig

n sa

les

data

are

cal

cula

ted

by a

pply

ing

the

shar

e of

for

eign

sal

es in

tot

al s

ales

of

the

prev

ious

yea

r to

tot

al s

ales

of

2003

.j

Fore

ign

asse

ts d

ata

are

calc

ulat

ed b

y ap

plyi

ng t

he s

hare

of

fore

ign

asse

ts in

tot

al a

sset

s of

the

pre

viou

s ye

ar t

o to

tal a

sset

s of

200

3.k

Dat

a fo

r ou

tsid

e H

ong

Kon

g (C

hina

) an

d m

ainl

and

Chi

na.

lD

ata

for

outs

ide

Asi

a.m

Fore

ign

empl

oym

ent

data

are

cal

cula

ted

by a

pply

ing

the

shar

e of

for

eign

ass

ets

in t

otal

ase

ts t

o to

tal e

mpl

oym

ent.

nFo

reig

n em

ploy

men

t dat

a ar

e ca

lcul

ated

by

appl

ying

the

aver

age

of th

e sh

ares

of f

orei

gn e

mpl

oym

ent i

n to

tal e

mpl

oym

ent o

f all

com

pani

es in

the

sam

e in

dust

ry (

omitt

ing

the

extr

emes

) to

tota

l em

ploy

men

t.o

Fore

ign

asse

ts a

re c

alcu

late

d by

app

lyin

g th

e sh

are

of f

orei

gn e

mpl

oym

ent

in t

otal

em

ploy

men

t to

the

bal

ance

of

tota

l ass

ets.

272 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.I.1

1.

Th

e t

op

10

no

n-f

ina

nc

ial

TN

Cs

fro

m S

ou

th-E

as

t E

uro

pe

an

d t

he

CIS

,a r

an

ke

d b

y f

ore

ign

as

se

ts,

20

03

(Mill

ion

s o

f d

olla

rs a

nd

nu

mb

er

of

em

plo

yee

s)

Ran

king

by

Fore

ign

Ass

ets

Sal

esE

mpl

oym

ent

TNI

b

asse

ts

TNIb

Cor

pora

tion

Hom

e co

untr

yIn

dust

ryFo

reig

nTo

tal

Fore

ign

Tota

lFo

reig

nTo

tal

(Per

cen

t)

14

Luko

il JS

CR

ussi

an F

eder

atio

nP

etro

leum

and

nat

ural

gas

7 24

726

574

16 2

6022

118

13 9

29c

150

000

36.7

210

Nor

ilsk

Nic

kel,

OJS

C M

MC

Rus

sian

Fed

erat

ion

Min

ing

1 51

85

916

1 51

811

253

1 56

996

520

13.6

33

Nov

oshi

p C

o.R

ussi

an F

eder

atio

nTr

ansp

orta

tion

1 10

71

213

317

395

65

4 78

257

.64

2P

liva

d.d.

Cro

atia

Pha

rmac

eutic

als

925

1 62

990

81

078

3 50

06

780

64.2

55

Rus

alR

ussi

an F

eder

atio

nM

etal

min

ing

serv

ices

691

6 08

53

660

4 50

95

490

63 4

5833

.76

1P

rim

orsk

Shi

ppin

g C

orpo

ratio

nR

ussi

an F

eder

atio

nTr

ansp

orta

tion

382

d44

210

4e

134

1 30

5f

2 61

1f

71.3

77

Mec

hel

Rus

sian

Fed

erat

ion

Met

als

and

met

al p

rodu

cts

121

1 83

51

048

2 05

012

578

84 9

8224

.28

6P

odra

vka

Gro

upC

roat

iaFo

od a

nd b

ever

ages

/ ph

arm

aceu

tical

s10

457

121

048

01

241

7 37

626

.39

8Fa

r E

aste

rn S

hipp

ing

Co.

Rus

sian

Fed

erat

ion

Tran

spor

t52

d16

057

e18

016

6g

4 00

022

.810

9A

lros

aR

ussi

an F

eder

atio

nN

onm

etal

lic M

iner

al M

inin

g46

4 63

088

61

955

8g

46 9

9815

.4

So

urc

e:

UN

CT

AD

.a

Ba

sed

on

su

rve

y re

spo

nse

s a

nd

an

nu

al

rep

ort

s.b

Th

e T

NI

is c

alc

ula

ted

as

the

ave

rag

e o

f th

e f

ollo

win

g t

hre

e r

ati

os:

fo

reig

n a

sse

ts t

o t

ota

l a

sse

ts,

fore

ign

sa

les

to t

ota

l sa

les

an

d f

ore

ign

em

plo

yme

nt

to t

ota

l e

mp

loym

en

t.c

Fo

reig

n e

mp

loym

en

t d

ata

are

ca

lcu

late

d b

y a

pp

lyin

g t

he

sh

are

of

fore

ign

em

plo

yme

nt

in t

ota

l e

mp

loym

en

t o

f 2

00

1 t

o t

ota

l e

mp

loym

en

t o

f 2

00

3.

dF

ore

ign

ass

ets

da

ta a

re c

alc

ula

ted

by

ap

ply

ing

th

e s

ha

re o

f fo

reig

n a

sse

ts i

n t

ota

l a

sse

ts o

f th

e p

revi

ou

s ye

ar

to t

ota

l a

sse

ts o

f 2

00

3.

eF

ore

ign

sa

les

da

ta a

re c

alc

ula

ted

by

ap

ply

ing

th

e s

ha

re o

f fo

reig

n s

ale

s in

to

tal

sale

s o

f th

e p

revi

ou

s ye

ar

to t

ota

l sa

les

of

20

03

.f

20

02

da

ta.

gF

ore

ign

em

plo

yme

nt

da

ta a

re c

alc

ula

ted

by

ap

ply

ing

th

e s

ha

re o

f fo

reig

n e

mp

loym

en

t in

to

tal

em

plo

yme

nt

of

the

pre

vio

us

yea

r to

to

tal

em

plo

yme

nt

of

20

03

.

273ANNEX A

Annex table A.I.12. The world’s top 50 financial TNCs ranked by total assets, 2003(Millions of dollars, number of employees)

Ranking by Assets Employment No. of afiliates Numberof host

Assets Corporation Home economy (Total) (Total) Foreign Total IIa countries

1 Citigroup United States 1 264 032 275 000 320 601 53.2 772 UBS Switzerland 1 221 066 65 929 344 410 83.9 483 Allianz Group Germany 1 179 298 173 750 606 852 71.1 484 Mizuho Financial Group Japan 1 115 081 b 27 900 41 87 47.1 155 Crédit Agricole SA France 1 102 800 63 140 196 447 43.8 416 HSBC Bank plc United Kingdom 1 034 216 218 000 573 971 59.0 487 Deutsche Bank Germany 1 012 554 67 682 469 679 69.1 408 Mitsubishi Tokyo Financial Group Japan 995 403 b 37 000 49 82 59.8 379 BNP Paribas SA France 986 675 89 071 351 641 54.8 48

10 ING Group Netherlands 981 740 114 344 429 1 098 39.1 3411 Sumitomo Mitsui Financial Group Japan 967 978 b 22 431 27 59 45.8 1412 The Royal Bank of Scotland Group United Kingdom 813 030 120 900 166 968 17.1 2613 Barclays Bank PLc United Kingdom 791 754 74 800 117 507 23.1 3714 UFJ Holdings Japan 782 330 b 17 565 25 63 39.7 1115 Credit Suisse Switzerland 777 525 60 837 296 356 83.1 3716 JP Morgan Chase Group United States 770 912 93 453 209 411 50.8 2717 Bank of America NA United States 764 132 133 549 40 273 14.6 1418 HBOS United Kingdom 729 344 66 200 215 560 38.4 1019 ABN AMRO Netherlands 706 150 97 000 441 1 031 42.8 4820 Société Générale France 679 630 87 920 338 525 64.3 4721 AIG Group United States 678 350 86 000 141 349 40.4 3822 Fortis Group Belgium/Netherlands 659 295 64 454 83 598 13.9 1523 Industrial & Commercial Bank of China China 637 823 389 000 71 22 000 .. 1224 HVB Group Germany 626 850 60 214 572 935 61.2 3025 Morgan Stanley United States 602 843 51 196 102 174 58.6 1626 Axa Group France 567 250 117 113 340 432 78.7 2827 GE Capital Services United States 554 526 87 000 1 068 1 398 76.4 5028 Rabobank Nederland Netherlands 508 164 57 055 146 459 31.8 2229 Merril l Lynch United States 496 316 48 100 140 181 77.3 2130 Commerzbank AG Germany 480 797 32 377 102 217 47.0 2031 Caisse des Dépôts et Consignations France 478 178 4 620 58 354 16.4 1832 Lloyds TSB Bank plc United Kingdom 450 043 71 600 122 607 20.1 2133 Groupe Crédit Mutuel France 447 306 55 690 22 159 13.8 634 Bank of China China 442 598 188 716 543 12 090 4.5 2635 Dexia Belgium 440 850 23 865 9 54 16.7 536 Nissay (Nippon Life) Japan 424 700 70 073 3 11 27.3 337 Grupo Santander Spain 421 608 103 038 310 436 71.1 2738 DZ Bank Group Germany 417 970 25 313 48 232 20.7 1539 China Construction Bank China 409 438 410 000 9 21 000 .. 840 LB-BW Germany 406 722 12 648 16 274 5.8 1041 Goldman Sachs United States 403 799 19 476 37 86 43.0 1242 Wachovia Bank United States 401 032 86 670 6 165 3.6 443 Bayern LB Germany 394 923 9 061 28 141 19.8 1044 Wells Fargo Bank United States 387 798 140 000 13 358 3.6 545 Resona Holdings Japan 377 342 b 16 090 12 1 533 0.8 546 Aviva United Kingdom 375 623 60 740 235 424 55.4 1947 Grupo BBVA Spain 361 809 86 197 78 149 52.3 2248 Nordea Bank Sweden 330 360 8 165 64 119 53.8 1249 Banca Intesa Italy 327 870 60 040 51 103 49.5 1450 Gruppo Assicurazioni Generali Italy 327 360 60 638 311 359 86.6 38

Source: UNCTAD.

Notes: Two large mortgage companies in the United States, Fannie Mae and Freddie Mac, are excluded from thislist since they only operate in their home country. Similarly, the largest cooperative financial group in Japan,Zenkyoren, is excluded from the list.

a The Internationalization Index (II) is calculated as the number of foreign affiliates divided by the number of all affiliates(note: affiliates counted in this table refer to only majority-owned affiliates).

b Data refer to March 2004.

274 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.I.13. Inward FDI Performance and Potential Index rankings, 1990-2004a

Inward FDI Performance Index Inward FDI Potential Index

Economy 1990 1995 2000 2001 2002 2003 2004 1990 1995 2000 2001 2002 2003 2004

Albania .. 30 85 67 55 50 42 .. 116 101 95 78 80 ..Algeria 102 130 110 103 96 92 95 48 97 82 77 74 71 ..Angola 105 10 3 2 4 4 4 73 102 100 97 81 76 ,,Argentina 39 58 42 48 82 98 82 59 48 46 53 71 66 ..Armenia .. 65 17 31 30 31 22 .. 112 112 102 89 78 ..Australia 15 46 92 91 72 84 40 12 12 20 23 21 19 ..Austria 77 87 77 75 79 83 89 18 18 24 24 24 24 ..Azerbaijan .. 3 10 35 13 3 1 .. 104 124 107 100 82 ..Bahamas 66 52 48 56 52 62 52 28 41 48 46 48 48 ..Bahrain 24 45 43 55 75 56 27 23 29 33 30 28 29 ..Bangladesh 103 128 125 127 127 132 122 102 118 107 117 113 115 ..Belarus .. 121 89 90 104 102 99 .. 64 65 62 57 52 ..Belgium and Luxembourg 8 23 1 1 1 1 2 10 11 9 7 8 9 ..Benin 16 107 95 87 97 101 97 113 136 134 133 133 134 ..Bolivia 29 27 12 12 16 22 43 87 89 75 80 82 83 ..Botswana 23 138 103 115 66 34 41 32 47 68 67 62 65 ..Brazil 78 103 46 37 38 52 62 52 71 66 71 72 70 ..Brunei Darussalam 90 19 7 7 5 2 3 29 32 37 42 44 45 ..Bulgaria .. 96 29 26 24 16 12 .. 40 67 65 63 61 ..Burkina Faso 93 104 121 124 122 121 115 85 122 127 128 129 129 ..Cameroon 114 131 134 133 136 137 137 80 132 116 114 111 110 ..Canada 38 70 33 33 34 71 94 2 2 5 5 5 4 ..Chile 10 25 20 18 31 30 21 45 38 45 45 47 50 ..China 46 14 52 57 50 42 45 41 61 44 44 41 38 ..Colombia 42 66 81 82 73 69 69 58 82 86 96 99 103 ..Congo 83 12 14 16 45 29 10 72 110 98 98 96 99 ..Congo, Dem. Rep. of 110 133 119 117 100 75 20 105 139 139 139 140 140 ..Costa Rica 18 36 65 73 68 53 51 51 62 63 64 66 69 ..Côte d’Ivoire 80 56 78 85 87 89 87 91 115 109 110 124 122 ..Croatia .. 91 32 22 26 21 33 .. 84 56 49 49 49 ..Cyprus 27 40 22 15 12 10 14 34 36 38 40 42 42 ..Czech Republic .. 34 18 14 14 19 28 .. 39 40 39 40 39 ..Denmark 56 41 11 10 9 46 139 16 15 16 18 17 18 ..Dominican Republic 26 49 50 44 56 51 58 57 59 53 63 58 63 ..Ecuador 32 35 54 38 35 27 34 66 96 104 101 104 109 ..Egypt 14 57 106 112 116 124 108 70 83 70 72 73 75 ..El Salvador 88 117 55 96 84 82 73 97 49 80 83 90 96 ..Estonia .. 7 19 21 20 13 16 .. 58 36 38 36 32 ..Ethiopia 99 118 79 74 49 26 24 112 128 114 120 123 121 ..Finland 65 77 24 43 32 48 55 9 14 10 10 12 13 ..France 45 68 71 64 62 59 80 7 8 13 16 15 12 ..Gabon 35 140 137 139 138 109 57 55 79 85 84 91 92 ..Gambia 9 29 15 11 10 9 6 60 106 102 100 105 106 ..Georgia .. 113 39 61 44 25 13 .. 130 136 130 114 104 ..Germany 86 115 49 40 37 90 118 4 6 8 9 10 8 ..Ghana 89 38 83 77 90 94 91 81 100 113 108 110 100 ..Greece 37 80 123 114 119 122 129 33 37 34 34 33 33 ..Guatemala 22 95 94 100 102 108 120 103 109 92 94 98 101 ..Guinea 61 126 112 116 124 104 85 84 124 119 119 119 124 ..Guyana 60 1 21 19 19 28 31 107 66 71 73 86 94 ..Haiti 82 137 122 125 128 134 133 117 134 135 134 136 137 ..Honduras 33 63 60 53 63 58 53 88 99 99 99 106 118 ..Hong Kong, China 3 15 2 3 3 8 7 20 16 12 12 13 15 ..Hungary 49 4 26 25 28 39 46 50 60 43 41 39 40 ..Iceland 84 132 99 97 94 76 76 14 19 17 15 14 14 ..India 98 110 120 121 121 118 112 76 92 91 88 84 85 ..Indonesia 57 59 138 138 139 139 136 44 67 77 82 87 91 ..Iran, Islamic Rep. of 111 125 133 132 132 131 130 49 46 57 59 59 57 ..Ireland 52 48 4 5 2 5 5 27 22 15 11 9 10 ..Israel 76 81 70 66 70 67 83 31 26 22 21 23 23 ..Italy 64 111 116 109 103 99 98 17 24 25 26 26 26 ..Jamaica 25 37 30 24 23 15 17 64 68 76 75 76 84 ..Japan 104 129 128 129 133 136 134 13 7 11 14 16 16 ..Jordan 75 134 37 50 57 79 48 61 57 60 60 61 60 ..Kazakhstan .. 8 23 13 11 7 11 .. 72 84 74 64 59 ..Kenya 74 122 118 119 117 125 127 86 101 115 125 126 123 ..Korea, Republic of 81 119 93 98 109 116 109 21 17 19 19 19 20 ..Kuwait 101 127 129 134 137 138 138 47 30 31 31 38 41 ..Kyrgyzstan .. 22 66 108 134 105 77 .. 135 117 118 120 111 ..Latvia .. 20 34 52 65 70 47 .. 86 58 52 46 44 ..Lebanon 94 116 101 99 98 91 90 74 65 49 57 60 64 ..Libyan Arab Jamahiriya 68 135 135 135 131 133 116 46 50 42 36 37 34 ..

/...

275ANNEX A

Lithuania .. 79 35 59 43 61 59 .. 91 59 56 51 47 ..Macedonia, TFYR .. 102 68 29 25 32 72 .. 111 108 112 117 120 ..Madagascar 72 112 102 89 101 115 123 100 131 126 123 131 132 ..Malawi 43 85 96 88 113 120 119 93 121 122 126 130 133 ..Malaysia 5 9 51 71 74 77 56 38 33 30 32 32 35 ..Mali 85 47 105 80 33 33 37 108 108 111 122 116 112 ..Malta 21 21 5 6 48 107 84 37 34 35 37 34 37 ..Mexico 34 42 72 65 67 63 79 43 54 50 48 50 51 ..Moldova, Rep. of .. 33 31 20 15 20 26 .. 76 129 113 107 102 ..Mongolia .. 86 63 45 29 17 9 42 81 88 93 94 90 ..Morocco 63 62 100 54 59 35 65 68 88 95 92 92 87 ..Mozambique 87 53 27 23 21 12 23 111 127 120 111 101 98 ..Myanmar 28 84 107 123 125 127 117 118 120 96 89 80 79 ..Namibia 79 31 75 34 18 23 32 96 73 78 81 83 86 ..Nepal 97 123 131 130 135 135 135 109 133 131 132 132 135 ..Netherlands 13 39 9 8 7 18 68 8 10 7 8 11 11 ..New Zealand 7 13 61 63 71 81 74 25 25 27 28 29 30 ..Nicaragua 96 54 25 28 36 38 30 114 125 118 106 112 113 ..Niger 58 124 130 122 123 123 128 104 129 123 129 128 130 ..Nigeria 4 28 82 72 60 41 44 62 94 90 90 97 97 ..Norway 51 61 59 69 92 103 103 5 4 4 2 2 2 ..Oman 36 99 126 113 115 93 110 35 52 55 51 52 53 ..Pakistan 71 89 117 120 118 113 102 92 113 130 131 127 125 ..Panama 116 26 16 32 51 40 29 65 44 47 50 54 56 ..Papua New Guinea 2 11 45 42 89 86 92 89 56 97 109 122 128 ..Paraguay 59 71 84 105 112 119 107 69 80 93 103 108 105 ..Peru 91 17 76 83 80 68 64 79 95 79 78 79 89 ..Phil ippines 30 44 87 95 95 96 100 83 69 61 58 56 58 ..Poland 100 43 47 46 61 72 75 53 55 41 43 43 43 ..Portugal 12 72 67 49 42 44 81 39 35 32 35 35 36 ..Qatar 109 69 97 101 86 66 63 19 20 18 13 6 7 ..Romania .. 83 64 76 76 57 35 .. 87 94 86 77 81 ..Russian Federation .. 101 104 106 111 97 88 .. 31 39 33 30 27 ..Rwanda 62 120 127 128 126 128 124 115 140 138 138 137 131 ..Saudi Arabia 113 108 132 131 130 130 121 30 28 28 29 31 31 ..Senegal 67 93 90 93 105 110 101 94 123 105 104 103 107 ..Sierra Leone 48 136 91 86 88 126 131 101 137 140 140 139 139 ..Singapore 1 2 6 4 6 6 8 15 3 2 4 4 5 ..Slovakia .. 64 41 27 8 14 25 .. 51 51 47 45 46 ..Slovenia .. 88 113 110 58 49 60 .. 43 29 27 27 28 ..South Africa 107 109 114 84 81 78 126 54 63 74 70 75 73 ..Spain 19 55 53 39 27 37 49 24 27 26 25 25 25 ..Sri Lanka 73 74 108 107 108 100 96 99 107 106 116 109 114 ..Sudan 108 114 62 58 41 24 18 116 138 125 121 118 116 ..Suriname 117 139 140 140 140 140 140 75 93 89 91 95 88 ..Sweden 53 24 8 9 22 54 93 6 9 6 6 7 6 ..Switzerland 31 97 36 36 39 45 61 11 13 14 17 18 17 ..Syrian Arab Republic 54 16 57 51 47 43 39 77 77 83 85 93 95 ..Taiwan Province of China 50 100 111 102 106 117 125 22 21 21 20 20 21 ..Tajikistan .. 60 88 94 85 87 19 .. 98 137 137 134 127 ..Thailand 17 75 44 60 83 88 106 40 42 52 55 53 55 ..Togo 44 76 86 68 53 55 66 95 126 110 115 115 119 ..Trinidad and Tobago 20 6 13 17 17 11 15 67 78 64 61 55 54 ..Tunisia 55 32 69 78 64 64 67 71 75 69 68 68 67 ..Turkey 69 105 124 111 110 106 111 63 74 73 79 70 72 ..Uganda 106 50 80 79 69 60 54 106 117 103 105 102 108 ..Ukraine .. 98 98 92 91 74 71 .. 53 81 76 69 62 ..United Arab Emirates 92 92 136 136 107 95 104 26 23 23 22 22 22 ..United Kingdom 11 67 28 30 40 85 78 3 5 3 3 3 3 ..United Rep. of Tanzania 95 51 58 41 46 36 36 90 114 128 127 125 126 ..United States 41 94 74 81 93 111 114 1 1 1 1 1 1 ..Uruguay 70 90 109 104 99 80 70 56 70 62 66 88 93 ..Uzbekistan .. 106 115 118 120 114 105 .. 90 121 124 121 117 ..Venezuela 40 78 56 70 77 73 86 36 45 54 54 65 74 ..Viet Nam 47 5 38 47 54 47 50 78 85 72 69 67 68 ..Yemen 115 73 139 137 114 112 132 110 103 87 87 85 77 ..Zambia 6 18 40 62 78 65 38 98 119 132 135 135 136 ..Zimbabwe 112 82 73 126 129 129 113 82 105 133 136 138 138 ..

Source: UNCTAD.

Note: Covering 140 economies. The potential index is based on 12 economic and policy variables.

a Three-year moving averages, using data for the three previous years, including the year in question.

Annex table A.I.13. Inward FDI Performance and Potential Index rankings, 1990-2004a(concluded)

Inward FDI Performance Index Inward FDI Potential Index

Economy 1990 1995 2000 2001 2002 2003 2004 1990 1995 2000 2001 2002 2003 2004

276 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.I.14. Outward FDI Performance Index rankings, 1990-2004a

Economy 1990 1995 2000 2001 2002 2003 2004 Economy 1990 1995 2000 2001 2002 2003 2004

Albania .. 39 68 73 87 106 103 Korea, Republic of 30 36 30 38 39 43 41Algeria 75 88 98 89 81 83 64 Kuwait 14 132 132 80 124 132 132Angola 79 117 59 60 60 60 60 Kyrgyzstan .. 105 40 48 59 73 131Argentina 65 42 47 53 84 85 70 Latvia .. 131 51 61 69 61 44Armenia .. 104 102 99 56 53 53 Lebanon 105 30 60 62 54 56 49Australia 20 31 48 31 23 21 18 Libyan Arab Jamahiriya 35 129 43 45 67 68 117Austria 22 38 21 24 20 22 17 Lithuania .. 93 92 86 80 71 42Azerbaijan .. 107 18 21 24 6 4 Macedonia, TFYR .. 109 122 103 105 105 98Bahamas 111 100 105 108 107 63 67 Madagascar 108 65 109 115 113 112 110Bahrain 23 17 26 26 27 11 6 Malawi 83 106 76 113 112 111 109Bangladesh 81 99 113 97 100 100 102 Malaysia 27 8 23 28 28 30 23Belarus .. 82 110 107 131 129 127 Mali 96 102 52 58 57 64 91Belgium and Luxembourg 9 9 1 1 1 1 1 Malta 84 70 34 36 50 55 69Benin 99 78 78 74 77 88 96 Mexico 54 75 57 49 51 46 51Bolivia 61 83 100 93 93 95 92 Moldova, Rep. of .. 35 125 101 101 102 80Botswana 49 40 91 19 18 17 19 Morocco 57 74 79 65 65 75 82Brazil 43 60 53 75 68 94 37 Mozambique 72 97 114 109 116 107 105Brunei Darussalam 85 25 70 54 49 54 65 Namibia 68 119 111 127 130 127 126Bulgaria .. 125 90 71 71 67 125 Netherlands 3 3 3 3 3 4 11Burkina Faso 78 61 83 98 92 89 87 New Zealand 10 12 22 42 62 128 43Cameroon 38 69 97 68 82 76 95 Nicaragua 98 96 58 59 58 51 66Canada 21 13 14 14 14 16 13 Niger 45 51 65 125 127 124 119Chile 59 19 17 15 19 25 27 Nigeria 15 34 54 52 52 49 47China 36 41 62 57 53 59 72 Norway 19 14 19 20 22 37 29Colombia 51 52 50 64 48 41 38 Oman 62 77 124 105 117 117 115Congo 55 120 127 69 66 69 78 Pakistan 52 116 108 96 91 91 86Congo, Dem. Rep. of 94 95 126 121 121 119 116 Panama 1 5 10 16 9 2 2Costa Rica 53 79 99 88 75 70 56 Papua New Guinea 37 128 131 43 44 42 89Côte d’Ivoire 42 29 67 95 118 96 88 Paraguay 103 53 86 82 83 82 76Croatia .. 72 56 51 33 34 32 Peru 46 118 74 70 88 79 83Cyprus 56 58 28 22 17 14 12 Philippines 60 43 104 128 125 90 50Czech Republic .. 54 66 66 61 57 48 Poland 63 85 84 118 95 86 62Denmark 17 11 8 6 7 15 129 Portugal 40 44 15 12 16 19 15Dominican Republic 92 67 80 90 86 123 121 Qatar 107 55 75 79 97 120 120Ecuador 101 68 72 110 108 108 106 Romania .. 92 123 120 120 97 77Egypt 58 73 96 92 94 99 74 Russian Federation .. 50 38 35 34 27 24El Salvador 87 122 71 76 126 121 104 Rwanda 66 121 115 114 111 110 108Estonia .. 59 36 23 21 23 20 Saudi Arabia 24 63 85 124 123 122 85Ethiopia 93 112 32 130 119 114 112 Senegal 48 64 64 91 64 66 58Finland 11 10 4 8 5 20 36 Sierra Leone 97 81 121 111 109 109 107France 13 16 9 10 10 12 16 Singapore 8 2 12 4 4 3 7Gabon 31 49 63 56 122 130 128 Slovakia .. 57 130 132 79 78 124Gambia 32 15 31 33 32 26 30 Slovenia .. 127 61 47 45 35 28Georgia .. 108 103 104 90 81 68 South Africa 47 23 37 131 132 131 46Germany 18 21 16 18 26 39 81 Spain 28 32 11 11 12 9 10Ghana 90 113 45 40 41 45 61 Sri Lanka 64 76 82 87 96 80 75Greece 77 91 42 37 36 50 52 Sweden 2 7 7 9 8 7 8Guatemala 104 124 77 72 74 87 93 Switzerland 4 4 6 5 6 8 9Guinea 67 90 93 78 72 74 79 Syrian Arab Republic 95 115 120 117 115 116 114Guyana 82 62 117 100 76 101 101 Taiwan Province of China 6 22 25 25 25 24 21Haiti 109 94 119 119 102 104 99 Tajikistan .. 103 116 112 110 52 55Honduras 106 126 87 84 98 118 118 Thailand 44 46 73 63 73 58 57Hong Kong, China 5 1 2 2 2 5 3 Togo 41 48 44 126 128 126 123Hungary 70 66 39 39 38 33 35 Trinidad and Tobago 88 111 29 29 40 28 33Iceland 50 47 20 17 15 13 5 Tunisia 71 86 106 102 104 103 97India 80 87 94 67 63 62 54 Turkey 100 80 55 50 55 65 59Indonesia 73 24 88 81 78 84 84 Uganda 89 26 129 129 129 113 111Iran, Islamic Rep. of 102 123 81 85 103 125 122 Ukraine .. 89 112 94 99 98 100Ireland 12 28 13 13 13 10 26 United Arab Emirates 69 84 49 44 43 48 63Israel 33 27 27 30 30 36 22 United Kingdom 7 6 5 7 11 18 14Italy 26 37 35 32 29 29 34 United Rep. of Tanzania 86 110 118 116 114 115 113Jamaica 34 18 33 34 35 32 31 United States 25 20 24 27 31 31 25Japan 16 45 46 41 37 40 40 Uruguay 76 114 107 106 89 77 71Jordan 110 130 95 83 70 72 73 Venezuela 29 33 41 46 47 38 39Kazakhstan .. 101 101 122 42 47 130 Yemen 91 98 128 123 106 92 90Kenya 74 71 89 55 46 44 45 Zimbabwe 39 56 69 77 85 93 94

Source: UNCTAD.

Note: Covering 132 economies.a Three-year moving averages, using data for the three previous years, including the year in question.

277ANNEX A

A.I.15. International investment agreements (other than BITs and DTTs) concluded, 2004-2005

Agreement Year Geographical scope

Framework Agreement on the BIMSTEC Free Trade Area 2004 Regional (1group)Free Trade Agreement between the EFTA States and the Republic of Tunisia 2004 Interregional (1 group + 1 country)Free Trade Agreement between the Government of the United Mexican States

and the Republic of Uruguay 2004 Regional (bilateral)Framework Agreement on the South Asia Free Trade Area (SAARC) 2004 Regional (1group)Free Trade Agreement between the Kingdom of Morocco and the

United Sates of America 2004 Interregional (bilateral)Free Trade Agreement between Australia and Thailand 2004 Regional (bilateral)Free Trade Agreement between Australia and the United States of America 2004 Interregional (bilateral)Agreement Concerning the Development of Trade and Investment Relations

between the Government of the United States of America and theGovernment of the State of Qatar 2004 Interregional (bilateral)

Agreement Concerning the Development of Trade and Investment Relationsbetween the Government of the United States of America and theGovernment of the United Arab Emirates 2004 Interregional (bilateral)

Agreement Concerning the Development of Trade and Investment Relationsbetween Mongolia and the United Sates of America 2004 Interregional (bilateral)

Agreement Concerning the Development of Trade and Investment Relationsbetween the Government of the United States of America and theGovernment of the State of Kuwait 2004 Interregional (bilateral)

Agreement Concerning the Development of Trade and Investment Relationsbetween Malaysia and the United States of America 2004 Interregional (bilateral)

Agreement Concerning the Development of Trade and Investment Relationsbetween the Government of the United States of America and

Government of the Republic of Yemen 2004 Interregional (bilateral)Free Trade Agreement between the Kingdom of Bahrain and the

United States of America 2004 Interregional (bilateral)Free Trade Agreement between the Republic of Albania and the

Republic of Romania 2004 Regional (bilateral)Free Trade Agreement between the Republic of Albania and the

Republic of Serbia and Montenegro 2004 Regional (bilateral)Free Trade Agreement between the Hashemite Kingdom Jordan and the

Republic of Singapore 2004 Interregional (bilateral)Cooperation Agreement between the European Community and Pakistan 2004 Interregional (1 group + 1 country)Free Trade Agreement between the EFTA States and Lebanon 2004 Interregional (1 group + 1 country)Free Trade Agreement between the Republic of Bosnia Herzegovina and

the Republic of Moldova 2004 Regional (bilateral)Free Trade Agreement between the Republic of Bosnia Herzegovina

and the Republic of Romania 2004 Regional (bilateral)Free Trade Agreement between Central America, the Dominican Republic

and the United States of America (CAFTA) 2004 Regional (1 group + 2 country)Partial Reach Agreement for Economic, Trade and Investment Promotion

between the Republic of Argentina and the Republic of Bolivia 2004 Regional (bilateral)Economic Complementation General Agreement on Integration, Economic

and Social Cooperation for the Establishment of a Common Marketbetween the Republic of Bolivia and the Republic of Peru 2004 Regional (bilateral)

Framework Agreement Between the Government of the United Statesof America, the Government of the Republic of Kazakhstan,The Government of the Kyrgyz Republic, the Government of theRepublic of Tajikistan, the Government of Turkmenistan, and theGovernment of the Republic of Uzbekistan Concerning theDevelopment of Trade and Investment Relations 2004 Interregional (1 group + 1 country)

Free Trade Agreement between the Caribbean Community (CARICOM)and Costa Rica 2004 Regional (1 group + 1 country)

Interim Free Trade Agreement between the Republic of Turkey and thePalestinian Authority 2004 Regional (bilateral)

Framework Agreement between MERCOSUR and the Arab Republic of Egypt 2004 Interregional (1 group + 1 country)Agreement for the Establishment of a Free Trade Area between the Gulf

Cooperation Council and Lebanon 2004 Regional (1 group + 1 country)Framework Agreement on Economic Cooperation Agreement between the 2004 Interregional (1 group + 1 country)

Gulf Cooperation Council (GCC) and IndiaComprehensive Economic Cooperation Agreement between the

Republic of India and the Republic of Chile 2005 Interregional (bilateral)Agreement between Japan and the United Mexican States for the

Strengthening of Economic Partnership 2005 Interregional (bilateral)Agreement on Closer Economic Partnership between New Zealand

and Thailand 2005 Interregional (bilateral)Comprehensive Economic Cooperation Agreement between India and Singapore 2005 Regional (bilateral)

Source: UNCTAD.

278 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

A.I.16. International investment agreements (other than BITs and DTTs) under negotiationor consultation, as of end 2004

Agreement Geographical scope

Agreement Establishing an Association between the European Communities and TheirMember States, of the One Part, and Syria, of the Other Parta Interregional (1 group + 1 country)

Closer Economic Partnership Agreement between Hong Kong (China) and New Zealand Regional (bilateral)Comprehensive Economic Cooperation Agreement between China and the Republic of India Interregional (bilateral)Comprehensive Economic Cooperation Agreement between India and Singapore Regional (bilateral)Comprehensive Economic Cooperation Agreement between the Republic of India and the

Republic of Mauritius Interregional (bilateral)Economic Framework Agreement between Canada and Japan Interregional (bilateral)Economic Partnership Agreement between CARICOM and the European Union Interregional (2 groups)Economic Partnership Agreement between India and Sri Lanka Comprehensive Regional (bilateral)Economic Partnership Agreement between Japan and the Kingdom of Thailand Regional (bilateral)Economic Partnership Agreement between Japan and the Philippines Regional (bilateral)Economic Partnership Agreement between the European Union and CEMAC Inter-regional (2 groups)Economic Partnership Agreement between the European Union and Eastern and

Southern Africa (ESA) Interregional (2 groups)Economic Partnership Agreement between the European Union and ECOWAS Interregional (2 groups)Economic Partnership Agreement between the European Union and SADC Interregional (2 groups)Free Trade Agreement between ASEAN, Australia, New Zealand Regional (1 group + 2 country)Free Trade Agreement between Australia and China Regional (bilateral)Free Trade Agreement between Canada and Central America Interregional (1 group + 1 country)Free Trade Agreement between Canada and the Republic of Korea Interregional (bilateral)Free Trade Agreement between Canada and the Republic of Singapore Interregional (bilateral)Free Trade Agreement between Canada the Dominican Republic Interregional (bilateral)Free Trade Agreement between CARICOM and Canada Interregional (1 group + 1 country)Free Trade Agreement between CARICOM and the United States of America Interregional (1 group + 1 country)Free Trade Agreement between China and New Zealand Regional (bilateral)Free Trade Agreement between China and the Republic of Chile Interregional (bilateral)Free Trade Agreement between EFTA and Canada Interregional (1 group + 1 country)Free Trade Agreement between EFTA and CARICOM Interregional (2 groups)Free Trade Agreement between EFTA and SACU Interregional (2 groups)Free Trade Agreement between EFTA and the Kingdom of Thailand Interregional (1 group + 1 country)Free Trade Agreement between EFTA and the Republic of Korea Interregional (1 group + 1 country)Free Trade Agreement between Japan and the Republic of Korea Regional (bilateral)Free Trade Agreement between Japan and the Republic of Malaysia Regional (bilateral)Free Trade Agreement between SACU and the United States of America Interregional (1 group + 1 country)Free Trade Agreement between the Andean Community and Canada Interregional (1 group + 1 country)Free Trade Agreement between the Andean Community and the United States of America Interregional (1 group + 1 country)Free Trade Agreement between the Arab Republic of Egypt and the Republic of Singapore Interregional (bilateral)Free Trade Agreement between the Gulf Cooperation Council (GCC) and China Regional (1 group + 1 country)Free Trade Agreement between the Kingdom of Bahrain and the Republic of Singapore Regional (bilateral)Free Trade Agreement between the Kingdom of Kuwait and the Republic of Singapore Regional (bilateral)Free Trade Agreement between the Republic of Chile and Japan Regional (bilateral)Free Trade Agreement between the Republic of Chile and the Republic of Ecuador Regional (bilateral)Free Trade Agreement between the Republic of Chile and the Republic of Peru Regional (bilateral)Free Trade Agreement between the Republic of Costa Rica and the Republic of Panama Regional (bilateral)Free Trade Agreement between the Republic of Guatemala and Taiwan (Province of China) Inter-regional (bilateral)Free Trade Agreement between the Republic of Korea and the Republic of Singapore Regional (bilateral)Free Trade Agreement between the Republic of Korea and the United States of America Interregional (bilateral)Free Trade Agreement between the Republic of Nicaragua and Taiwan (Province of China) Interregional (bilateral)Free Trade Agreement between the Republic of Panama and the Republic of Singapore Interregional (bilateral)Free Trade Agreement between the Republic of Peru and the Kingdom of Thailand Interregional (bilateral)Free Trade Agreement between the Republic of Singapore and Qatar Regional (bilateral)Free Trade Agreement between the Republic of Sri Lanka and the Republic of Singapore Regional (bilateral)Free Trade Agreement between the Republic of Thailand and the United States of America Interregional (bilateral)Free Trade Agreement between the United Mexican States and the Republic of Singapore Interregional (bilateral)Free Trade Agreement between the United States of America and Colombia Regional (bilateral)Free Trade Agreement between the United States of America and Ecuador Regional (bilateral)Free Trade Agreement between the United States of America and Oman Interregional (bilateral)Free Trade Agreement between the United States of America and Panama Regional (bilateral)Free Trade Agreement between the United States of America and Peru Regional (bilateral)Free Trade Agreement between the United States of America and the United Arab Emirates Interregional (bilateral)Free Trade Agreement between the United States of America and Uruguay Interregional (bilateral)Free Trade Area between ASEAN and the Republic of Korea Regional (1 group + 1 country)Free Trade Area of the Americas (FTAA) Regional (1 group)Inter-Regional Association Agreement between the European Union and MERCOSUR Interregional (2 groups)Pacific Three Free Trade Agreement between Chile, New Zealand and Singapore Regional (3 countries)Partial Scope Trade Agreement between Belize and the Republic of Guatemala Regional (bilateral)SAARC Agreement for the Promotion and Protection of Investment Regional (1 group)Trade and Investment Enhancement Agreement between Canada and the European Union Interregional (1 group + 1 country)Trans-Regional EU-ASEAN Trade Initiative Interregional (2 groups)

Source: UNCTAD.a Negotiations on the EC-Syria association agreement are formally completed.

279ANNEX A

An

ne

x t

ab

le A

.II.

1.

Cro

ss

-bo

rde

r M

&A

de

als

wit

h v

alu

es

of

ov

er

$1

00

mil

lio

n c

on

clu

de

d i

n d

ev

elo

pin

g a

nd

tra

ns

itio

n e

co

no

mie

s,

20

04

Valu

eR

ank

($ m

illio

n)A

cqui

red

com

pany

Hos

t ec

onom

yIn

dust

ry o

f th

e ac

quir

ed c

ompa

nyA

cqui

ring

com

pany

Hom

e ec

onom

yIn

dust

ry o

f th

e ac

quir

ing

com

pany

Afr

ica

11

504

Ash

anti

Gol

dfie

lds

Co

Ltd

Gha

naG

old

ores

Ang

logo

ld L

tdS

outh

Afr

ica

Gol

d or

es2

1 20

5G

old

Fiel

ds L

tdS

outh

Afr

ica

Gol

d or

esN

orim

et L

tdU

nite

d K

ingd

omM

etal

s se

rvic

e ce

nter

s an

d of

fices

3 5

00E

nerg

y A

fric

a Lt

dS

outh

Afr

ica

Cru

de p

etro

leum

and

nat

ural

gas

Tullo

w O

il P

lcU

nite

d K

ingd

omC

rude

pet

role

um a

nd n

atur

al g

as4

250

Ken

Cel

l Com

mun

icat

ions

Ltd

Ken

yaR

adio

tele

phon

e co

mm

unic

atio

nsC

elte

l Uga

nda

Uga

nda

Tele

phon

e co

mm

unic

atio

ns,

exce

pt r

adio

tele

phon

e5

235

Ros

etta

Con

cess

ion

Offs

hore

Egy

ptC

rude

pet

role

um a

nd n

atur

al g

asB

G G

roup

Plc

Uni

ted

Kin

gdom

Cru

de p

etro

leum

and

nat

ural

gas

6 1

37M

auri

tani

a H

oldi

ngs

BV

Mau

rita

nia

Cru

de p

etro

leum

and

nat

ural

gas

BG

Gro

up P

lcU

nite

d K

ingd

omC

rude

pet

role

um a

nd n

atur

al g

as7

102

Blo

ck 5

AS

udan

Cru

de p

etro

leum

and

nat

ural

gas

ON

GC

Vid

esh

Ltd

Indi

aC

rude

pet

role

um a

nd n

atur

al g

as

Lati

n A

mer

ica

and

the

Car

ibbe

an1

3 97

4B

raco

SA

Bra

zil

Mal

t be

vera

ges

Inte

rbre

w S

AB

elgi

umM

alt

beve

rage

s2

3 88

8G

rupo

Fin

anci

ero

BB

VA B

anco

mer

Mex

ico

Ban

ksB

BVA

Spa

inB

anks

31

304

Tele

foni

ca M

ovil

Chi

le S

AC

hile

Rad

iote

leph

one

com

mun

icat

ions

Tele

foni

ca M

ovile

s S

AS

pain

Rad

iote

leph

one

com

mun

icat

ions

41

199

Ban

k of

Ber

mud

a Lt

dB

erm

uda

Ban

ksH

SB

C H

oldi

ngs

PLC

{HS

BC

}U

nite

d K

ingd

omB

anks

51

195

Telc

el C

elul

ar S

AVe

nezu

ela

Rad

iote

leph

one

com

mun

icat

ions

Tele

foni

ca M

ovile

s S

AS

pain

Rad

iote

leph

one

com

mun

icat

ions

61

050

Bel

lsou

th-C

olom

bian

Ope

ratio

nsC

olom

bia

Rad

iote

leph

one

com

mun

icat

ions

Tele

foni

ca M

ovile

s S

AS

pain

Rad

iote

leph

one

com

mun

icat

ions

7 8

33B

ells

outh

-Ecu

ador

Ope

ratio

nsE

cuad

orTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Tele

foni

ca M

ovile

s S

AS

pain

Rad

iote

leph

one

com

mun

icat

ions

8 7

15E

l Pas

o C

orp-

Aru

ba R

efin

ery

Aru

baP

etro

leum

ref

inin

gVa

lero

Ene

rgy

Cor

pU

nite

d S

tate

sP

etro

leum

ref

inin

g9

657

Bel

lSou

th d

e P

anam

aP

anam

aTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Tele

foni

ca M

ovile

s S

AS

pain

Rad

iote

leph

one

com

mun

icat

ions

10 5

91H

olci

m A

pasc

o S

A de

CV

Mex

ico

Cem

ent,

hyd

raul

icH

olci

m L

tdS

witz

erla

ndC

emen

t, h

ydra

ulic

11 5

25G

rupo

Im

sa-M

exic

an B

atte

ry O

psM

exic

oE

lect

rica

l ind

ustr

ial a

ppar

atus

, ne

cJo

hnso

n C

ontr

ols

Inc

Uni

ted

Sta

tes

Pro

cess

con

trol

inst

rum

ents

12 4

42M

ova

Pha

rmac

eutic

als

Cor

pP

uert

o R

ico

Pha

rmac

eutic

al p

repa

ratio

nsP

athe

on I

ncC

anad

aP

harm

aceu

tical

pre

para

tions

13 4

00E

mbr

atel

Par

ticip

acoe

s S

AB

razi

lTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Tele

fono

s de

Mex

ico

SA

de C

VM

exic

oTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

14 3

75H

ipot

ecar

ia N

acio

nal S

A d

e C

VM

exic

oP

erso

nal c

redi

t in

stitu

tions

Gru

po F

inan

cier

o B

BVA

Ban

com

erM

exic

oB

anks

15 3

64A

vian

ca S

AC

olom

bia

Air

tra

nspo

rtat

ion,

sch

edul

edS

iner

gyB

razi

lIn

vest

ors,

nec

16 3

40C

arib

bean

Res

taur

ants

Inc

Pue

rto

Ric

oE

atin

g pl

aces

Cas

tle H

arla

n In

cU

nite

d S

tate

sS

ecur

ity b

roke

rs,

deal

ers,

and

flo

tatio

n co

mpa

nies

17 3

24A

lpar

t Ja

mai

caJa

mai

caM

isce

llane

ous

met

al o

res,

nec

Hyd

ro A

lum

iniu

m A

SN

orw

ayP

rim

ary

prod

uctio

n of

alu

min

um18

309

Pla

za C

arol

ina,

San

Jua

n,P

uert

oP

uert

o R

ico

Ope

rato

rs o

f no

n-re

side

ntia

l bui

ldin

gsS

imon

Pro

pert

y G

roup

Inc

Uni

ted

Sta

tes

Rea

l est

ate

inve

stm

ent

trus

ts19

305

Tele

Cen

tro

Oes

te C

elul

ar P

art

Bra

zil

Tele

phon

e co

mm

unic

atio

ns,

exce

pt r

adio

tele

phon

eTe

lesp

Cel

ular

Par

ticip

acoe

sB

razi

lTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

20 3

00B

ompr

eco

SA

Sup

erm

erca

dos

Bra

zil

Gro

cery

sto

res

Wal

-Mar

t S

tore

s In

cU

nite

d S

tate

sVa

riet

y st

ores

21 2

95C

ia d

e Te

leco

mun

icac

ione

sE

l Sal

vado

rTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Am

eric

a M

ovil

SA

de

CV

Mex

ico

Rad

iote

leph

one

com

mun

icat

ions

22 2

75A

ccio

n-B

uild

ings

Por

tfol

ioM

exic

oO

pera

tors

of

non-

resi

dent

ial b

uild

ings

Inve

stor

Gro

upU

nite

d S

tate

sIn

vest

ors,

nec

23 2

66B

unge

Bra

sil S

AB

razi

lIn

vest

ors,

nec

Bun

ge L

tdU

nite

d S

tate

sFl

our

and

othe

r gr

ain

mill

pro

duct

s24

260

Rio

Par

acat

u M

iner

acao

(R

PM

)B

razi

lM

isce

llane

ous

met

al o

res,

nec

Kin

ross

Gol

d C

orp

Can

ada

Gol

d or

es25

235

Cop

amex

-Tis

sue

Bus

ines

sM

exic

oS

anita

ry p

aper

pro

duct

sS

vens

ka C

ellu

losa

AB

Sw

eden

San

itary

pap

er p

rodu

cts

26 2

10B

ells

outh

Per

u S

AP

eru

Rad

iote

leph

one

com

mun

icat

ions

Tele

foni

ca M

ovile

s S

AS

pain

Rad

iote

leph

one

com

mun

icat

ions

27 2

10R

efin

eria

de

Caj

amar

quill

a S

AP

eru

Lead

and

zin

c or

esVo

tora

ntim

Met

ais

Bra

zil

Rol

ling,

dra

win

g, &

ext

rudi

ng o

f no

nfer

rous

met

als

28 2

07AT

&T

Latin

-Sou

th A

mer

ican

Uni

tB

razi

lTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Tele

fono

s de

Mex

ico

SA

de C

VM

exic

oTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

29 2

00P

lusP

etro

l Nor

teP

eru

Cru

de p

etro

leum

and

nat

ural

gas

CN

PC

Chi

naC

rude

pet

role

um a

nd n

atur

al g

as30

195

FLA

G T

elec

om G

roup

Ltd

Ber

mud

aTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Rel

ianc

e G

atew

ay N

et P

vt L

tdIn

dia

Tele

phon

e co

mm

unic

atio

ns,

exce

pt r

adio

tele

phon

e31

180

Sco

ttis

h R

E G

roup

Ltd

Ber

mud

aLi

fe in

sura

nce

Cyp

ress

Gro

up L

LCU

nite

d S

tate

sIn

vest

ors,

nec

32 1

75B

ells

outh

Gua

tem

ala

Gua

tem

ala

Tele

phon

e co

mm

unic

atio

ns,

exce

pt r

adio

tele

phon

eTe

lefo

nica

Mov

iles

SA

Spa

inR

adio

tele

phon

e co

mm

unic

atio

ns33

155

Cen

tenn

ial P

uert

o R

ico

Cab

leP

uert

o R

ico

Cab

le a

nd o

ther

pay

tel

evis

ion

serv

ices

Hic

ks M

use

Tate

& F

urst

Inc

Uni

ted

Sta

tes

Inve

stor

s, n

ec34

150

Bel

lsou

th-N

icar

agua

n C

ellu

lar

Nic

arag

uaR

adio

tele

phon

e co

mm

unic

atio

nsTe

lefo

nica

Mov

iles

SA

Spa

inR

adio

tele

phon

e co

mm

unic

atio

ns35

124

Indu

sval

-Con

sum

er F

inan

ce O

psB

razi

lB

anks

HS

BC

Ban

k B

rasi

l SA

Bra

zil

Ban

ks

/...

280 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.II.

1.

Cro

ss

-bo

rde

r M

&A

de

als

wit

h v

alu

es

of

ov

er

$1

00

mil

lio

n c

on

clu

de

d i

n d

ev

elo

pin

g a

nd

tra

ns

itio

n e

co

no

mie

s,

20

04

(c

on

tin

ue

d)

Valu

eR

ank

($ m

illio

n)A

cqui

red

com

pany

Hos

t ec

onom

yIn

dust

ry o

f th

e ac

quir

ed c

ompa

nyA

cqui

ring

com

pany

Hom

e ec

onom

yIn

dust

ry o

f th

e ac

quir

ing

com

pany

36 1

20P

osVe

n C

AVe

nezu

ela

Pri

mar

y m

etal

pro

duct

s, n

ecM

AS

ISA

Vene

zuel

aG

ray

and

duct

ile ir

on f

ound

ries

37 1

19C

elul

ar C

RT

Par

ticip

acoe

s S

AB

razi

lTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Bra

silc

el N

VB

razi

lIn

vest

ors,

nec

38 1

14Te

chte

l Tel

ecom

unic

acio

nes

Arg

entin

aTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

Tele

fono

s de

Mex

ico

SA

de C

VM

exic

oTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

39 1

10U

nile

ver-

Mex

ican

Oil

Bra

nds

Mex

ico

Edi

ble

fats

and

oils

, ne

cA

CH

Foo

d C

ompa

nies

Inc

Uni

ted

Sta

tes

Bre

ad a

nd o

ther

bak

ery

prod

ucts

, ex

cept

coo

kies

40 1

10C

orpo

raci

on D

igite

l(Te

leco

m)

Vene

zuel

aC

omm

unic

atio

ns s

ervi

ces,

nec

TIM

Int

erna

tiona

lIt

aly

Inve

stor

s, n

ec41

100

Sup

erga

sbra

s In

ds e

Com

erci

oB

razi

lG

as p

rodu

ctio

n an

d/or

dis

trib

utio

nS

HV

Hol

ding

s N

VN

ethe

rlan

dsG

roce

ry s

tore

s

Asi

a an

d O

cean

ia1

2 00

0P

roct

er &

Gam

ble-

Hut

chis

on L

tdC

hina

Soa

p &

oth

er d

eter

gent

s, e

xcep

t sp

ecia

lty c

lean

ers

Pro

cter

& G

ambl

e C

oU

nite

d S

tate

sS

oap

& o

ther

det

erge

nts,

exc

ept

spec

ialty

cle

aner

s2

1 74

9B

oCO

MM

Chi

naB

anks

HS

BC

Hol

ding

s P

LC{H

SB

C}

Uni

ted

Kin

gdom

Ban

ks3

1 63

7K

oram

Ban

kK

orea

, R

epub

lic o

fB

anks

Citi

grou

p In

cU

nite

d S

tate

sN

atio

nal c

omm

erci

al b

anks

4 9

84K

oram

Ban

kK

orea

, R

epub

lic o

fB

anks

Citi

grou

p In

cU

nite

d S

tate

sN

atio

nal c

omm

erci

al b

anks

5 8

24H

ynix

-Non

Mem

ory

Chi

p O

pK

orea

, R

epub

lic o

fS

emic

ondu

ctor

s an

d re

late

d de

vice

sC

itigr

oup

Ven

ture

Cap

ital

Uni

ted

Sta

tes

Inve

stor

s, n

ec6

562

NA

PO

CO

R-M

asin

loc

Pow

er P

lant

Phi

lippi

nes

Ele

ctri

c se

rvic

esY

NN

Pac

ific

Con

sort

ium

Inc

Aus

tral

iaIn

vest

ors,

nec

7 5

43B

ank

of A

sia

PC

LTh

aila

ndB

anks

UO

BS

inga

pore

Ban

ks8

529

Guo

co G

roup

Ltd

Hon

g K

ong,

Chi

naIn

vest

men

t ad

vice

Guo

line

Ove

rsea

s Lt

dH

ong

Kon

g, C

hina

Inve

stor

s, n

ec9

460

Har

bin

Bre

wer

y G

rp L

tdH

ong

Kon

g, C

hina

Mal

t be

vera

ges

Anh

euse

r-B

usch

Cos

Inc

Uni

ted

Sta

tes

Mal

t be

vera

ges

10 4

15In

tern

atio

nal B

ank

of A

sia

Hon

g K

ong,

Chi

naB

anks

Fubo

n Fi

nanc

ial H

oldi

ng C

o Lt

dTa

iwan

Pro

vinc

e of

Chi

naIn

vest

ors,

nec

11 3

91H

abib

Ban

k Lt

dP

akis

tan

Ban

ksA

ga K

han

Fund

for

Eco

nom

icS

wed

enIn

vest

men

t of

fices

, ne

c12

375

Hyu

ndai

Cap

ital C

o Lt

dK

orea

, R

epub

lic o

fP

erso

nal c

redi

t in

stitu

tions

GE

Con

sum

er F

inan

ceU

nite

d S

tate

sIn

vest

men

t ad

vice

13 3

62Ti

ngyi

-Bre

wer

ies

Op(

13)

Hon

g K

ong,

Chi

naB

ottle

d &

can

ned

soft

dri

nks

& c

arbo

nate

d w

ater

sA

-I C

hina

Bre

wer

ies

Ltd

Japa

nB

ottle

d &

can

ned

soft

dri

nks

& c

arbo

nate

d w

ater

s14

326

Inte

rnet

Auc

tion

Co

Ltd

Kor

ea,

Rep

ublic

of

Bus

ines

s se

rvic

es,

nec

eBay

Inc

Uni

ted

Sta

tes

Cat

alog

and

mai

l-or

der

hous

es15

317

Glo

balS

anta

Fe-L

and

Dri

ll A

sts

Kuw

ait

Cru

de p

etro

leum

and

nat

ural

gas

Pre

cisi

on D

rilli

ng C

orp

Can

ada

Dri

lling

oil

and

gas

wel

ls16

309

Dig

ital G

loba

lSof

t Lt

dIn

dia

Pre

pack

aged

Sof

twar

eH

ewle

tt P

acka

rd L

eide

n B

VN

ethe

rlan

dsIn

vest

ors,

nec

17 3

05B

ank

Per

mat

a Tb

k P

TIn

done

sia

Ban

ksIn

vest

or G

roup

Uni

ted

Kin

gdom

Inve

stor

s, n

ec18

305

Hyu

ndai

Inv

estm

ent T

rust

& S

ecK

orea

, R

epub

lic o

fIn

vest

men

t ad

vice

Pru

dent

ial F

inan

cial

Inc

Uni

ted

Sta

tes

Inve

stm

ent

advi

ce19

304

Fort

is B

ank

Asi

a H

KH

ong

Kon

g, C

hina

Ban

ksIC

BC

(Asi

a)H

ong

Kon

g, C

hina

Ban

ks20

267

Suc

cess

ful R

oad

Cor

pH

ong

Kon

g, C

hina

Inve

stor

s, n

ecC

hina

Mer

chan

t H

ldg(

Pac

)Ltd

Sin

gapo

reH

otel

s an

d m

otel

s21

260

Hym

all C

omm

erci

al R

etai

l Gro

upC

hina

Gro

cery

sto

res

Tesc

o P

LCU

nite

d K

ingd

omG

roce

ry s

tore

s22

227

Hug

hes

Sof

twar

e S

yste

ms

Ltd

Indi

aC

ompu

ter

rela

ted

serv

ices

,nec

Flex

tron

ics

Inte

rnat

iona

l Ltd

Sin

gapo

reP

rint

ed c

ircu

it bo

ards

23 2

09In

dust

rial

Ban

k C

o Lt

dC

hina

Ban

ksH

ang

Sen

g B

ank

Ltd

Hon

g K

ong,

Chi

naB

anks

24 2

03S

cott

s S

hopp

ing

Cen

tre,

Asc

ott

Sin

gapo

reO

pera

tors

of

nonr

esid

entia

l bui

ldin

gsM

arco

Pol

o D

evel

opm

ents

Ltd

Sin

gapo

reH

otel

s an

d m

otel

s25

195

Sia

m N

issa

n A

utom

obile

Co

Ltd

Thai

land

Mot

or v

ehic

les

and

pass

enge

r ca

r bo

dies

Nis

san

Mot

or C

o Lt

dJa

pan

Mot

or v

ehic

les

and

pass

enge

r ca

r bo

dies

26 1

90Id

ea C

ellu

lar

Ltd

Indi

aR

adio

tele

phon

e co

mm

unic

atio

nsIn

vest

or G

roup

Sin

gapo

reIn

vest

ors,

nec

27 1

88C

hina

Hua

rong

Ass

et M

gmt-

Ass

etC

hina

Inve

stm

ent

advi

ceIn

vest

or G

roup

Uni

ted

Sta

tes

Inve

stor

s, n

ec28

170

Dak

sh e

Ser

vice

s P

vt L

tdIn

dia

Info

rmat

ion

retr

ieva

l ser

vice

sIB

M C

orp

Uni

ted

Sta

tes

Com

pute

r pr

ogra

mm

ing

serv

ices

29 1

59In

tern

et A

uctio

n C

o Lt

dK

orea

, R

epub

lic o

fB

usin

ess

serv

ices

, ne

ceB

ay I

ncU

nite

d S

tate

sC

atal

ogue

and

mai

l-or

der

hous

es30

149

She

nzhe

n D

vlp

Ban

k C

o Lt

dC

hina

Ban

ksN

ewbr

idge

Asi

a A

IV I

II L

PU

nite

d S

tate

sM

anag

emen

t in

vest

men

t of

fices

, op

en-e

nd31

144

Ban

k Li

ppo

Tbk

PT

Indo

nesi

aB

anks

Inve

stor

Gro

upS

witz

erla

ndIn

vest

ors,

nec

32 1

39Ja

wa

Pow

er P

TIn

done

sia

Ele

ctri

c se

rvic

esY

TL P

ower

Int

erna

tiona

l Bhd

Mal

aysi

aE

lect

ric

serv

ices

33 1

39G

loba

l Con

duit

Hol

ding

s Lt

dH

ong

Kon

g, C

hina

Inve

stor

s, n

ecA

nheu

ser-

Bus

ch C

os I

ncU

nite

d S

tate

sM

alt

beve

rage

s34

134

PTP

Gro

upC

hina

Rec

onst

itute

d w

ood

prod

ucts

Car

ter

Hol

t H

arve

y Lt

dN

ew Z

eala

ndLo

ggin

g35

132

Chi

na L

ion

Bre

win

g G

roup

Chi

naM

alt

beve

rage

sIn

terb

rew

SA

Bel

gium

Mal

t be

vera

ges

/...

281ANNEX A

An

ne

x t

ab

le A

.II.

1.

Cro

ss

-bo

rde

r M

&A

de

als

wit

h v

alu

es

of

ov

er

$1

00

mil

lio

n c

on

clu

de

d i

n d

ev

elo

pin

g a

nd

tra

ns

itio

n e

co

no

mie

s,

20

04

(c

on

clu

de

d)

Valu

eR

ank

($ m

illio

n)A

cqui

red

com

pany

Hos

t ec

onom

yIn

dust

ry o

f th

e ac

quir

ed c

ompa

nyA

cqui

ring

com

pany

Hom

e ec

onom

yIn

dust

ry o

f th

e ac

quir

ing

com

pany

36 1

32C

hina

Lio

n B

rew

ing

Gro

upC

hina

Mal

t be

vera

ges

Inte

rbre

w S

AB

elgi

umM

alt

beve

rage

s37

131

Jaya

Hol

ding

s Lt

dS

inga

pore

Dee

p se

a fo

reig

n tr

ansp

orta

tion

of f

reig

htS

ime

Dar

by E

aste

rn L

tdS

inga

pore

Insu

ranc

e ag

ents

, br

oker

s, a

nd s

ervi

ce38

129

e-S

erve

Int

erna

tiona

l Ltd

Indi

aIn

form

atio

n re

trie

val s

ervi

ces

Citi

bank

Ove

rsea

s In

vest

Cor

pU

nite

d S

tate

sIn

vest

ors,

nec

39 1

26A

dvan

tage

Ltd

Hon

g K

ong,

Chi

naIn

vest

ors,

nec

Sta

ndar

d C

hart

ered

Lin

ks(H

K)

Hon

g K

ong,

Chi

naIn

vest

ors,

nec

40 1

22P

arks

on V

entu

re P

te L

tdS

inga

pore

Inve

stor

s, n

ecLi

on D

iver

sifie

d H

oldi

ngs

Bhd

Mal

aysi

aIn

vest

ors,

nec

41 1

21C

&M

Com

mun

icat

ions

Co

Ltd

Kor

ea,

Rep

ublic

of

Cab

le a

nd o

ther

pay

tel

evis

ion

serv

ices

GS

Cap

ital P

artn

ers

2000

LP

Uni

ted

Sta

tes

Inve

stm

ent

advi

ce42

120

UM

Ci P

te L

tdS

inga

pore

Sem

icon

duct

ors

and

rela

ted

devi

ces

UM

CTa

iwan

Pro

vinc

e of

Chi

naS

emic

ondu

ctor

s an

d re

late

d de

vice

s43

120

3721

Net

wor

k S

oftw

are

Co

Ltd

Hon

g K

ong,

Chi

naIn

form

atio

n re

trie

val s

ervi

ces

Yaho

o! H

oldi

ngs(

Hon

g K

ong)

Ltd

Hon

g K

ong,

Chi

naIn

vest

ors,

nec

44 1

20S

tarw

ay M

anag

emen

t Lt

dH

ong

Kon

g, C

hina

Ele

ctri

c eq

uipm

ent,

nec

Rim

Hol

ding

s In

cU

nite

d S

tate

sC

ompu

ter

mai

nten

ance

and

rep

air

45 1

15P

T B

ank

Bua

na I

ndon

esia

Indo

nesi

aB

anks

UO

BS

inga

pore

Ban

ks46

110

Ctr

ip.c

om I

nter

natio

nal L

tdC

hina

Bus

ines

s se

rvic

es,

nec

Rak

uten

Inc

Japa

nIn

form

atio

n re

trie

val s

ervi

ces

47 1

07B

aeky

ang

Tunn

el L

tdK

orea

, R

epub

lic o

fB

ridg

e, t

unne

l, an

d el

evat

ed h

ighw

ay c

onst

ruct

ion

Kor

ean

Roa

d In

fras

truc

ture

Kor

ea,

Rep

ublic

of

Inve

stm

ent

advi

ce48

107

Bri

zay

Pro

pert

y P

te L

tdS

inga

pore

Land

sub

divi

ders

and

dev

elop

ers,

exc

ept

cem

eter

ies

TC S

ervi

ces

@ W

ilby

Pte

Ltd

Sin

gapo

reIn

vest

ors,

nec

49 1

05M

utur

i PS

C,W

est

Pap

uaIn

done

sia

Cru

de p

etro

leum

and

nat

ural

gas

CN

OO

C M

utur

i Ltd

Indo

nesi

aIn

vest

ors,

nec

50 1

03C

hina

Cin

da-N

on-P

erfo

rmin

gC

hina

Inve

stm

ent

advi

ceS

ilver

Gra

nt I

ntl A

sts

Mng

t C

oH

ong

Kon

g, C

hina

Inve

stor

s, n

ec51

102

Thai

Am

arit-

Thai

land

Ass

ets

Thai

land

Mal

t be

vera

ges

San

Mig

uel C

orp

Phi

lippi

nes

Mal

t52

101

Dae

woo

Com

mer

cial

Veh

icle

Co

Kor

ea,

Rep

ublic

of

Mot

or v

ehic

les

and

pass

enge

r ca

r bo

dies

Tata

Mot

ors

Ltd

Indi

aM

otor

veh

icle

s an

d pa

ssen

ger

car

bodi

es53

101

Adi

ra D

inam

ika

Mul

tifin

ance

Indo

nesi

aP

erso

nal c

redi

t in

stitu

tions

Ban

k D

anam

on T

bk P

TIn

done

sia

Ban

ks54

101

Ban

k of

Asi

a P

CL

Thai

land

Ban

ksU

OB

Sin

gapo

reB

anks

55 1

00G

lovi

s C

o Lt

dK

orea

, R

epub

lic o

fTr

ucki

ng,

exce

pt lo

cal

Wilh

elm

Wilh

elm

sen

AS

AN

orw

ayD

eep

sea

fore

ign

tran

spor

tatio

n of

fre

ight

Sou

th-E

ast

Eur

ope

and

CIS

11

454

Mob

ilTel

AG

Bul

gari

aR

adio

tele

phon

e co

mm

unic

atio

nsB

idC

o A

DA

ustr

iaS

peci

al p

urpo

se f

inan

ce c

ompa

ny2

1 43

0V

impe

lcom

OJS

CR

ussi

an F

eder

atio

nR

adio

tele

phon

e co

mm

unic

atio

nsA

lfa T

elec

om L

tdB

ritis

h V

irgi

n Is

land

sTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

31

350

NK

Sla

vnef

tR

ussi

an F

eder

atio

nO

il an

d ga

s fie

ld e

xplo

ratio

n se

rvic

esB

P P

LCU

nite

d K

ingd

omC

rude

pet

role

um a

nd n

atur

al g

as4

1 00

4P

etro

m S

AR

oman

iaC

rude

pet

role

um a

nd n

atur

al g

asO

MV

AG

Aus

tria

Cru

de p

etro

leum

and

nat

ural

gas

5 8

09P

etro

m S

AR

oman

iaC

rude

pet

role

um a

nd n

atur

al g

asO

MV

AG

Aus

tria

Cru

de p

etro

leum

and

nat

ural

gas

6 6

25S

vyaz

inve

st J

SC

Rus

sian

Fed

erat

ion

Tele

phon

e co

mm

unic

atio

ns,

exce

pt r

adio

tele

phon

eA

cces

s In

dust

ries

Inc

Uni

ted

Sta

tes

Ele

vato

rs a

nd m

ovin

g st

airw

ays

7 4

25K

ar-T

el L

tdK

azak

hsta

nR

adio

tele

phon

e co

mm

unic

atio

nsV

impe

lcom

OJS

CR

ussi

an F

eder

atio

nR

adio

tele

phon

e co

mm

unic

atio

ns8

349

Bul

gari

an T

elec

omm

unic

atio

nsB

ulga

ria

Rad

iote

leph

one

com

mun

icat

ions

Viv

a Ve

ntur

es H

oldi

ng G

mbH

Aus

tria

Inve

stor

s, n

ec9

345

Cap

ital E

lect

rici

ty C

oB

ulga

ria

Ele

ctri

c se

rvic

esC

EZ

asC

zech

Rep

ublic

Ele

ctri

c se

rvic

es10

326

Plo

vdiv

EA

DB

ulga

ria

Ele

ctri

c se

rvic

esE

VN

AG

Aus

tria

Ele

ctri

c se

rvic

es11

279

Mob

iFon

SA

Rom

ania

Rad

iote

leph

one

com

mun

icat

ions

Tele

syst

em I

nt W

irel

ess

Inc

Can

ada

Rad

iote

leph

one

com

mun

icat

ions

12 2

05B

alka

n S

tar

JSC

Rus

sian

Fed

erat

ion

Cig

ars

Alta

dis

SA

Spa

inC

igar

ette

s13

179

Gor

na O

yaho

vitz

a E

AD

Bul

gari

aE

lect

ric

serv

ices

E O

N E

nerg

ie A

GG

erm

any

Ele

ctri

c se

rvic

es14

150

Del

taB

ank

CJS

CR

ussi

an F

eder

atio

nB

anks

GE

Con

sum

er F

inan

ceU

nite

d S

tate

sIn

vest

men

t ad

vice

15 1

26A

lban

ian

Sav

ings

Ban

kA

lban

iaB

anks

Rai

ffeis

en Z

entr

alba

nk A

GA

ustr

iaB

anks

16 1

21U

zdun

robi

taU

zbek

ista

nTe

leph

one

com

mun

icat

ions

, ex

cept

rad

iote

leph

one

OA

O M

obile

Tel

esys

tem

sR

ussi

an F

eder

atio

nR

adio

tele

phon

e co

mm

unic

atio

ns17

103

Wel

or K

eram

aR

ussi

an F

eder

atio

nC

eram

ic w

all a

nd f

loor

tile

sM

araz

zi G

rupp

o C

eram

iche

SpA

Ital

yC

eram

ic w

all a

nd f

loor

tile

s

So

urc

e:

UN

CT

AD

, cr

oss

-bo

rde

r M

&A

da

tab

ase

.

No

te:

M&

A d

ea

ls w

ith

in t

he

sa

me

eco

no

my,

bu

t w

he

re t

he

ult

ima

te p

are

nt

com

pa

ny

is d

iffe

ren

t, a

re s

till

co

nsi

de

red

cro

ss-b

ord

er

M&

As.

282 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.II.2. West Asia: selected FDI-related liberalization, 2004

Country Content

Iraq The maximum marginal tax rate on corporate income is limited to 15%.

Revised regulations by the Iraqi Central Bank oblige the prospective foreign bank to present feasibil itystudies of their planned activities in Iraq on how they could enhance the economy, especially in the areaof investment through loans, respecting laws that forbid money laundering and dealing in internationalterrorism money; and stipulate that the number of Iraqi employees in those banks should not be lessthan 80% of total staff, and that each bank should open at least three branches.

Order 64 amending Company Law No. (21) of 1997 states that in order to implement foreign investmentusing freely convertible currencies and Iraqi legal tender, foreign investors may establish a whollyforeign-owned company or economic establishment, including a branch or an office in Iraq, in allindustries except oil and mineral extraction. The Company Law now allows the incorporation of Iraqicompanies by foreign individuals and corporate entities.

Kuwait An amendment to the banking law has opened the banking sector to foreign participation.

A new taxation law on corporate taxation has reduced the maximum rate from 55% to 25%.

Oman A Royal Decree has increased the permitted level of foreign ownership in privatization projects to 100%.Privatization includes the conversion of a State-owned or mixed enterprise into a private sector firm andthe establishment of any new firm providing a commercial service that had previously been provided bythe State (e.g. electricity).

A Ministerial Decision allows foreign nationals to own real estate in tourist complexes in Oman.

Qatar Law No. 31/2004 allows foreign investment in the banking and insurance industries.

Law No. 17/2004 allows foreigners to own residential property in selected projects of the Pearl of theGulf Real Estate Development Project.

Saudi Arabia A new income tax law has reduced the previous graduated corporate tax rate to a flat 20%. Investmentsin certain strategic resources are stil l taxed at higher rates: 30% for gas and 85% for oil and hydrocarbons.The new executive bylaw to the new income tax has lowered the rate of taxation on foreign investors from45% to 20%. The law was imposed on foreign companies and individuals doing business in the Kingdom.

A new law for investment in the mining sector has simplified and streamlined the procedures forobtaining exploration and licences for mining and makes them more transparent.

Turkey Law No. 5035 amends some laws to accord tax incentives to the Technology Development ZonesManagement Company and the firms active in the zone.

Law 5084 revises the free zones law to effectively eliminate certain income and corporate tax immunitiesfor the zones.

Law 5177 abolishes the pre-licence period in the mining industry to reduce red tape. The amount oftaxation has been reduced by 50% on mining production that comes from domestic investors using theirown facilit ies and creating added value. The permission procedures in the mining industry shall beconcluded within three months.

Law 5189 removes the limit on foreign ownership of Türk Telekom. The privatization plan foresees ablock sale of 55% of company's shares.

Law No. 5228 amends Decree-law No.178 and some tax laws by expanding the scope of corporate taxexceptions.

United Arab The Dubai International Financial Centre (DIFC) has created a self-regulating financial free zone grantingEmirates 100% foreign ownership, zero tax rate and permission to repatriate capital and profits without

restrictions.

Source: UNCTAD, based on national sources.

283ANNEX A

Annex table A.II.3. New projects announced by TNCs in the non-oil mining and oil and gasindustries in Latin America, January 2004 - May 2005

Projectedamount

Company Home country Description of the project Host country ($ million)

Non-oil mining projects

Minmetals China Financing-based partnership with Codelco. Chile 2 000Barrick Canada Pascua-Lama copper and gold project,

scheduled to start production by 2009. Argentina/Chile 1 450Xstrata Switzerland Exploration and possible exploitation of

Las Bambas copper deposit.a Peru 1 163BHP Bill iton Australia Development of Spencer copper mine. Chile 1 000Rio Tinto Australia Expansion of its existing iron ore operations and

improvement of rail and water infrastructure. Brazil 1 000Minera Escondidab Australia Building of a copper cathode plant. Chile 870Phelps Dodge United States Expansion of Cerro Verde copper mine. Peru 850Northern Orion Canada Agua Rica Copper-gold-molybdenum project . Argentina 600CVRD Brazil Identification and evaluation of deposits of

potash in the province of Neuquen. Argentina ..

Oil and gas projects

Repsol-YPF Spain Investment plans for 2005-2009. Argentina 6 500Chevron Corp./Repsol-YPF United States Heavy oil in Orinoco Belt region. Venezuela 5 000Exxon Mobil United States Preliminary agreement for partnership Venezuela 2 500

with PDVSA to construct ethylene plantby the end of the decade.

Camisea Consortiumc Argentina Export of LNG to Mexico and the United States. Peru 2 000d

Repsol-YPF Spain Investment plans for 2005-2009. Bolivia and Brazil 2 000Petrobras Brazil Investment plans for 2004-2007. Argentina 1 600Repsol-YPF Spain Investment plans for 2005-2009. Trinidad & Tobago 1 250Repsol-YPF Spain Investment plans for 2005-2009. Venezuela 1 050Conoco Philips United States Duplication of production in

Corocoro oil f ield by 2009. Venezuela 850Chevron Corp. United States LNG liquefaction terminal.e Venezuela ..Total France Construction of a second Sincor synthetic

crude oil project by 2010.f Venezuela ..Chevron Corp. United States Natural gas import terminal. Mexico ..

Source: UNCTAD, based on press accounts.a Xstrata won a concession to develop the Las Bambas copper deposit with a $121 million bid. Bidding rules require

a minimum investment in exploration of $42 million over four years and $1 billion in the construction phase if reservesare found.

b Minera Escondida is controlled by the Australian BHP Billiton (57.5%) in partnership with the Australian Rio Tinto(30%), Japan’s Mitsubishi (10%) and the World Bank’s International Financial Corp (2.5%).

c Camisea Consortium, led by Pluspetrol (Argentina), includes Tecpetrol (Argentina), Hunt Oil (United States), SK(Republic of Korea) and Sonatrach (Algeria).

d This amount includes investments to be made in Mexico and the United States.e Chevron Corp. declared that this project will be launched if sufficient commercial gas is found in its two blocks in

Plataforma Deltana. At the end of 2004, the company announced that significant amounts of natural gas had beenfound in block 2 in Plataforma Deltana (www.chevrontexaco.com/news/press/2004/).

f Pending government approval. Total already operates the Sincor plant which has extra-heavy crude oil.

284 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.II.4. New projects announced by TNCs in the automobile industry in Argentina,Brazil and Mexico, January 2004 - May 2005

Projectedamount

Company Home country ($ million) Description of project

Argentina

Volkswagen Germany 200 Raise cars and parts production and produce new model, 95%destined for export.

PSA Peugeot-Citroen France 60 Begin assembling locally its 307 model currently imported fromFrance, 60% will be exported.

Daimler Chrysler Germany 38 New assembly line for the production of NCV3 (12,000 units peryear by 2007) — 100% destined for extraregional exports

Ford Motor United States 25 Launch a Mondeo model and expand dealership network.General Motors United States 20.5 Increase capacity.Daimler Chrysler Germany 12 Development of auto parts.Toyota Japan .. Introduce a new Hilux pickup truck (production started on

February 2005), and plans underway to produce an SUV modelfrom the second half of 2005; 70% of two models are destined forexport.

Brazil

Fiat Italy 490 Develop technologies, processes and new products.Bridgestone Firestone Japan 300 Build tyre plant near a Ford Motor plant.Continental Germany 270 Build tyre plant near a Ford Motor plant.General Motors United States 240 Expand car capacity for export.Hyundai Rep. of Korea 205 Set up new plant.Pirell i Italy 100 Increase production.Michellin France 98 Build a new earthmover tyre plant adjacent to its already existing

heavy truck site in Campo Grande.Deere United States 80 Raise production of tractors.Mitsubishi Motors Japan 44 ..Kia Rep. of Korea .. Build a factory to assemble light commercial vehicles.

Mexico

Volkswagen Germany 2 000 Investment plans 2003-2008.Ford Motor United States 1 200 Expand Hermosillo plant in 2004-2005; the new facility will open

during the second half of 2005.Bridgestone Firestone Japan 220 Build a plant in Nuevo Leon that makes high-performance radial

tyres for cars and vans.Toyota Japan 140 Build a plant in Baja California that makes trucks and truck beds.

It has been fired up in February 2005.Nissan Japan .. Expand Aguascalientes plant.

Source: UNCTAD, based on press accounts.

285ANNEX A

An

ne

x t

ab

le A

.II.

5.

Ind

us

try

co

mp

os

itio

n o

f F

DI

infl

ow

s i

n s

ele

cte

d S

ou

th-E

as

t E

uro

pe

an

co

un

trie

s a

nd

CIS

, 2

00

3,

20

04

(Mill

ion

s o

f d

olla

rs)

S

outh

-Eas

t E

urop

e

C

IS

T

FYR

Rus

sian

Bul

gari

aC

roat

iaa

M

aced

onia

Rom

ania

bK

azak

hsta

ncFe

dera

tion

dU

krai

nee

Sec

tor/

Indu

stry

2003

2004

2003

2004

2003

2004

2003

2004

2003

2004

2003

2004

2003

2004

Pri

mar

y24

2719

342

1232

192

190

5 43

01

768

2 71

788

7A

gric

ultu

re,

hunt

ing,

for

estr

y an

d fis

hing

31

....

26

3219

2-2

....

9421

Min

ing,

qua

rryi

ng a

nd p

etro

leum

2126

1934

-6

....

2 18

95

432

1 76

82

717

f-6

-13

Man

ufac

turi

ng59

139

256

5..

1654

849

2 00

61

000

516

2 34

23

830

621

426

Food

, be

vera

ges

and

toba

cco

8868

53..

42

....

4337

341

322

146

108

Text

iles,

clo

thin

g an

d le

athe

r84

43..

..3

1..

..2

1..

..20

20W

ood,

pub

lishi

ng,

pulp

and

pap

er78

-12

....

41

....

1010

200

330

6945

Cok

e, p

etro

leum

pro

duct

s an

d nu

clea

r fu

el17

0-3

g51

2..

21..

..19

720

....

f-1

128

Che

mic

als

and

chem

ical

pro

duct

s..

..g

....

16

....

625

....

5715

1R

ubbe

r an

d pl

astic

pro

duct

s..

..g

....

10.

2..

..6

9..

....

..N

on-m

etal

lic m

iner

al p

rodu

cts

7798

....

....

....

359

669

1 06

731

25M

etal

s an

d m

etal

pro

duct

s67

54..

..1

19..

..62

432

249

61

008

161

-32

Mac

hine

ry a

nd e

quip

men

t1

14..

....

-..

..3

825

652

7h

132

67E

lect

rica

l and

ele

ctro

nic

equi

pmen

t30

-1..

..0.

1-

....

7071

....

....

Mot

or v

ehic

les

and

othe

r tr

ansp

ort

equi

pmen

t2

-1..

..0.

2..

....

21

....

h..

..U

nspe

cifie

d m

anuf

actu

ring

-513

2..

..1

4..

..1

238

057

516

13S

ervi

ces

1 41

92

195

206

195

8297

568

1 12

51

418

2 47

82

665

2 81

062

91

064

Ele

ctri

city

, ga

s an

d w

ater

987

1..

..1

3..

..68

11..

..-3

5-1

4C

onst

ruct

ion

662

....

0.1

0.1

7-4

5115

485

138

2648

Trad

e49

739

071

935

828

235

616

426

810

5212

9919

642

3H

otel

s an

d re

stau

rant

s28

17..

177

761

-18

713

....

2669

Tran

spor

t, s

tora

ge a

nd c

omm

unic

atio

ns17

443

9..

1532

6328

176

7682

181

335

137

93Fi

nanc

e48

823

640

6131

9..

..53

6821

327

985

179

Rea

l est

ate

and

busi

ness

act

iviti

es19

117

595

105

7..

..99

51

792

1 13

440

515

419

3P

ublic

adm

inis

trat

ion

and

defe

nce

....

....

....

....

....

-35

4-0

.1-

Edu

catio

n, h

ealth

and

soc

ial s

ervi

ces

284

....

....

....

491

....

1718

Uns

peci

fied

serv

ices

....

....

10.

219

061

5..

....

..25

55U

nspe

cifi

ed55

1197

90-

0.3

-22

58..

..6

64-

-To

tal

2 08

92

625

888

319

9916

31

427

3 20

84

608

8 42

46

781

9 42

01

337

1 49

7

So

urc

e:

UN

CT

AD

, b

ase

d o

n n

ati

on

al

sou

rce

s.a

Eq

uit

y in

vest

me

nt

on

ly.

bC

alc

ula

tio

ns

ba

sed

on

th

e d

ata

of

the

Na

tio

na

l T

rad

e R

eg

iste

r O

ffic

e o

f R

om

an

ia;

the

to

tals

do

no

t n

ece

ssa

rily

co

rre

spo

nd

to

FD

I fl

ow

s re

po

rte

d i

n t

he

ba

lan

ce o

f p

aym

en

ts.

cG

ross

FD

I in

flo

ws.

dC

alc

ula

tio

n b

ase

d o

n t

he

Ru

ssia

n F

ed

era

tio

n,

Sta

te S

tati

stic

al

Se

rvic

e,

Cu

rre

nt

Sta

tist

ica

l S

urv

ey:

Qu

art

erl

y M

ag

azi

ne

, N

o.

1 (

52

), 2

00

5.

No

te t

ha

t a

s o

f 2

00

1 t

he

Sta

teS

tati

stic

al

Se

rvic

e s

top

pe

d p

rovi

din

g i

nd

ust

ry b

rea

kdo

wn

s o

f d

ire

ct a

nd

po

rtfo

lio

in

vest

me

nt

– t

he

est

ima

tes

for

FD

I a

re p

roje

cte

d f

rom

to

tal

fore

ign

in

vest

me

nt.

eC

alc

ula

ted

fro

m H

un

ya 2

00

5.

fR

uss

ian

sta

tist

ics

sho

w t

he

ext

ract

ion

of

pe

tro

leu

m a

nd

th

e p

etr

ole

um

pro

du

cts

ind

ust

rie

s to

ge

the

r u

nd

er

the

he

ad

ing

of

“fu

el

ind

ust

ry”.

gC

oke

an

d p

etr

ole

um

, ch

em

ica

ls a

nd

ru

bb

er

an

d p

last

ic p

rod

uct

s a

re t

ake

n t

og

eth

er.

hM

ach

ine

ry a

nd

eq

uip

me

nt

incl

ud

es

mo

tor

veh

icle

s a

nd

oth

er

tra

nsp

ort

eq

uip

me

nt.

286 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.III

.1.

Illu

str

ati

ve

ma

trix

of

tec

hn

olo

gic

al

an

d o

rga

niz

ati

on

al

ca

pa

bil

itie

s w

ith

in f

irm

s

Inve

stm

ent

Prod

uctio

n m

anag

emen

t & e

ngin

eerin

g

Tec

hnol

ogic

al li

nkag

es

Mar

ketin

g

Nat

ure

of c

apab

ility

Indu

stria

l eng

inee

ring

bui

ldin

g st

rate

gyPr

e-Pr

ojec

tPr

oces

sPr

oduc

tan

d hu

man

reso

uce

and

effo

rtin

vest

men

t e

xecu

tion

engi

neer

ing

eng

inee

ring

deve

lopm

ent

Dom

estic

Fore

ign

Dom

estic

Fore

ign

So

urc

e:

UN

CT

AD

, b

ase

d o

n L

all

(1

99

2)

an

d F

igu

eir

ed

o (

20

01

).

No

te:

Th

is i

s o

nly

an

ill

ust

rati

ve l

ist

of

cap

ab

ilit

ies

wit

hin

a m

an

ufa

ctu

rin

g f

irm

. It

do

es

no

t in

clu

de

se

vera

l ty

pe

s o

f ca

pa

bil

ity,

su

ch a

s fi

na

nci

al

ma

na

ge

me

nt,

la

bo

ur

rela

tio

ns

an

d l

og

isti

cs.

Con

tinuo

us li

nks

with

R&D

inst

itutio

ns a

ndun

iver

sitie

s;lic

ensi

ng o

wn

tech

nolo

gy to

othe

rs; d

eepe

ning

of in

nova

tive

links

with

oth

er fi

rms;

spec

ializ

atio

n in

cont

ext o

f net

wor

ksan

d cl

uste

rs

Sim

ple,

rout

ine:

bas

edm

ainl

y on

inte

rnal

effo

rtan

dex

perie

nce

Pre-

feas

ibilit

yan

d fe

asib

ility

stud

ies,

site

sele

ctio

n,sc

hedu

ling,

arra

ngin

gfin

anci

ng

Rou

tine

engi

neer

ing

ofci

vil w

orks

;an

cilla

ry s

ervi

ces;

erec

tion

and

com

mis

sion

ing

Deb

uggi

ng p

lant

;ro

utin

e pr

oces

sco

ordi

natio

n; q

ualit

ym

anag

emen

t; ro

utin

em

aint

enan

ce;

proc

ess

qual

ityce

rtific

atio

n

Assi

mila

tion

of b

asic

prod

uct d

esig

n;pr

oduc

t qua

lity

man

agem

ent a

ndce

rtific

atio

n; m

inor

adap

tatio

ns to

mee

tm

arke

t nee

ds

Wor

kflo

wsc

hedu

ling;

tim

e/m

otio

n st

udie

s;in

nova

tive

man

agem

ent a

ndop

timiz

atio

n; s

kills

upgr

adin

g an

dtra

inin

g

Loca

l pro

cure

men

tsy

stem

s an

dpr

oced

ures

, dra

win

gon

ava

ilabl

ekn

owle

dge

from

inst

itutio

ns

Fore

ign

sour

cing

;in

form

atio

n fro

msu

pplie

rs;

indu

stry

netw

orki

ng;

acce

ssin

g pu

blic

info

rmat

ion

Mar

ket r

esea

rch;

dist

ribut

ion

and

serv

icin

gsy

stem

s; s

ome

adve

rtisi

ng

Expo

rt m

arke

tan

alys

is; l

inks

with

buy

ers

and

othe

r exp

ort

chan

nels

;de

sign

/pa

ckag

ing

capa

bilit

y

Adap

tive,

dupl

icat

ive:

base

d on

sear

ch,

expe

rimen

tatio

nan

d in

ter-f

irman

d ot

her

coop

erat

ion

Sear

ch fo

rso

urce

s of

tech

nolo

gy,

equi

pmen

t;co

ntra

ctne

gotia

tion

Equi

pmen

tpr

ocur

emen

t;de

taile

den

gine

erin

g; s

taff

recr

uitm

ent;

train

ing

and

capa

city

bui

ldin

g

Cap

acity

stre

tchi

ng;

adap

t & im

prov

ete

chno

logy

; use

new

prod

uctio

n te

chni

ques

(Jus

t in

Tim

e, T

otal

Qua

lity

Man

agem

ent,

etc.

); ro

utin

e pr

oces

sen

gine

erin

g;pr

even

tive

mai

nten

ance

Prod

uct q

ualit

y/de

sign

impr

ovem

ent;

licen

sing

new

tech

nolo

gy;

reve

rse

engi

neer

ing;

cont

inuo

usm

onito

ring

of g

loba

lte

chno

logi

es

Con

tinuo

us a

ndsy

stem

atic

prod

uctiv

ity a

naly

sis

and

ben

chm

arki

ng;

skill

aud

it an

dfo

rmal

ized

trai

ning

;su

pply

-cha

in/

logi

stic

sm

anag

emen

t;ad

vanc

ed in

vent

ory

cont

rol

Tech

nolo

gy tr

ansf

erto

and

from

loca

lsu

pplie

rs/b

uyer

s;co

ordi

natio

n in

desi

gn a

ndm

anuf

actu

re; l

inki

ngw

ith te

chno

logy

and

othe

r ins

titut

ions

;de

vlop

ing

capa

city

to ta

ke c

olle

ctiv

eac

tion

Verti

cal

tech

nolo

gytra

nsfe

r;sy

stem

atic

coor

dina

tion

ofin

tern

atio

nal

know

ledg

eso

urce

s; li

nks

with

tech

nolo

gyin

stitu

tions

over

seas

Ded

icat

edm

arke

ting

depa

rtmen

t;sy

stem

atic

mon

itorin

g,fe

edba

ckan

alys

is;

bran

ding

and

diffe

rent

iatio

n

Syst

emat

icm

arke

t-bui

ldin

gan

d an

alys

is o

ffo

reig

n m

arke

ts;

allia

nces

and

netw

orks

abr

oad;

bran

din

trodu

ctio

n;O

EMar

rang

emen

ts

Inno

vativ

e,ris

ky: b

ased

on p

urpo

sive

effo

rt, R

&Dan

d ad

vanc

edfo

rms

ofco

llabo

ratio

n

Ow

n pr

ojec

tou

tline

and

desi

gnca

pabi

lity;

build

ing

ofw

orld

cla

sspr

ojec

tm

anag

emen

tca

pabi

litie

s

Basi

c pr

oces

sen

gine

erin

g,eq

uipm

ent

desi

gn a

nd s

tart

up; t

urnk

eyca

pabi

lity

Con

tinuo

us p

roce

ssim

prov

emen

t; pr

oces

sin

nova

tion;

bas

icre

sear

ch; u

se o

f new

proc

ess

desi

gnm

etho

ds;

enha

ncin

gor

gani

zatio

nal

capa

city

for

gene

ratin

g, c

odify

ing,

soci

aliz

ing

know

ledg

e

Mas

tery

of p

rodu

ctde

sign

met

hods

;pr

oduc

t inn

ovat

ion;

basi

c re

sear

ch;

stra

tegi

c al

lianc

es;

enha

ncem

ent o

for

gani

zatio

nal

capa

city

for

inno

vatio

n an

d ris

kta

king

Wor

ld-c

lass

indu

stria

len

gine

erin

g an

dsu

pply

-cha

inca

pabi

litie

s, tr

aini

ngsy

stem

s, in

vent

ory

man

agem

ent

AdvancedIntermediateBasic

Coo

pera

tive

R&D

;st

rate

gic

allia

nces

;ad

vanc

edle

vera

ging

stra

tegi

es fo

r new

tech

nolo

gies

;fo

reig

nac

quis

ition

s; d

irect

inve

stm

ent

Adva

nced

bra

ndcr

eatio

n;co

ordi

natio

nw

ith r

etai

lers

/bu

yers

;ad

vanc

eddi

strib

utio

nsy

stem

s

Bran

d de

epen

ing;

orig

inal

des

ign

man

agem

ent

and

orga

niza

tiona

lbe

havi

our

man

agem

ent

arra

ngem

ents

;ow

n m

arke

ting

and

desi

gnch

anne

ls a

ndaf

filia

tes

abro

ad

287ANNEX A

Annex table A.III.2. Gross domestic expenditure on R&D (GERD) andbusiness enterprise R&D (BERD), 1991-2003

(Millions of dollars)

GERD BERD

Region/economy 1991 1996 2001 2002 a 2003 1991 1996 2001 2002 a 2003

World 438 092 575 612 661 473 676 514 .. 291 485 376 343 437 459 449 818 ..

Developed countries 426 958 531 128 604 914 619 403 .. 289 450 355 914 416 107 417 881 ..Western Europe 147 761 174 709 169 200 184 421 .. 94 987 109 988 107 910 116 657 ..

European Union 139 274 163 920 159 926 174 651 .. 89 379 102 812 101 476 110 005 ..EU-15 138 157 161 427 156 877 171 279 .. 88 691 101 658 100 098 108 651 121 875Austria 2 608 3 664 3 931 4 506 5 532 1 508 c 2 422 e .. .. ..Belgium 3 442 4 743 4 935 5 471 7 038 2 289 3 395 3 635 4 012 5 212Denmark 2 204 3 390 3 823 4 346 .. 1 290 2 065 2 627 3 010 ..Finland 2 120 3 179 4 133 4 546 5 665 1 208 2 104 2 939 3 176 3 982France 30 810 35 344 29 429 32 495 .. 18 942 21 752 18 597 20 553 23 988Germany 46 899 52 274 46 534 50 222 61 296 32 522 34 551 32 511 34 775 42 786Greece 216 558 d 762 .. .. 56 136 249 270 351Ireland 435 969 1 150 1 351 .. 277 685 806 930 1 214Italy 11 300 12 562 12 145 13 740 .. 6 306 6 720 5 960 6 641 8 082Luxembourg .. .. 335 f .. .. .. .. 310 f .. ..Netherlands 6 250 8 056 7 239 .. .. 3 104 4 244 4 217 4 276 ..Portugal .. 654 d 929 1 132 .. 113 b 147 d 296 390 ..Spain 3 570 4 892 5 572 6 770 9 269 1 999 2 365 2 918 3 695 5 015Sweden 6 905 8 776 d 9 371 .. 12 010 4 729 6 569 d 7 274 .. 8 899United Kingdom 21 396 22 367 26 588 29 328 .. 14 347 14 505 17 758 19 649 22 347New EU members 1 117 2 493 3 049 3 372 3 450 688 1 154 1 379 1 354 1 452Cyprus .. 21 e 25 32 43 .. .. .. .. ..Czech Republic 516 599 745 903 1 143 358 359 448 551 697Estonia .. 25 44 52 76 .. 6 e 15 16 26Hungary 358 294 491 665 784 148 127 197 236 288Latvia .. 24 34 39 42 .. .. 12 16 15Lithuania .. 41 82 94 .. .. 1 24 16 ..Malta .. .. 3 f .. .. .. .. .. .. ..Poland .. 1 024 1 187 1 108 1 172 .. 419 425 225 321Slovakia 244 193 134 140 191 182 108 90 90 105Slovenia .. 272 306 339 .. .. 133 167 203 ..

Other Western Europe 8 486 10 790 9 273 9 770 .. 5 608 7 175 6 433 6 652 ..Iceland 78 136 e 234 263 .. 17 .. 138 150 ..Norway 1 944 2 571 d 2 718 3 186 .. 1 061 1 463 d 1 624 1 830 ..Switzerland 6 464 b 8 082 6 321 f .. .. 4 530 b 5 712 4 672 f .. ..

North America 170 291 207 421 287 845 290 015 300 608 119 349 148 235 207 446 202 320 204 922Canada 9 400 10 133 14 280 13 830 16 024 4 674 5 864 8 941 7 890 8 810United States 160 891 197 288 273 565 276 185 284 584 114 675 142 371 198 505 194 430 196 112

Other developed countries 108 906 148 998 147 869 144 966 .. 75 115 97 691 100 752 98 904 ..Australia 4 761 6 881 5 997 f .. .. 1 842 3 314 2 868 .. ..Israel 1 499 2 883 5 376 .. .. 835 1 745 3 512 .. ..Japan 102 233 138 623 136 000 132 988 144 947 72 328 92 466 94 225 92 328 101 429New Zealand 412 611 496 f 605 822 111 165 147 f 196 304

Developing economies 10 893 39 519 51 877 51 616 .. 2 035 17 561 18 656 28 760 ..Africa .. 1 001 1 217 1 083 .. .. .. .. .. ..

Burkina Faso .. 4 .. .. .. .. .. .. .. ..Cape Verde .. - d - - .. .. .. .. .. ..Egypt .. 144 189 f .. .. .. .. .. .. ..Madagascar .. 8 d 5 f .. .. .. .. .. .. ..Mauritius .. 13 d .. .. .. .. .. .. .. ..Seychelles .. .. .. 1 .. .. .. .. .. ..South Africa .. 742 e 871 710 .. .. .. .. .. ..Tunisia .. 60 106 132 .. .. .. .. .. ..Uganda .. 30 46 .. .. .. .. .. .. ..Zambia .. - .. .. .. .. .. .. .. ..

Latin America and the Caribbean 1 265 9 383 10 942 9 114 .. 205 3 464 3 564 2 960 ..South America 244 8 181 8 186 6 079 .. 58 3 218 2 785 2 182 ..

Argentina .. 1 137 1 141 397 532 .. 294 260 103 154Bolivia 21 b 24 24 22 .. .. 6 6 5 ..Brazil .. 6 004 5 855 f .. 4 647 .. 2 733 2 389 f .. 1 876Chile 208 400 366 473 .. 58 89 99 165 ..Colombia .. 291 136 81 .. .. 87 25 .. ..

/...

288 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.III.2. Gross domestic expenditure on R&D (GERD) andbusiness enterprise R&D (BERD), 1991-2003 (concluded)

(Millions of dollars)

GERD BERD

Region/economy 1991 1996 2001 2002 a 2003 1991 1996 2001 2002 a 2003

Ecuador .. 19 .. .. .. .. 1 .. .. ..Paraguay .. .. 6 5 .. .. .. .. .. ..Peru .. 49 d 58 58 .. .. 7 d 6 6 ..Uruguay 15 54 48 f 32 .. 1 - - f .. ..Venezuela .. 204 553 362 .. .. .. .. .. ..

Other Latin America and theCaribbean 1 021 1 201 2 756 3 035 .. 147 246 779 779 ..

Costa Rica .. 35 62 f .. .. .. 8 14 f .. ..Cuba 111 87 179 190 .. .. .. .. .. ..El Salvador .. 10 e .. .. .. .. - e .. .. ..Honduras .. .. 3 f .. .. .. .. .. .. ..Jamaica .. .. 5 6 .. .. .. .. .. ..Mexico 887 c 1 030 2 453 2 719 .. 147 c 236 763 .. ..Nicaragua .. 3 d .. 2 .. .. .. .. .. ..Panama 22 31 45 44 44 .. 1 .. .. ..Saint Vincent and the Grenadines .. .. - 1 .. .. .. .. .. ..Trinidad and Tobago .. 6 9 .. .. .. 1 1 .. ..

Asia and Oceania 9 628 29 135 39 717 41 419 .. 1 829 14 097 15 092 25 799 ..West Asia 837 903 1 286 1 378 .. 176 229 385 381 ..

Kuwait 39 57 232 155 158 7 b 17 29 30 ..Syrian Arab Republic .. 27 d .. .. .. .. .. .. .. ..Turkey 798 819 1 054 1 223 .. 168 213 356 351 ..

South, East andSouth-East Asia 8 791 28 232 38 432 40 041 41 600 1 654 13 868 14 707 25 418 23 920

China .. 4 865 12 595 15 556 18 601 .. .. .. 9 520 11 601Hong Kong, China .. 723 e 909 967 .. .. 206 267 321 ..India .. 2 112 3 743 .. .. .. .. .. .. ..Korea, Republic of 5 670 13 522 12 479 13 848 16 002 .. 9 899 9 507 10 371 12 177Malaysia .. 218 440 f 658 .. .. .. .. .. ..Mongolia .. 2 d 3 3 .. .. .. .. .. ..Nepal .. .. .. 36 .. .. .. .. .. ..Pakistan .. 92 d 113 164 .. .. .. .. .. ..Philippines 72 158 .. 107 .. 19 52 .. 71 ..Singapore .. 1 271 1 804 1 901 .. .. 804 1 141 1 168 ..Sri Lanka .. 26 .. .. .. .. .. .. .. ..Taiwan Province of China 3 049 5 024 6 064 6 491 6 997 1 635 2 906 3 792 3 966 ..Thailand .. 218 282 309 .. .. .. .. .. 143

South-East Europe and CIS 241 4 965 4 683 5 496 6 000 .. 2 868 2 696 3 177 ..South-East Europe 241 404 446 515 340 .. 215 111 119 154

Bulgaria .. 51 63 76 100 .. 30 13 14 20Croatia .. 89 213 255 .. .. .. .. .. ..Macedonia, TFYR .. 14 d 11 10 10 .. 2 d 1 - -Romania 241 c 249 158 174 230 .. 183 97 105 134Serbia and Montenegro .. - d - .. .. .. .. .. .. ..

CIS .. 4 561 4 237 4 981 5 660 2 652 2 585 3 058 ..Armenia .. 1 3 6 6 .. .. .. .. ..Azerbaijan .. 7 19 19 .. .. .. .. .. ..Belarus .. 103 d 88 91 109 .. 54 d 46 46 49Georgia .. 10 8 10 .. .. .. .. .. ..Kazakhstan .. 73 49 .. .. .. .. .. .. ..Kyrgyzstan .. 3 d 3 3 4 .. .. 1 2 3Moldova, Republic of .. 12 6 7 6 .. 2 1 1 4Russian Federation .. 3 753 3 609 4 307 5 534 .. 2 597 2 536 3 009 ..Ukraine .. 598 d 453 490 .. .. .. .. .. ..

Source: UNCTAD, based on national sources, OECD, Main Science and Technology Indicators, various issues, WorldBank, World Development Indicators, 2004, data from the Iberoamerican Web of Science and TechnologyIndicators (RICYT), and data from the UNESCO Institute of Statistics.

a Regional totals for 2002 have been complemented by data from 2001 or 2000 (and 2003 for Brazil) for countriesthat did not report R&D spending in 2002.

b 1992.c 1993.d 1997.e 1998.f 2000.

289ANNEX A

Annex table A.III.3. Patent applications from developing countries and South-East Europe andCIS in the United States, by residence of inventor, 1991-2003

(Period averages)

Share of Share of ChangeAverage foreign Average foreign between

Region/economy 1991-1993 (%) 2001-2003 (%) periods (%)

Developing economies 5 121 6.63 25 322 16.78 394.5Africa 221 0.29 257 0.17 16.1

Egypt 6 0.01 13 0.01 ..Kenya 2 - 12 0.01 ..South Africa 213 0.28 232 0.15 8.9

Latin America and the Caribbean 347 0.45 670 0.44 93.2Argentina 58 0.08 119 0.08 105.2Brazil 114 0.15 240 0.16 111.4Chile 12 0.02 31 0.02 168.6Colombia 7 0.01 22 0.01 ..Costa Rica 4 - 7 - ..Ecuador 3 - 8 0.01 ..Mexico 98 0.13 179 0.12 83.6Panama 2 - 7 - ..Peru 3 - 8 0.01 ..Uruguay 4 0.01 9 0.01 ..Venezuela 43 0.06 39 0.03 -7.8

Asia and Oceania 4 553 5.89 24 395 16.17 435.8West Asia 18 0.02 69 0.05 294.3

Saudi Arabia 13 0.02 32 0.02 137.5Turkey 3 - 31 0.02 ..United Arab Emirates 2 - 7 - ..

South, East and South-East Asia 4 536 5.87 24 326 16.12 436.3China 130 0.17 849 0.56 553.3Hong Kong, China 146 0.19 679 0.45 365.9India 56 0.07 909 0.60 1513.0Indonesia 10 0.01 13 0.01 37.9Korea, Republic of 1 472 1.91 8 356 5.54 467.6Malaysia 19 0.03 165 0.11 753.4Phil ippines 10 0.01 50 0.03 420.7Singapore 85 0.11 788 0.52 823.4Sri Lanka 10 0.01 64 0.04 536.7Taiwan Province of China 2 598 3.36 12 453 8.25 379.4

South-East Europe and CIS 157 0.20 480 0.32 205.5Belarus 4 0.01 8 0.01 ..Bulgaria 7 0.01 10 0.01 ..Croatia 2 - 20 0.01 ..Romania 3 - 11 0.01 ..Russian Federation 112 0.14 384 0.25 242.6Serbia and Montenegro 23 0.03 7 - -70.6Ukraine 7 0.01 40 0.03 ..

Memorandum:New EU members a 114 0.15 273 0.18 139.9Czech Republic - - 65 0.04 ..Hungary 83 0.11 116 0.08 40.7Lithuania - - 5 - ..Poland 21 0.03 48 0.03 125.0Slovakia - - 6 - ..Slovenia 9 0.01 31 0.02 ..

Developed countries a 71 805 92.94 124 905 82.77 73.9All foreign applications 77 263 100.00 150 899 100.00 95.3Domestic applications 93 445 .. 183 566 .. 96.4All applications 170 708 .. 334 465 .. 95.9

Source: United States Patent and Trademark Office, Information Products Division, Technology Assessment and ForecastBranch, special tabulations, Washington, DC, February 2005.

a In the new United Nations classification, the total for developed countries includes the new EU members under EU(box I.2).

290 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table A.III.4. Technological Activity Index

Rank 1995 2001 Rank 1995 2001

1 Sweden 0.981 Sweden 0.976 60 Zimbabwe 0.405 Egypt 0.3872 United States 0.963 Finland 0.973 61 Malaysia 0.401 Thailand 0.3613 Japan 0.949 Switzerland 0.955 62 Morocco 0.396 Kenya 0.3584 Switzerland 0.947 United States 0.948 63 China 0.390 Iran, Islamic Rep. of 0.3365 Finland 0.932 Japan 0.935 64 Qatar 0.362 Morocco 0.3326 Denmark 0.931 Denmark 0.917 65 Moldova, Rep. of 0.342 Zimbabwe 0.3277 Canada 0.930 Taiwan Province of China 0.902 66 Bahrain 0.340 India 0.3238 Norway 0.905 Canada 0.900 67 Thailand 0.340 Kyrgyzstan 0.3239 Australia 0.900 Iceland 0.895 68 Peru 0.332 Jamaica 0.31510 Taiwan Province of China 0.890 Germany 0.891 69 India 0.328 Bahrain 0.31111 Germany 0.887 Norway 0.890 70 Kazakhstan 0.320 Colombia 0.31112 United Kingdom 0.877 Singapore 0.875 71 Sri Lanka 0.304 Uruguay 0.29813 Netherlands 0.875 Netherlands 0.872 72 Honduras 0.296 Sri Lanka 0.29814 France 0.867 Australia 0.870 73 United Arab Emirates 0.294 United Arab Emirates 0.29015 Israel 0.858 Belgium 0.863 74 Tajikistan 0.288 Peru 0.28916 Belgium 0.848 United Kingdom 0.861 75 Colombia 0.288 Tunisia 0.28517 Iceland 0.843 France 0.849 76 Philippines 0.264 Syrian Arab Rep. 0.28118 Singapore 0.803 Israel 0.846 77 Dominican Rep. 0.255 Algeria 0.27819 Austria 0.798 Austria 0.830 78 Jordan 0.253 Qatar 0.27720 New Zealand 0.793 Korea, Rep. of 0.812 79 Iran, Islamic Rep. of 0.242 Moldova, Rep. of 0.27521 Russian Federation 0.792 New Zealand 0.802 80 Mongolia 0.238 Philippines 0.26522 Ireland 0.783 Ireland 0.781 81 Kyrgyzstan 0.237 Botswana 0.26123 Slovenia 0.766 Slovenia 0.764 82 Botswana 0.231 Mauritius 0.25724 Korea, Rep of. 0.762 Russian Federation 0.759 83 Tunisia 0.225 Ecuador 0.23525 Italy 0.753 Spain 0.744 84 Kenya 0.210 Tajikistan 0.23126 Estonia 0.734 Estonia 0.730 85 Indonesia 0.203 Viet Nam 0.23127 Spain 0.728 Italy 0.703 86 Pakistan 0.199 Tanzania, United Rep. of 0.22728 Belarus 0.721 Hungary 0.692 87 Namibia 0.185 Mongolia 0.22129 Hungary 0.696 Greece 0.681 88 El Salvador 0.181 El Salvador 0.20430 Greece 0.660 Czech Rep. 0.680 89 Oman 0.178 Madagascar 0.19531 Ukraine 0.653 Portugal 0.678 90 Viet Nam 0.162 Uganda 0.18532 Georgia 0.643 Lithuania 0.674 91 Benin 0.159 Namibia 0.18533 Poland 0.635 Hong Kong (China) 0.632 92 Algeria 0.155 Oman 0.17634 Lithuania 0.629 South Africa 0.621 93 Malawi 0.151 Indonesia 0.17535 Portugal 0.621 Belarus 0.618 94 Zambia 0.143 Pakistan 0.16936 Bulgaria 0.619 Jordan 0.606 95 Paraguay 0.127 Nigeria 0.16137 Hong Kong (China) 0.613 Argentina 0.603 96 Senegal 0.126 Bolivia 0.15538 Armenia 0.611 Bulgaria 0.602 97 Ghana 0.126 Ghana 0.13939 Argentina 0.609 Ukraine 0.600 98 Bolivia 0.122 Malawi 0.13040 Saudi Arabia 0.601 Poland 0.598 99 Ecuador 0.116 Benin 0.12241 Czech Rep. 0.597 Slovakia 0.588 100 Cameroon 0.113 Senegal 0.12042 Cyprus 0.597 Georgia 0.567 101 Nicaragua 0.111 Cameroon 0.10243 South Africa 0.588 Kuwait 0.564 102 Syrian Arab Rep. 0.111 Zambia 0.10144 Kuwait 0.576 Latvia 0.563 103 Guatemala 0.105 Côte d’Ivoire 0.09745 Chile 0.560 Cyprus 0.555 104 Tanzania, United Rep. of 0.105 Nicaragua 0.08146 Uruguay 0.558 Chile 0.544 105 Nigeria 0.104 Honduras 0.07647 Costa Rica 0.551 Armenia 0.543 106 Côte d’Ivoire 0.092 Paraguay 0.07548 Romania 0.539 Saudi Arabia 0.538 107 Uganda 0.079 Bangladesh 0.06349 Slovakia 0.504 Costa Rica 0.526 108 Djibouti 0.071 Ethiopia 0.05950 Venezuela 0.499 Romania 0.522 109 Bangladesh 0.069 Guatemala 0.05551 Uzbekistan 0.493 Lebanon 0.507 110 Ethiopia 0.063 Mozambique 0.04252 Lebanon 0.483 Brazil 0.478 111 Mauritania 0.038 Mauritania 0.03853 Mexico 0.474 Uzbekistan 0.472 112 Madagascar 0.033 Dominican Rep. 0.02954 Brazil 0.459 Mexico 0.461 113 Mozambique 0.021 Yemen 0.02155 Mauritius 0.457 Malaysia 0.446 114 Eritrea 0.017 Eritrea 0.01756 Egypt 0.430 Venezuela 0.438 115 Yemen 0.013 Angola 0.01357 Jamaica 0.419 Turkey 0.425 116 Haiti 0.008 Haiti 0.00858 Turkey 0.415 China 0.417 117 Angola 0.000 Djibouti 0.00059 Latvia 0.412 Kazakhstan 0.404

Sources: UNCTAD.

Note: Each component of the Index has equal weights, the Index value being the simple average of the normalizedvalue of the three variables: R&D manpower, patents in the United States and scientific journal articles.

Med

ium

-hig

h in

nova

tion

Hig

h in

nova

tion

Low

inno

vatio

n

M

ediu

m in

nova

tion

291ANNEX A

An

ne

x t

ab

le A

.III

.5.

Th

e H

um

an

Ca

pit

al

Ind

ex

Hig

hM

ediu

m

Low

1995

20

01

19

95

2001

199

5

2001

1A

ustr

alia

0.98

9Fi

nlan

d0.

982

41A

rgen

tina

0.67

0U

zbek

ista

n0.

655

81In

done

sia

0.34

9Ir

an,

Isla

mic

Rep

. of

0.35

52

Bel

gium

0.97

5S

wed

en0.

982

42S

lova

kia

0.65

7K

azak

hsta

n0.

646

82S

yria

n A

rab

Rep

.0.

340

El S

alva

dor

0.35

43

Can

ada

0.96

4A

ustr

alia

0.97

143

Mol

dova

, R

ep.

of0.

653

Bah

rain

0.62

283

Par

agua

y0.

339

Par

agua

y0.

351

4Fi

nlan

d0.

963

Nor

way

0.95

744

Phi

lippi

nes

0.64

1S

inga

pore

0.62

184

Nam

ibia

0.33

7In

done

sia

0.34

75

New

Zea

land

0.95

6N

ew Z

eala

nd0.

955

45S

inga

pore

0.63

5G

eorg

ia0.

619

85M

auri

tius

0.32

4A

lger

ia0.

347

6N

orw

ay0.

954

Uni

ted

Kin

gdom

0.95

146

Bah

rain

0.62

9Th

aila

nd0.

615

86C

hina

0.31

8S

ri L

anka

0.33

77

Uni

ted

Kin

gdom

0.95

1D

enm

ark

0.93

447

Tajik

ista

n0.

620

Chi

le0.

609

87N

icar

agua

0.31

3C

hina

0.29

88

Net

herl

ands

0.94

9B

elgi

um0.

924

48C

hile

0.60

1Le

bano

n0.

602

88A

lban

ia0.

308

Om

an0.

288

9D

enm

ark

0.93

8C

anad

a0.

914

49Le

bano

n0.

593

Rom

ania

0.58

689

Zim

babw

e0.

298

Nic

arag

ua0.

277

10Fr

ance

0.93

6U

nite

d S

tate

s0.

905

50Q

atar

0.58

0Jo

rdan

0.58

490

Bot

swan

a0.

297

Hon

dura

s0.

272

11S

wed

en0.

933

Net

herl

ands

0.90

451

Hon

g K

ong

(Chi

na)

0.57

3P

hilip

pine

s0.

581

91O

man

0.29

0N

amib

ia0.

251

12U

nite

d S

tate

s0.

929

Spa

in0.

895

52R

oman

ia0.

569

Bra

zil

0.57

992

Vie

t N

am0.

275

Indi

a0.

247

13A

ustr

ia0.

907

Fran

ce0.

877

53S

outh

Afr

ica

0.56

9B

oliv

ia0.

578

93H

ondu

ras

0.26

2Ye

men

0.23

914

Spa

in0.

900

Aus

tria

0.87

554

Cyp

rus

0.56

4C

ypru

s0.

577

94M

oroc

co0.

251

Zim

babw

e0.

229

15G

erm

any

0.89

2P

olan

d0.

867

55P

eru

0.56

3E

gypt

0.56

295

Indi

a0.

247

Mor

occo

0.22

216

Kor

ea,

Rep

of.

0.87

9K

orea

, R

ep.

of0.

866

56C

osta

Ric

a0.

558

Mon

golia

0.56

296

Gua

tem

ala

0.22

4G

uate

mal

a0.

215

17Ir

elan

d0.

875

Icel

and

0.85

757

Arm

enia

0.53

8P

eru

0.56

197

Nep

al0.

173

Syr

ian

Ara

b R

ep.

0.21

218

Japa

n0.

863

Irel

and

0.84

858

Vene

zuel

a0.

509

Mol

dova

, R

ep.

of0.

550

98N

iger

ia0.

169

Ban

glad

esh

0.18

019

Icel

and

0.82

6La

tvia

0.84

659

Kyr

gyzs

tan

0.50

4Q

atar

0.52

899

Côt

e d’

Ivoi

re0.

166

Nep

al0.

170

20B

elar

us0.

819

Slo

veni

a0.

838

60Th

aila

nd0.

485

Hon

g K

ong

(Chi

na)

0.51

410

0Za

mbi

a0.

157

Cam

eroo

n0.

167

21E

ston

ia0.

815

Japa

n0.

835

61E

gypt

0.46

9A

rmen

ia0.

509

101

Cam

eroo

n0.

152

Ken

ya0.

161

22Ta

iwan

Pro

vinc

e of

Chi

na0.

813

Taiw

an P

rovi

nce

of C

hina

0.82

962

Dom

inic

an R

ep.

0.45

9Ta

jikis

tan

0.49

310

2B

angl

ades

h0.

148

Côt

e d’

Ivoi

re0.

157

23It

aly

0.80

9E

ston

ia0.

820

63Ir

an,

Isla

mic

Rep

. of

0.45

6M

alay

sia

0.48

810

3Ye

men

0.14

6N

iger

ia0.

153

24G

reec

e0.

806

Rus

sian

Fed

erat

ion

0.81

764

Kuw

ait

0.45

4Ve

nezu

ela

0.48

210

4K

enya

0.13

7G

hana

0.14

825

Ukr

aine

0.80

4P

ortu

gal

0.81

465

Col

ombi

a0.

447

Mex

ico

0.47

710

5G

hana

0.13

6Za

mbi

a0.

130

26R

ussi

an F

eder

atio

n0.

802

Lith

uani

a0.

811

66Tu

rkey

0.44

6C

olom

bia

0.47

610

6P

akis

tan

0.12

2P

akis

tan

0.10

427

Pol

and

0.80

0U

krai

ne0.

810

67B

oliv

ia0.

443

Jam

aica

0.47

510

7M

auri

tani

a0.

103

Mau

rita

nia

0.09

828

Sw

itzer

land

0.79

4G

erm

any

0.81

068

Ecu

ador

0.44

2S

outh

Afr

ica

0.47

510

8H

aiti

0.10

2U

gand

a0.

095

29P

ortu

gal

0.78

7S

witz

erla

nd0.

799

69M

exic

o0.

433

Tuni

sia

0.44

510

9M

adag

asca

r0.

097

Eri

trea

0.09

230

Slo

veni

a0.

760

Gre

ece

0.79

470

Jord

an0.

426

Alb

ania

0.43

511

0S

eneg

al0.

083

Ben

in0.

090

31Is

rael

0.75

8It

aly

0.78

971

Mon

golia

0.40

5C

osta

Ric

a0.

419

111

Uga

nda

0.08

3H

aiti

0.08

332

Bul

gari

a0.

723

Bel

arus

0.77

672

Uni

ted

Ara

b E

mir

ates

0.39

9S

audi

Ara

bia

0.14

011

2B

enin

0.07

7S

eneg

al0.

081

33G

eorg

ia0.

722

Arg

entin

a0.

767

73S

audi

Ara

bia

0.39

1D

omin

ican

Rep

.0.

413

113

Eri

trea

0.07

7M

alaw

i0.

080

34K

azak

hsta

n0.

722

Isra

el0.

762

74M

alay

sia

0.38

5E

cuad

or0.

404

114

Mal

awi

0.06

6M

adag

asca

r0.

071

35U

zbek

ista

n0.

717

Hun

gary

0.75

875

Bra

zil

0.38

3M

auri

tius

0.38

911

5Ta

nzan

ia, U

nite

d R

ep. o

f0.

056

Tanz

ania

, Uni

ted

Rep

. of

0.06

336

Hun

gary

0.71

3B

ulga

ria

0.72

976

Tuni

sia

0.37

9K

uwai

t0.

383

116

Ang

ola

0.04

4D

jibou

ti0.

055

37Li

thua

nia

0.70

1U

rugu

ay0.

715

77E

l Sal

vado

r0.

371

Bot

swan

a0.

370

117

Djib

outi

0.04

3E

thio

pia

0.04

438

Cze

ch R

ep.

0.70

0C

zech

Rep

.0.

701

78Ja

mai

ca0.

369

Uni

ted

Ara

b E

mir

ates

0.36

311

8E

thio

pia

0.02

8A

ngol

a0.

025

39La

tvia

0.69

7K

yrgy

zsta

n0.

676

79S

ri L

anka

0.36

8V

iet

Nam

0.35

811

9M

ozam

biqu

e0.

015

Moz

ambi

que

0.01

940

Uru

guay

0.67

5S

lova

kia

0.66

480

Alg

eria

0.35

9Tu

rkey

0.35

5

So

urc

e:

UN

CT

AD

.

No

te:

Th

e H

um

an

Ca

pit

al

Ind

ex

is c

alc

ula

ted

fro

m t

he

lit

era

cy r

ate

(w

eig

ht

of

1),

se

con

da

ry e

nro

lme

nts

(w

eig

ht

of

2)

an

d t

ert

iary

en

rolm

en

ts i

n a

ll su

bje

cts

(we

igh

t o

f 3

).

292 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.IV

.1.

R&

D e

xp

en

dit

ure

by

fo

reig

n a

ffil

iate

s i

n s

ele

cte

d e

co

no

mie

s,

19

93

-20

03

(Mill

ion

s o

f d

olla

rs a

nd

pe

r ce

nt

of

bu

sin

ess

R&

D)

Eco

nom

yC

ateg

ory

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Arg

entin

aE

xpen

ditu

re..

2122

4243

5626

3843

24..

Sha

re (

%)

....

..14

.312

.015

.17.

111

.816

.523

.2..

Aus

tral

iaE

xpen

ditu

re..

..97

8..

....

1 09

0..

....

..S

hare

(%

)..

..30

.3..

....

41.1

....

....

Bra

zil

Exp

endi

ture

....

....

....

..1

145

....

898

Sha

re (

%)

....

....

....

..48

.0..

..47

.9C

anad

aE

xpen

ditu

re1

582

1 64

61

732

1 86

62

187

2 16

82

241

2 43

92

650

2 65

83

070

Sha

re (

%)

....

29.8

31.8

34.6

33.2

32.0

29.3

29.6

33.7

34.8

Chi

leE

xpen

ditu

re..

215

67

64

118

6..

Sha

re (

%)

..2.

314

.16.

710

.79.

36.

210

.88.

13.

6..

Chi

naE

xpen

ditu

re..

....

....

....

....

2 09

82

748

Sha

re (

%)

....

....

..18

.019

.221

.621

.722

.023

.7C

zech

Rep

ublic

Exp

endi

ture

....

7165

8514

111

815

220

323

932

5S

hare

(%

)..

..20

.918

.022

.130

.727

.436

.945

.343

.446

.6Fi

nlan

dE

xpen

ditu

re..

..25

0..

305

358

449

388

427

476

..S

hare

(%

)..

..13

.9..

14.0

14.2

15.9

13.4

14.5

15.0

..Fr

ance

Exp

endi

ture

..2

793

3 72

13

633

..3

238

....

4 00

63

986

..S

hare

(%

)..

14.2

17.1

16.7

..16

.4..

..21

.519

.4..

Ger

man

yE

xpen

ditu

re4

065

..4

554

..4

744

..5

501

..7

170

....

Sha

re (

%)

13.4

..13

.0..

14.5

..15

.4..

22.1

....

Gre

ece

Exp

endi

ture

6..

6..

5..

10..

....

..S

hare

(%

)6.

4..

3.7

..3.

8..

4.5

....

....

Hun

gary

Exp

endi

ture

1529

3156

9065

7111

314

115

518

0S

hare

(%

)12

.422

.621

.844

.465

.352

.753

.268

.471

.465

.562

.5In

dia

Exp

endi

ture

....

..48

5984

103

....

....

Sha

re (

%)

....

1.7

2.3

2.4

3.2

3.4

....

....

Irel

and

Exp

endi

ture

266

320

407

452

454

504

532

498

521

639

875

Sha

re (

%)

67.1

66.8

66.7

65.9

65.4

64.4

63.7

64.2

64.6

68.7

72.1

Isra

elE

xpen

ditu

re..

9697

169

208

141

389

630

726

889

..S

hare

(%

)..

7.9

6.7

9.7

10.0

6.1

14.3

17.5

20.7

....

Ital

yE

xpen

ditu

re..

....

....

....

..1’

964

....

Sha

re (

%)

....

....

....

....

33.0

....

Japa

nE

xpen

ditu

re70

21

319

1 36

586

21

140

1 38

63

666

3 63

63

197

....

Sha

re (

%)

0.9

1.5

1.4

0.9

1.3

1.7

3.9

3.6

3.4

....

Kor

ea,

Rep

.E

xpen

ditu

re..

1729

3441

2910

114

315

716

7..

Sha

re (

%)

..0.

20.

30.

30.

40.

51.

41.

61.

71.

6.. /...

293ANNEX A

An

ne

x t

ab

le A

.IV

.1.

R&

D e

xp

en

dit

ure

by

fo

reig

n a

ffil

iate

s i

n s

ele

cte

d e

co

no

mie

s,

19

93

-20

03

(c

on

clu

de

d)

(Mill

ion

s o

f d

olla

rs a

nd

pe

r ce

nt

of

bu

sin

ess

R&

D)

Eco

nom

yC

ateg

ory

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Mex

ico

Exp

endi

ture

..18

358

121

126

191

238

303

248

284

..S

hare

(%

)..

52.7

29.3

51.3

46.9

38.6

39.9

45.9

32.5

....

The

Net

herl

ands

Exp

endi

ture

....

....

857

885

983

1 07

11

042

....

Sha

re (

%)

....

....

20.4

21.2

21.7

26.1

24.7

....

Pol

and

Exp

endi

ture

....

....

4261

9752

6243

61S

hare

(%

)..

....

..10

.312

.720

.213

.114

.619

.219

.1P

ortu

gal

Exp

endi

ture

....

....

....

35..

91..

..S

hare

(%

)..

....

....

..17

.9..

30.9

....

Sin

gapo

reE

xpen

ditu

re..

....

....

....

..65

861

871

5S

hare

(%

)..

....

....

....

..57

.652

.959

.8S

lova

kia

Exp

endi

ture

..3

45

43

313

1619

20S

hare

(%

)..

4.1

3.9

4.4

2.3

2.9

3.2

15.2

18.1

20.7

19.0

Spa

inE

xpen

ditu

re74

2..

673

..79

8..

934

..98

11

223

1 37

1S

hare

(%

)39

.6..

30.0

..35

.7..

33.8

..33

.633

.127

.3S

wed

enE

xpen

ditu

re58

2..

1 19

3..

1 22

5..

2 50

8..

2 95

7..

4 03

2S

hare

(%

)13

.3..

19.3

..18

.7..

36.4

..40

.7..

45.3

Thai

land

Exp

endi

ture

....

....

....

....

....

40S

hare

(%

)..

....

....

....

....

..28

.1Tu

rkey

Exp

endi

ture

....

....

4526

3245

....

..S

hare

(%

)..

....

..14

.88.

47.

310

.6..

....

Uni

ted

Kin

gdom

Exp

endi

ture

..3

939

4 25

84

226

5 13

15

104

5 70

05

457

7 20

57

468

10 0

49S

hare

(%

)..

29.1

29.6

29.1

32.8

30.4

31.2

31.3

40.6

38.0

45.0

Uni

ted

Sta

tes

Exp

endi

ture

14 1

9915

566

17 5

4217

984

19 4

2825

373

24 0

2726

180

26 4

6327

508

..S

hare

(%

)12

.313

.313

.512

.612

.515

.213

.313

.213

.314

.1..

Mem

oran

dum

ite

ms:

Dev

elop

ed c

ount

ries

aE

xpen

ditu

re28

973

32 3

0336

778

37 7

0440

116

47 0

5551

304

56 3

4959

400

62 3

42..

Sha

re (

%)

10.6

11.2

11.4

11.3

11.9

13.6

13.6

13.8

14.9

15.7

..D

evel

opin

g co

untr

ies

Exp

endi

ture

223

223

172

295

321

392

1 64

92

446

4 40

24

135

..S

hare

(%

)2.

32.

31.

52.

32.

54.

111

.814

.318

.317

.7..

Eco

nom

ies

in t

rans

ition

bE

xpen

ditu

re18

104

106

167

220

269

288

331

422

455

..S

hare

(%

)9.

319

.018

.316

.820

.022

.925

.631

.136

.441

.3..

Est

imat

ed t

otal

Exp

endi

ture

29 2

1432

630

37 0

7538

166

40 6

5747

716

53 2

4159

125

64 2

2366

933

..S

hare

(%

)10

.311

.011

.111

.011

.613

.313

.513

.915

.215

.9..

So

urc

e:

UN

CT

AD

, b

ase

d o

n n

ati

on

al

sou

rce

s a

nd

da

ta p

rovi

de

d f

rom

th

e O

EC

D A

FA

da

tab

ase

.a

Exc

lud

ing

ne

w E

U m

em

be

rs.

bP

roxi

ed

by

da

ta f

or

fou

r n

ew

EU

me

mb

ers

: th

e C

zech

Re

pu

bli

c, H

un

ga

ry,

Po

lan

d a

nd

Slo

vaki

a.

No

te:

Th

e a

nn

ua

l to

tals

ha

ve b

ee

n e

stim

ate

d u

sin

g t

he

da

ta a

vail

ab

le f

or

the

giv

en

ye

ar;

wh

ere

n

o d

ata

we

re a

vail

ab

le,

the

da

ta o

f th

e p

rece

din

g,

or

sub

seq

ue

nt

yea

r,in

th

at

ord

er

of

pre

fere

nce

, h

ave

be

en

use

d.

294 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.IV

.2.

R&

D p

erf

orm

ed

by

fo

reig

n a

ffil

iate

s o

f U

nit

ed

Sta

tes

TN

Cs

by

co

un

try

an

d N

AIC

S i

nd

us

try,

20

02

(Mill

ion

s o

f d

olla

rs)

Com

pute

rsE

lect

rica

lP

rofe

ssio

nal,

Pri

mar

y an

dan

deq

uipm

ent,

Tran

spor

-sc

ient

ific

and

All

Tota

lfa

bric

ated

elec

tron

icap

plia

nces

,ta

tion

tec

hnic

alR

egio

n/ec

onom

yin

dust

ries

Man

ufac

turi

ngFo

odC

hem

ical

s m

etal

sM

achi

nery

prod

ucts

and

com

pone

nts

equ

ipm

ent

ITse

rvic

es

All

coun

trie

s21

151

18 6

9633

84

819

181

642

5 27

841

85

898

507

1 23

7D

evel

oped

eco

nom

ies

b15

528

213

4 40

614

654

63

505

b5

471

bb

Can

ada

2 34

52

272

2743

814

2551

013

1 17

029

16E

U-1

5b

11 4

5318

03

234

127

462

2 12

224

94

276

258

601

Japa

n1

433

1 28

34

732

550

375

b25

bb

Isra

el88

952

02

2-

949

8-

-56

b

Dev

elop

ing

econ

omie

s2

703

2 61

385

274

..65

1 69

9..

....

..A

sia

and

Oce

ania

2 11

31

955

990

..30

1 62

8..

....

..C

hina

646

609

133

-2

bb

1b

b

Hon

g K

ong

(Chi

na)

bb

-3

1-

b1

-b

b

Indi

a80

19-

8-

35

a3

bb

Indo

nesi

a3

3a

2-

-a

-a

--

Kor

ea,

Rep

ublic

of

167

149

210

a11

902

278

6M

alay

sia

bb

-1

-1

b-

--

-P

hilip

pine

s50

482

4-

a41

-a

--

Sin

gapo

re58

957

8-

11-

a55

05

b1

5Ta

iwan

Pro

vinc

e of

Chi

na70

b1

16b

925

-b

b1

Thai

land

2221

22

a4

3-

6-

-La

tin A

mer

ica

and

the

Car

ibbe

anb

633

7517

2b

3371

b18

9b

b

Bra

zil

306

298

3668

b28

30b

bb

3M

exic

o28

418

533

49-

52

1b

-b

Sub

-Sah

aran

Afr

ica

b25

112

-2

--

ba

a

Sou

th A

fric

ab

191

9-

a-

-b

aa

Uns

peci

fied

b55

540

139

3431

7414

720

141

363

7

So

urc

e:

UN

CT

AD

, b

ase

d o

n U

nit

ed

Sta

tes

Bu

rea

u o

f E

con

om

ic A

na

lysi

s, S

urv

ey

of

U.S

. D

ire

ct I

nve

stm

en

t A

bro

ad

, a

nn

ua

l se

rie

s, w

ww

.be

a.g

ov/

be

a.

aL

ess

th

an

ha

lf a

mil

lio

n.

bW

ith

he

ld t

o a

void

dis

clo

sin

g o

pe

rati

on

s o

f in

div

idu

al

com

pa

nie

s.

295ANNEX A

An

ne

x t

ab

le A

.IV

.3.

Se

lec

ted

ca

se

s o

f R

&D

by

fo

reig

n T

NC

s i

n S

ing

ap

ore

, M

arc

h 2

00

5

Year

TNC

Hom

e co

untr

yes

tabl

ishe

dR

&D

act

iviti

es in

Sin

gapo

re

AB

BS

witz

erla

nd20

02In

dust

rial

IT

R&

D c

entr

e ($

15 m

illio

n in

vest

ed).

Agi

lent

Tec

hnol

ogie

sU

nite

d S

tate

s..

R&

D c

entr

e on

pho

toni

cs (

150

desi

gn &

R&

D e

ngin

eers

em

ploy

ed;

half

of t

hem

exp

atri

ates

) .

Dai

mle

rChr

ysle

rG

erm

any

..E

nvir

onm

enta

l tes

ting

for

fuel

cel

l car

s .

Del

phi

Uni

ted

Sta

tes

1989

Sin

gapo

re D

esig

n an

d E

ngin

eeri

ng C

entr

e (S

DE

C)

desi

gns

adva

nced

ele

ctro

nic

com

pone

nts

(140

des

igne

rs f

rom

Sin

gapo

re a

nd t

he r

egio

n).

Eli

Lilly

Uni

ted

Sta

tes

2001

Cen

tre

for

Sys

tem

s B

iolo

gy (

LSB

) de

velo

ps c

ompu

tatio

nal t

ools

for

dru

g di

scov

ery

($14

0 m

illio

n to

be

inve

sted

in 2

005-

2009

).E

mer

son

Pro

cess

Man

agem

ent

Uni

ted

Sta

tes

..R

&D

on

adva

nced

pro

cess

con

trol

inst

rum

ents

, in

col

labo

ratio

n w

ith t

he p

aren

t en

gine

erin

g ce

ntre

($1

mill

ion

inve

sted

).E

rics

son

Sw

eden

..E

rics

son

Cyb

erla

b de

velo

ps,

for

exam

ple,

Chi

nese

Lan

guag

e S

MS

and

mes

sagi

ng (

15 e

mpl

oyee

s).

Ess

ilor

Fran

ce..

Opt

omet

ry R

&D

cen

tre

in S

inga

pore

is a

n in

tegr

al n

ode

in E

ssilo

r’s

glob

al R

&D

net

wor

k (w

ith F

ranc

e, t

he U

nite

d S

tate

s an

d Ja

pan)

(8

.em

ploy

ees,

to

grow

to

25 b

y 20

07).

Fujit

suJa

pan

..R

&D

Cen

tre

in S

inga

pore

, se

t up

in 1

989,

han

dles

res

earc

h in

tel

ecom

mun

icat

ion

soft

war

e, r

egio

nal t

elec

omm

unic

atio

ns s

uppo

rt a

s in

crea

tion

of s

witc

h co

nfig

urat

ion

data

base

, co

rpor

ate

netw

ork

syst

em f

or v

oice

, da

ta a

nd v

ideo

conf

eren

cing

. It

em

ploy

s ab

out

130

peop

le,

ofw

hich

80

are

engi

neer

s.G

empl

usFr

ance

2002

R&

D c

entr

e co

ncen

trat

es o

n se

curi

ty t

echn

olog

ies

and

the

adop

tion

of s

mar

t ca

rd a

pplic

atio

ns in

Asi

a ($

100

mill

ion

inve

sted

; 10

0 R

&D

sta

ff)G

laxo

Sm

ithK

line

Uni

ted

Kin

gdom

2004

Neu

rode

gene

rativ

e di

seas

e R

&D

cen

tre

(fir

st A

sian

pre

-clin

ical

res

earc

h fa

cilit

y; $

62 m

illio

n pl

anne

d in

vest

men

t).

Hew

lett

-Pac

kard

Uni

ted

Sta

tes

1991

Glo

bal p

rint

er a

nd in

kjet

pro

duct

de

velo

pmen

t (4

8 pa

tent

s re

gist

ered

in t

he U

nite

d S

tate

s; 1

80 e

ngin

eers

em

ploy

ed,

two-

thid

s fr

omS

inga

pore

).IB

MU

nite

d S

tate

s20

04Tw

o ra

dio

freq

uenc

y id

entif

icat

ion

cent

res,

in c

olla

bora

tion

with

Nan

yang

Pol

ytec

hnic

($1

2-m

illio

n pr

ojec

t).

Isis

Pha

rmac

eutic

als

Uni

ted

Sta

tes

..B

iote

chno

logy

res

earc

h .

Milt

enyi

Bio

tec

Ger

man

y..

App

licat

ions

in s

tem

cel

l res

earc

h, h

uman

gen

ome

stud

ies

and

cellu

lar

ther

apy.

Mot

orol

aU

nite

d S

tate

s..

Thre

e R

&D

cen

tres

in S

inga

pore

: S

inga

pore

Des

ign

Cen

tre

(150

eng

inee

rs),

Glo

bal S

oftw

are

Cen

tre,

Int

egra

ted

Cir

cuits

Des

ign

Cen

tre.

Nes

tléS

witz

erla

nd..

One

of

the

25 d

evel

opm

ent

cent

res

spre

ad o

ver

10 c

ount

ries

; fo

cuse

s on

pro

duct

spe

cific

atio

ns,

man

ufac

turi

ng p

roce

sses

and

env

iron

men

t-fr

iend

ly p

est

man

agem

ent

(ow

ns o

ver

100

pate

nts;

90

staf

f).

Nov

artis

Sw

itzer

land

2001

Nov

artis

Ins

titut

e of

Tro

pica

l Dis

ease

s de

velo

ps n

ovel

dru

gs f

or t

uber

culo

sis

and

deng

ue f

ever

.O

ptim

er P

harm

aceu

tical

sU

nite

d S

tate

s..

R&

D la

bora

tory

in S

inga

pore

dev

elop

s th

erap

eutic

s ba

sed

on c

arbo

hydr

ate

chem

istr

y.P

hilip

sN

ethe

rlan

ds..

Phi

lips

Inno

vatio

n C

ampu

s is

the

fir

m’s

larg

est

R&

D c

entr

e ou

tsid

e E

urop

e (m

ore

than

$13

0 m

illio

n in

vest

ed;

mor

e th

an 1

,000

eng

inee

rsfr

om m

ore

than

20

coun

trie

s).

Rho

dia

Fran

ce20

00N

ine

labs

at

Sci

ence

Par

k 2;

R&

D f

or a

pplic

atio

n in

bak

ing,

sau

ces,

sou

ps,

dres

sing

and

bev

erag

es,

chem

ical

s.R

olls

Roy

ceU

nite

d K

ingd

om20

04A

dvan

ced

tech

nolo

gy c

entr

e on

mat

eria

ls f

or a

ircr

aft

engi

nes,

sol

id o

xide

fue

l cel

ls.

Sie

men

sG

erm

any

..R

&D

on

info

rmat

ion

and

com

mun

icat

ions

tec

hnol

ogy

(300

jobs

will

be

crea

ted)

.S

ony

Sem

icon

duct

orJa

pan

1982

Thre

e R

&D

fac

ilitie

s in

Sin

gapo

re:

com

pone

nts

R&

D D

ivis

ion

(199

6),

Des

ign

Cen

tre

Asi

a (1

993)

, S

emic

ondu

ctor

Eng

inee

ring

Cen

tre

(199

4)(1

50 R

&D

em

ploy

ees)

.S

un M

icro

syst

ems

Uni

ted

Sta

tes

..R

&D

cen

tres

run

by

Sun

Mic

rosy

stem

s in

clud

e th

e Ja

va W

irel

ess

Com

pete

ncy

Cen

tre

at T

eleT

ech

Par

k, S

cien

ce P

ark

II,

and

the

Sun

Mic

rosy

stem

s A

sia

Pac

ific

Sci

ence

and

Tec

hnol

ogy

Cen

ter.

Was

eda-

Oly

mpu

s B

iosc

ienc

eJa

pan

..R

esea

rch

Inst

itute

foc

uses

on

high

er b

rain

fun

ctio

ns s

uch

as in

telle

ct a

nd a

war

enes

s .

Wye

th P

harm

aceu

tical

sU

nite

d S

tate

s..

R&

D c

entr

e ta

ps in

to lo

cal p

harm

aceu

tical

tec

hnol

ogie

s.

So

urc

e:

UN

CT

AD

, b

ase

d o

n T

oh

20

05

.

296 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

An

ne

x t

ab

le A

.V.1

. N

um

be

r o

f s

tud

en

ts e

nro

lle

d a

t th

e t

ert

iary

le

ve

l in

all

su

bje

cts

an

d t

ec

hn

ica

l s

ub

jec

ts (

sc

ien

ce

, e

ng

ine

eri

ng

,m

ath

em

ati

cs

, c

om

pu

tin

g),

20

00

-20

01

(Th

ou

san

ds)

Tota

l ter

tiary

Tert

iary

tech

nica

lTo

tal t

ertia

ryTe

rtia

ry te

chni

cal

Tota

l ter

tiary

Tert

iary

tech

nica

lTo

tal t

ertia

ryTe

rtia

ry te

chni

cal

1U

nite

d S

tate

s13

595

.6C

hina

2 58

0.4

27A

ustr

alia

845.

1R

oman

ia17

6.3

53C

zech

Rep

.26

0.0

Bel

gium

48.6

79P

arag

uay

83.1

Arm

enia

14.9

2C

hina

12 1

43.7

Rus

sian

Fed

erat

ion

2 38

8.0

28P

eru

824.

0A

lger

ia17

6.2

54B

ulga

ria24

7.0

Geo

rgia

46.4

80Ta

jikis

tan

78.5

Gha

na14

.0

3In

dia

9 83

4.0

Indi

a1

913.

029

Viet

Nam

749.

9B

elar

us14

4.5

55Tu

nisi

a20

7.4

Irel

and

43.9

81A

rmen

ia68

.7K

enya

14.0

4R

ussi

an F

eder

atio

n7

224.

0U

nite

d S

tate

s1

718.

530

Vene

zuel

a66

0.0

Kaz

akhs

tan

133.

756

Den

mar

k19

1.6

Sw

itzer

land

42.7

82C

amer

oon

68.5

Uru

guay

13.2

5Ja

pan

3 97

2.5

Kor

ea, R

ep. o

f1

000.

431

Sou

th A

fric

a64

4.8

Can

ada

131.

057

Kyr

gyzs

tan

190.

5S

audi

Ara

bia

41.6

83S

ri La

nka

65.0

Yem

en10

.4

6In

done

sia

3 01

7.9

Japa

n81

7.1

32M

alay

sia

549.

2C

hile

127.

058

Nor

way

189.

9Tu

nisi

a39

.684

Gha

na64

.1Zi

mba

bwe

10.4

7K

orea

, Rep

. of

3 00

3.5

Ukr

aine

643.

833

Rom

ania

533.

2G

reec

e12

1.0

59Ye

men

173.

1H

ong

Kon

g (C

hina

)39

.585

Uga

nda

62.6

Nic

arag

ua9.

1

8B

razi

l2

781.

3G

erm

any

636.

634

Net

herla

nds

504.

0B

angl

ades

h11

8.8

60N

ew Z

eala

nd17

2.0

Jord

an36

.886

Cos

ta R

ica

61.7

Cos

ta R

ica

8.2

9E

gypt

2 44

7.1

Indo

nesi

a58

5.6

35G

reec

e47

8.2

Per

u10

6.1

61Ir

elan

d16

6.6

Den

mar

k35

.487

Est

onia

57.8

Jam

aica

7.5

10P

hilip

pine

s2

432.

0M

exic

o57

6.8

36A

lger

ia45

6.4

Por

tuga

l94

.462

Sw

itzer

land

163.

4H

unga

ry35

.388

Zim

babw

e48

.9M

adag

asca

r6.

8

11G

erm

any

2 15

9.7

Uni

ted

Kin

gdom

496.

237

Chi

le45

2.2

Sw

eden

93.4

63Jo

rdan

162.

7N

orw

ay31

.989

Jam

aica

42.5

Tanz

ania

, Uni

ted

Rep

.of

6.2

12Th

aila

nd2

095.

7B

razi

l46

7.8

38K

azak

hsta

n44

5.7

Finl

and

92.7

64S

lova

kia

143.

9La

tvia

30.8

90A

lban

ia40

.9U

gand

a5.

9

13U

nite

d K

ingd

om2

067.

3S

pain

460.

739

Bel

arus

438.

0N

ethe

rland

s82

.165

Geo

rgia

140.

6M

oldo

va, R

ep. o

f30

.891

Nic

arag

ua36

.0Za

mbi

a5.

4

14M

exic

o2

047.

9Ir

an, I

slam

ic R

ep.

455.

640

Sau

di A

rabi

a43

2.3

Sou

th A

fric

a75

.366

Lith

uani

a13

5.9

New

Zea

land

25.4

92M

adag

asca

r31

.4S

eneg

al5.

4

15Fr

ance

2 03

1.7

Taiw

an P

rov.

of C

hina

368.

941

Por

tuga

l38

7.7

Mal

aysi

a74

.967

Leba

non

134.

0Ta

jikis

tan

22.0

93S

eneg

al29

.5A

lban

ia5.

1

16U

krai

ne1

950.

8Ita

ly36

8.2

42B

elgi

um35

9.3

Mor

occo

71.7

68H

ong

Kon

g (C

hina

)12

8.1

Est

onia

20.2

94Za

mbi

a24

.6C

ypru

s3.

0

17S

pain

1 83

3.5

Pol

and

360.

443

Sw

eden

358.

0P

akis

tan

71.3

69E

l Sal

vado

r11

4.7

Leba

non

19.5

95Ta

nzan

ia, U

nite

d R

ep. o

f22

.0O

man

2.4

18Ita

ly1

812.

3P

hilip

pine

s34

8.1

44P

akis

tan

340.

0M

oroc

co70

.670

Nep

al10

3.3

Hon

dura

s18

.996

Om

an19

.3B

otsw

ana

1.7

19P

olan

d1

775.

0Fr

ance

346.

145

Nig

eria

336.

0C

zech

Rep

.69

.271

Latv

ia10

2.8

Eth

iopi

a17

.497

Mau

ritiu

s12

.5M

aurit

ius

1.1

20A

rgen

tina

1 76

6.9

Egy

pt28

9.4

46H

unga

ry33

0.5

Bol

ivia

64.7

72M

oldo

va, R

ep. o

f10

2.8

Cam

eroo

n17

.398

Cyp

rus

11.9

Erit

rea

1.1

21Tu

rkey

1 60

7.4

Turk

ey27

2.4

47M

oroc

co31

0.3

Nig

eria

63.6

73U

rugu

ay97

.5S

love

nia

17.3

99M

ozam

biqu

e9.

3Q

atar

0.9

22Ir

an, I

slam

ic R

ep.

1 56

9.8

Arg

entin

a26

8.1

48Fi

nlan

d27

9.6

Kyr

gyzs

tan

57.2

74K

enya

95.1

Par

agua

y16

.910

0M

aurit

ania

9.0

Mau

ritan

ia0.

7

23C

anad

a1

212.

2C

olom

bia

265.

549

Bol

ivia

278.

8B

ulga

ria56

.075

Slo

veni

a91

.5N

epal

16.3

101

Qat

ar7.

8M

alaw

i0.

4

24Ta

iwan

Pro

v. o

f Chi

na1

008.

2A

ustr

alia

227.

550

Isra

el27

1.0

Aus

tria

51.0

76H

ondu

ras

90.6

Sin

gapo

re15

.910

2B

otsw

ana

7.7

Moz

ambi

que

0.3

25C

olom

bia

934.

1Th

aila

nd18

6.0

51A

ustr

ia26

4.7

Isra

el50

.877

Eth

iopi

a87

.4S

ri La

nka

15.5

103

Erit

rea

5.5

Djib

outi

-

26B

angl

ades

h87

8.5

Vene

zuel

a18

4.8

52A

ustr

ia26

4.7

Slo

vaki

a50

.078

Sin

gapo

re85

.0E

l Sal

vado

r15

.010

4M

alaw

i3.

2

105

Djib

outi

0.5

So

urc

es:

UN

CT

AD

, b

ase

d o

n U

NE

SC

O a

nd

Ta

iwa

n P

rovi

nce

of

Ch

ina

sta

tist

ics.

No

te:

Da

ta o

n t

ert

iary

en

rolm

en

ts b

y su

bje

ct a

re n

ot

ava

ilab

le a

fte

r 1

99

7.

It i

s a

ssu

me

d h

ere

th

at

the

sh

are

of

tech

nic

al

sub

ject

s in

to

tal

tert

iary

en

rolm

en

ts w

as

the

sa

me

in 2

00

0/0

1 a

s in

19

97

. F

or

a n

um

be

r o

f e

con

om

ies

in S

ou

th-E

ast

Eu

rop

e a

nd

th

e C

IS,

da

ta a

re n

ot

ava

ila

ble

at

all

. In

th

ese

ca

ses,

it

is a

ssu

me

d t

ha

t th

e s

ha

res

of

tech

nic

al

to t

ota

l te

rtia

ry s

tud

en

ts w

ere

th

e s

am

e a

s in

th

e e

con

om

ies

in S

ou

th-E

ast

Eu

rop

e a

nd

th

e C

IS f

or

wh

ich

da

ta a

re a

vail

ab

le.

DEFINITIONS AND SOURCES

A. General definitions

1. Transnational corporations

Transnational corporations (TNCs) are incorporated or unincorporated enterprises comprising parententerprises and their foreign affiliates. A parent enterprise is defined as an enterprise that controls assetsof other entities in countries other than its home country, usually by owning a certain equity capital stake.An equity capital stake of 10 per cent or more of the ordinary shares or voting power for an incorporatedenterprise, or its equivalent for an unincorporated enterprise, is normally considered as the threshold forthe control of assets.1 A foreign affiliate is an incorporated or unincorporated enterprise in which an investor,who is a resident in another economy, owns a stake that permits a lasting interest in the management ofthat enterprise (an equity stake of 10 per cent for an incorporated enterprise, or its equivalent for anunincorporated enterprise). In WIR, subsidiary enterprises, associate enterprises and branches – definedbelow – are all referred to as foreign affiliates or affiliates.

• A subsidiary is an incorporated enterprise in the host country in which another entity directly ownsmore than a half of the shareholder’s voting power, and has the right to appoint or remove a majorityof the members of the administrative, management or supervisory body.

• An associate is an incorporated enterprise in the host country in which an investor owns a total ofat least 10 per cent, but not more than half, of the shareholders’ voting power.

• A branch is a wholly or jointly owned unincorporated enterprise in the host country which is oneof the following: (i) a permanent establishment or office of the foreign investor; (ii) an unincorporatedpartnership or joint venture between the foreign direct investor and one or more third parties; (iii)land, structures (except structures owned by government entities), and /or immovable equipment andobjects directly owned by a foreign resident; or (iv) mobile equipment (such as ships, aircraft, gas-or oil-drilling rigs) operating within a country, other than that of the foreign investor, for at leastone year.

2. Foreign direct investment

Foreign direct investment (FDI) is defined as an investment involving a long-term relationship andreflecting a lasting interest and control by a resident entity in one economy (foreign direct investor orparent enterprise) in an enterprise resident in an economy other than that of the foreign direct investor(FDI enterprise or affiliate enterprise or foreign affiliate).2 FDI implies that the investor exerts a significantdegree of influence on the management of the enterprise resident in the other economy. Such investmentinvolves both the initial transaction between the two entities and all subsequent transactions between themand among foreign affiliates, both incorporated and unincorporated. FDI may be undertaken by individualsas well as business entities.

Flows of FDI comprise capital provided (either directly or through other related enterprises) bya foreign direct investor to an FDI enterprise, or capital received from an FDI enterprise by a foreigndirect investor. FDI has three components: equity capital, reinvested earnings and intra-company loans.

• Equity capital is the foreign direct investor’s purchase of shares of an enterprise in a country otherthan its own.

• Reinvested earnings comprise the direct investor’s share (in proportion to direct equity participation)of earnings not distributed as dividends by affiliates, or earnings not remitted to the direct investor.Such retained profits by affiliates are reinvested.

• Intra-company loans or intra-company debt transactions refer to short- or long-term borrowing andlending of funds between direct investors (parent enterprises) and affiliate enterprises.

298 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

FDI stock is the value of the share of their capital and reserves (including retained profits) attributableto the parent enterprise, plus the net indebtedness of affiliates to the parent enterprise. FDI flow and stockdata used in WIR are not always defined as above, because these definitions are often not applicable todisaggregated FDI data. For example, in analysing geographical and industrial trends and patterns ofFDI, data based on approvals of FDI may also be used because they allow a disaggregation at the countryor industry level. Such cases are denoted accordingly.

3. Non-equity forms of investment

Foreign direct investors may also obtain an effective voice in the management of another businessentity through means other than acquiring an equity stake. These are non-equity forms of investment,and they include, inter alia, subcontracting, management contracts, turnkey arrangements, franchising,licensing and product-sharing. Data on these forms of transnational corporate activity are usually notseparately identified in the balance-of-payments statistics. These statistics, however, usually present dataon royalties and licensing fees, defined as “receipts and payments of residents and non-residents for: (i)the authorized use of intangible non-produced, non-financial assets and proprietary rights such as trademarks,copyrights, patents, processes, techniques, designs, manufacturing rights, franchises, etc., and (ii) the use,through licensing agreements, of produced originals or prototypes, such as manuscripts, films, etc.”3

B. Availability, limitations and estimates of FDI datapresented in WIR

FDI data have a number of limitations. This sections therefore spells out how UNCTAD collectsand reports such data. These limitations need to be kept in mind also when dealing with the size of TNCactivities and their impact.

1 . FDI flows

Data on FDI flows in annex table B.1, as well as in most of the tables in the text, are on a net basis(capital transactions’ credits less debits between direct investors and their foreign affiliates). Net decreasesin assets (outward FDI) or net increases in liabilities (inward FDI) are recorded as credits (recorded witha positive sign in the balance of payments), while net increases in assets or net decreases in liabilitiesare recorded as debits (recorded with a negative sign in the balance of payments). In the annex tables,as well as in the tables in the text, the negative signs are reversed for practical purposes in the case ofFDI outflows. Hence, FDI flows with a negative sign in WIR indicate that at least one of the three componentsof FDI (equity capital, reinvested earnings or intra-company loans) is negative and is not offset by positiveamounts of the other components. These are instances of reverse investment or disinvestment.

UNCTAD regularly collects published and unpublished national official FDI flows data directlyfrom central banks, statistical offices or national authorities on an aggregated and disaggregated basisfor its FDI/TNC database(www.unctad.org/fdistatistics). These data constitute the main source for reporteddata on FDI flows. These data are further complemented by data obtained from: (i) other internationalorganizations such as the International Monetary Fund (IMF), the World Bank and the Organisation forEconomic Co-operation and Development (OECD); (ii) regional organizations such as the ASEAN Secretariatand the European Bank for Reconstruction and Development (EBRD); (iii) Banque Centrale de l’Afriquede l’Ouest; (iv) Banque Centrale des Etats de l’Afrique Centrale and (vi) UNCTAD’s own estimates.

For those economies for which data were not available from national official sources, or for thosefor which data were not available for the entire period covered in the World Investment Report 2005 (WIR05),data from the IMF were obtained using the IMF’s CD-ROM on International Financial Statistics and Balanceof Payments, June 2005. If the data were not available from the above IMF data source, data from theIMF’s Country Report, under Article IV of the IMF’s Articles of Agreements, were used.

For those economies for which data were not available from national official sources and the IMF,or for those for which data were not available for the entire period, data from the World Bank’s WorldDevelopment Indicators Online were used. This report covers data up to 2003 and reports data on netFDI flows (FDI inflows less FDI outflows) and inward FDI flows only. Consequently, data on FDI outflows,which are reported as World Bank data, are estimated by subtracting inward FDI flows from net FDI flows.

299DEFINITIONS AND SOURCES

Data from the EBRD were utilized for those economies in the Commonwealth of Independent Statesfor which data were not available from one of the above-mentioned sources.

Furthermore, data on the FDI outflows of the OECD, as presented in its publication, GeographicalDistribution of Financial Flows to Developing Countries, and as obtained from its online databank, wereused as a proxy for FDI inflows. As these OECD data are based on FDI outflows to developing economiesfrom the member countries of the Development Assistance Committee (DAC) of OECD,4 inflows of FDIto developing economies may be underestimated. In some economies, FDI data from large recipients andinvestors are also used as proxies.

Finally, in those economies for which data were not available from either of the above-mentionedsources, or only partial data (quarterly or monthly) were available, estimates were made by: annualizingthe data, if they are only partially available (monthly or quarterly) from either the IMF or national officialsources; and using data on cross-border mergers and acquisitions (M&As) and their growth rates.

2. FDI stocks

Annex table B.2, as well as some tables in the text, present data on FDI stocks at book value orhistorical cost, reflecting prices at the time when the investment was made except for countries that reportstock at market value (e.g. Australia, Hong Kong (China)).

UNCTAD regularly collects published and unpublished national official FDI stock data directly fromcentral banks, statistical offices and/or national authorities on an aggregated and disaggregated basis forits FDI/TNC database. These data constitute the main source for the reported data on FDI stocks. Theyare further complemented by the data obtained from the IMF.

As for economies for which data were not available from national official sources, or for those forwhich data were not available for the entire period, data on international investment position assets andliabilities from the IMF’s CD-ROMs on International Financial Statistics and Balance of Payments, June2005, were used instead.

For a large number of economies, FDI stocks were estimated by either adding up FDI flows overa period of time, or adding or subtracting flows to an FDI stock that had been obtained for a particularyear from national official sources, or the IMF data series on assets and liabilities of direct investment,or by using the mirror data of FDI stock of major economies as proxy.

Details of how data on FDI flows and stocks were obtained for each economy used in the Report,are given in the WIR website (www.unctad.org/wir).

3. Special notes on recent changes in the methodology

a. FDI inflows

- Bahrain. FDI data cover only the financial sector.

- Belgium and Luxembourg Economic Union. Up to 2001, the Belgium National Bank reported FDIdata for the Belgium and Luxembourg Economic Union. As of 2002, this economic union is no longerin effect. Consequently, FDI data are reported separately by the respective national authorities.Therefore, data for 2002 onwards are not comparable to the combined flows as reported in previousyears because of different methodologies.

- China. Data from the Ministry of Commerce (MOFCOM) were used for FDI inflows in that country.These data are reported on a gross basis (or do not take into account debits of FDI inward transactions).FDI outflows data were obtained from State Administration of Foreign Exchange (SAFE).

- Egypt. FDI inflows started to include investment in the petroleum sector in the third quarter of2004.

- Republic of Korea. Data from the Ministry of Commerce, Industry and Energy (MOCIE) were usedfor FDI inflows in that country for the entire period 1980-2004, instead of those from the Bank ofKorea. The MOCIE’s data series include equity, long-term loans, investment in kind (i.e. provisionof technology and capital goods) and conversion of convertible bonds.

- Lesotho. The Lesotho Highland Water Project, is excluded from its FDI statistics as it is not consideredas foreign investment.

300 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

- Luxembourg. FDI flows data exclude investment by and from SPEs (holding companies and otherfinancial vehicles). However, data include transactions made by these SPEs.

- Macao (China). The data covers only eight main industries, namely: Industrial production; construction;wholesale and retail; hotels and restaurants; transport, storage and communications; financial services;cultural, recreational, gambling and other services.

- Malta. The direct reporting system was installed by the National Statistics Office and the Centralbank of Malta in 2003 for all sectors of its economy. This methodology is applied to data from 1995onwards. Consequently, FDI statistics record a break in the series since 1995.

- Netherlands. The new direct reporting system was introduced in April 2003 to improve the methodto record intra-company transactions in such a way that the Dutch National Bank (DNB) was ableto clearly differentiate between loans taken by or lent abroad by TNCs (including the parent, subsidiary,sister etc.).

- Oman, Saudi Arabia and Syrian Arab Republic. For the first time in 2004, after technical cooperationwas given by the Economic and Social Commission for Western Asia (ESCWA) and UNCTAD, a surveyon inward FDI was undertaken. Flow data from this survey were used.

- Philippines. The 5th edition of the Balance of Payments Manual (BPM5) was adopted in 2000 coveringdata starting 1999. There is a difference in coverage of data on direct investment flows from 1999onwards compared to those of prior years. In particular, the change in coverage pertains to inter-company loans. From 1999 onwards, direct investment flows include intra-company loans under the“other capital” component of direct investment, as spelled out in the BPM5 manual. Previously, intra-company loans were not part of direct investment but classified under the medium-and long-termloan accounts.

- United States. Data on FDI used in this Report do not include current cost adjustments, in otherwords, they are on a historical-cost basis.

b. FDI stocks

- Belgium. Stock data are estimated by subtracting the reported stock of Luxembourg in 2001 fromthe stock reported for Belgium and Luxembourg Economic Union for the same year. Flows are addedto this estimated stock thereafter.

- The data on Chinese FDI stock during the period 1994-2004 are revised as reported by the Ministryof Commerce. The previous data in the past WIRs were also reported by the same Ministry, but theywere the accumulation of FDI inflows. The revision was made on the basis of the China's own FDIstatistical methodology and accounting rules, as well as the following assumptions: FDI inflowsto China were mainly greenfield investment that accounted for 95% of total flows, 95% of whichwere transferred into fixed assets.

- Hong Kong (China). Data are in accordance with international standards and practices and are basedon market value. Thus, the inward FDI stock for 1997 onward are not directly comparable to thatof previous years.

- Republic of Korea. Data were obtained from the Ministry of Commerce, Industry and Energy. Inwardstock refers to implemented FDI inflows less withdrawals accumulated since 1962, whereas outwardstock refers to actual investment outflows less withdrawals, accumulated since 1968.

- Luxembourg. Stock data have been derived from the annual survey on FDI since 1995. The bankingand insurance sectors are covered fully, while only the larger companies are included in the othersectors so as to ensure a high level of significance of the statistics. Stock data on Luxembourg excludesassets and liabilities of SPEs (holding companies and other financial vehicles). The population ofcompanies surveyed has been progressively extended over time.

- Oman and Saudi Arabia. For the first time in 2004, after technical cooperation was given by theEconomic and Social Commission for Western Asia (ESCWA) and UNCTAD, a survey on inward FDIwas undertaken. Stock data from this survey were used.

- Philippines. Stock data of FDI started only in 2002 when the Bangko Sentral ng Pilipinas (BSP)compiled the international investment position statistics in compliance with the Special DataDissemination Standard (SDDS) requirement of the IMF.

- Singapore. In the case of FDI stock, data are collected through the FDI survey, in line with therecommendations of the BPM5, conducted twice a year since 2001 for the purpose of IIP publication.The survey is based on purposive sampling method and covers all economic sectors. The total respondent

301DEFINITIONS AND SOURCES

is around 900, comprising companies/enterprises, banks and non-bank financial institutions - on averagethe response rate of the survey is around 50%.

- United States. Data on FDI used in this Report do not include current cost adjustments, in otherwords, they are on a historical-cost basis. The Bureau of Economic Analysis prepares estimates ofthe positions that are valued on three bases—historical cost, current cost, and market value. Unlikethe positions on a current-cost and market-value basis, the historical-cost position is not ordinarilyadjusted to account for changes in the replacement cost of the tangible assets of affiliates or in themarket value of foreign parent companies’ equity in United States’ affiliates.

C. Data revisions and updates

All FDI data and estimates in WIR are continuously revised. Because of ongoing revisions, FDIdata reported in WIR may differ from those reported in earlier Reports or other publications of UNCTAD.In particular, recent FDI data are being revised in many economies according to the fifth edition of theBalance-of-Payments Manual of the IMF. Because of this, the data reported in last year’s Report may becompletely or partly changed in this Report.

D. Data verificationIn compiling data for this year’s Report, requests were made to national official sources of all economies

for verification and confirmation of the latest data revisions and accuracy. In addition, websites of nationalofficial sources were consulted. This verification process continued until 15 June 2005. Any revisionsmade after this process are not reflected in the Report.

E. Definitions and sources of the data in annex table B.3

This annex table shows the ratios of inward and outward FDI flows to gross fixed capital formationand inward and outward FDI stock to GDP. All of these data are in current prices.

The data on GDP were obtained from the UNCTAD Secretariat, the IMF’s CD-ROM on InternationalFinancial Statistics, June 2005 and the IMF’s World Economic Outlook, April 2005. For some economies,such as Taiwan Province of China, data are complemented by official sources.

The data on gross fixed capital formation were obtained from the IMF’s CD-ROM on InternationalFinancial Statistics, June 2005. For some economies, for which data are not available, or part of it, dataare complemented by data on gross capital formation. These data are further complemented by data obtainedfrom: (i) national official sources; and (ii) World Bank data on gross fixed capital formation or grosscapital formation, obtained from World Development Indicators Online.

Figures exceeding 100 per cent may result from the fact that, for some economies, the reported dataon gross fixed capital formation do not necessarily reflect the value of capital formation accurately, andFDI flows do not necessarily translate into capital formation.

Data on FDI are from annex tables B.1-B.2.

F. Definitions and sources of the data on cross-border M&Asin annex tables B.4-B.5

FDI is a balance-of-payments concept involving the cross-border transfer of funds. Cross-borderM&A statistics shown in the Report are based on information reported by Thomson Financial. In somecases, these include M&As between foreign affiliates and firms located in the same host economy. SuchM&As conform to the FDI definition as far as the equity share is concerned. However, the data also includepurchases via domestic and international capital markets, which should not be considered as FDI flows.Although it is possible to distinguish types of financing used for M&As (e.g. syndicated loans, corporatebonds, venture capital), it is not possible to trace the origin or country-sources of the funds used. Therefore,the data used in the Report include the funds not categorized as FDI.

302 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

FDI flows are recorded on a net basis (capital account credits less debits between direct investorsand their foreign affiliates) in a particular year. On the other hand, M&A data are expressed as the totaltransaction amount of particular deals, and not as differences between gross acquisitions and divestmentabroad by firms from a particular country. Transaction amounts recorded in the UNCTAD M&A statisticsare those at the time of closure of the deals, and not at the time of announcement. The M&A values arenot necessarily paid out in a single year.

Cross-border M&As are recorded in both directions of transactions. That is, when a cross-borderM&A takes place, it registers as both a sale in the country of the target firm, and as a purchase in thehome country of the acquiring firm. Data showing cross-border M&A activities on an industry basis arealso recorded as sales and purchases. Thus, if a food company acquires a chemical company, this transactionis recorded in the chemical industry in the table on M&As by industry of seller, it is also recorded inthe food industry in the table on M&As by industry of purchaser.

Notes1 In some countries, an equity stake of other than 10% is still used. In the United Kingdom, for example, a stake of

20% or more was the threshold used until 1997.2 This general definition of FDI is based on OECD, Detailed Benchmark Definition of Foreign Direct Investment, third

edition (OECD 1996) and International Monetary Fund, Balance of Payments Manual, fifth edition (IMF 1993).3 International Monetary Fund, op. cit., p. 40.4 Includes Australia, Austria, Belgium, Canada, Denmark, European Commission, Finland, France, Germany, Greece,

Ireland, Italy, Japan, Luxembourg, the Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland,United Kingdom and United States.

303ANNEX B

Annex table B.1. FDI flows, by region and economy, 2002-2004 (Millions of dollars)

FDI inflows FDI outflows

Region/economy 2002 2003 2004 2002 2003 2004

World 716 128 632 599 648 146 652 181 616 923 730 257Developed economies 547 778 442 157 380 022 599 895 577 323 637 360

Europe 427 560 359 369 223 400 396 868 390 021 309 498European Union 420 433 338 678 216 440 384 549 372 400 279 830

Austria 356 7 352 4 865 5 807 6 776 7 164Belgium and Luxembourg .. .. .. .. .. ..Belgium 15 626 32 098 34 366 12 693 36 900 26 125Cyprus 1 057 1 011 1 146 461 524 630Czech Republic 8 483 2 101 4 463 207 206 546Denmark 6 630 2 595 - 10 722 5 686 1 126 - 10 363Estonia 284 891 926 132 148 257Finland 7 919 3 296 4 648 7 622 - 2 590 - 1 028France 49 035 42 498 24 318 50 441 53 147 47 802Germany 50 516 27 265 - 38 557 15 171 - 3 570 - 7 267Greece 50 661 1 351 655 47 607Hungary 2 994 2 162 4 167 278 1 647 538Ireland 28 981 26 888 9 120 10 332 3 543 - 7 400Italy 14 545 16 415 16 815 17 123 9 071 19 262Luxembourg 117 218 91 055 57 000 126 098 101 044 59 008Latvia 254 300 647 4 36 109Lithuania 732 179 773 18 37 263Malta - 426 294 421 - 9 19 9Netherlands 25 038 19 331 - 4 605 33 901 37 778 1 458Poland 4 131 4 123 6 159 230 196 806Portugal 1 767 6 558 1 112 155 7 326 6 178Slovakia 4 094 669 1 122 5 22 - 155Slovenia 1 686 337 516 153 466 498Spain 43 696 29 013 18 361 36 454 30 807 54 246Sweden 11 738 1 288 - 371 10 633 21 238 15 147United Kingdom 24 029 20 298 78 399 50 300 66 457 65 391

Other developed Europe 7 127 20 691 6 961 12 319 17 621 29 668Gibraltar 83 a 7 a 15 a .. .. ..Iceland 91 318 308 323 370 2 594Norway 677 3 801 2 159 4 138 2 139 1 866Switzerland 6 276 16 564 4 478 7 859 15 112 25 207

North America 92 838 63 183 102 152 161 704 140 859 276 747Canada 21 507 6 349 6 293 26 758 21 453 47 453United States 71 331 56 834 95 859 134 946 119 406 229 294

Other developed countries 27 379 19 604 54 469 41 323 46 443 51 115Australia 15 632 6 955 42 594 7 876 15 277 16 288Israel 1 770 3 880 1 619 982 2 067 3 037Japan 9 239 6 324 7 816 32 281 28 800 30 951New Zealand 738 2 445 2 441 185 299 839

Developing economies 155 528 166 337 233 227 47 775 29 016 83 190Africa 12 994 18 005 18 090 427 1 215 2 824

North Africa 3 872 5 262 5 270 22 115 514Algeria 1 065 634 882 100 14 258Egypt 647 237 1 253 28 21 159Libyan Arab Jamahiriya 145 143 131 - 136 63 62Morocco 481 2 314 853 28 12 31Sudan 713 1 349 1 511 .. .. ..Tunisia 821 584 639 2 5 4

Other Africa 9 122 12 743 12 821 404 1 100 2 310West Africa 2 928 3 117 3 562 649 274 325

Benin 14 45 60 a 1 - ..Burkina Faso 15 29 35 a 2 2 1 a

Cape Verde 12 14 20 - .. ..Côte d’Ivoire 213 165 360 a - 4 a 21 a ..

/...

304 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.1. FDI flows, by region and economy, 2002-2004 (continued) (Millions of dollars)

FDI inflows FDI outflows

Region/economy 2002 2003 2004 2002 2003 2004

Gambia 43 25 a 60 a 5 7 a 1 a

Ghana 59 137 139 44 a .. ..Guinea 30 79 100 a 7 a .. ..Guinea-Bissau 4 4 5 a 1 1 1 a

Liberia 3 1 a 20 a 386 a 80 a 60 a

Mali 244 132 180 a 2 1 1 a

Mauritania 118 214 300 a .. - 1 a ..Niger 2 11 20 a - 2 - ..Nigeria 2 040 2 171 2 127 172 167 261Saint Helena .. .. .. .. .. ..Senegal 78 52 70 a 34 3 4 a

Sierra Leone 2 3 5 .. .. ..Togo 53 34 60 a 2 - 6 - 3 a

Central Africa 3 212 6 346 6 122 9 - 32 35Angola 1 672 3 505 2 048 29 24 30 a

Cameroon - - - 7 a .. ..Central African Republic 6 3 - 13 1 - ..Chad 924 713 478 - - ..Congo 137 323 668 6 2 ..Congo, Dem. Rep. of 117 158 900 a - 2 a .. ..Equatorial Guinea 323 1 431 1 664 - - ..Gabon 30 206 323 - 32 - 57 5 a

São Tomé and Principe 3 7 a 54 a

East Africa 1 521 2 013 2 098 108 74 87Burundi - - 3 a - - ..Comoros - 1 2 a .. .. ..Djibouti 4 11 33 .. .. ..Eritrea 20 22 30 a .. .. ..Ethiopia 255 465 545 .. .. ..Kenya 52 81 46 86 24 49Madagascar 8 13 45 a .. .. ..Malawi 6 10 a 16 a .. .. ..Mauritius 33 70 65 9 41 33Mayotte .. .. .. .. .. ..Mozambique 348 337 132 - a - -Reunion - a .. .. .. .. ..Rwanda 7 5 11 .. .. ..Seychelles 48 58 60 a 9 8 5 a

Somalia - a - a 9 a .. .. ..Uganda 203 211 237 .. .. ..United Rep. of Tanzania 430 527 470 .. .. ..Zambia 82 172 334 .. .. ..Zimbabwe 26 30 a 60 a 3 .. ..

Southern Africa 1 460 1 267 1 038 - 362 783 1 863Botswana 405 418 47 43 206 274Lesotho 27 42 52 - - -Namibia 181 149 286 - 5 - 10 - 21South Africa 757 720 585 - 399 577 1 606Swaziland 90 - 61 69 - 1 10 4

Latin America and the Caribbean 50 492 46 908 67 526 11 351 10 562 10 943South and Central America 45 359 37 906 57 437 7 040 9 887 14 381

South America 28 463 24 357 37 872 4 099 5 246 10 587Argentina 2 149 1 887 4 254 - 627 774 319Bolivia 677 197 117 3 3 3Brazil 16 590 10 144 18 166 2 482 249 9 471Chile 2 550 4 385 7 603 343 1 884 943Colombia 2 115 1 793 2 739 857 938 142Ecuador 1 275 1 555 1 241 - - a ..

/...

305ANNEX B

Annex table B.1. FDI flows, by region and economy, 2002-2004 (continued) (Millions of dollars)

FDI inflows FDI outflows

Region/economy 2002 2003 2004 2002 2003 2004

Falkland Islands (Malvinas) .. .. .. .. .. ..French Guiana .. .. .. .. .. ..Guyana 44 26 48 a - a .. ..Paraguay 6 36 119 2 6 6Peru 2 156 1 335 1 816 - 60 40 a

Suriname - 74 - 76 - 60 a .. .. ..Uruguay 194 416 311 14 15 11Venezuela 782 2 659 1 518 1 026 1 318 - 348

Central America 16 896 13 548 19 565 2 940 4 641 3 794Belize 50 58 170 a - - - a

Costa Rica 658 574 618 34 27 62El Salvador 470 173 466 - 26 19 7Guatemala 111 131 155 15 a 2 a ..Honduras 176 247 293 2 a - 6 a ..Mexico 15 129 11 373 16 602 930 1 784 2 240Nicaragua 204 201 250 12 a 10 a ..Panama 99 792 1 012 1 974 a 2 804 a 1 485 a

Caribbean and other America 5 133 9 002 10 089 4 311 675 - 3 438Anguilla 38 33 104 .. .. ..Antigua and Barbuda 80 179 106 15 a .. ..Aruba 306 188 131 3 18 - 1Bahamas 153 147 206 a - 28 a - 6 a

Barbados 17 58 50 a - 1 ..Bermuda 2 155 a 1 908 a 3 800 a 1 754 a - 3 808 a - 1 006 a

British Virgin Islands 178 a 12 a 100 a 8 501 a 2 362 a - 2 364 a

Cayman Islands - 242 a 4 084 a 3 000 a - 6 157 a 1 773 a - 205 a

Cuba 3 a - 9 a 2 a - a .. ..Dominica 12 20 19 .. .. ..Dominican Republic 917 613 645 12 a - 38 a ..Grenada 61 85 42 .. .. ..Guadeloupe .. .. .. .. .. ..Haiti 6 8 7 a 1 a .. ..Jamaica 481 721 650 a 74 116 90 a

Martinique .. .. .. .. .. ..Montserrat 2 2 2 .. .. ..Netherlands Antil les 8 - 81 - 30 1 - 2 25Puerto Rico - 5 a - a 24 a .. .. ..Saint Kitts and Nevis 81 67 62 .. .. ..Saint Lucia 55 102 111 .. .. ..Saint Vincent and the Grenadines 37 55 56 .. .. ..Trinidad and Tobago 791 808 1 001 106 225 29Turks and Caicos Islands - a 1 a .. .. .. ..

Asia and Oceania 92 042 101 424 147 611 35 998 17 239 69 423Asia 92 009 101 278 147 545 35 994 17 231 69 422

West Asia 5 691 6 522 9 840 910 - 3 954 - 6Bahrain 217 517 865 190 741 1 036Iran, Islamic Rep. of 548 482 500 a 39 a - 356 a - 114 a

Iraq - 2 a 5 a 300 a .. .. ..Jordan 64 424 620 25 3 -Kuwait 7 - 67 - 20 - 155 - 4 982 - 1 873Lebanon 257 358 288 a 96 a 17 a 45 a

Oman 26 528 - 18 - a - 1 a - a

Palestinian Territory - 5 a .. .. .. .. ..Qatar 624 a 625 a 679 a - 21 a - 2 a - 2 a

Saudi Arabia 453 778 1 867 143 a 83 a 73 a

Syrian Arab Republic 1 030 1 084 1 206 .. .. ..Turkey 1 063 1 753 2 733 175 499 859United Arab Emirates 1 307 a 30 a 840 a 407 a 43 a - 30 a

/...

306 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.1. FDI flows, by region and economy, 2002-2004 (continued) (Millions of dollars)

FDI inflows FDI outflows

Region/economy 2002 2003 2004 2002 2003 2004

Yemen 102 6 - 21 11 a .. ..South, East and South-East Asia 86 318 94 755 137 705 35 083 21 186 69 429

East Asia 67 282 72 060 105 037 27 555 14 442 53 521China 52 743 53 505 60 630 2 518 - 152 1 805Hong Kong, China 9 682 13 624 34 035 17 463 5 492 39 753Korea, Dem. People’s Rep. of - 15 a 158 a 40 a - a 1 a ..Korea, Rep. of 2 975 3 785 7 687 2 617 3 426 4 792Macao, China 375 403 600 a 71 - 5 25 a

Mongolia 78 132 147 a .. .. ..Taiwan Province of China 1 445 453 1 898 4 886 5 682 7 145

South Asia 4 528 5 331 7 005 1 149 962 2 288Afghanistan 1 a 2 a 1 a .. .. ..Bangladesh 52 268 460 a 3 3 4 a

Bhutan - a 1 a 1 a .. .. ..India 3 449 4 269 5 335 1 107 913 2 222Maldives 12 14 13 a .. .. ..Nepal - 6 15 10 a .. .. ..Pakistan 823 534 952 28 19 56Sri Lanka 197 229 233 11 27 6

South-East Asia 14 507 17 364 25 662 6 379 5 781 13 620Brunei Darussalam 1 035 2 009 103 27 a - 1 a ..Cambodia 145 84 131 6 10 10Indonesia 145 - 597 1 023 182 a 15 a 107 a

Lao People’s Dem. Rep. 25 19 17 .. - a ..Malaysia 3 203 2 473 4 624 1 905 1 369 2 061Myanmar 191 e 291 e 556 a .. .. ..Philippines 1 792 347 469 59 197 412Singapore 5 822 9 331 16 060 4 095 3 705 10 667Thailand 947 1 952 1 064 106 486 362Timor-Leste 1 a 5 a 4 a .. .. ..Viet Nam 1 200 1 450 1 610 .. .. ..

Oceania 33 146 67 5 8 1Cook Islands - a .. .. 2 a .. ..Fij i 18 23 - 9 a 2 4 - a

French Polynesia - 2 a - 11 a .. .. .. ..Kiribati .. .. .. .. .. ..Marshall Islands - 47 a 5 a 10 a .. .. ..Micronesia, Federated States of .. .. .. .. .. ..Nauru 1 a 1 a .. .. .. ..New Caledonia 2 a - 2 a 5 a .. .. ..Niue 9 a .. .. .. .. ..Northern Mariana Islands .. .. .. .. .. ..Palau 1 a 3 a 5 a .. .. ..Papua New Guinea 18 101 25 1 3 -Samoa - a 1 a 1 a .. .. ..Solomon Islands - 1 - 2 - 5 a .. .. ..Tokelau - a - a .. .. .. ..Tonga 1 a 12 a 4 a .. .. ..Tuvalu 25 a - a 9 a .. .. ..Vanuatu 9 15 22 a 1 1 1 a

Wallis and Futuna Islands .. .. .. .. .. ..South-East Europe and the CIS 12 821 24 106 34 897 4 511 10 584 9 707

South-East Europe 3 790 8 365 10 778 589 140 158Albania 135 178 426 a 1 a .. ..Bosnia and Herzegovina 265 381 497 .. .. 1 a

Bulgaria 905 2 097 2 488 28 27 - 228Croatia 1 126 2 042 1 076 539 108 314Macedonia, TFYR 78 95 151 - - 1

/...

307ANNEX B

Annex table B.1. FDI flows, by region and economy, 2002-2004 (concluded) (Millions of dollars)

FDI inflows FDI outflows

Region/economy 2002 2003 2004 2002 2003 2004

Romania 1 144 2 213 5 174 16 41 70Serbia and Montenegro 137 1 360 966 5 - 35 a ..

CIS 9 032 15 741 24 119 3 921 10 443 9 549Armenia 144 157 235 19 a - 2Azerbaijan 1 392 3 285 4 769 a 326 933 1 383 a

Belarus 247 172 169 - 206 2 - 1Georgia 165 338 499 4 4 10Kazakhstan 2 590 2 088 4 269 426 - 121 - 1 279Kyrgyzstan 5 46 77 a - - - 173 a

Moldova, Rep. of 132 71 151 - - 3Russian Federation 3 461 7 958 11 672 3 533 9 727 9 601Tajikistan 36 32 272 .. 12 a ..Turkmenistan 100 100 150 a - 176 a - 126 a ..Ukraine 693 1 424 1 715 - 5 13 4Uzbekistan 65 70 140 a .. .. ..

Memorandum

Least developed countriesb 6 327 10 351 10 702 488 123 110Major petroleum exportersc 12 162 15 767 15 994 2 095 - 2 705 - 482All developing economies, excluding China 102 785 112 832 172 597 45 257 29 168 81 385EU-15 397 145 326 611 196 099 383 072 369 099 276 330

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics).a Estimates. For details, see “Methodological notes: definitions and sources” (www.unctad.org/wir).b Least developed countries include: Afghanistan, Angola, Bangladesh, Benin, Bhutan, Burkina Faso, Burundi, Cambodia, Cape Verde, Central

African Republic, Chad, Comoros, Democratic Republic of Congo, Djibouti, Equatorial Guinea, Eritrea, Ethiopia, Gambia, Guinea, Guinea-Bissau, Haiti, Kiribati, Lao People’s Democratic Republic, Lesotho, Liberia, Madagascar, Malawi, Maldives, Mali, Mauritania, Mozambique,Myanmar, Nepal, Niger, Rwanda, Samoa, Sao Tome and Principe, Senegal, Sierra Leone, Solomon Islands, Somalia, Sudan, Timor-Leste,Togo, Tuvalu, Uganda, United Republic of Tanzania, Vanuatu, Yemen and Zambia.

c Major pertoleum exporters include: Algeria, Angola, Bahrain, Brunei Darussalam, Congo, Gabon, Indonesia, Islamic Republic of Iran, Iraq,Kuwait, Libyan Arab Jamahiriya, Nigeria, Oman, Qatar, Saudi Arabia, Syrian Arab Republic, Trinidad and Tobago, United Arab Emirates, Venezuelaand Yemen.

Note: Data on FDI inflows in China as reported by China’s State Administration of Foreign Exchange are the following: $49,308 million for2002, $47,077 for 2003 and $54,936 for 2004.

308 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.2. FDI stock, by region and economy, 1990, 2000, 2004a

(Millions of dollars)

FDI inward stock FDI outward stock

Region/economy 1990 2000 2004 1990 2000 2004

World 1 768 589 5 780 846 8 895 279 1 785 264 6 148 284 9 732 233Developed economies 1 404 411 3 976 356 6 469 832 1 637 760 5 257 261 8 610 146

Europe 800 751 2 292 922 4 258 547 882 899 3 324 128 5 658 814European Union 753 707 2 174 834 4 023 935 805 851 3 046 301 5 189 738

Austria 10 972 30 431 62 657 4 747 24 821 67 424Belgium and Luxembourg 58 388 195 219 .. 40 636 179 773 ..Belgium .. .. 258 875 b .. .. 248 367 b

Cyprus .. a,c 2 910 a 8 132 a 8 a 560 a 2 684 a

Czech Republic 1 363 a 21 644 56 415 .. 738 3 061Denmark 9 192 73 574 98 172 7 342 73 106 99 570Estonia .. 2 645 9 530 .. 259 1 398Finland 5 132 24 272 55 946 11 227 52 109 80 936France 86 845 259 796 535 201 a 110 126 445 059 769 353 a

Germany 111 231 271 611 347 957 a 151 581 541 861 833 651 a

Greece 5 681 a 14 113 27 213 2 882 a 6 094 13 056Hungary 569 22 870 60 328 197 1 280 4 472Ireland 42 058 a 127 088 229 241 12 779 a 27 925 95 955Italy 59 998 121 170 220 720 60 184 180 275 280 481Luxembourg .. 23 492 d 182 894 a .. 7 927 d 176 499 a

Latvia .. 2 084 4 493 .. 241 226Lithuania .. 2 334 6 389 .. 29 423Malta 465 a 2 385 3 557 a .. 203 361 a

Netherlands 68 731 243 733 428 803 a 106 899 305 462 545 808 a

Poland 109 34 227 61 427 a 408 a 1 018 2 661 a

Portugal 10 571 28 696 65 213 900 17 256 45 555Slovakia 81 3 733 14 501 .. 325 618Slovenia 665 a 2 894 4 962 a 258 768 2 450 a

Spain 65 916 154 806 346 676 15 652 166 064 332 655Sweden 12 636 93 970 162 973 50 720 123 230 203 943United Kingdom 203 905 438 631 771 658 229 307 897 845 1 378 130

Other developed Europe 47 045 118 088 234 612 77 047 277 827 469 076Gibraltar 263 a 529 a 646 a .. .. ..Iceland 146 490 1 807 76 664 3 948Norway 12 391 30 265 51 126 a 10 884 43 793 72 109 a

Switzerland 34 245 86 804 181 033 66 087 233 370 393 019North America 507 754 1 469 583 1 777 678 515 328 1 553 886 2 387 982

Canada 112 843 212 716 303 818 84 807 237 639 369 777United States 394 911 1 256 867 1 473 860 a 430 521 1 316 247 2 018 205 a

Other developed countries 95 906 213 852 433 608 239 533 379 247 563 350Australia 73 641 111 141 253 620 30 506 85 387 167 541Israel 4 476 24 319 33 081 1 188 9 353 16 010Japan 9 850 50 322 96 984 201 441 278 442 370 544New Zealand 7 938 28 070 49 922 6 398 a 6 065 9 256

Developing economies 364 057 1 734 543 2 225 994 147 313 868 920 1 035 676Africa 59 445 151 246 219 277 19 919 45 406 45 600

North Africa 24 542 44 264 70 213 1 836 3 380 4 346Algeria 1 561 a 3 647 a 7 423 a 183 a 346 a 727 a

Egypt 11 043 a 18 254 a 20 902 a 163 a 655 a 875 a

Libyan Arab Jamahiriya 678 a 472 a 758 a 1 321 a 1 943 a 2 107 a

Morocco 3 591 a 8 825 a 17 959 a 155 a 403 a 591 a

Sudan 55 a 1 398 a 5 545 a .. .. ..Tunisia 7 615 11 668 17 626 15 33 46

Other Africa 34 903 106 982 149 064 18 082 42 025 41 254West Africa 13 822 33 528 45 587 1 862 7 094 8 164

Benin 159 a 213 291 a 2 a 36 a 40 a

Burkina Faso 39 a 28 87 a 4 a 20 a 25 a

Cape Verde 4 a 173 a 228 a 1 a 7 a 7 a

Côte d’ Ivoire 975 a 2 483 3 932 a 38 a 641 a 652 a

/...

309ANNEX B

Gambia 157 216 349 a 22 44 54 a

Ghana 319 a 1 493 a 1 917 a .. 271 a 355 a

Guinea 69 a 263 a 474 a .. 7 a 19 a

Guinea-Bissau 8 a 38 a 51 a .. .. 1 a

Liberia 2 454 a 2 968 a 3 001 a 453 a 1 524 a 1 737 a

Mali 229 a 132 863 a 22 a 63 a 84 a

Mauritania 59 a 140 a 864 a 3 a 5 a 4 a

Niger 286 a 295 367 a 54 a 144 a 139 a

Nigeria 8 539 a 23 786 a 31 402 a 1 207 a 4 132 a 4 826 a

Senegal 258 a 832 a 1 065 a 49 a 121 a 154 a

Sierra Leone .. a,c 40 a 59 a .. .. ..Togo 268 a 427 a 637 a 8 a 79 a 65 a

Central Africa 4 769 13 133 32 443 373 701 758Angola 1 025 a 7 977 a 17 347 a 1 a 49 a 146 a

Cameroon 1 044 a 1 053 a 1 054 a 150 a 260 a 294 a

Central African Republic 95 a 104 a 106 a 18 a 43 a 45 a

Chad 250 a 577 a 3 152 a 37 a 70 a 70 a

Congo 575 a 1 893 a 3 098 a .. .. ..Congo, Dem. Rep. of 546 a 617 a 1 874 a .. .. ..Equatorial Guinea 25 a 1 128 a 5 491 a - a .. a,c 3 a

Gabon 1 208 a .. a,c 242 a 167 a 280 a 200 a

São Tomé and Principe - a 11 a 79 a .. .. ..East Africa 3 315 12 941 20 437 255 1 239 1 600

Burundi 30 a 48 a 51 a - a 2 a 2 a

Comoros 17 a 21 a 25 a 1 a 1 a 1 a

Djibouti 6 a 34 a 85 a .. .. ..Eritrea .. 337 a 422 a .. .. ..Ethiopia 124 a 933 a 2 547 a .. 435 a 435 a

Kenya 668 a 984 a 1 223 a 99 a 134 a 370 a

Madagascar 107 a 354 a 513 a 1 a 11 a 11 a

Malawi 198 a 328 a 379 a .. 8 a 8 a

Mauritius 169 a 687 a 887 a 1 a 132 a 219 a

Mozambique 42 a 1 094 a 2 166 a 1 a 2 a 2 a

Rwanda 213 a 252 a 279 a 2 a 4 a 4 a

Seychelles 204 a 577 a 808 a 61 a 136 a 167 a

Somalia .. a,c 4 a 13 a .. .. ..Uganda 6 a 807 1 613 .. 133 a 133 a

United Rep. of Tanzania 388 a 3 038 5 203 .. .. ..Zambia 1 022 a 2 360 a 3 019 a .. .. ..Zimbabwe 124 a 1 085 a 1 204 a 88 a 241 a 249 a

Southern Africa 12 996 47 379 50 596 15 593 32 992 30 732Botswana 1 309 a 1 821 a 1 382 a 447 517 1 814Lesotho 83 a 330 a 479 a - a 2 a 2 a

Namibia 2 047 1 230 1 527 a 80 45 10 a

South Africa 9 221 43 462 46 283 a 15 027 32 333 28 790 a

Swaziland 336 537 926 38 95 116Latin America and the Caribbean 118 133 514 634 723 752 58 950 210 921 271 690

South and Central America 96 491 399 746 563 947 54 667 107 775 143 311South America 68 017 284 498 353 969 49 295 95 934 115 456

Argentina 8 778 a 67 601 53 697 a 6 057 a 21 141 21 819 a

Bolivia 1 026 5 10 7 a 29 40Brazil 37 243 103 015 150 965 a 41 044 a 51 946 a 64 363 a

Chile 10 067 45 753 54 464 154 a 11 154 14 447Colombia 3 500 10 992 22 278 402 2 989 4 284Ecuador 1 626 7 081 12 482 16 a 152 a 152 a

Falkland Islands (Malvinas) - a 58 a 76 a .. .. ..Guyana 42 a 759 a 933 a .. 1 a 1 a

310 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.2. FDI stock, by region and economy, 1990, 2000, 2004a (continued)(Millions of dollars)

FDI inward stock FDI outward stock

Region/economy 1990 2000 2004 1990 2000 2004

Peru 1 330 11 062 13 310 112 505 874Suriname .. a,c .. a,c .. a,c .. .. ..Uruguay 671 a 2 088 2 110 a 186 a 126 a 123 a

Venezuela 3 865 35 480 43 575 1 221 7 676 9 204Central America 28 474 115 248 209 978 5 372 11 841 27 855

Belize 89 a 296 a 693 a 20 a 43 a 44 a

Costa Rica 1 309 a 2 709 4 815 a 44 a 90 219 a

El Salvador 212 2 001 3 686 a 56 a 74 154 a

Guatemala 1 734 3 420 4 441 a .. 71 a 106 a

Honduras 383 a 1 482 a 2 390 a .. .. ..Mexico 22 424 97 170 182 536 a 1 064 a 7 540 a 15 885 a

Nicaragua 126 a 1 395 a 2 201 a .. 19 a 57 a

Panama 2 198 a 6 775 9 217 4 188 a 4 004 a 11 391 a

Caribbean and other America 21 642 114 888 159 806 4 284 103 146 128 379Anguilla 11 a 230 a 441 a .. .. ..Antigua and Barbuda 290 a 644 a 1 121 a .. .. ..Aruba 145 a 934 a 1 294 a 490 a 694 a 728 a

Bahamas 586 a 1 587 a 2 195 a 614 a 1 385 a 1 407 a

Barbados 171 308 451 a 23 41 43 a

Bermuda 13 849 a 56 393 a 77 602 a 1 550 a 14 942 a 8 533 a

British Virgin Islands 126 a 11 363 a 11 876 a 875 a 64 531 a 97 041 a

Cayman Islands 1 749 a 24 973 a 36 172 a 648 a 20 423 a 18 737 a

Cuba 2 a 74 a 74 a .. .. ..Dominica 66 a 275 a 341 a .. .. ..Dominican Republic 572 5 214 a 8 468 a .. 113 a 59 a

Grenada 70 a 364 a 613 a .. - a - a

Haiti 149 a 215 a 240 a .. 3 a 4 a

Jamaica 790 a 3 317 a 5 783 a 42 a 709 a 1 079 a

Montserrat 40 a 77 a 85 a .. .. ..Netherlands Antil les 408 a 78 a .. a,c 21 a 11 a 36 a

Saint Kitts and Nevis 160 a 505 a 805 a .. - a - a

Saint Lucia 316 a 825 a 1 157 a .. - a - a

Saint Vincent and the Grenadines 48 a 500 a 669 a - a - a - a

Trinidad and Tobago 2 093 7 008 a 10 443 a 21 a 293 a 711 a

Turks and Caicos Islands 2 a 4 a 5 a .. .. ..Asia and Oceania 186 479 1 068 663 1 282 964 68 444 612 594 718 387

Asia 183 849 1 064 078 1 278 608 68 178 612 305 717 997West Asia 32 010 64 391 100 141 7 585 10 717 14 604

Bahrain 552 5 906 7 585 a 719 1 752 3 935 a

Iran, Islamic Rep. of 2 039 a 2 474 a 4 065 a .. 411 a .. a,c

Iraq .. a,c .. a,c 273 a .. .. ..Jordan 615 a 2 272 a 3 501 a 16 a .. a,c .. a,c

Kuwait 37 a 608 a 381 a 3 662 1 427 .. a,c

Lebanon 53 a 1 116 a 2 269 a 43 a 430 a 611 a

Oman 1 706 a 2 506 a 3 432 a 10 a 33 a 32 a

Palestinian Territory .. 932 a 947 a .. .. ..Qatar 71 a 1 920 a 4 144 a .. 74 a 67 a

Saudi Arabia 14 467 a 16 851 a 20 454 1 873 a 2 204 a 1 892 a

Syrian Arab Republic 374 a 8 224 a 12 491 a .. .. ..Turkey 11 194 19 209 35 188 a 1 157 a 3 668 6 997 a

United Arab Emirates 751 a 1 061 a 4 422 a 99 a 819 a 1 440 a

Yemen 180 1 336 990 a 5 a .. a,c 9 a

South, East and South-East Asia 151 839 999 687 1 178 467 60 593 601 588 703 394East Asia 84 065 707 616 802 657 49 032 509 636 575 468

China 20 691 a 193 348 245 467 4 455 a 27 768 a 38 825 a

Hong Kong, China 45 073 a 455 469 456 833 11 920 a 388 380 405 589Korea, Dem. People’s Rep. of 572 a 1 046 a 1 225 a .. .. ..Korea, Rep. of 5 186 37 189 55 327 2 301 26 833 39 319 a

/...

311ANNEX B

Annex table B.2. FDI stock, by region and economy, 1990, 2000, 2004a (continued)(Millions of dollars)

FDI inward stock FDI outward stock

Region/economy 1990 2000 2004 1990 2000 2004

Macao, China 2 809 a 2 801 a 4 195 a .. .. 497 a

Mongolia - a 182 a 581 a .. .. ..Taiwan Province of China 9 735 a 17 581 39 029 a 30 356 a 66 655 91 237 a

South Asia 4 602 28 706 52 221 422 2 501 7 556Afghanistan 12 a 17 a 22 a .. .. ..Bangladesh 324 a 2 429 3 433 a 45 a 68 100 a

Bhutan 2 a 12 a 15 a .. .. ..India 1 657 a 17 517 38 676 124 a 1 859 6 592Maldives 25 a 119 a 169 a .. .. ..Nepal 12 a 97 a 135 a .. .. ..Pakistan 1 892 6 919 7 596 a 245 489 733 a

Sri Lanka 679 a 1 596 2 175 a 8 a 86 a 131 a

South-East Asia 63 171 263 365 323 588 11 138 89 450 120 369Brunei Darussalam 39 a 3 874 a 7 548 a .. 447 a 481 a

Cambodia 38 a 1 580 2 090 .. 193 256Indonesia 8 855 a 24 780 a 11 352 a 86 a 6 940 a .. a,c

Lao People’s Dem. Rep. 13 a 556 a 641 a .. 28 a 28 a

Malaysia 10 318 52 747 a 46 291 a 2 671 21 276 13 796 a

Myanmar 281 3 865 4 679 .. .. ..Philippines 3 268 12 810 12 685 a 155 1 597 1 606 a

Singapore 30 468 112 571 160 422 a 7 808 56 766 100 910 a

Thailand 8 242 29 915 48 598 a 418 2 203 3 393 a

Timor-Leste - a 72 a 166 a .. .. ..Viet Nam 1 650 a 20 596 29 115 a .. .. ..

Oceania 2 630 4 585 4 356 267 288 390Cook Islands 14 a 34 a 35 a .. .. ..Fij i 394 a 805 a 269 a 241 a 25 a 55 a

French Polynesia 69 a 139 a 106 a .. .. ..Kiribati - a 1 a 1 a .. - a - a

New Caledonia 76 a 146 a 150 a .. .. ..Niue .. - a 8 a .. .. ..Northern Mariana Islands 304 a 767 a 767 a .. .. ..Palau .. 97 a 117 a .. .. ..Papua New Guinea 1 582 2 007 a 2 214 a 26 a 263 a 322 a

Samoa 9 a 53 a 56 a .. .. ..Solomon Islands 70 a 150 a 130 a .. .. ..Tokelau .. - a 1 a .. .. ..Tonga 1 a 21 a 39 a .. .. ..Tuvalu .. .. a,c 34 a .. .. ..Vanuatu 110 366 430 .. .. 13 a

South-East Europe and the CIS 121 69 947 199 453 191 22 103 86 410South-East Europe 112 15 000 46 863 191 1 220 2 773

Albania .. 568 a 1 514 a .. 82 a 83 a

Bosnia and Herzegovina .. 398 a 1 660 a .. 40 a 41 a

Bulgaria 112 a 2 257 7 569 a 124 a 87 .. a,c

Croatia .. 3 568 12 989 .. 875 2 426Macedonia, TFYR .. 410 a 1 175 a .. - a 2 a

Romania - 6 480 18 009 66 136 301Serbia and Montenegro .. 1 319 a 3 947 a .. .. ..

CIS 9 54 947 152 590 .. 20 883 83 637Armenia 9 a 632 1 004 .. 3 a 25 a

Azerbaijan .. 3 735 13 408 a .. 474 a 2 642 a

Belarus .. 1 305 2 057 .. 24 8Georgia .. 423 a 1 536 a .. .. ..Kazakhstan .. 10 078 22 399 .. 16 .. c

Kyrgyzstan .. 447 568 a .. 33 .. a,c

Moldova, Rep. of .. 459 940 a .. 23 26 a

Russian Federation .. 32 204 98 444 a .. 20 141 81 874 a

/...

312 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Tajikistan .. 146 a 495 a .. .. ..Turkmenistan .. 944 a 1 464 a .. .. ..Ukraine .. 3 875 9 217 a .. 170 168 a

Uzbekistan .. 699 a 1 057 a .. .. ..Memorandum

Least developed countriese 9 444 38 384 71 953 729 3 099 3 601Major pertoleum exportersf 48 992 149 630 191 397 10 596 28 836 25 396All developing economies, excluding China 343 366 1 541 195 1 980 527 142 858 841 152 996 851EU-15 751 256 2 077 108 3 794 201 804 981 3 040 879 5 171 384

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics).a Estimates. For details, see “Methodological notes: definitions and sources” (www.unctad.org/wir).b Estimated by UNCTAD. The stock data for Luxembourg, as reported by the official national source, are subtracted from the stock data of

the Belgium and Luxembourg Monetary Union for 2001, the last year for which the data of the latter are available. Flows are added thereafterto arrive at the stock data for each year.

c Negative stock value. However, this value is included in the regional and global totals.d This value is not included in the regional and global totals to avoid double counting as Luxembourg was covered under the Belgium and Luxembourg

Monetary Union whose data were reported until 2001.e Least developed countries include: Afghanistan, Angola, Bangladesh, Benin, Bhutan, Burkina Faso, Burundi, Cambodia, Cape Verde, Central

African Republic, Chad, Comoros, Democratic Republic of Congo, Djibouti, Equatorial Guinea, Eritrea, Ethiopia, Gambia, Guinea, Guinea-Bissau, Haiti, Kiribati, Lao People’s Democratic Republic, Lesotho, Liberia, Madagascar, Malawi, Maldives, Mali, Mauritania, Mozambique,Myanmar, Nepal, Niger, Rwanda, Samoa, Sao Tome and Principe, Senegal, Sierra Leone, Solomon Islands, Somalia, Sudan, Timor-Leste,Togo, Tuvalu, Uganda, United Republic of Tanzania, Vanuatu, Yemen and Zambia.

f Major petroleum exporters include: Algeria, Angola, Bahrain, Brunei Darussalam, Congo, Gabon, Indonesia, Islamic Republic of Iran, Iraq,Kuwait, Libyan Arab Jamahiriya, Nigeria, Oman, Qatar, Saudi Arabia, Syrian Arab Republic, Trinidad and Tobago, United Arab Emirates, Venezuelaand Yemen.

Annex table B.2. FDI stock, by region and economy, 1990, 2000, 2004a (concluded)(Millions of dollars)

FDI inward stock FDI outward stock

Region/economy 1990 2000 2004 1990 2000 2004

313ANNEX B

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Worldinward 10.6 8.3 7.5 8.4 18.3 21.7outward 9.7 8.2 8.7 8.7 19.7 24.0

Developed economiesinward 10.9 7.9 6.1 8.2 16.3 20.5outward 12.0 10.3 10.3 9.6 21.5 27.3

Europeinward 22.9 16.1 8.6 10.8 26.5 32.0outward 21.2 17.5 12.0 12.0 38.4 42.5

European Unioninward 23.7 16.0 8.8 10.7 26.4 31.7outward 21.7 17.6 11.4 11.5 37.0 40.9

Austriainward 0.8 12.8 7.7 6.8 16.0 21.6outward 12.8 11.8 11.3 2.9 13.0 23.3

Belgium and Luxembourginward .. .. .. 27.8 78.8 ..outward .. .. .. 19.4 72.5 ..

Belgiuminward 32.6 55.7 52.8 .. .. 73.5 a

outward 26.5 64.0 40.1 .. .. 70.6 a

Cyprusinward 55.6 44.2 40.2 .. b 33.0 52.7outward 24.2 22.9 22.1 0.2 6.4 17.4

Czech Republicinward 43.2 8.7 15.4 3.9 38.9 52.7outward 1.1 0.9 1.9 .. 1.3 2.9

Denmarkinward 18.6 6.2 - 22.3 6.9 46.4 40.5outward 16.0 2.7 - 21.6 5.5 46.1 41.1

Estoniainward 14.1 34.5 29.6 .. 51.4 85.1outward 6.5 5.7 8.2 .. 5.0 12.5

Finlandinward 31.7 11.3 13.4 3.8 20.2 30.1outward 30.5 - 8.9 - 3.0 8.2 43.5 43.5

Franceinward 17.6 12.6 6.2 7.1 19.9 26.5outward 18.1 15.7 12.1 9.1 34.0 38.1

Germanyinward 13.7 6.4 - 8.3 6.6 14.5 12.9outward 4.1 - 0.8 - 1.6 9.0 29.0 30.8

Greeceinward 0.2 1.5 2.6 6.8 12.4 13.2outward 2.1 0.1 1.2 3.4 5.4 6.4

Hungaryinward 19.7 11.7 18.6 1.7 49.0 60.7outward 1.8 8.9 2.4 0.6 2.7 4.5

Irelandinward 106.2 74.9 20.1 88.9 134.1 126.3outward 37.9 9.9 - 16.3 27.0 29.5 52.9

Italyinward 6.2 5.8 5.2 5.4 11.3 13.1outward 7.3 3.2 5.9 5.5 16.8 16.7

Luxembourginward 2 580.3 1 594.6 918.3 .. .. 575.4outward 2 775.8 1 769.5 950.7 .. .. 555.3

Latviainward 11.4 11.2 16.7 .. 29.1 32.9outward 0.2 1.4 2.8 .. 3.4 1.7

/...

314 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Lithuaniainward 25.5 4.6 15.8 .. 20.9 28.8outward 0.6 0.9 5.4 .. 0.3 1.9

Maltainward - 68.0 30.6 37.0 20.1 67.1 66.0outward - 1.5 2.0 0.8 .. 5.7 6.7

Netherlandsinward 28.7 18.7 - 3.9 23.3 65.8 74.2outward 38.9 36.5 1.2 36.3 82.4 94.4

Polandinward 11.4 10.8 14.5 0.2 20.9 25.4outward 0.6 0.5 1.9 0.7 0.6 1.1

Portugalinward 5.6 19.5 2.8 14.8 27.0 39.0outward 0.5 21.8 15.7 1.3 16.2 27.2

Slovakiainward 61.1 8.0 11.1 0.5 18.4 35.3outward 0.1 0.3 - 1.5 .. 1.6 1.5

Sloveniainward 32.7 5.1 6.5 3.8 15.3 15.1outward 3.0 7.0 6.3 1.5 4.0 7.5

Spaininward 26.5 13.5 7.0 12.8 27.6 34.9outward 22.1 14.4 20.6 3.0 29.6 33.5

Swedeninward 29.2 2.7 - 0.7 5.3 39.2 47.0outward 26.4 44.7 27.5 21.3 51.4 58.9

United Kingdominward 9.3 6.9 21.9 20.6 30.5 36.3outward 19.5 22.7 18.2 23.2 62.4 64.8

Other developed Europeinward 7.4 19.2 5.6 13.4 28.0 37.4outward 12.9 16.3 24.1 22.0 66.3 74.9

Icelandinward 6.2 15.2 11.5 2.3 5.8 14.1outward 21.8 17.7 96.5 1.2 7.9 30.9

Norwayinward 2.0 9.9 4.8 10.7 18.1 20.4outward 12.0 5.6 4.1 9.4 26.2 28.8

Switzerlandinward 10.5 24.5 5.9 15.0 36.1 50.6outward 13.2 22.4 33.4 28.9 97.1 109.8

North America

inward 4.5 2.9 4.4 8.0 14.0 14.0outward 7.8 6.4 12.0 8.1 14.8 18.8

Canadainward 14.9 3.8 12.5 19.6 29.8 30.5outward 18.5 12.7 94.4 14.7 33.3 37.1

United Statesinward 3.7 2.8 4.2 6.9 12.9 12.6outward 7.0 5.9 10.1 7.5 13.5 17.2

Other developed countriesinward 2.5 1.6 4.2 2.8 4.0 7.9outward 3.8 3.9 3.9 6.9 7.1 10.2

Australiainward 16.5 5.6 28.2 23.7 28.6 41.1outward 8.3 12.3 10.8 9.8 22.0 27.1

Israelinward 9.2 20.0 8.1 8.5 20.2 28.4outward 5.1 10.6 15.2 2.3 7.8 13.8

/...

315ANNEX B

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Japaninward 1.0 0.6 0.7 0.3 1.1 2.1outward 3.4 2.8 2.8 6.6 5.8 7.9

New Zealandinward 6.0 14.0 11.1 18.2 54.3 51.5outward 1.5 1.7 3.8 14.7 11.7 9.5

Developing economiesinward 9.5 8.8 10.5 9.8 26.2 26.4outward 2.8 1.6 4.2 4.3 13.6 12.7

Africainward 13.0 15.0 12.5 12.7 26.5 27.8outward - 1.1 2.4 4.8 8.5 6.2

North Africainward 8.0 10.5 9.2 13.3 17.3 24.4outward - 0.2 1.0 1.1 1.4 1.6

Algeriainward 7.7 4.0 4.5 2.5 6.7 9.1outward 0.7 0.1 1.3 0.3 0.6 0.9

Egyptinward 4.3 2.0 9.9 25.6 17.7 27.1outward 0.2 0.2 1.3 0.4 0.6 1.1

Libyan Arab Jamahiriyainward 5.5 4.8 3.9 2.4 1.4 2.6outward - 5.1 2.1 1.9 4.6 5.7 7.3

Moroccoinward 5.8 22.5 7.5 13.9 26.5 36.1outward 0.3 0.1 0.3 0.6 1.2 1.2

Sudaninward 23.8 37.0 41.4 0.4 12.1 26.1outward .. .. .. .. .. ..

Tunisiainward 15.3 10.0 9.9 62.0 60.0 61.7outward - 0.1 0.1 0.1 0.2 0.2

Other Africainward 17.7 18.2 14.7 12.3 34.0 29.7outward - 1.8 3.5 7.6 14.5 8.8

West Africainward 26.8 20.3 17.4 20.1 42.6 34.7outward 2.5 1.5 1.8 3.2 9.1 6.3

Benininward 2.6 6.2 7.3 8.6 9.5 7.1outward 0.3 - .. 0.1 1.6 1.0

Burkina Fasoinward 2.3 3.8 3.1 1.4 1.3 1.7outward 0.3 0.2 0.1 0.1 0.9 0.5

Cape Verdeinward 9.4 6.5 8.4 1.1 31.9 23.1outward - .. .. 0.4 1.2 0.7

Côte d’ Ivoireinward 18.3 12.8 24.7 9.0 23.2 24.7outward - 0.4 1.6 .. 0.4 6.0 4.1

Gambiainward 54.6 32.9 69.9 49.4 51.3 85.9outward 6.1 8.9 1.2 6.9 10.4 13.3

Ghanainward 5.1 8.2 7.0 5.4 30.0 21.7outward 3.8 .. .. .. 5.5 4.0

Guineainward 7.3 22.1 27.1 2.4 8.6 12.6outward 1.7 .. .. .. 0.2 0.5

/...

316 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Guinea-Bissauinward 17.4 12.9 9.7 3.3 16.9 18.1outward 4.9 1.7 1.0 .. .. 0.4

Liberiainward .. .. .. 194.9 548.7 483.8outward .. .. .. 36.0 281.7 280.1

Maliinward 38.7 17.2 20.7 9.5 5.4 17.5outward 0.2 0.2 0.1 0.9 2.6 1.7

Mauritaniainward 35.9 43.7 62.6 5.8 15.6 64.2outward .. - 0.2 .. 0.2 0.6 0.3

Nigerinward 1.0 4.0 6.1 11.5 16.4 11.7outward - 0.7 - .. 2.2 8.0 4.4

Nigeriainward 49.2 32.4 20.4 30.0 56.3 44.0outward 4.1 2.5 2.5 4.2 9.8 6.8

Senegalinward 6.7 3.6 4.4 4.5 19.0 14.0outward 2.9 0.2 0.3 0.9 2.8 2.0

Sierra Leoneinward 2.9 2.3 3.3 .. b 6.2 5.5outward .. .. .. .. .. ..

Togoinward 23.5 9.9 15.5 16.5 32.1 31.4outward 1.0 - 1.9 - 0.8 0.5 5.9 3.2

Central Africainward 32.0 54.1 46.5 11.1 37.4 51.8outward 0.1 - 0.4 0.5 1.2 2.6 1.5

Angolainward 46.1 82.6 42.7 10.0 87.4 88.8outward 0.8 0.6 0.6 - 0.5 0.7

Camerooninward - - - 9.4 11.4 7.3outward 0.4 .. .. 1.3 2.8 2.0

Central African Republicinward 3.9 2.0 - 6.8 6.4 11.5 7.9outward 0.9 - .. 1.2 4.8 3.3

Chadinward 73.6 49.7 45.2 14.4 44.3 72.9outward - - .. 2.1 5.4 1.6

Congoinward 15.9 33.8 54.3 20.6 58.8 66.7outward 0.7 0.2 .. .. .. ..

Congo, Dem. Rep. ofinward 29.0 20.4 75.8 5.8 12.4 28.7outward - 0.5 .. .. .. .. ..

Equatorial Guineainward 62.6 247.7 254.8 19.2 90.0 123.7outward - - .. 0.2 .. b 0.1

Gaboninward 2.1 14.2 20.1 20.3 .. b 3.3outward - 2.3 - 3.9 0.3 2.8 5.6 2.7

São Tomé and Principeinward 17.3 38.7 .. 0.7 24.9 123.4outward .. .. .. .. .. ..

East Africainward 13.0 15.2 14.0 6.4 20.8 26.6outward 2.1 1.8 2.0 0.9 2.7 2.8

/...

317ANNEX B

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Burundiinward - - 3.2 2.7 6.7 7.5outward - - .. - 0.3 0.4

Comorosinward 1.6 2.9 5.0 6.8 11.2 6.9outward .. .. .. 0.4 0.6 0.3

Djiboutiinward 4.8 14.1 36.0 1.5 6.1 12.8outward .. .. .. .. .. ..

Eritreainward 11.9 13.1 14.9 .. 44.7 67.7outward .. .. .. .. .. ..

Ethiopiainward 20.5 34.2 32.7 1.8 15.5 31.0outward .. .. .. .. 7.2 5.3

Kenyainward 3.3 4.5 2.3 7.8 9.4 7.8outward 5.5 1.4 2.4 1.2 1.3 2.4

Madagascarinward 1.4 1.3 4.5 3.5 9.1 11.8outward .. .. .. - 0.3 0.2

Malawiinward 3.3 6.0 8.4 10.5 18.8 20.4outward .. .. .. .. 0.5 0.5

Mauritiusinward 3.1 5.5 4.6 6.4 15.1 15.0outward 0.9 3.2 2.3 - 2.9 3.7

Mozambiqueinward 47.5 44.9 15.5 1.7 29.7 39.0outward - - - - - -

Rwandainward 2.4 1.5 3.0 8.2 14.6 15.1outward .. .. .. 0.1 0.2 0.2

Seychellesinward 22.5 41.8 43.2 55.4 96.3 114.7outward 4.2 5.8 3.6 16.6 22.7 23.8

Somaliainward .. .. .. .. b 0.2 0.8outward .. .. .. .. .. ..

Ugandainward 16.7 15.1 16.3 0.1 14.1 23.6outward .. .. .. .. 2.3 1.9

United Rep. of Tanzaniainward 23.2 27.7 21.9 9.1 33.4 48.0outward .. .. .. .. .. ..

Zambiainward 10.3 16.0 27.7 31.1 72.9 55.8outward .. .. .. .. .. ..

Zimbabweinward 1.6 1.7 3.0 1.4 15.1 20.7outward 0.2 .. .. 1.0 3.3 4.3

Southern Africainward 7.7 4.2 2.7 10.9 34.2 21.9outward - 1.9 2.6 4.8 13.0 23.8 13.3

Botswanainward 33.1 23.7 2.3 34.8 36.6 15.1outward 3.5 11.7 13.5 11.9 10.4 19.9

Lesothoinward 8.8 9.6 14.6 13.5 38.2 31.6outward - - - - 0.2 0.1

/...

318 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Namibiainward 32.4 15.8 38.6 80.9 35.6 32.6outward - 0.9 - 1.0 - 2.8 3.1 1.3 0.2

South Africainward 4.5 2.7 1.7 8.2 33.9 21.7outward - 2.4 2.2 4.6 13.4 25.3 13.5

Swazilandinward 42.7 - 25.7 24.9 39.9 38.6 39.2outward - 0.3 4.1 1.3 4.5 6.8 4.9

Latin America and the Caribbeaninward 15.4 12.9 15.5 10.5 24.7 34.1outward 2.3 3.2 3.9 5.5 10.3 13.1

South and Central Americainward 15.0 12.4 15.2 9.0 20.8 29.1outward 2.3 3.2 4.0 5.2 5.6 7.4

South Americainward 17.4 14.3 17.0 8.7 22.3 30.1outward 2.5 3.1 4.9 6.3 7.5 9.8

Argentinainward 17.6 9.6 14.5 6.2 23.8 35.3outward - 5.1 3.9 1.1 4.3 7.4 14.4

Boliviainward 54.4 19.0 10.5 21.1 0.1 0.1outward 0.2 0.2 0.3 0.1 0.4 0.4

Brazilinward 19.6 11.3 15.3 8.0 17.1 25.2outward 2.9 0.3 8.0 8.8 8.6 10.7

Chileinward 17.8 28.2 39.2 33.2 61.1 58.2outward 2.4 12.1 4.9 0.5 14.9 15.4

Colombiainward 17.6 16.1 20.8 8.7 13.1 23.4outward 7.1 8.4 1.1 1.0 3.6 4.5

Ecuadorinward 23.0 25.1 18.9 15.2 44.4 41.8outward - - .. 0.2 1.0 0.5

Guyanainward 30.2 16.2 26.5 10.6 106.5 120.9outward 0.1 .. .. .. 0.1 0.2

Paraguayinward 0.6 3.2 9.8 7.6 17.2 14.6outward 0.2 0.5 0.5 1.8 2.8 2.1

Peruinward 21.6 12.3 14.7 5.1 20.8 19.6outward - 0.6 0.3 0.4 1.0 1.3

Surinameinward - 12.0 - 11.1 - 7.8 .. b .. b .. b

outward .. .. .. .. .. ..Uruguay

inward 15.6 39.4 20.6 7.2 10.4 17.5outward 1.1 1.4 0.7 2.0 0.6 1.0

Venezuelainward 3.8 20.4 7.9 8.0 29.3 40.5outward 5.0 10.1 - 1.8 2.5 6.3 8.6

Central Americainward 12.2 10.0 12.7 9.8 17.7 27.5outward 2.1 3.4 2.6 1.9 1.8 3.7

Belizeinward 23.9 31.3 81.7 22.1 39.1 66.2outward - 0.2 - 5.0 5.7 4.2

/...

319ANNEX B

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Costa Ricainward 20.7 17.2 16.0 22.9 17.0 26.0outward 1.1 0.8 1.6 0.8 0.6 1.2

El Salvadorinward 20.0 6.9 18.9 4.4 15.2 23.3outward - 1.1 0.7 0.3 1.2 0.6 1.0

Guatemalainward 3.1 3.6 3.8 22.7 18.1 17.0outward 0.4 0.1 .. .. 0.4 0.4

Hondurasinward 12.0 15.2 16.0 12.6 25.1 32.2outward 0.1 - 0.4 .. .. .. ..

Mexicoinward 12.1 9.4 12.2 8.5 16.7 27.0outward 0.7 1.5 1.6 0.4 1.3 2.3

Nicaraguainward 20.4 19.8 21.1 12.4 35.3 49.7outward 1.2 1.0 .. .. 0.5 1.3

Panamainward 5.9 35.9 28.8 41.4 67.6 66.8outward 118.6 127.0 42.3 78.8 40.0 82.6

Caribbean and other Americainward 22.9 23.9 21.8 41.8 71.9 87.7outward 1.7 3.2 1.5 20.7 78.6 86.1

Anguillainward 121.1 93.9 264.5 19.9 215.2 369.1outward .. .. .. .. .. ..

Antigua and Barbudainward 21.8 45.9 24.1 74.0 108.3 139.6outward 4.1 .. .. .. .. ..

Arubainward 71.6 39.2 24.1 16.8 50.3 63.4outward 0.6 3.8 - 0.1 56.6 37.3 35.7

Bahamasinward 9.9 8.5 10.5 18.9 36.8 39.9outward - 1.6 - 0.3 19.8 32.1 25.6

Barbadosinward 4.2 12.7 9.7 10.0 11.9 15.9outward 0.1 0.1 .. 1.4 1.6 1.5

Bermudainward .. .. .. 869.7 1 660.1 1 793.5outward .. .. .. 97.3 439.9 197.2

British Virgin Islandsinward .. .. .. 8.0 1 644.4 1 195.4outward .. .. .. 55.9 9 338.8 9 767.7

Cayman Islandsinward .. .. .. 353.3 1 840.3 2 195.8outward .. .. .. 131.0 1 505.0 1 137.5

Cubainward .. .. .. - 0.3 0.2outward .. .. .. .. .. ..

Dominicainward 22.8 30.9 24.9 39.7 101.4 130.7outward .. .. .. .. .. ..

Dominican Republicinward 18.9 16.3 15.1 8.1 20.8 43.5outward 0.2 - 1.0 .. .. 0.5 0.3

Grenadainward 45.2 45.2 20.0 31.8 104.4 141.2outward .. .. .. .. - -

/...

320 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Haitiinward 0.7 0.9 0.7 5.0 6.1 6.7outward 0.1 .. .. .. 0.1 0.1

Jamaicainward 18.0 29.6 24.0 18.6 43.0 66.4outward 2.8 4.8 3.3 1.0 9.2 12.4

Montserratinward 10.1 10.5 9.3 55.7 220.6 225.8outward .. .. .. .. .. ..

Netherlands Antil lesinward .. .. .. 22.4 2.8 .. b

outward .. .. .. 1.2 0.4 1.2Saint Kitts and Nevis

inward 47.9 38.7 31.4 100.6 153.5 202.9outward .. .. .. .. - -

Saint Luciainward 35.8 67.5 65.1 79.5 122.0 162.9outward .. .. .. .. - -

Saint Vincent and the Grenadinesinward 34.4 43.2 38.7 24.3 148.8 169.8outward .. .. .. - - -

Trinidad and Tobagoinward 52.5 39.7 43.0 41.3 85.8 83.3outward 7.1 11.0 1.2 0.4 3.6 5.7

Turks and Caicos Islandsinward .. .. .. .. 2.1 2.3outward .. .. .. .. .. ..

Asia and Oceaniainward 7.7 7.3 9.1 8.7 26.9 23.2outward 3.1 1.3 4.4 3.6 16.0 13.4

Asiainward 7.7 7.3 9.1 8.7 26.9 23.2outward 3.1 1.3 4.4 3.6 16.1 13.4

West Asiainward 3.7 4.2 4.9 6.5 9.1 9.9outward 0.7 - 2.7 - 2.3 1.6 1.5

Bahraininward 14.9 27.8 41.1 13.0 74.1 70.5outward 13.0 39.8 49.2 17.0 22.0 36.6

Iran, Islamic Rep. ofinward 1.4 1.2 1.0 2.2 2.4 2.4outward 0.1 - 0.9 - 0.2 .. 0.4 -

Iraqinward .. .. .. .. b .. b 1.8outward .. .. .. .. .. ..

Jordaninward 3.5 20.1 27.6 15.3 26.8 31.9outward 1.4 0.1 - 0.4 .. b .. b

Kuwaitinward 0.2 - 1.9 - 0.5 0.2 1.7 0.7outward - 4.8 - 138.0 - 45.9 19.9 4.0 .. b

Lebanoninward 8.3 11.3 8.9 1.9 6.8 11.6outward 3.1 0.5 1.4 1.5 2.6 3.1

Omaninward 1.0 15.5 - 0.5 16.2 12.6 14.0outward - - - 0.1 0.2 0.1

Palestinian Territoryinward - 4.1 .. .. .. 20.1 26.9outward .. .. .. .. .. ..

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

/...

321ANNEX B

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Qatarinward 15.5 13.9 13.4 1.0 10.8 14.6outward - 0.5 - - .. 0.4 0.2

Saudi Arabiainward 1.3 2.0 4.3 13.8 8.9 8.2outward 0.4 0.2 0.2 1.8 1.2 0.8

Syrian Arab Republicinward 5.8 21.3 22.7 3.0 33.3 52.6outward .. .. .. .. .. ..

Turkeyinward 3.5 4.7 5.1 7.4 9.6 11.7outward 0.6 1.3 1.6 0.8 1.8 2.3

United Arab Emiratesinward 9.0 0.2 4.6 2.2 2.0 4.6outward 2.8 0.3 - 0.2 0.3 1.5 1.5

Yemeninward 6.4 0.3 - 1.0 3.7 15.7 7.7outward 0.7 .. .. 0.1 - 0.1

South, East and South-East Asiainward 8.2 7.7 9.7 9.3 30.7 26.2outward 3.4 1.8 5.0 3.9 19.1 16.1

East Asiainward 8.9 8.1 10.1 9.7 34.8 28.4outward 3.7 1.6 5.1 5.8 25.3 20.5

Chinainward 10.4 8.6 8.2 5.8 17.9 14.9outward 0.5 - 0.2 1.3 2.6 2.4

Hong Kong, Chinainward 26.4 39.4 92.1 60.3 275.4 277.6outward 47.6 15.9 107.6 15.9 234.9 246.5

Korea, Dem. People’s Rep. ofinward .. .. .. 3.4 9.9 10.7outward .. .. .. .. .. ..

Korea, Rep. ofinward 1.9 2.1 3.8 2.1 8.1 8.1outward 1.6 1.9 2.4 0.9 5.8 5.8

Macao, Chinainward 51.9 36.0 35.9 86.4 45.2 52.1outward 9.8 - 0.5 1.5 .. .. 6.2

Mongoliainward 23.9 30.2 30.0 - 19.2 45.1outward .. .. .. .. .. ..

Taiwan Province of Chinainward 2.9 0.9 3.1 6.1 5.7 12.8outward 9.8 11.4 11.6 19.0 21.5 29.9

South Asiainward 3.2 3.2 3.7 1.1 4.7 6.3outward 0.8 0.6 1.2 0.1 0.4 0.9

Afghanistaninward 0.1 0.3 0.1 0.1 0.8 0.5outward .. .. .. .. .. ..

Bangladeshinward 0.5 2.2 3.5 1.1 5.0 6.1outward - - - 0.1 0.1 0.2

Bhutaninward 0.1 0.3 0.2 0.7 2.5 2.1outward .. .. .. .. .. ..

Indiainward 3.0 3.2 3.4 0.5 3.7 5.9outward 1.0 0.7 1.4 - 0.4 1.0

/...

322 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Maldivesinward 7.6 5.8 5.0 12.6 19.0 22.5outward .. .. .. .. .. ..

Nepalinward - 0.6 1.3 0.8 0.3 1.8 2.1outward .. .. .. .. .. ..

Pakistaninward 7.2 4.3 6.2 4.7 10.9 9.2outward 0.2 0.2 0.4 0.6 0.8 0.9

Sri Lankainward 5.6 5.6 5.1 8.5 9.8 10.8outward 0.3 0.7 0.1 0.1 0.5 0.6

South-East Asiainward 9.7 9.9 14.2 18.7 43.0 38.2outward 5.1 4.2 8.4 3.4 16.6 16.2

Brunei Darussalaminward .. .. .. 1.1 89.8 135.9outward .. .. .. .. 10.3 8.7

Cambodiainward 16.0 9.4 12.6 3.4 46.9 47.2outward 0.7 1.1 1.0 .. 5.7 5.8

Indonesiainward 0.4 - 1.3 1.9 7.7 16.5 4.4outward 0.5 - 0.2 0.1 4.6 -

Lao People’s Dem. Rep.inward 7.2 4.5 3.5 1.5 32.1 26.6outward .. - .. .. 1.6 1.2

Malaysiainward 14.5 10.8 19.1 23.4 58.6 39.3outward 8.6 6.0 8.5 6.1 23.6 11.7

Myanmarinward .. .. .. .. 9.3 7.9outward .. .. .. .. .. ..

Philippinesinward 13.3 2.6 3.3 7.4 16.9 14.9outward 0.4 1.5 2.9 0.3 2.1 1.9

Singaporeinward 25.6 41.7 62.7 83.1 123.1 150.2outward 18.0 16.5 41.6 21.3 62.1 94.5

Thailandinward 3.3 5.7 2.5 9.7 24.4 29.7outward 0.4 1.4 0.9 0.5 1.8 2.1

Timor-Lesteinward .. .. .. .. 22.3 50.6outward .. .. .. .. .. ..

Viet Naminward 11.0 11.6 11.3 25.5 65.7 66.3outward .. .. .. .. .. ..

Oceaniainward 0.6 16.5 5.2 28.9 30.3 21.2outward 0.5 0.9 0.1 5.8 3.2 3.2

Cook Islandsinward .. .. .. .. 42.5 25.4outward .. .. .. .. .. ..

Fijiinward 7.6 8.7 - 3.1 28.5 48.7 10.1outward 0.7 1.4 - 17.4 1.5 2.0

French Polynesiainward .. .. .. .. 4.2 2.4outward .. .. .. .. .. ..

/...

323ANNEX B

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (continued)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Kiribatiinward .. .. .. 1.2 1.9 1.3outward .. .. .. .. 0.1 -

New Caledoniainward .. .. .. 3.0 5.4 3.8outward .. .. .. .. .. ..

Palauinward .. .. .. .. 82.2 90.6outward .. .. .. .. .. ..

Papua New Guineainward 2.9 21.2 4.7 49.1 58.7 56.0outward 0.1 0.6 - 0.8 7.7 8.2

Samoainward .. .. .. 8.1 22.6 17.6outward .. .. .. .. .. ..

Solomon Islandsinward - 2.6 - 3.4 - 7.4 33.0 44.5 50.3outward .. .. .. .. .. ..

Tongainward 2.0 36.4 12.2 0.8 14.8 18.7outward .. .. .. .. .. ..

Tuvaluinward .. .. .. .. .. b 137.0outward .. .. .. .. .. ..

Vanuatuinward 16.6 25.9 32.6 71.8 165.4 136.0outward 1.1 1.1 1.2 .. .. 4.0

South-East Europe and the CISinward 11.6 17.1 19.1 0.2 15.8 21.5outward 4.2 7.8 5.6 0.3 5.4 9.9

South-East Europeinward 16.0 26.8 27.6 0.2 16.5 27.1outward 2.6 0.5 0.5 0.3 1.5 1.9

Albaniainward 10.9 13.0 26.0 .. 14.8 20.2outward 0.1 .. .. .. 2.1 1.1

Bosnia and Herzegovinainward 24.1 27.5 29.7 .. 9.0 20.1outward .. .. 0.1 .. 0.9 0.5

Bulgariainward 31.8 54.3 49.2 0.5 17.9 31.7outward 1.0 0.7 - 4.5 0.6 0.7 .. b

Croatiainward 20.1 25.7 11.4 .. 19.4 39.1outward 9.6 1.4 3.3 .. 4.7 7.3

Macedonia, TFYRinward 12.4 11.9 16.2 .. 11.4 24.8outward - - 0.1 .. - -

Romaniainward 11.7 17.4 31.7 - 17.5 25.2outward 0.2 0.3 0.4 0.2 0.4 0.4

Serbia and Montenegroinward 5.5 43.4 24.6 .. 12.0 16.4outward 0.2 - 1.1 .. .. .. ..

CISinward 10.4 14.3 16.8 .. 15.6 20.2outward 4.6 9.7 6.9 .. 6.3 11.5

Armeniainward 28.8 23.2 29.9 .. 33.0 28.3outward 3.8 0.1 0.3 .. 0.1 0.7

/...

324 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.3. FDI flows as a percentage of gross fixed capital formation (GFCF), 2002-2004 and FDI stocksas a percentage of gross domestic product (GDP), 1990, 2000, 2004, by region and economy (concluded)

(Per cent)

FDI flows as a percentage of GFCF FDI stocks as a percentage of GDP

Region/economy 2002 2003 2004 1990 2000 2004

Azerbaijaninward 65.5 90.6 105.5 .. 70.8 157.0outward 15.3 25.7 30.6 .. 9.0 30.9

Belarusinward 7.7 3.8 2.8 .. 12.5 9.0outward - 6.4 - - .. 0.2 -

Georgiainward 23.1 36.3 47.3 .. 13.9 34.5outward 0.6 0.4 0.9 .. .. ..

Kazakhstaninward 43.8 29.4 46.6 .. 55.1 55.0outward 7.2 - 1.7 - 14.0 .. 0.1 .. b

Kyrgyzstaninward 1.8 15.8 28.0 .. 32.6 26.2outward - - - 63.1 .. 2.4 .. b

Moldova, Rep. ofinward 48.8 21.2 42.6 .. 35.6 36.4outward 0.2 - 0.9 .. 1.8 1.0

Russian Federationinward 5.6 10.1 11.2 .. 12.4 16.9outward 5.7 12.4 9.2 .. 7.8 14.0

Tajikistaninward 27.2 18.8 173.7 .. 16.8 23.9outward .. 6.9 .. .. .. ..

Turkmenistaninward 8.1 6.3 8.4 .. 19.1 12.0outward - 14.3 - 8.0 .. .. .. ..

Ukraineinward 8.2 14.5 13.3 .. 12.4 14.2outward - 0.1 0.1 - .. 0.5 0.3

Uzbekistaninward 3.1 3.2 6.0 .. 5.1 10.9outward .. .. .. .. .. ..

MemorandumLeast developed countriesc

inward 16.2 23.0 20.8 5.8 18.5 24.4outward 0.4 0.2 0.2 0.9 2.5 2.1

Major pertoleum exportersd

inward 5.5 6.7 6.2 7.5 16.4 14.9outward 1.1 - 1.4 - 0.2 2.2 3.3 2.1

All developing economies, excluding Chinainward 9.1 8.8 11.8 10.2 27.8 29.1outward 3.9 2.5 6.4 4.7 15.8 15.3

EU-15inward 23.6 16.3 8.4 10.9 26.3 31.3outward 22.8 18.4 11.8 11.7 38.5 42.7

Source: UNCTAD, FDI/TNC database (www.unctad.org/fdistatistics).a Based on UNCTAD estimates for FDI stock.b Negative stock value. However, this value is included in the regional and global totals.c Least developed countries include: Afghanistan, Angola, Bangladesh, Benin, Bhutan, Burkina Faso, Burundi, Cambodia, Cape Verde, Central

African Republic, Chad, Comoros, Democratic Republic of Congo, Djibouti, Equatorial Guinea, Eritrea, Ethiopia, Gambia, Guinea, Guinea-Bissau, Haiti, Kiribati, Lao People’s Democratic Republic, Lesotho, Liberia, Madagascar, Malawi, Maldives, Mali, Mauritania, Mozambique,Myanmar, Nepal, Niger, Rwanda, Samoa, Sao Tome and Principe, Senegal, Sierra Leone, Solomon Islands, Somalia, Sudan, Timor-Leste,Togo, Tuvalu, Uganda, United Republic of Tanzania, Vanuatu, Yemen and Zambia.

d Major pertoleum exporters include: Algeria, Angola, Bahrain, Brunei Darussalam, Congo, Gabon, Indonesia, Islamic Republic of Iran, Iraq,Kuwait, Libyan Arab Jamahiriya, Nigeria, Oman, Qatar, Saudi Arabia, Syrian Arab Republic, Trinidad and Tobago, United Arab Emirates, Venezuelaand Yemen.

325ANNEX B

Annex table B.4. Cross-border M&As, by region/economy of seller/purchaser, 2002-2004(Mill ions of dollars)

Sales Purchases

Region/economy 2002 2003 2004 2002 2003 2004

World 369 789 296 988 380 598 369 789 296 988 380 598

Developed economies 322 502 244 426 315 851 341 548 256 935 339 799Europe 215 453 142 152 185 809 231 284 129 371 176 095

European Union 208 785 126 018 178 772 214 293 121 208 164 677Austria 38 2 115 1 787 1 848 1 744 5 810Belgium 5 449 3 182 2 345 5 474 3 166 9 309Cyprus .. 19 - 36 5 -Czech Republic 5 204 1 756 558 30 141 360Denmark 2 014 1 384 5 893 2 012 2 724 4 703Estonia 15 14 18 - 11 -Finland 8 206 3 557 3 232 5 304 600 2 712France 30 122 17 495 20 132 33 865 8 777 14 994Germany 46 605 25 158 35 868 45 110 19 669 18 613Greece 65 943 1 455 139 371 74Hungary 1 278 1 109 453 242 949 317Ireland 5 241 185 2 878 4 027 1 702 3 554Italy 11 608 15 259 10 953 8 242 4 662 5 167Luxembourg 2 952 958 72 3 683 613 558Latvia 4 12 - .. - -Lithuania 225 135 102 - - 5Malta 134 34 431 - - 52Netherlands 11 037 9 180 13 321 14 947 8 506 9 130Poland 3 131 802 1 275 58 529 216Portugal 1 132 1 732 1 233 1 481 107 3 105Slovakia 3 350 160 432 4 - 232Slovenia 1 502 1 168 63 15 59Spain 8 903 5 110 7 143 6 276 5 538 32 492Sweden 7 614 4 321 10 916 12 231 4 428 5 906United Kingdom 52 958 31 397 58 107 69 220 56 953 47 307

Other developed Europe 6 668 16 134 7 038 16 992 8 163 11 418Andorra - - - - - 38Gibraltar - - 92 - - -Guernsey 136 17 - - 339 775Iceland 229 142 365 358 289 1 952Isle of Man 52 - 4 .. 3 3Jersey 225 43 - 236 - 5Liechtenstein - - - - 159 -Monaco 8 382 198 .. 77 -Norway 2 162 5 579 1 603 6 823 303 3 080Switzerland 3 856 9 970 4 776 9 575 6 993 5 564

North America 89 549 74 827 101 574 91 419 98 436 144 068Canada 16 317 5 157 19 635 12 990 16 041 34 047United States 73 233 69 670 81 939 78 429 82 395 110 022

Other developed countries 17 499 27 448 28 467 18 845 29 128 19 636Australia 10 653 9 713 15 128 8 799 14 549 10 492Israel 466 808 171 544 1 357 4 003Japan 5 689 10 948 8 875 8 661 8 442 3 787New Zealand 692 5 979 4 292 840 4 780 1 354

Developing economies 44 410 40 166 54 700 27 549 31 060 39 809Africa 4 684 6 427 4 595 1 999 1 067 2 718

North Africa 598 4 594 443 5 433 111Algeria - 3 25 - - -Egypt 335 2 200 254 - 3 61Libyan Arab Jamahiriya - - - - 430 50Morocco 47 1 624 25 - - -Sudan 25 768 136 - - -Tunisia 191 - 3 5 - -

Other Africa 4 086 1 832 4 153 1 994 634 2 607West Africa 52 56 1 685 - 37 -

Burkina Faso - - 4 - - -Ghana 50 55 1 509 - - -Guinea - 1 - - - -Liberia - - - - 37 -Mali 2 - 13 - - -Mauritania - - 147 - - -

/...

326 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Nigeria - - 10 - - -Sierra Leone - - 2 - - -

Central Africa 993 - 65 - - -Equatorial Guinea 993 - - - - -Gabon - - 65 - - -

East Africa 30 127 350 47 9 272Kenya - - 265 .. 2 -Madagascar - 5 - - - -Malawi 6 - - - - -Mauritius - 32 19 40 - 22Mayotte - - 1 - - -Mozambique - 88 - - - -Rwanda - - 9 - - -Seychelles - - - - 7 -Uganda 20 - - - - 250United Rep. of Tanzania 1 2 - - - -Zambia - - 48 - - -Zimbabwe 4 - 7 7 - -

Southern Africa 3 011 1 650 2 053 1 947 588 2 334Botswana 78 20 70 .. 20 -Namibia - 67 16 - - 14South Africa 2 933 1 563 1 935 1 947 568 2 320Swaziland - - 33 - - -

Latin America and the Caribbean 22 433 12 085 25 284 11 701 11 460 16 487South and Central America 20 313 10 162 21 067 8 557 9 293 11 551

South America 12 395 8 566 13 148 3 643 3 879 9 488Argentina 1 207 2 467 285 4 679 103Bolivia 80 - - 4 - -Brazil 5 897 5 271 6 639 1 302 3 065 9 124Chile 3 783 95 1 720 1 744 39 95Colombia 830 37 1 421 530 2 28Ecuador 70 273 848 - - -Peru 461 247 710 59 91 18Uruguay 56 12 60 .. 3 -Venezuela 10 164 1 465 - - 120

Central America 7 918 1 595 7 919 4 914 5 414 2 063Belize - - 57 - - 5Costa Rica 229 23 20 - 13 81El Salvador .. 417 295 - - -Guatemala - - 175 .. - -Mexico 7 137 1 155 6 403 4 664 5 282 1 973Nicaragua 53 - 206 - - -Panama 499 - 763 249 120 4

Caribbean and other America 2 120 1 924 4 218 3 145 2 166 4 936Antigua and Barbuda - 47 40 - - -Aruba - - 715 - - -Bahamas 28 55 4 44 825 810Barbados 814 44 33 671 - -Bermuda 241 1 414 1 580 1 750 428 1 883British Virgin Islands 230 150 237 464 127 1 527Cayman Islands - 126 9 83 156 13Jamaica 214 - 324 - - -Netherlands Antil les 301 - - - 624 332Puerto Rico 250 - 1 251 133 7 370Saint Lucia - - 6 - - -Trinidad and Tobago 40 87 18 - - -United States Virgin Islands 1 - - .. - -

Asia and Oceania 17 293 21 654 24 820 13 849 18 533 20 604Asia 17 265 21 572 24 768 13 816 18 533 20 598

West Asia 458 1 404 575 3 038 1 555 1 280Bahrain - 9 - 646 432 -Iran, Islamic Rep. of 18 - 77 .. - 9Iraq - - 9 - - -Jordan - 990 - - - -Kuwait - - 317 114 441 845

Annex table B.4. Cross-border M&As, by region/economy of seller/purchaser, 2002-2004 (concluded)(Mill ions of dollars)

Sales Purchases

Region/economy 2002 2003 2004 2002 2003 2004

/...

327ANNEX B

Lebanon - 98 - .. - 7Oman 4 - 20 9 125 -Qatar .. - - - 15 192Saudi Arabia .. - - 2 020 473 78Syrian Arab Republic .. - 7 - - -Turkey 427 282 132 38 7 108United Arab Emirates 9 26 14 10 62 40Abu Dhabi - - - 201 - -

South, East and South-east Asia 16 807 20 167 24 193 10 778 16 978 19 319South Asia 1 923 1 461 2 218 336 1 362 877

Bangladesh - 437 60 - - -India 1 698 949 1 760 270 1 362 863Pakistan 222 - 398 63 - 14Sri Lanka 3 76 - 3 - -

East Asia 9 991 14 105 16 743 6 280 6 730 5 207China 2 072 3 820 6 768 1 047 1 647 1 125Hong Kong, China 1 865 6 098 3 936 5 062 4 168 2 963Korea, Dem. People’s Rep. of 90 - - - - -Korea, Rep. of 5 375 3 757 5 638 98 662 409Macao, China 109 - - - - -Mongolia 0 7 3 - - -Taiwan Province of China 480 422 398 74 253 710

South-East Asia 4 893 4 601 5 232 4 163 8 886 13 235Brunei Darussalam - - 5 - - -Cambodia - - 1 - - -Indonesia 2 790 2 031 1 269 197 2 491Lao People’s Dem. Rep. 266 - 85 - - -Malaysia 485 84 638 930 3 685 816Myanmar .. 417 - - - -Philippines 544 230 733 2 1 105Singapore 556 1 766 1 190 2 946 5 018 11 638Thailand 247 55 1 236 87 176 185Viet Nam 6 18 74 .. 4 -

Oceania 28 83 53 33 - 5Fiji - 1 - - - 4Marshall Islands - - 6 - - -New Caledonia - - 1 - - -Northern Mariana Islands - - 33 - - -Papua New Guinea 28 82 13 28 - 2Samoa - - - 5 - -

South-East Europe and CIS 2 877 12 395 10 047 691 8 992 991South-East Europe 1 429 2 355 5 294 85 56 36

Albania - 2 126 - - -Bosnia and Herzegovina 19 - 110 - - -Bulgaria 138 383 2 685 8 - 30Croatia 875 613 51 42 32 6Macedonia, TFYR 5 - 4 16 - -Romania 124 493 2 200 19 1 -Serbia and Montenegro - - 38 - - -Yugoslavia (former) 268 863 80 - 23 -

CIS 1 447 10 040 4 753 606 8 936 954Armenia 52 25 - - - -Azerbaijan 52 1 387 - - - -Belarus - 2 5 - - -Georgia - 1 - - - -Kazakhstan 1 507 428 - 170 5Kyrgyzstan 1 5 3 - - -Moldova, Rep. of - 19 16 - - -Russian Federation 1 252 7 880 4 062 606 8 763 949Tajikistan 4 - - - - -Ukraine 74 194 41 - 3 -Uzbekistan 11 21 199 - - -

Source: UNCTAD, cross-border M&A database (www.unctad.org/fdistatistics).

Note: The data cover the deals involving the acquisition of an equity stake of more than 10 per cent.

Annex table B.4. Cross-border M&As, by region/economy of seller/purchaser, 2002-2004 (continued)(Mill ions of dollars)

Sales Purchases

Region/economy 2002 2003 2004 2002 2003 2004

328 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

Annex table B.5. Cross-border M&As, by sector/industry, 2002-2004(Millions of dollars)

Sales Purchases

Sector/industry 2002 2003 2004 2002 2003 2004

Total 369 789 296 988 380 598 369 789 296 988 380 598

Primary 12 751 7 714 6 978 9 309 4 227 4 766Agriculture, hunting, forestry, and fishing 265 1 350 1 245 37 228 648Mining, quarrying and petroleum 12 486 6 363 5 733 9 272 4 000 4 119

Manufacturing 137 414 129 713 134 975 115 460 112 758 119 674Food, beverages and tobacco 32 072 29 597 23 870 20 996 23 307 22 735Textiles, clothing and leather 915 676 1 585 549 681 256Wood and wood products 7 325 2 765 3 769 5 258 2 671 3 916Publishing, and printing 2 986 11 886 8 965 5 731 11 370 4 578Coke, petroleum and nuclear fuel 33 018 24 267 15 108 28 201 20 260 13 138Chemicals and chemical products 20 370 22 927 41 788 20 958 16 927 31 290Rubber and plastic products 2 257 1 582 570 819 893 747Non-metallic mineral products 3 183 2 688 5 178 2 186 1 867 6 032Metals and metal products 10 034 8 083 4 579 9 015 11 390 4 541Machinery and equipment 2 564 4 332 6 688 3 432 1 932 4 722Electrical and electronic equipment 8 556 5 409 12 998 8 678 7 817 18 216Precision instruments 5 064 8 046 5 871 2 689 7 072 4 799Motor vehicles and other transport equipment 8 590 5 760 3 639 6 516 6 322 4 010Other manufacturing 479 1 694 367 432 250 696

Services 219 623 159 561 238 645 243 771 180 002 256 156Electricity, gas and water 61 572 15 909 24 799 57 866 13 440 17 596Construction 1 465 1 089 3 324 1 041 1 048 610Trade 16 710 13 183 26 445 22 886 10 761 13 087Hotels and restaurants 3 860 4 142 4 618 1 433 5 496 1 268Transport, storage and communications 30 824 34 724 36 214 37 115 21 598 24 634Finance 41 903 54 790 81 809 90 787 114 150 174 096Business services 47 248 23 565 55 261 29 805 9 090 22 387Public administration and defence 76 55 18 318 604 -Education 7 77 79 .. 41 88Health and social services 781 1 115 2 726 710 541 321Community, social and personal service activities 15 169 10 911 3 349 1 809 3 231 2 068Other services 7 2 3 .. 2 -

Unknowna - - - 1 248 - 2

Source: UNCTAD, cross-border M&A database (www.unctad.org/fdistatistics).a Including non-classified establishments.

������������ ������������������������� �

I. WORLD INVESTMENT REPORT PASTISSUES

UNCTAD, World Investment Report 2004. The ShiftTowards Services (New York and Geneva, 2004). 468pages. Document symbol: UNCTAD/WIR/2004. SalesNo. E.04.II.D.36. $75.

UNCTAD, World Investment Report 2004. The ShiftTowards Services. Overview. 54 pages (A, C, E, F,R, S). Document symbol: UNCTAD/WIR/2004 (Over-view). Available free of charge.

UNCTAD, World Investment Report 2003. FDI Poli-cies for Development: National and InternationalPerspectives (New York and Geneva, 2003). 303pages. Sales No. E.03.II.D.8.

UNCTAD, World Investment Report 2003. FDI Poli-cies for Development: National and InternationalPerspectives. Overview. 42 pages (A, C, E, F, R, S).Document symbol: UNCTAD/WIR/2003 (Overview).Available free of charge.

UNCTAD, World Investment Report 2002:Transnational Corporations and Export Competi-tiveness (New York and Geneva, 2002). 350 pages.Sales No. E.02.II.D.4.

UNCTAD, World Investment Report 2002:Transnational Corporations and Export Competi-tiveness. Overview. 66 pages (A, C, E, F, R, S). Docu-ment symbol: UNCTAD/WIR/2002 (Overview).Available free of charge.

UNCTAD, World Investment Report 2001: PromotingLinkages (New York and Geneva, 2001). 354 pages.Sales No. E.01.II.D.12.

UNCTAD, World Investment Report 2001: PromotingLinkages. Overview . 63 pages (A, C, E, F, R, S).Document symbol: UNCTAD/WIR/2001 (Overview).Available free of charge.

UNCTAD, World Investment Report 2000: Cross-border Mergers and Acquisitions and Development(New York and Geneva, 2000). 337 pages. Sales No.E.00.II.D.20.

UNCTAD, World Investment Report 2000: Cross-border Mergers and Acquisitions and Development.Overview . 65 pages (A, C, E, F, R, S). Documentsymbol: UNCTAD/WIR/2000 (Overview). Availablefree of charge.

UNCTAD, World Investment Report 1999: ForeignDirect Investment and the Challenge of Develop-ment (New York and Geneva, 1999). 541 pages. SalesNo. E.99.II.D.3.

UNCTAD, World Investment Report 1999: ForeignDirect Investment and the Challenge of Develop-ment. Overview. 75 pages (A, C, E, F, R, S). Docu-

ment symbol: UNCTAD/WIR/1999 (Overview).Available free of charge.

UNCTAD, World Investment Report 1998: Trendsand Determinants (New York and Geneva, 1998). 463pages. Sales No. E.98.II.D.5.

UNCTAD, World Investment Report 1998: Trendsand Determinants. Overview. 72 pages (A, C, E, F,R, S). Document symbol: UNCTAD/WIR/1998 (Over-view). Available free of charge.

UNCTAD, World Investment Report 1997:Transnational Corporations, Market Structure andCompetition Policy (New York and Geneva, 1997).416 pages. Sales No. E.97.II.D. 10.

UNCTAD, World Investment Report 1997:Transnational Corporations, Market Structure andCompetition Policy. Overview. 76 pages (A, C, E,F, R, S). Document symbol: UNCTAD/ITE/IIT/5(Overview). Available free of charge.

UNCTAD, World Investment Report 1996: Invest-ment, Trade and International Policy Arrangements(New York and Geneva, 1996). 364 pages. Sales No.E.96.11.A. 14.

UNCTAD, World Investment Report 1996: Invest-ment, Trade and International Policy Arrangements.Overview . 22 pages (A, C, E, F, R, S). Documentsymbol: UNCTAD/DTCI/32 (Overview). Availablefree of charge.

UNCTAD, World Investment Report 1995:Transnational Corporations and Competitiveness(New York and Geneva, 1995). 491 pages. Sales No.E.95.II.A.9.

UNCTAD, World Investment Report 1995:Transnational Corporations and Competitiveness.Overview . 68 pages (A, C, E, F, R, S). Documentsymbol: UNCTAD/DTCI/26 (Overview). Availablefree of charge.

UNCTAD, World Investment Report 1994:Transnational Corporations, Employment and theWorkplace (New York and Geneva, 1994). 482 pages.Sales No.E.94.11.A.14.

UNCTAD, World Investment Report 1994:Transnational Corporations, Employment and theWorkplace. An Executive Summary. 34 pages (C, E,also available in Japanese). Document symbol:UNCTAD/DTCI/10 (Overview). Available free ofcharge.

UNCTAD, World Investment Report 1993:Transnational Corporations and Integrated Inter-national Production (New York and Geneva, 1993).290 pages. Sales No. E.93.II.A.14.

330 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

UNCTAD, World Investment Report 1993:Transnational Corporations and Integrated Interna-tional Production. An Executive Summary. 31 pages(C, E). Document symbol: ST/CTC/159 (ExecutiveSummary). Available free of charge.

DESD/TCMD, World Investment Report 1992:Transnational Corporations as Engines of Growth(New York, 1992). 356 pages. Sales No. E.92.II.A.24.

DESD/TCMD, World Investment Report 1992:Transnational Corporations as Engines of Growth:An Executive Summary. 26 pages. Document symbol:ST/CTC/143 (Executive Summary). Available free ofcharge.

UNCTC, World Investment Report 1991: The Triadin Foreign Direct Investment (New York, 1991). 108pages. Sales No. E.9 1.II.A. 12. $25.

II. OTHER PUBLICATIONS(2002-2005)

A. Studies on Trends in FDI and the Activitiesof TNCs

UNCTAD, Prospects for Foreign Direct Investmentand the Strategies of Transnational Corporations2004-2007 (Geneva, 2004). 61 pages. Sales No.E.05.II.D.3. $12.

UNCTAD, FDI in Landlocked Developing Countriesat a Glance (Geneva, 2003). Document symbol:UNCTAD/ITE/IIA/2003/5. Available free of charge.

UNCTAD, Foreign Direct Investment in the Worldand Poland: Trends, Determinants and EconomicImpact. (Warsaw, 2002). ISBN 83-918182-0-9.

UNCTAD, FDI in ACP Economies: Recent Trends andDevelopment (Geneva, 2002). 36 pages. Documentsymbol: UNCTAD/ITE/IIA/Misc.2.

UNCTAD, FDI in Least Developed Countries at aGlance: 2002 (Geneva, 2002). Document symbol:UNCTAD/ITE/IIA/6. 150 pages. Available free ofcharge.

B. Development Issues and FDI

Transnational Corporations. A refereed journal pub-lished three times a year. (Supersedes the CTC Reporteras of February 1992). Annual subscription (3 issues):$45. Single issue: $20.

UNCTAD, Investment and Technology Policies forCompetitiveness: Review of Successful Country Ex-periences (Geneva, 2003). Document symbol:UNCTAD/ITE/ICP/2003/2.

UNCTAD, The Development Dimension of FDI:Policy and Rule-Making Perspectives (Geneva, 2003).Sales No. E.03.II.D.22. $35.

UNCTAD, FDI and Performance Requirements: NewEvidence from Selected Countries (Geneva, 2003).Sales No. E.03.II.D.32. 318 pages. $ 35.

C. Sectoral Studies

UNCTAD, TNCs and the Removal of Textiles andClothing Quotas (New York and Geneva, 2005). SalesNo. E.05.II.D.20.

UNCTAD, Tradability of Consulting Services and ItsImplications for Developing Countries (New York andGeneva, 2002).189 pages. UNCTAD/ITE/IPC/Misc.8.

D. TNCs, Technology Transfer and IntellectualProperty Rights

UNCTAD, Science, Technology and Innovation PolicyReview: The Islamic Republic of Iran (Geneva, 2005).118 pages. Document symbol: UNCTAD/ITE/IPC/2005/7.

UNCTAD, Facilitating Transfer of Technology to De-veloping Countries: A Survey of Home-Country Meas-ures (New York and Geneva, 2004). 52 pages. Docu-ment symbol: UNCTAD/ITE/IPC/2004/5.

UNCTAD, The Biotechnology Promise – Capacity-Building for Participation of Developing Countriesin the Bio Economy (Geneva, 2004). 141 pages. Docu-ment symbol: UNCTAD/ITE/IPC/2004/2.

UNCTAD, Investment and Technology Policies forCompetitiveness: Review of Successful Country Ex-periences (Geneva, 2003). 79 pages. Document symbol:UNCTAD/ITE/IPC/2003/2.

UNCTAD, Africa’s Technology Gap: Case Studies onKenya, Ghana, Tanzania and Uganda (Geneva, 2003).123 pages. Document symbol: UNCTAD/ITE/IPC/Misc.13.

UNCTAD, Transfer of Technology for Successful In-tegration into the Global Economy (New York andGeneva, 2003). Sales No. E.03.II.D.31. 206 pages.

E. International Arrangements andAgreements

1. Series on Issues in InternationalInvestment Agreements (IIAs)

UNCTAD, State Contracts (New York and Geneva,2005). 84 pages. Document symbol: UNCTAD/ITE/IIT/2004/11. Sales No. E.05.II.D.5. $15.

UNCTAD, Competition (New York and Geneva, 2004).112 pages. Document symbol: UNCTAD/ITE/IIT/2004/6. Sales No. E.04.II.D.44. $15.

UNCTAD, Glossary of Key Concepts Used in IIAs.UNCTAD Series on Issues in International InvestmentAgreements (New York and Geneva, 2003).

331Selected UNCTAD Publications on TNCs and FDI

UNCTAD, Incentives UNCTAD Series on Issues inInternational Investment Agreements (New York andGeneva, 2003). Sales No. E.04.II.D.6. $15.

UNCTAD, Transparency. UNCTAD Series on Issuesin International Investment Agreements (New Yorkand Geneva, 2003). Sales No. E.03.II.D.7. $15.

UNCTAD, Dispute Settlement: Investor-State.UNCTAD Series on Issues in International Invest-ment Agreements (New York and Geneva, 2003). 128pages. Sales No. E.03.II.D.5. $15.

UNCTAD, Dispute Settlement: State-State. UNCTADSeries on Issues in International Investment Agree-ments (New York and Geneva, 2003). 109 pages. SalesNo. E.03.II.D.6 $16.

2. Series on International InvestmentPolicies for Development

UNCTAD, A Wave of South-South Cooperation inthe Area of International Investment Policies (NewYork and Geneva, 2005). 64 pages. Document symbol:UNCTAD/ITE/IIT/2005/3.

UNCTAD, The REIO Exception in MFN TreatmentClauses (New York and Geneva, 2004). 92 pages.Document symbol: UNCTAD/ITE/IIT/2004/7. SalesNo. E.05.II.D.1. $15.

3. Other studies

UNCTAD, International Investment Agreements: KeyIssues Vols. I, II and III, Sales No. E.05.II.D.6.

UNCTAD, International Investment Instruments:A Compendium (New York and Geneva). Vol. VII:Sales No. E.02.II .D.14. Vol. VIII: Sales No.E.02.II.D.15. Vol. IX: Sales No. E.02.II.D.16. Vol.X: Sales No. E.02.II .D.21. Vol. XI: Sales No.E.04.II .D.9. Vol. XII: Sales No. E.04.II .D.10.Vol. XIII: Sales No. E.05.II.D.7. Vol. XIV: Sales No.E.05.II.D.8.

F. National Policies, Laws, Regulations andContracts Relating to TNCs

1. Investment Policy Reviews

UNCTAD, Investment Policy Review of Kenya (Ge-neva, 2005). 126 pages. UNCTAD/ITE/IPC/2005/8.Sales No. E.05.II.D.21.

UNCTAD, Investment Policy Review of Benin (Ge-neva, 2005). 147 pages. UNCTAD/ITE/IPC/2003/4.Sales No. F.04.II.D.43.

UNCTAD, Investment Policy Review of Algeria (Ge-neva, 2004). 110 pages. UNCTAD/ITE/IPC/2003/9.

UNCTAD, Investment Policy Review of Sri Lanka(Geneva, 2003). 89 pages. UNCTAD/ITE/IPC/2003/8.

UNCTAD, Investment Policy Review of Lesotho (Ge-neva, 2003). 105 pages. Sales No. E.03.II.D.18.

UNCTAD, Investment Policy Review of Nepal. (Ge-neva, 2003). 89 pages. Sales No.E.03.II.D.17.

UNCTAD, Investment Policy Review of Ghana (Ge-neva, 2002). 103 pages. Sales No. E.02.II.D.20.

UNCTAD, Investment Policy Review of Botswana(Geneva, 2003). 107 pages. Sales No. E.03.II.D.1.

UNCTAD, Investment Policy Review of Tanzania(Geneva, 2002). 109 pages. Sales No. E.02.II.D.6.

2. Investment Guides

UNCTAD and ICC, An Investment Guide to Mau-ritania (Geneva, 2004). Document symbol: UNCTAD/IIA/2004/4. Free of charge.

UNCTAD and ICC, An Investment Guide to Cam-bodia (Geneva, 2003). 89 pages. Document symbol:UNCTAD/IIA/2003/6. Free of charge.

UNCTAD and ICC, An Investment Guide to Nepal(Geneva, 2003). 97 pages. Document symbol:UNCTAD/IIA/2003/2. Free of charge.

UNCTAD and ICC, An Investment Guide to Mozam-bique (Geneva, 2002). 109 pages. Document sym-bol: UNCTAD/IIA/4. Free of charge.

G. International Standards of Accountingand Reporting

UNCTAD, International Accounting and ReportingIssues:

2003 Review (Geneva, 2003). UNCTAD/ITE/TEB/2003/9.

2002 Review (Geneva, 2002). UNCTAD/ITE/TEB/2003/4.

UNCTAD, Accounting and Financial ReportingGuidelines for Small and Medium-Sized Enterprises(SMEGA)): Level 3 Guidance (Geneva, 2004). 20pages. Document symbol: UNCTAD/ITE/TEB/2003/6. Sales No. E.04.II.D.15. $10.

UNCTAD, Accounting and Financial ReportingGuidelines for Small and Medium-Sized Enterprises(SMEGA)): Level 2 Guidance (Geneva, 2004). 72pages. Document symbol: UNCTAD/ITE/TEB/2003/5. Sales No. E.04.II.D.14. $15.

UNCTAD, A Manual for the Preparers and Usersof Eco-efficiency Indicators (New York and Geneva,2004). 126 pages. Document symbol: UNCTAD/ITE/IPC/2003/7. Sales No. E.04.II.D.13. $28.

UNCTAD, Selected Issues in Corporate Governance:Regional and Country Experiences (New York andGeneva, 2003). Sales No. E.03.II.D.26.

332 World Investment Report 2005: Transnational Corporations and the Internationalization of R&D

H. Data and Information Sources

UNCTAD, World Investment Directory.

Volume IX: Latin America and the Caribbean(New York and Geneva, 2004). Sales No.E.03.II.D.12. $25.

HOW TO OBTAIN THE PUBLICATIONS

The sales publications may be purchased from distributors of United Nations publicationsthroughout the world. They may also be obtained by writing to:

United Nations Publications or United Nations PublicationsSales and Marketing Section, DC2-853 Sales and Marketing Section, Rm. C. 113-1United Nations Secretariat United Nations Office at GenevaNew York, N.Y. 100 17 Palais des NationsU.S.A. CH-1211 Geneva 10Tel.: ++1 212 963 8302 or 1 800 253 9646 SwitzerlandFax: ++1 212 963 3489 Tel.: ++41 22 917 2612E-mail: publications@un.org Fax: ++4122 917 0027

E-mail: unpubli@unog.ch

INTERNET: www.un.org/Pubs/sales.htm

For further information on the work on foreign direct investment and transnational corporations,please address inquiries to:

Khalil HamdaniOfficer-in-ChargeDivision on Investment, Technology and Enterprise DevelopmentUnited Nations Conference on Trade and DevelopmentPalais des Nations, Room E-10052CH-1211 Geneva 10 SwitzerlandTelephone: ++41 22 907 4533Fax: ++41 22 907 0498E-mail: khalil.hamdani@unctad.org

INTERNET: www.unctad.org/en/subsites/dite

Volume VIII: Central and Eastern Europe (NewYork and Geneva, 2003). Sales No. E.03.II.D.12.$25.

A list of publications on FDI and TNCs (1973-2003) is available in the Digital l ibrary of UNCTAD atwww.unctad.org, and in hard copy (document symbol: UNCTAD/ITE/2004).

QUESTIONNAIRE

World Investment Report 2005:Transnational Corporations and the

Internationalization of R&D

In order to improve the quality and relevance of the work of the UNCTAD Division on Investment,Technology and Enterprise Development, it would be useful to receive the views of readers on this andother similar publications. It would therefore be greatly appreciated if you could complete the followingquestionnaire and return it to:

Readership SurveyUNCTAD, Division on Investment,Technology and Enterprise DevelopmentPalais des NationsRoom E-10054CH-1211 Geneva 10SwitzerlandOr by Fax to: (+41 22) 907 04 98

1. Name and professional address of respondent (optional):

2. Which of the following best describes your area of work?

Government Public enterprisePrivate enterprise institution Academic or researchInternational organization MediaNot-for-profit organization Other (specify)

3. In which country do you work?

4. What is your assessment of the contents of this publication?

Excellent AdequateGood Poor

5. How useful is this publication to your work?

Very useful Of some use Irrelevant

6. Please indicate the three things you liked best about this publication and how are they useful foryour work:

7. Please indicate the three things you liked least about this publication:

This questionnaire is alsoavailable to be filled out online at: www.unctad.org/wir.

8. On the average, how useful is this publication to you in your work?

Very useful Of some use Irrelevant

9. Are you a regular recipient of Transnational Corporations, UNCTAD’s tri-annual refereed journal?

Yes No

If not, please check here if you would like to receive a sample copy sent to the name and addressyou have given above. Other title you would like to receive instead (see list of publications):

10. How and where did you obtain this publication:

I bought it In a seminar/workshopI requested a courtesy copy Direct mailingOther

11. Would you like to receive information on UNCTAD’s work in the areas of investment, technology andenterprise development through e-mail? If yes, please provide us with your e-mail address: