divergent paths in post-communist transformation: capitalism for all or capitalism for the few?

Divergent Pathsin Post-Communist

TransformationCapitalism for All or Capitalism for the Few?

Oleh Havrylyshyn

Studies in Economic Transition

General Editors: Jens Höischer, Reader in Economics, University of Brighton; andHorst Tomann, Professor of Economics, Free University Berlin

This series has been established in response to a growing demand for a greater under-standing of the transformation of economic systems. It brings together theoretical andempirical studies on economic transition and economic development. The post-com-munist transition from planned to market economies is one of the main areas ofapplied theory because in this field the most dramatic examples of change and eco-nomic dynamics can be found. The series aims to contribute to the understandingof specific major economic changes as well as to advance the theory of economicdevelopment. The implications of economic policy will be a major point of focus.

Titles include:

Irwin Collier, Herwig Roggemann, Oliver Scholz and Horst Tomann (editors)WELFARE STATES IN TRANSITIONEast and West

Hella EngererPRIVATIZATION AND ITS LIMITS IN CENTRAL AND EASTERN EUROPEProperty Rights in Transition

Hubert Gabrisch and Rüdigger Pohl (editors)EU ENLARGEMENT AND ITS MACROECONOMIC EFFECTS IN EASTERN EUROPECurrencies, Prices, Investment and Competitiveness

Oleh HavrylyshynDIVERGENT PATHS IN POST-COMMUNIST TRANSFORMATIONCapitalism for All or Capitalism for the Few?

Jens Hölscher (editor)FINANCIAL TURBULENCE AND CAPITAL MARKETS IN TRANSITION COUNTRIES

Jens Hölscher and Anja Hochberg (editors)EAST GERMANY’S ECONOMIC DEVELOPMENT SINCE UNIFICATION Domestic and Global Aspects

Mihaela Kelemen and Monika Koestra (editors)CRITICAL MANAGEMENT RESEARCH IN EASTERN EUROPEManaging the Transition

Emil J. Kirchner (editor)DECENTRALIZATION AND TRANSITION IN THE VISEGRADPoland, Hungary, the Czech Republic and Slovakia

Julie PellegrinTHE POLITICAL ECONOMY OF COMPETITIVENESS IN AN ENLARGED EUROPE

Stanislav Poloucek (editor)REFORMING THE FINANCIAL SECTOR IN CENTRAL EUROPEAN COUNTRIES

Gregg S. RobinsBANKING IN TRANSITIONEast Germany after Unification

Studies in Economic TransitionSeries Standing Order ISBN 0–333–73353–3(Outside North America Only)

You can receive future titles in this series as they are published by placing a standing order. Pleasecontact your bookseller or, in case of difficulty, write to us at the address below with your nameand address, the title of the series and the ISBN quoted above.

Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, HampshireRG21 6XS, England

Johannes StephanECONOMIC TRANSITION IN HUNGARY AND EAST GERMANYGraudualism and Shock Therapy in Catch-up Development

Hans van ZonTHE POLITICAL ECONOMY OF INDEPENDENT UKRAINE

Adalbert Winkler (editor)FINANCIAL DEVELOPMENT IN EASTERN EUROPEThe First Ten Years

Also by Oleh Havrylyshyn

A DECADE OF TRANSITION (co-edited with Saleh Nsouli)

FUNDAMENTALS OF ECONOMIC THEORY (in Ukrainian)

LIBERALIZING FOREIGN TRADE IN DEVELOPING COUNTRIES: Vol. 3, The Experienceof Yugoslavia

DEVELOPMENT OF TRADE POLICY IN POLAND (in Polish with Dariusz Rosati)

CANADA’S TRADE RELATIONS WITH DEVELOPING COUNTRIES (with Vittorio Corbo)

PLANNING FOR ECONOMIC DEVELOPMENT: The Construction and Use of a Multi-Sectoral Model for Tunisia (with André Martens, Robert Pindyck and Mouheb-EddineHamza)

Divergent Paths in Post-CommunistTransformationCapitalism for All or Capitalism for the Few?

Oleh Havrylyshyn

© International Monetary Fund 2006Foreword © International Monetary Fund 2006

All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission.

No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP.

Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages.

The author has asserted his right to be identified as the author of this work in accordance with the Copyright,Designs and Patents Act 1988.

First published 2006 byPALGRAVE MACMILLANHoundmills, Basingstoke, Hampshire RG21 6XS and175 Fifth Avenue, New York, N.Y. 10010Companies and representatives throughout the world.

PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd.Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries.

ISBN-13: 978–1–4039–9634–3ISBN-10: 1–4039–9634–2

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources.

A catalogue record for this book is available from the British Library.

Library of Congress Cataloging-in-Publication DataHavrylyshyn, Oleh.

Divergent paths in post-communist transformation : capitalism for all orcapitalism for the few? / Oleh Havrylyshyn.

p. cm.—(Studies in economic transition)Includes bibliographical references and index.ISBN 1–4039–9634–2 (cloth)1. Europe, Eastern – Economic conditions – 1989– 2. Europe,

Central – Economic conditions. 3. Former Soviet republics – Economic conditions. 4. Post-communism. 5. Comparative economics. I. Title. II. Series.

HC244.H379 2005330.947�0009�049—dc22 2005049930

10 9 8 7 6 5 4 3 2 115 14 13 12 11 10 09 08 07 06

Printed and bound in Great Britain byAntony Rowe Ltd, Chippenham and Eastbourne

To my wife, Natalia Ingrid, who supported in all ways my effortsfrom the beginning of transition to the completion of this volume,and provided numerous insights to help see more clearly throughthe post-communist fog

This page intentionally left blank

Contents

List of Tables and Figures x

Preface and Acknowledgements xii

Foreword by Anne O. Krueger xv

Introduction and Overview 1

Part I Background: Transition Debates and Statistical Facts

1 Key Debates on Transition 15The search for a transition model 15Principal debates of the 1990s 21Main issues looking forward 38The transitology paradigm in political science literature 42

2 Measuring Progress in Transition 47How to measure progress 47The EBRD transition indicator: how good a measure of progress? 48Patterns of reform progress over time 58Main determinants of growth in transition 62Appendix: unrecorded economic activity and measures of GDP 70

3 Economic and Social Costs of Transition: A Hard Look at Soft Facts 78Introduction 78Were negative effects inevitable? 79Aggregate economic costs: lost output and

unemployment output losses 81Impact on individuals as measured by

social indicators 91Summing up the evidence on social costs 115Appendix: estimates of output loss under different assumptions 118

Part II An Ex Post Transition Paradigm: Uncharted Waters, Pirate Raids and Safe Havens

4 The SS Transition Navigation Model 123Introduction 123The value of comprehensive explanations 125

vii

A parsimonious approach: the navigation model 133Testable hypotheses 143What is new in the navigation model 146Appendix: an illustrative formalization of the navigation model 149

5 The Search for a Navigation Chart: Legitimate Debates, Vested Interests and Reformist Commitments 151The release from communist captivity: carpe diemor follow the money? 151Availability of navigation charts 152A timeline of economic transformation events 158Commitment to a liberal market economy 167

6 From Rent-Seeking to Oligarchy to State Capture 177Overview 177Partial reforms and the evolution of rent-seeking 178Other forces affecting vulnerability to rent-seeking 185From rent-seeking to state capture 191Appendix: rent-seeking, vested interests and

entrenchment in non-transition economies 199

7 Safe Havens for Market Reforms: Membership in the EU and Other International Organizations 203Courtship games of the EU and post-communist countries 203EU membership incentives: sometimes a carrot,

sometimes a stick 208Unrequited desires for membership 212Other international beacons 216EU membership prospects and reform delays 220Appendix: EU membership as a factor in

political and economic reform in the Baltic republics 225

Part III A Summing Up

8 Future Prospects for Captured States 233A fork in the transition road 233Is further transition inevitable or is it frozen? 234Consequences of a frozen transition 242Can anything be done to free the captured states? 248

9 Diverse Outcomes: Liberal Societies, Captured States and Undetermined Polities 255Introduction 255Variation in transition progress 255

viii Contents

Contents ix

Why such divergent outcomes? 264Future prospects and policy implications 272

Notes 277

Bibliography 295

Index 309

List of Tables and Figures

Tables

1.1 Quality of institutions in transition economies 1997–8 372.1 Recovery: estimated index of GDP, 2003 532.2 Inflation performance 542.3 Cumulative FDI per captia, 1989–2003 542.4 Transition indicator values by year and type 592.5 Growth of GDP since 1998 692.6 Year that the TPI growth threshold (2.55) was reached

for CISM countries 69A2.1 Transitional indicator averages 73

3.1 Official and adjusted index of output 813.2 Summary estimates of cumulative output loss 853.3 Unemployment rates in the transition 883.4 Average human development indicator values 933.5 Range of Gini estimates in the literature 973.6 Probable changes in Gini values 1003.7 Range of Gini values by country groups 1023.8 Poverty ratio estimates: various sources 1053.9 Range of poverty ratios by country group and period 106

3.10 Male life-expectancy trends in transition 1123.11 Gross educational enrolment ratios 114A3.1 Estimates of cumulative loss of output 120

5.1 Chronology of events: post-communist countries 1605.2 Milestones of transition progress 1625.3 Type of first government and reform strategy 1695.4 Liberal commitment at start and reform strategy 1705.5 Countries with reversals in government 1716.1 State-capture index, 1999 1927.1 Strength of EU ‘offer’ and reform delay 221

A7.1 Contractual relations between the EU and the Baltic republics up to 1995 230

9.1 Key indicators of transformation outcomes 2589.2 Final transformation outcomes and principal determinants 265

x

List of Tables and Figures xi

Figures

1.1 An illustration of the Washington Consensus 182.1 EBRD transition progress indicator (TPI), 2004 502.2 Constitutional liberalism and progress in transition 56

A2.1 CE transition progress indicators 74A2.2 CE average liberalization index per year 74A2.3 Baltics transition progress indicators 74A2.4 Baltics average liberalization index per year 75A2.5 SEE transition progress indicators 75A2.6 SEE average liberalization index per year 75A2.7 CISM transition progress indicators 76A2.8 CISM average liberalization index per year 76A2.9 CISL transition progress indicators 76

A2.10 CISL average liberalization index per year 77A3.1 Actual GDP and socialist counterfactuals 119A3.2 Schema for calculating GDP loss under various

counterfactuals 1194.1 An encyclopedic approach to explaining diverse

transition results 1264.2 Transition progress and continuity of communist-era judges 1304.3 Vicious circle of delayed reform and oligarchic development 1384.4 Virtuous circle of timely reform and development of

competitive market and rule of law 1394.5 Testable hypotheses of the navigation model 1475.1 Uncharted waters or too many charts? 1545.2 Transition countries and type of reform strategy 1596.1 Rent-seeking type A: gradual or aborted big-bang

reformers, high vulnerability 1836.2 Rent-seeking, types A and B 1846.3 Delay facilitates state capture 1936.4 State capture and delay in stabilization 1946.5 State capture and delay in reform progress 1946.6 State capture freezes transition 1988.1 State capture leads to frozen transition 240

Preface and Acknowledgements

Many years have now passed since Europe’s annus mirabilis, 1989, certainlyenough time to allow more than a preliminary assessment of the transitionfrom communist central-planning to a market democracy in the 27 countriesof the former Socialist bloc. While the voyage is by no means complete in allof the countries, it is nearly so in many, and reasonably clear why not in theothers. This book summarizes my efforts to describe the various paths fol-lowed by different countries, and to explain these differences in a systematicpolitical-economy framework. The book also reflects for me a culmination ofprofessional involvement in this issue since 1989, an involvement with astrong dose of personal connections, devotion and passion. This the readershould know, and judge my attempt to balance the personal advantage of‘insider’ insights, and its risks of intellectual biases. As a Ukrainian-bornrefugee of the Second World War, I was blessed with the opportunity to livein many free societies and to be educated in a ‘Western’ professional para-digm. Nevertheless, my origins, parents and diaspora community ensuredthe indigenous weltanschaung was not completely lost. This did not fullyequate with the experience of living in the pre-1989 communist societies,but it helped a great deal to understand them better when my professionalinterest, like that of many Western economists, carried me into the field oftransition economies. This combined background considerably facilitatedmy work in the area as an academic, Ukrainian government official, andofficial of the World Bank and International Monetary Fund.

Before I express due acknowledgements two prefatory comments on thebook itself may be useful here. First, the methodological approach I havetaken to seek the simplest systematic explanation of how events turned out,may to some appear counter to the historical complexity and uniqueness ofthis transformation. I believe a parsimonious but comprehensive explana-tion is indeed compatible with the unquestionably complex reality of all thechanges to be made in a post-communist economy, polity and society. I donot argue that the simplified paradigm of the book is a substitute for thedetailed analysis of individual countries or component issues such as priva-tization or development of market institutions. In fact, such detailed studieswere in my own research for this volume the most useful way to condensethe varied experience of many countries into a more focused understandingof the key driving forces.

Second, the navigation paradigm of the book tells a fable of the S.S. Transition voyage, whose course, schedule and final port of call weredetermined by the ability to sail expeditiously through the ‘unchartedwaters’ of market reforms avoiding extensive delays, the capacity to mitigate

xii

Preface and Acknowledgements xiii

effects of ‘pirate raids’ (rent-seeking activities), and commitment to followthe beacon of EU standards to the safe haven of eventual membership. Themost successful voyages have ended for those countries which acceded tothe EU in May 2004. The most troublesome voyages were, and remain, thoseof the so-called captured states where all three of the above forces had strongnegative tendencies. But there is an even simpler way of telling the storyfocusing on the states where an oligarch clique developed and captured notonly the levers of the economy but those of the polity as well.

This alternative story centres on the partiality of economic reforms, and isthe story often told by frustrated reformers of the early 1990s in Russia(Gaidar, Fyodorov, Yavlinsky) and Ukraine (Yuschenko, Pynzenyk). In thesocialist period, private ownership was largely illegal, and the disciplinaryrole of a market was largely substituted by central-plan directives backed byparty and judicial discipline. In the early transition period, private owner-ship become legal, and the discipline of the Central Plan, the Party and thecourts became so loose or ineffective that it was virtually non-existent. Intheory, transition to market operation was meant to provide a substitute, thediscipline of competition. But reforms that were partial not only createdhuge price distortions attracting rent-seeking pirate–raids, they also failedto ensure an open competition environment. Hence, oligarchs becameincreasingly powerful, and for now at least find it in their interest to retainthe status quo of partial transformation, blocking further moves to completeliberalization.

A vast number of individuals have influenced my thinking on transitionand thus contributed importantly to this book. Only a few can be namedhere and I will note first a special subgroup: those independent thinkers andfree souls – and inevitably patriots – who suffered the ignominy of Sovietsuppression of ideas. I wish to thank specifically the many government offi-cials in the region who struggled with early reform efforts and with whomI had the honour to work and thus learn: Leszek Balcerowicz, Yegor Gaidar,Vadym Hetman, Andrei Ilarianov, Pero Jurkovic, Grzegorz Kolodko,Volodymyr Lanovey, Andrzej Olechowsky, Bozo Prka, Hryhoriy Pyatachenko,Viktor Pynzenyk, Dariusz Rosati, Marko Skreb, Serhiy Teriokhin and ViktorYuschenko, now President of Ukraine.

There are also many who contributed directly to the preparation of thisvolume. I thank first the International Monetary Fund and its senior manage-ment not only for the opportunity to work with so many different transitioncountries, but also for the sabbatical year granted to me in 2004 to bringtogether all these experiences in a compact written form. I am most gratefulfor the academic home provided to me for one year by the Centre for Russianand East European Studies a the University of Toronto, and its Director PeterSolomon. I extend my appreciation to the many colleagues and studentsthere who comprised its vibrant intellectual environment. I am also mostgrateful to Karlo Basta who prepared the first versions of the Chapter 1 section

on transitology and the Appendix on the Baltics’ efforts to move towards EUmembership; he deserves full credit for the value added by these items, butthe responsibility for any errors or misinterpretations is mine. The researchunderlying Section 5.3 on the timeline of transformation and preparation ofthe index were done by Natalia Ingrid Havrylyshyn.

A number of friends and colleagues read parts of the manuscript at variousstages and their comments were highly beneficial: Robert Austin, JulianBerengaut, Albert Berry, Laszlo Czaba, Randal Filer, Natalia Ingrid Havrylyshyn,Gerald Helleiner, Simon Johnson, Jeffrey Kopstein, John Odling-Smee, JacquesPolak, Tapio Saavalainen, Peter Solomon, Lucio Vinhas de Souza, Paul Wachteland Bernard Yeung. The analysis, interpretations and views presented hereremain the sole responsibility of the author, and do not necessarily reflect theviews of the above, nor do they represent the position of the InternationalMonetary Fund or its Board of Directors.

Last but not least, the preparation of the typescript and publication, alwaysa large and complicated task, would have been impossible without thepatient assistance of many people, including Joy Zhou, Doris Ugaz, SusanaOtero, Rachel Stazi, Faiza Mohamed, Oksana Polyuga, Olive Stone, LeahTaylor and Sean Willett. At Palgrave Macmillan I am very grateful to KeithPovey for a thorough editing which much improved the text, to AmandaHamilton for general guidance, and Katie Button for efficient administrativesupport. Jeanette Morrison, Sean Culhane and James McEuen of the IMF’sExternal Relations Department were most patient and helpful with finaleditorial and publications logistics.

OLEH HAVRYLYSHYN

xiv Preface and Acknowledgements

Foreword to Havrylyshyn Book

It is less than 20 years since the collapse of communism in Europe. Yet, fromthe perspective of the early twenty-first-century, the events triggered by thefall of the Berlin Wall in 1989 seem much more distant. Even for thosedirectly involved it can be hard to recall what life was like in the centrallyplanned economies – and correspondingly easy to overlook the scale of thechallenge that faced those charged with transforming those countries intomarket economies.

This is at least in part because so much has been achieved so fast. Many ofthe countries crippled by economic policies that wasted resources on a hugescale, stifled growth and held down living standards are now in theEuropean Union. The transformation of many parts of the former Sovietempire has been remarkable.

Turning centrally run economies into well-functioning market economiesthat foster competition and growth had never before attempted. The earlyyears of the process were punctuated by intense debates among economistsand policy-makers about the best way to achieve the desired objective: inparticular, there were arguments about whether speed or caution would bethe more effective route.

The process is not complete, of course, and has further to go in some partsof the former communist world than others. But, given what has alreadybeen achieved, it is appropriate to take stock, and that is what this new bookby Oleh Havrylyshyn seeks to do. In part, this is a historical exercise, tounderstand more clearly what happened and why. This volume presents adetailed survey of more than 20 countries, providing statistical informationand analysis of policies pursued and their results.

But this is also very much a forward-looking exercise. Increasing ourunderstanding of the past couple of decades, and finding out more aboutwhich policy choices worked best, can also have a prescriptive value. Thereis now a sufficiently large database for us to attempt judgements about therelative merits of competing approaches. Havrylyshyn concludes, for exam-ple, that those countries where reforms were introduced most rapidly andcomprehensively have experienced the most rapid progress – on all fronts.They have grown most rapidly; they have been most successful at integrat-ing with the global economy and attracting foreign investment; they havebeen more successful at mitigating the social costs of transition; and theyhave the most mature democratic institutions.

The issue of transition costs was at the heart of the reform debate. Thosefavouring a slower and more gradual transition consistently argued that this wasessential to avoid imposing high social costs, especially on the poorest members

xv

of society. But Havrylyshyn’s findings suggest that delaying reforms did notreduce social costs: rapid reforms led to more rapid and sustained growth and somade possible more rapid and more significant reductions is poverty.

It is tempting to note at this point that it was ever thus and this, indeed, isone of the most interesting conclusions Havrylyshyn’s detailed analysisthrows up. The initial challenge for the transition policy-makers was differentfrom anything that had gone before both in nature and degree. But as theprocess of turning these countries into properly functioning marketeconomies, integrated with the global economy, got under way, the similari-ties with other reform efforts in other parts of the world became more striking.Many of the lessons learned from the experience of the transition economiescould be usefully applied to many countries where reforms have been hesitantor stumbling. It is essential to create a virtuous circle, emphasizing macroeco-nomic stability, liberalization and market-friendly institutions: and equallyessential to avoid a vicious circle in which vested interests undermine thecommitment to reform, stifle competition and undermine growth.

In that sense this book has a relevance well beyond the former Soviet bloc.Its conclusions are a further reminder that cultural and national differencesneed not detract from more widely applicable lessons about economic behav-iour. Economic growth – rapid and sustained – is a prerequisite for raisingliving standards and reducing poverty. Reforms need to be consistentlyapplied with the aim of making accelerated growth possible. Speed, compre-hensiveness and consistency play an important part in determining successfuloutcomes. That is true in the transition economies. But it also holds true moregenerally, as the experience of countries as different as Chile and Korea shows.

The International Monetary Fund’s relationship with the transitioneconomies is perhaps relevant in this context. The Fund was intimatelyinvolved in supporting the reform efforts in these countries: Fund staff, likeall those involved in the transition process, faced a very steep learning curve.A special department, known as European II, was established to handle thetransition economies. It is a measure of the achievement of so many of thesecountries that by the end of 2003 it had been decided to close downEuropean II and to integrate the work on the transition economies into therest of the Fund’s work. They are now treated like any other Fund membercountry. This also ensures that the lessons learned by Fund staff in support-ing economic reforms efforts can also be disseminated more widely.

This volume is also an exercise in dissemination. It is underpinned bydetailed statistical analysis. But it is aimed at the more general reader inter-ested in knowing what happened, when and why; and a such it is a timelyreminder of how far we have come.

ANNE O. KRUEGER

First Deputy Managing Director, International Monetary Fund

xvi Foreword to Havrylyshyn Book

Introduction and Overview

1

On 9 November 1989 the Berlin Wall fell, marking the end of the communistexperiment and the beginning of the end of the Soviet Empire. This waswidely considered as a momentous historical event, not only because it wasso unexpected, but because it symbolized an end to the Cold War, the free-ing of several hundred million individuals from an authoritarian state whichhad kept them closed off from the rest of the world, and a liberation of pri-vate economic initiatives from the constraints of the socialist central-planning regime. People around the world joined in welcoming citizens ofthe Socialist bloc, and euphoria would not be an exaggerated characteriza-tion of the latter’s emotional state. However, the immediate impact was notyet clear, as liberation from socialism would be implemented at varyingspeed for different countries over the next few years. In particular, for theindividual Republics of the Soviet Union, the political independence theysought was by no means assured in November 1989. But the direction ofchange was assumed by all to be greater openness and freedom – personal,political and economic. The subject of this book is to review how far suchchanges have gone in 15 years, and to explain why some countries haveprogressed more, others less.

Citizens elsewhere in the world echoed a sympathetic euphoria. For eightymillion Germans, now reunified, it was most immediate and arguablystrongest; other Europeans felt a kinship for their German neighbours, and ahuman sympathy sharing the joy of their eastern neighbours who could nowrejoin European society. Americans whose country was the major protago-nist in the Cold War rejoiced in the victory and in their contribution to free-ing the captured states. Many attributed this to the open and tough stance ofPresident Reagan, symbolized in his remark ‘Mr. Gorbachev, tear down thiswall!’ With time, it was understood that other forces were also at play under-mining the Soviet bloc, and the possibility that had the addressee not beenGorbachev, the choice might have been to call Reagan’s bluff and not teardown the wall. At the least, Gorbachev, too, has earned historical recogni-tion for his decision. It is also virtually axiomatic that Pope John Paul II takes

place of honour as the first, if not the major, figure leading the effort to endthe communist captivity.

The outcome

It has now been a decade and a half since the bloodless breaching of theBerlin Wall, and of course the euphoria has passed. But has it passed becausethat is the nature of euphoria, or because the euphoric expectations of thatmoment have evolved into disillusionment? No doubt some of both, thoughthis book takes a positive view that, despite important shortfalls, overallthere has been a lot of economic progress and social improvement. It issignificant and starkly evident in most of Central Europe and the Baltics, butfor others, especially the former republics of the USSR, so far it has beenlimited and even minimal, and has come at considerable cost to parts of thepopulation. This relatively positive view is not shared by all observers.Numerous critics including politicians, citizens of the region and outsideanalysts argue that the outcome of the transition is at best mixed and insome cases disastrous; that the economic decline following transition hasbarely been recovered in a few countries and is far from this elsewhere; thathuge increases in poverty and inequality are not an acceptable cost for anyof the gains of the new capitalist endeavour; that the availability of goods atunaffordable prices is perhaps worse than the queuing for scarce but cheapgoods; that the virtual theft of state assets by a new class of ‘oligarchs’ andtheir capture of the polity cannot be considered an improvement over thesocialist period of domination by a communist nomenklatura.

I do not propose to give a response to all of the above criticisms, but willpresent updated facts and analysis which, while confirming there weresignificant economic costs especially in the first half-decade, demonstratethat the outcome falls well short of the very negative picture depicted bysome critics. After nearly 15 years, the initial costs have been more thanovercome for the most successful countries in Central Europe, and at aminimum are on the way to being overcome in most of the others.

Putting the above in the jargon of economists, 15 years of transition haveyielded many benefits, and produced significant costs, but with a great diver-gence of the benefit–cost ratio across the region. The judgment of this bookis that for about a dozen countries in Central Europe and the Baltics, thebenefits have been large, the costs relatively small, though not all citizenswere winners, and the losers even in these societies have not always beenproperly compensated. For countries in South-East Europe the balance seemsto be moving in the same direction but it is still early to judge, perhapsbecause the costly conflicts in this region are so recent. In the rest, that is thenon-Baltic part of the former USSR, the story is indeed more mixed. A hand-ful of winners within the society have won hugely – the so-called ‘oligarchs’ –a large number of losers remain poorly compensated if at all, while the

2 Introduction and Overview


majority of the population in the middle is by now somewhat better-off,and certainly no worse-off, but even they suffered in the early years of thetransformation.

Another important divergence in the outcomes is the nature of thepolitical economic system that has emerged. Eight countries in CentralEurope and the Baltics have become EU members as of 1 May 2004, at leastthree more in South-East Europe are formally on track for membership in thenear future, and the rest in this group have some hope for future EU mem-bership. All of these countries have become, or are likely soon to become,reasonably well-functioning middle-income market economies and increas-ingly liberal democracies. In contrast, the countries of the CIS are highlyunlikely to move in that direction in the foreseeable future, with the possi-ble exceptions of Georgia, Ukraine and even Kyrgyz Republic, though theirrevolutions of colour are too recent to assess. For this group, state capture byeconomic oligarchs is the order of the day, bringing in parallel increasinglyautocratic political regimes. EU membership prospects are virtually nil, againwith the possible exception of Ukraine and less so Georgia.

The history of this oligarch phenomenon varied greatly. It was mostimportant in the CIS group, more moderate and now waning in the middle-group of South-East Europe, and very short-lived and never fully evolved inthe Central Europe–Baltic group. Many now say that for all practical purposestransition is over and there’s not much more of interest to analyse in post-communist transformation. For the new and prospective EU membersanchored to the EU and its economic-democratic liberal standards, there isindeed very little more transformation to do. Most of the Commonwealth ofIndependent States (CIS) are locked out of the EU path and are approachinga stable equilibrium of distorted capitalist economies dominated by a smalloligarch clique, without competitive liberal markets, and polities that are atbest only superficially democratic. Hence their transition is also ‘over’ in thesense of being frozen for now

But the real story is neither so simple, nor so uninteresting. Certainly, thefuture direction for the captured states remains unclear, and this poses animportant, intellectually challenging question. Even if one believes thetransition issues of the 1990s are all in the past, either resolved or renderedirrelevant, a new transition debate has arisen which merits very close attention:in the oligarch regimes is further transformation indeed frozen, or is itjust temporarily stalled and bound to move forward as the new capitalists(yesterday’s thieves as many label them fairly or unfairly) seek legitimation,demand secure property rights under transparent rule-of-law, and pushinevitably forward the final stages of economic and political liberalization?Another way forward for them is the possibility of radical regime change anda renewed democratic opportunity, as may be happening with the Rose,Orange and Tulip revolutions. This debate is surely of enough practical andacademic interest to justify continued analysis of the transition, at least for

the oligarch regimes. To inform this debate, it is surely important to under-stand exactly why some countries went ‘to the West’ – to borrow a phrasefrom Egor Gaidar – with its dual liberal vision of democracy and freemarkets, and some went ‘to the East’ with oligarchic regimes in autocraticstates with formal but largely meaningless elections.1

There are many fascinating questions related to this divergence in thetransition explored in detail by this volume; I flag here only three. Was thecoincidence of movement towards the EU and away from state capture onlya coincidence, or were there systemic behavioural forces behind the differentoutcomes? If this was systemic and not coincidental, was it simply histori-cally predetermined by geographic location or were there other factors atwork as well? And finally, what was the direction of causation: was it that aprior ‘invitation’ to Central European countries for EU membership preventedthe excesses of oligarch dominance and state capture, or the reverse, thatoligarch development and the accompanying non-liberal institutionaldevelopment precluded oligarch-regimes from wanting to be serious candi-dates for EU membership?

It seems reasonable to conclude the transition is not fully over for allcountries, and from those where it is we can learn much of how it came tosucceed so quickly in those countries. This understanding will not onlysatisfy historical curiosity, but will be critical to analysing the future ofcountries where the transition has not been completed.

Objectives of the book

The book has three broad objectives: to describe for different groups ofcountries the differences in what happened, to explain why the differences inoutcome, and to assess the probable future directions, that is what next?The 27 non-Asian transition countries analysed here cover a wide range ofgeographical, historical and cultural phenomena, hence it is tempting to lookto the numerous country-specific characteristics like years under communism,distance from Western European influences, or the degree of internal socialharmony as the explanations for each country’s path. While these and otherinitial conditions cannot be ignored, such a case-study approach leads to anoverwhelmingly complex picture with a different explanation and story-linefor each country. The book proposes instead a cross-country comparativeframework simplified enough to avoid losing sight of the forest for the trees,but still powerful enough to explain the main forces which have driven thedifferent outcomes of post-communist transformation in the region.

The central analytical paradigm

There are two common ways for social scientists to analyse and explaindifferent developments across countries: case study or comparative analysis.



Each has its well-known advantages and disadvantages, and I choosecomparative analysis because I believe in this case its advantage – not losingsight of the forest by coming too close to the trees – is overwhelming. I showlater in the book that the existing literature of post-communist transforma-tion contains such a vast number of plausible explanations that its sum givestoo many explanations (in jargon: a hugely over-determined system), henceone is quickly lost in the thicket of alternative and often contradictory expla-nations. Admittedly, the risk of simplification is oversimplification andreduced applicability; my approach to minimize this risk is two-fold. First,I will make country-specific references throughout. Second, I propose threeexplanatory factors common to all countries, but do so not in a formal math-ematical model (though I do suggest how this might be done), rather as aframework or paradigm in which each of these three key determinants maythemselves be explained by other factors taken from the long list of country-specific factors. As a precursor of this framework, let me outline it in a fewsentences.

Consider as a mnemonic for the countries in this region after the termina-tion of the socialist central-plan regime, the figurative paradigm of a shipsetting out on a voyage; the SS Transition sailing towards a new destination,a democratic market economy. Though the nature of the final destination iswell-known and there are many illustrations of it, there is no precise naviga-tion chart based on previous sailings from socialism to capitalism, no consen-sus of how to get there avoiding the perils any sailing voyage or large socialchange encounters. Numerous country-specific and comparative analyseshave been written by observers of how well this voyage has proceeded, howfar along the ship has come, how much damage it has suffered in the stormsof change. While accepting that for individual countries a large number ofcountry-specific factors have played a role, I propose a parsimonious frame-work which, like any useful paradigm, is not a full explanation but goes along way to explain the experience for most of this fleet of 27. The threefactors are expressed in the form of hypotheses as follows:

● the greater the intensity and time spent on debating about the rightnavigation chart the worse the final outcome;

● high proclivity of a country to fall prey to economic vested-interestsresults in a pirate raid by rent-seekers, and capture of the state, again aworse outcome;

● the accessibility of a safe-haven to escape the pirate raids or any otherthreats on the voyage, such as EU membership, leads to better outcomes.

The framework can be made into a formal model in the sense that the threefactors individually contribute to the explanation of greater success: lessdebate means getting started sooner and achieving results sooner; greaterrent-seeking results in less liberalization; and external commitment to a safe

haven like the EU pushes liberalization faster and further. The three factorsare interrelated: for example chances of a pirate raid are increased by longerdebates, even as they are influenced by exogenous or historical factors suchas whether the first government largely excludes the former communist eliteand is more committed to the dual liberal vision of democracy and openmarkets. Even without mathematical formalization, the framework providesat a minimum a simplifying paradigm which helps focus discussion but stillallows many other factors to be included. Thus, as an example of the latter,just how early EU membership became an explicit factor in the policies ofBaltic governments is not entirely clear because they certainly did not enjoythe ‘invitations’ offered early on to Central Europe until the mid-1990s. Buttheir prior history of very reluctant adherence to socialism and the empire,as well as the fact that their incorporation into the USSR was never formallyrecognized by many Western powers, doubtless led to a strong desire toanchor their independence in some institutional form to Europe from theoutset. That is there was a demand for membership before there was anoffer. This then pointed the way to undertake early and resolute policy deci-sions on market liberalization; the EU ‘safe haven’ may not have been aninstitutional certainty, but it existed as a mental construct or vision guidingthe way.

Some remarks on methodology and data

A few words about the methodological approach taken in this study. My ownexpertise as an economist involved in the process of economic reforms intransition countries means I will naturally give most weight to economicpolicies, statistics, results and debates among economists. But it will benecessary to adduce analysis from the political scientists and historians whospecialize in this region; it may be obvious why this is so, nevertheless let meillustrate with one point. A key strand of the heated discussions on the out-come has been whether the so-called ‘shock-therapy’ approach, or as I preferthe ‘big-bang’ approach, worked better than a gradual, piecemeal approach.In retrospect, it is clear the majority of analysts on either side of the debatedo not dispute the theoretical logic of the opposing view, but rather adducepolitical-historical factors to argue that their approach is more realistic. Thusthe critics of big-bang said that it will cause pain and a political counter-reaction inimical to both democratization and economic liberalization,hence proceeding gradually would be better. The proponents of big-bangpoint to the role of vested interests opposed to reform who exercise theirpolitical influence while hiding behind scientific arguments of gradualism,to undermine the economic and political liberalization and diverting stateassets and subsidies to their own benefit. Any analysis that ignores thepolitical-economy of this process will surely miss the most powerful forcesand explanations of the outcomes.



A few clarifications are also in order concerning the coverage and approachof the book. The first broad objective, to describe the outcome in detail andassess the benefits and costs, is not as easy as it sounds, given the many sharpdifferences in view about the economic performance and social impacts sofar. Resolving hard debates will require a lot of detailed and seemingly arcanequantitative analysis, with conclusions which will in the end remain uncon-vincing to some because the data is very soft and leaves room for differentinterpretation.

Somewhat the same problem applies to the second objective. The thesis ofthis book employs three key factors and their interplay as the explanatoryframework for different outcomes: the choice of reform strategy at the firstopportunity or ‘historical moment’; the pull of EU membership; and theinfluence of rent-seeking interests. These are not concepts that are easilyquantifiable and therefore only soft evidence of their role can be provided.Furthermore, this analytical framework faces several diametrically opposedbut deeply-rooted convictions. On one hand are those – including themajority of the population in the less successful states – who believe thewhole story is simple: the red nomenklatura used their historical position ofpower to steal from the state and become the new capitalist oligarchs – nomore, no less. On the other hand are those that argue providing the safehaven of democracy and free markets of EU membership only to geographi-cally proximate states in Central Europe is the whole story. The book willshow that while there is a lot to both these views, many exceptions exist andthe causation is not linear or unidimensional. Rather both these effects plusothers worked in some combination to give the final result observed for eachcountry. The analysis relies largely on cross-country comparison rather thanlengthy case studies, but many episodes from specific countries will be usedboth to illustrate the central tendency, and to note deviations from it.

The country coverage in this analysis is limited to the non-Asian transi-tion economies, more precisely the 27 countries covered by the mandate ofthe European Bank for Reconstruction and Development (EBRD). Thisexcludes the former GDR, or East Germany, because when it became part ofthe Federal Republic of Germany it no longer conducted independent eco-nomic policy. The reason for excluding Asian countries is not primarilybecause EBRD statistical data is unavailable for easy comparison, butbecause I believe the circumstances are vastly different and common analy-sis or transfer of lessons is of limited value, for two reasons. First, China andVietnam have remained polities under control of a communist partyauthority; the EBRD group all experienced a political revolution, albeit abloodless one in most cases, with the controlling communist party initiallyceding to democratic elections of varying purity. Second, the Asianeconomies were largely agricultural economies with huge reserves ofsurplus labour in that sector; almost all of the EBRD group were highlyindustrialized in comparison.

These differences allowed China to proceed on its gradual and piecemealpath to market reforms, a path – or luxury some have said – that wouldhave been impossible for the European socialist bloc to follow and ineffec-tive if it had been. Nevertheless, the high-profile attained by China’s strongeconomic performance in contrast to the output decline seen in the othernon-Asian countries, plus the central role China played as an example ofsuccessful gradualism in the arguments of many analysts, makes it impossi-ble to completely avoid a discussion of China. Therefore, in many placesthroughout the book, comparisons with China’s experience are made.

Judging success and failure

How does one judge whether a country has been successful or has failed inthe transformation process; that is, whether the benefits have exceeded thecosts or vice versa? First, this cannot be a simple two-fold categorization, butrather a continuum of more and less success. Second, there is no clear-cuttime horizon or end-point in which to make this judgment; transitionstarted 15 years ago for perhaps only the four Visegrad countries in CentralEurope, but began much later elsewhere either because civil unrest, politicaluncertainty or policy choice led to delays. The Soviet Republics were notindependent until late 1991 and most of them did not make a significantbeginning until about 1994–95. Yugoslav Federation Republics ended anymilitary–civil conflicts at various times in the 1990s, and some countries likeBelarus, Turkmenistan and arguably Uzbekistan to the present day havebarely begun a serious transformation to the market. These different startingpoints must be considered in any assessment.

A third consideration is how to weigh up the benefits and the costs. Theserious data shortcomings are discussed in the main body of the book;I focus here on the conceptual approach to such a judgment. Measuringthe benefits of renewed growth and greater efficiency raises many questions,but even more problematic is the fact that no simple metric exists to valuethe social costs such as unemployment, increased poverty, possible impactson health and education delivery. Furthermore, there is no agreed-uponmethod for dealing with distributional questions, such as the hypothesisthat the transition violated the old social contract, because some members ofsociety were made significantly worse-off, certainly in relative terms, but alsoin absolute levels of material well-being.

One possible approach for the last issue is to use the so-called ParetoOptimality Criterion of economists: a change results in improved well-beingfor society if those who gain from it gain enough to compensate those who lose. Intheoretical academic analysis of economists, knowledge is generallyadvanced by some degree of compartmentalization of disciplines and abstrac-tions from reality, so in the literature it often matters little if actual compen-sation took place as long as potential compensation to losers is possible. This



potential compensation is almost certainly feasible by now for manytransition countries and approaching that state for others. But that is smallcomfort to losers who were not in fact compensated. In practice, to be com-pelling, a favourable judgment on the transition must argue compensationactually did occur or is taking place. This is again not so simple; quantitativecost–benefit analysis for the society as a whole is impossible, thus one mustresort to sensible judgment. Such a judgment must also allow for delayedrather than immediate compensation, and must further be flexible (and adhoc) in judging whether the economic losers who are not yet compensatedeven after a long time, may nevertheless value future economic opportuni-ties for themselves and their children. One should also recognize that peoplealso value the personal freedoms gained with the end of an autocratic, policestate. The fact that many still vote for communist parties, which propose torevert to a socialist society with renationalization of property, suggests somecitizens may not place a high value on these freedoms. In contrast, theevidence of the Rose Revolution in Georgia, the Orange one in Ukraine, andthe Kyrgyz Tulip events, suggests many people do value democratic freedomshighly. In the end, I know of no other way to proceed on this assessmentthan sensible and well-informed personal judgment, and leave it to thereader to make a judgment as well.

The main conclusions on benefit–cost results are three. First by now forcountries in Central Europe and the Baltics most losers have probably beensufficiently compensated in an absolute sense by a combination of transfersand new economic opportunities. Second, and notwithstanding the firstpoint, in some of that region, the continuing high unemployment suggeststhat some portion of the population has not yet received adequate compen-sation. Third, for countries of the CIS, the number of losers were many andthey have only recently begun to enjoy meaningful compensation, thoughthe substantially increased poverty incidence probably remains somewhat oreven substantially higher than it was at the outset. Fourth, relative countryperformance on the Compensated Pareto Optimal Criterion tends to varyalong a spectrum from very good in Central Europe and the Baltics, mixed inSouth-East Europe, and generally poor in the CIS group of countries.

Structure of the book

Let me finally outline the contents of the book and how it addresses thethree main objectives I have set. Part I, Background, addresses the firstobjective: describing what happened with available statistical evidence.Chapter 1 sets the stage for what is most important to seek from the statis-tics, with a review of the key debates in the literature about the transitionprocess. This covers largely the economics literature, but incorporates themost relevant elements of the related literature in political science, the so-called transitology debate. Chapter 2 goes on to measure the extent of

progress towards a market economy using the annual index of progress onmarket reforms compiled by the EBRD, which is best thought of as a policy‘input’ intended to improve economic performance. It is shown that despitethe shortcomings and many possible subjectivities of this index, it correlatesvery well with other ‘policy input’ measures such as the degree of democracy,media freedom, and development of institutions. It is also strongly corre-lated with most indicators of ‘output’ success, not only economic perfor-mance, but also and surprisingly most social progress indicators. The highdegree of correlation justifies using the index in the rest of the book as aproxy indicator of progress in post-communist transformation broadlyunderstood. The index allows a categorization of the 27 countries into fivedistinct groups with limited overlap and considerable within-group homo-geneity. In order of transformation progress and success achieved these are:Central Europe, Baltics, South-East Europe, CIS moderate reformers, and CISlimited reformers. This greatly simplifies the further analysis without dis-torting it, as the broad trends can be observed and analysed for five groupsrather than 27 individual countries.

Chapter 3 measures the costs of transition, including output and employ-ment losses, broad measures of social well-being, increased inequality andpoverty, deterioration in health and education standards. These costs havebeen and remain substantial, but a hard look at these soft facts also shows thatmany early assessments showing dramatic deterioration were somewhat exag-gerated, partly because they were premature, being done at the bottom of thetransition cycle in the mid-1990s. There was virtually no disagreement in theearly 1990s that the transition from a distorted and inflationary economywould create disruption, dislocation and unemployment, that is things wouldget worse before they got better. The many studies of economic and social con-ditions done in the mid-1990s showed substantial deterioration in the five toseven years of the transitional recession, but this should not have been sur-prising as they covered the downturn phase of the transition cycle. The pre-sent work benefits from having data for the five to seven years of the upturn,and it shows an improvement in economic and social conditions, in severalcases very substantial and far exceeding the losses in the first half of the cycle.

Part II of the book, An Ex Post Transition Paradigm, addresses the secondobjective, asking: why the different outcomes for different countries and regions,using the navigation paradigm described above. Chapter 4 elaborates on thisparadigm and presents several empirically testable hypotheses that can bederived from it. Chapters 5, 6 and 7 contain the comparative empiricalanalysis of the relevant hypotheses for each of the three elements of themodel: the debate on how to proceed, the dynamic of rent-seeking vestedinterests first lobbying for privileges and then in some cases succeeding incapturing the state; and the influence of potential safe-havens such as EUmembership, but also others such as NATO, WTO, the IMF, the World Bankand the EBRD.



The analysis in Parts I and II forms the basis for addressing the thirdobjective of the book in Part III, A Summing-Up, which asks what willhappen next? The critical dividing line here is between those states that havebeen strongly captured by large economic vested interests, known popularlyas oligarchies, and those that have not. This dividing line coincides closelywith the one separating those countries with assured or prospective mem-bership in the EU from those that have low or near-zero likelihood. A keyquestion investigated is whether this is indeed just a coincidence or not.Chapter 8 focuses mostly on the captured states, considering their prospects.Most importantly Chapter 8 addresses perhaps the last but certainly not leastimportant transition debate: for the captured states is further transformationinevitable, or has it been frozen in an equilibrium part-way to the marketand democracy?

Finally, Chapter 9 sums up the findings in the book, first noting theimportant new divide between the most successful countries where transi-tion is virtually complete or on track, and the less successful ones where ithas become stalled due to capture of the state by the new oligarchic restoredinterests. Two other conclusions merit emphasis: one is the overwhelmingrole of three factors in explaining this bifurcation: reform delays, a strongertendency to rent-seeking under partial reforms, and the disciplining roleof the quest for EU membership. Another is that any social costs were,contrary to early concerns, much lower in countries that moved on reformsearly and rapidly, than in countries taking a more deliberate gradualapproach. In the simplest of words, the bottom line is that the first group isalready experiencing the dual vision of liberal markets and liberal democracywhere capitalism is open to all, whereas the second group is characterized byrestricted markets and superficial managed democracy, with capitalismbeing accessible only to the few.

Part I

Background: Transition Debates and Statistical Facts

1Key Debates on Transition

15

The search for a transition model

When the World Bank undertook what was perhaps the first retrospectivereview of the transition process in 1996, the economics literature on transitionalone was enough to result in a ‘Selective Bibliography’ exceeding 700 items.Given the lead time for academic research it is not surprising that the subse-quent years led to an explosion of writings; a June 2004 internet search under‘post-communist transition’ produced an impossible 50,000 � items, andeven a narrower academic works search at the University of Toronto libraryyielded over a hundred books, and thousands of journal articles. Anythingapproaching a comprehensive review has become nearly impossible, thus theaim of this chapter is much more modest but hopefully more usefully focused.On the basis of about 10 recent economic studies by recognized experts andinstitutions reviewing or assessing transition progress and performance, I willsummarize briefly a few of the most important debates in the 1990s, and thendraw attention to a handful of the most important issues looking forward.It turns out that how privatization was accomplished (regardless of how it wasplanned) provides an excellent link between past debates and future issues.

The rest of this section discusses whether a ‘model’ or ‘paradigm’ fortransition exists, as opposed to models for the beginning point – socialism –and the end-point – capitalism.1 The next section assesses the state of themain debates of the past, before moving on to focus on the handful thatremain most relevant for the future. It is notable that looking-forward, anissue that was not part of previous debates, has become perhaps more impor-tant than any of the previously unresolved ones: that is whether thosecountries that are lagging in progress to the market will inevitably continueprogressing, or remain in a sustainable equilibrium part-way to a well-functioning market economy. Interestingly, an analogous question has cometo front-stage in the debates of political scientists about the inevitabilityof further progress towards democracy; the final section provides a shortaccount of that literature.

Going back to early discussions about the ‘road map’ or ‘chart’ showingthe way from socialist central-planning to a liberal (and simultaneously demo-cratic) market economy, there was a great deal of debating on the detailsof how to proceed with transition. Three illustrative examples are Kemmeand Gordon (1990) (a Western think-tank), Zecchini (1991) (the OECD), andClague and Reuser (1992), (an academic conference). It was very commonlysaid that there is no theory to guide the practical process of transition, onlytheories of capitalism and socialism.2 This may still be true in the sense thata new consensus paradigm has not emerged from the vast literature on tran-sition, but I propose here that a formal unified theory is not needed tounderstand the main developments. To the extent it is useful to have acompact rather than complex analytical framework, it is not that difficult tocobble together from selected key writings a workable ‘model’ of transition.

There are two seminal writings which when combined, suffice to providethe nucleus of a transition model: Kornai (1994), and Blanchard (1997).Kornai did not set out to posit a model, but only to describe the specialcircumstances of the inevitable ‘transformational’ recession compared witha market economy recession. In so doing he highlights two key actions thatare needed:

● forcing a move from a sellers’ to a buyers’ market (via price liberalization);● enforcing a hard budget constraint (via privatization and elimination of

various government support mechanisms such as budget subsidies,directed low-cost credits, and tax exemptions).

Whether intended or not, this provides the two principal changes in regime,and new behavioural incentives for profit-maximizing as needed for a marketeconomy.

Blanchard (1997) also addresses the problem of the transition recessionand tries to explain it by focusing on the precise mechanisms of resourcereallocation from old inefficient activities to new efficient ones. He definesthe core process of actual change as comprising two elements:

● reallocation of resources from old to new activities (via closures andbankruptcies combined with establishment of new enterprises);

● restructuring of surviving firms (via labour rationalization, product linechange, and new investment).

These two changes, which are very reminiscent of the Schumpeterianconcept of ‘creative destruction’, should be stimulated by the new incentivesof Kornai. In the end, this process of first establishing new incentives foreconomic behaviour, then having economic decision-makers react to themby efficient reallocation of resources, brings the economy to a new marketequilibrium with an optimum allocation in line with comparative advantage.

16 Background

Key Debates on Transition 17

The combination of these four components of transition do not of courseprovide step-by-step guidelines for the details policy-makers have to imple-ment, but a useful paradigm must be simple and only need provide a centralframework of thinking around which the details can be logically constructed.The Kornai–Blanchard paradigm (KB) serves this purpose very well. The twomost important follow-up questions are :

● Is the process of resource–allocation and creative destruction necessarily‘a thorny road’ to borrow from the title of Kolodko’s (2000b) assessment,or can it be made smoothly?

● How do we delineate a road-map or chart of the detailed policy stepsneeded; that is, how to implement the changes of KB?

The first issue led very naturally to the intense debate which rages to thepresent day about speed and sequencing, the choice between big-bangreforms and gradualism. An elaboration comes later in the chapter, but giventhe importance of this issue for social welfare (that is, how to reduce anypotential pain and costs of transition) it will resurface in several placesthroughout this volume. Here, I will just note one point made cleverly, butI believe usefully, by EBRD Transition Report (2003) rephrasing Schumpeter tospeak of ‘destructive creation’ in post-communist economies, therebyemphasizing that destruction did in practice precede creation. Could it havebeen otherwise, is a question addressed later.3

The KB paradigm I would consider more than sufficient on what needs tobe done to achieve the goals of economic transition though not a naviga-tional chart of how to get there. The details of a navigation chart for the‘S.S. Transition’ specifying the key policy actions needed to put in placeKornai’s new incentives and ensure the smoothest possible process ofBlanchard’s resource reallocations, are described in many works includingsome recommendations by Kornai (1994) and Blanchard (1997). Again thenumber of plans of action are too numerous to even mention, but I ventureto say that the summary plan of action of the World Bank presented byFischer and Gelb (1991) and reproduced here as Figure 1.1 serves well as acomprehensive illustration of the main elements found in most plans. Sinceit also provides a good illustration of the Washington Consensus (WC) ideas,it will at a minimum serve as the ‘straw-man’ for both criticism by propo-nents of gradualism and defence by the proponents of big-bang.

The main elements of policy change include: macroeconomic stabilization;price and market liberalization; liberalization of the exchange and trade system;privatization of state-owned firms; establishing a competitive environmentwith easy market entry and exit; and redefining the role of the state as theprovider of macro stability, a stable legal framework and enforceable propertyrights, and occasionally as a corrector of market imperfections.

The early debates on the appropriateness of the WC, on gradualism v. big-bang (or as the critics but not the proponents prefer to label it today ‘shocktherapy’), on the institutionalist v. market liberalization approach, and inretrospect I believe most important, the modalities of privatization – all hadto do with the how. This and some other debates are discussed more fully inthe next section. However, it may be useful to take off the table early thespecific charges about the WC being an erroneous approach with critics say-ing it promotes a too-rapid approach which entails huge costs and risks offailure, and in particular fails to take account of the institutional basisneeded before privatization can be successful (see for example Stiglitz, 1999,and Reddaway and Glinski, 2001. Proponents argue there was little choice inthe context of the economic and political crisis facing these countries, that

18 Background

Figure 1.1 An illustration of the Washington Consensus. Reproduced from StanleyFischer and Alan Gelb, ‘The Process of Socialist Economic Transformation’, Journal ofEconomic Perspectives, vol. 5, no. 4, Fall 1991, with the permission of the authors andthe editors of the journal.

0 2 4 6 8 10 12 Years

Macrostabilization

More intense reform Less intense reform

Price and market reformGoods and services price reform

Trade reform

Unemployment

Labour market

Finance and banking

Restructuring and privatization

Large scale privatization

Small scale privatization andprivate sector development

Redefining the role of the state

Legal reform


the WC provides for needed flexibility on both speed and sequencing, andin the event the outcome is generally favourable (IMF, 2000).

This has been in my view a debate that is so broad and so vague that itbecame sterile in the end. There are two main faults in the criticisms. First,the WC as a theoretical concept was very broadly cast, incorporating clearlyinstitutional developments and long implementation times for some elementslike privatization, as in Figure 1.1. Therefore, any criticism that equates theWC with shock therapy or big-bang and promotes undue speed, or that itignores legal-institutional development is easily defended and the debategoes nowhere. A part of the problem is overzealous rhetoric starting with theinsistent use of ‘shock therapy’ rather than rapid reform or at most ‘big bang’.Balcerowicz (1993) and again in (2002) clearly argues for a less-rhetoricaluse of words, and his main critic in Poland, Kolodko (2000a), is able to keepto the substance of their differences. An example of the sterility is the criti-cism by Stiglitz (1999, p. 22) that the Washington Consensus proposed ‘theuse of a “shock therapy” to install institutions … [an approach historicallyassociated with] Jacobinism … and Bolshevism’. Reddaway and Glinski(2001) in the same spirit use the term ‘Market Bolshevism’ in their title.4

Nothing in Figure 1.1 nor the detailed writings supporting it suggests animmediate or shock approach to institutions, and indeed it was always clearthe rapid, big-bang approach was primarily applicable to macro-stabilization –on which Stiglitz fully agrees – and to a lesser extent on legal freeing ofprivate ownership, and liberalization of most prices excepting some keyconsumer goods.

This is clear in the Fischer–Gelb (1991) article, and in writings by threeearly implementers of rapid reforms: Balcerowicz, Klaus, and Gaidar.Balcerowicz (1993) writes: ‘market-type mechanisms were not introducedduring the first stages in those fields where the necessary institutional con-ditions were clearly absent’. Klaus (1995) contends that gradualism v. shocktherapy is a false dilemma, since ‘various measures have different time require-ments e.g. price and foreign trade liberalization can be done overnight, pri-vatization takes years to be completed. Things have to be done wheneverthey can be done or at least to be prepared without any delay.’ SimilarlyGaidar in Havrylyshyn and Nsouli (2001) emphasizes the immediate impor-tance of moving to a hard-budget constraint both for the government bud-get and for enterprises but does not insist on all other elements of reform,and indeed even for the hard-budget policies speaks of the ‘forces that couldnudge [my italics] in the direction of tighter restraints on enterprises’. As theother rapid reform practitioners he clearly points not to overnight, instanta-neous changes, but has a varying time-dimension for each element, much asin Figure 1.1.

Apart from misstating the ‘straw-man’ position, both these major criticsshare another logical problem: their analysis is largely based on Russia andignores the much more successful outcomes in Central Europe and the

Baltics; the statement by Stiglitz (1999, p. 18) that ‘successful restructuring israre in post-socialist economies’ is so egregious an error factually that it onlyserves to point out the important omission in his assessment of the suc-cesses. I will argue later that the critics are right to worry about the outcomein Russia and similar outcomes in Ukraine and the Commonwealth ofIndependent States (CIS) generally – the unfair, undemocratic and non-competitive oligarchic regime with concentration of assets and power in avery few hands.5 However, their conclusion that this came from an applica-tion of ‘shock-therapy’ is a misreading: the major reason this happened inCIS countries and not in Central Europe was the insider-dominated processof privatization.

The second reason the WC debate was sterile is a very simple but powerfulone: virtually none of the critics argued the reforms proposed should not bedone, but rather that they should have been done more slowly (as in China)or in a different sequence. The debate was largely about implementation.Furthermore the issue is now of decreasing relevance since in many countriesthe distance travelled means few remain at an early stage of deciding to movefast or slowly.6

If there is any historical value to continuing this debate it concernsthe possible criticism that in practice the IFIs (International FinancialInstitutions) and other rapid reform proponents may have only paid lip-service to institution-building but pushed harder the privatization and liber-alization elements, increasing the likelihood of failures or capture of theprocess by rent-seeking interests. This point is made by both critics notedabove, and it should not be dismissed out of hand; even if one agrees thatreversal of privatization is not an option, there remain possibilities for cor-recting the errors of privatization. In this volume, I also argue that someerrors in the process of privatization contributed to the strength of oligarchicdevelopments in many CIS countries, but while the critics’ logic goes fromtoo-rapid reforms to ‘theft of the state’ in Russia and elsewhere, my logic isconverse: too-slow reforms especially in liberalizing markets for open entryand competition is what led to oligarchic developments.

On the sequencing of institution-building and privatization, the IMFWorld Economic Outlook (WEO) report (2000, p. 41) does put forth quietly amea culpa, noting that while ‘the need for institutional infrastructure wasrecognized from the beginning … in practice it was not always given theattention it required, particularly if the macroeconomic situation appearedstable’. This does not admit the causal effect upon oligarchic developments;but the World Bank’s latest Country Economic Memorandum on Russia willdoubtless be interpreted as implicitly going in that direction. It analysesquantitatively how much concentration of ownership has occurred and,rather surprisingly for an international organization report in which thecountry analysed is a member, it uses the journalistic label ‘oligarchs’ andnames the heads of the top 23 business groups.

20 Background


Principal debates of the 1990s

The EBRD produces an annual transition report assessing general progressand focusing on different specific issues. In addition, many reviews with alonger backward perspective can be found in the literature. At about themid-point of transition a number of reviews of progress were undertaken.The World Bank devoted its annual World Development Report of 1996 to theissue, an academic publication appeared in the same year, Murrell (1996),while the EBRD’s chief economist, in a personal statement, Stern (1997), alsodrew some lessons from the first half-decade. More recently, with 10 years ofexperience several other reviews have been produced. An influential piece bythe Nobel Prize-winner, Stiglitz (1999), was a personal statement, though hewas then Chief Economist of the World Bank. Buiter (2000) is also a personalstatement from the Chief Economist of the EBRD, while the IMF’s annualWorld Economic Outlook (2000) focused on transition in an official statement,as does the World Bank (2000). Two recent reviews by academics are Svejnar(2002) and Campos and Coricelli (2002) which look more narrowly at theissue of recovery and growth of output in transition economies. This list caneasily be made much longer,7 but the above are sufficiently representative ofvarying points of view (extremely critical: Stiglitz; mixed: Stern and Svejnar;and modestly positive: IMF), and certainly comprehensive enough in coverageto allow a distillation of what were the key issues of debate about transitionin the 1990s. While a literal cumulative reading of these selected reviews pro-duces a list of more than a dozen issues, it is easy enough to group these. Insome cases the difference is only preferred terminology (speed and sequenc-ing in World Bank (1996) is really the same as big-bang v. gradualism inStiglitz (1999)). In others it is a difference in focus in the same general area(rule of law, property rights, role of government all fall under the concept ofinstitutional development.)

I propose the following eight groups of issues that capture well the mostimportant debates, either by their intensity at the time or their still-recognized relevance:

● the debate on speed, or more evocatively big-bang/shock therapy/rapidreform on the one hand, v. gradualism/incrementalism on the other;

● the related debate on best approaches to privatization comparing vouchers,direct sales, insider sales; I argue below why it is critically important toinclude facilitation of new enterprise formation;

● reasons for the unexpectedly large output fall;● the large difference in reform and growth performance between Central

Europe and the Baltics on the one hand, and CIS countries on the other;● the large social costs, rise of inequality and increased poverty incidence;● the sequencing and adequacy of development of market institutions

and the rule of law;

● the relationship between market liberalization and democratization; and● comparison with China’s less disruptive transition.

In this section, I summarize the past discussions on the big-bang–gradualismdebate, the large output fall, approaches to privatization and insti-tutional development, while later in the chapter we look separately at thelink of democracy and market reforms. The remaining debates are addressedin other chapters of this volume: differential progress in transition andgrowth performance across the countries and in different regions is the sub-ject of Chapter 2; Chapter 3 provides a comprehensive and updated analysisof statistical indicators showing social costs, inequality and poverty; therelevance of the Chinese experience is considered elsewhere. Before elabo-rating on the above list of issues, let me mention briefly a few others that arediscussed in the literature but do not merit elaboration, either because theconsensus on them is widely shared, or because they have in the end notbeen central to the transition.

The need to achieve macroeconomic stability and reduce inflation quicklyis virtually universally agreed; the sharpest critic of the WashingtonConsensus, Stiglitz (1999), clearly says he has ‘no great quarrel with shocktherapy to quickly reset expectations in an anti-inflationary program’ (1999,p. 22). At the margin, some issues of degree remain, but are not central to theoverall agreement. Thus, political scientists as in Reddaway and Glinski(2001) will occasionally express ambivalence, supporting inflation control,but arguing that a better policy than closure of inefficient enterprises was tocontinue providing credit for old industries until they could be restructured.In the economics literature one sees an early debate stimulated by the Polishbig-bang, about whether monetary tightness was not too heavy-handed orlasted too long, causing a credit crunch (see for example Calvo and Coricelli,1993, and later Kolodko, 2000a).

The need for substantial integration into the global economy and diversi-fication outside the socialist camp through liberalization of trade andfinancial flows was commonly expressed, but again there was little debate onits general desirability. Furthermore, the reality was a liberalization far fasterthan previous developing country experience,8 and a geographic reorienta-tion of trade that far exceeded early expectations.9 The importance ofreforming the financial sector into a market-oriented intermediation mech-anism linking savings and investment is a similar issue drawing much atten-tion but limited debate on the broad substance. As with the previous two,there was a broad consensus on the importance of financial-sector reform,including a vast operational literature on dealing with non-performingloans, achieving international norms for bank-supervision standards anddeveloping the capacity of banks to assess business lending (for exampleBokros, in Havrylyshyn and Nsouli 2001). I do not wish to suggest therewere no differences on the details of how to privatize banks, the role of

22 Background


foreign investment, the balance between restructuring and bankruptcy;but these discussions never took on the intensity and importance of theother debates discussed here. Another reason beyond the high degree of con-sensus may be that the relatively shallow nature of the banking sector inmost economies meant that for now, ‘lack of bank-finance does not seem toprevent private-sector growth’.10

Big-bang v. gradualism

I have argued above that wide-sweeping condemnations of the WashingtonConsensus have not contributed usefully to the debate on transition policies;it is better to break down the issues into narrower and manageable ones. Thefirst of these concerns speed, or the choice between gradualism and big-bang, although even this is made difficult by differences in definition andinterpretation. While no-one to my knowledge proposed doing everythingat once, there is certainly a useful distinction between advocates of seizingthe moment of political opportunity and implementing the major feasiblechanges as quickly as possible, in a big-bang – Balcerowicz, 1993; Klaus,1995; and Sachs, 1994 – and those who caution a more gradual approach tominimize disruptions, output and job losses and consequent reversal ofpolitical will – Aghion and Blanchard, 1994; Dewatripont and Roland, 1992.The institutionalist view argued in Murrell, Clague and Reuser (1992) lies in-between, admitting the need and usefulness of moving rapidly on stabi-lization and liberalization but cautioning that institutional developmentshave to begin in parallel and be sufficient to ensure the intended benefits ofmarket reform. It is notable that the final aims of the different approachesare the same, as is the concern for minimizing disruption and avoidingpolitical opposition to reforms.

Much of the argument for a more gradual approach has been conductedusing theoretical models, and as remarked by Aslund, Boone and Johnson(1996) there is somewhat of a dichotomy in the literature: theoretical analy-sis tends to conclude gradualism is better, while empirical analysts find thatearly and quick reforms result in an earlier and stronger recovery. This latteris confirmed in a review of econometric studies of growth (Havrylyshyn,2001; Campos and Coricelli, 2002).11

In addition, as noted earlier, there were theoretical doubters of gradualismmodels, showing in effect that they are not very robust. Katz and Owen(1993) develop a model of optimal privatization speed which maximizes out-put under constraints on unemployment and re-employment, and concludethat it is never optimal to privatize instantaneously, but it may be optimal tobegin rapidly then taper off, or even the reverse, begin slowly and accelerate.The outcome of the model depends on various assumptions about elasticitiesof response to new incentives both by state firms and newly privatizedones. In a later work, Katz and Owen (2000) show explicitly that the typeof gradualist model exemplified by the seminal work of Dewatripont and

Roland (1992) ‘may be viewed as lending strong support to the big-bangstrategy’, depending on the values assigned to certain parameters of themodel.

The most balanced view comes in Blanchard (1997) who models theprocess of reallocation taking care to consider the unemployment effects ofshifting labour from the state to the private sector:

I have not come up with a simple list of good or bad policies or a simplejudgment as to whether big-bang policies are better than gradualism …Our models rarely generate such simple messages [because] analysis oftransition is very much an analysis of second-best and comes with typicalattendant ambiguities.

One such ambiguity Blanchard’s analysis comes to is very striking inretrospect: ‘the analysis strongly suggests that carefully designed [my italics]insider privatization is the best way to achieve restructuring of state firms inthe long run’. This is diametrically opposed to the most recent consensus oneffects of privatization which suggests that insider privatization increasedthe likelihood of concentrated ownership and the strengthening of rent-seeking interests; the cases of Russia and CIS oligarchs are the clearest exam-ples, but even the Czech privatization while appearing to be voucher-basedwas in fact highly dominated by insider procedures and is judged by manyobservers to be an explanation of the poorer performance of the Czechrepublic within Central Europe (Svejnar, 2002). The reason the model canproduce this result is twofold: its simplifications assume agency problems areavoided by insider privatization compared to the mass-voucher based process;and the potential for generating concentration and rent-seeking is not partof the model.

Attempts at empirical analysis asking specifically if big-bang cases tend tocome out better than gradualism cases are limited and with ambiguousresults, probably because it is not a simple matter to define these concepts inpractice. There are three possible definitions of big-bang: early start after thedissolution of communism; rapid reforms once started; or a cumulative levelof reforms reached after ‘x’ years. With GDP growth as the ‘outcome’ mea-sure, Heybey and Murrell (1999) use the last definition and find speed doesnot help, while Berg et al. (1999) use the first and find it does. The bulk ofeconometric evidence on growth in transition until the late 1990s concludesthat the cumulative level of reforms reached is a key determinant (see thereviews by Havrylyshyn, 2001, and Campos and Coricelli, 2002), withCentral Europe and the Baltics showing much earlier recovery and fastergrowth than the CIS. This is consistent with an early start and if ‘x’ is definedas about five years also consistent with a cumulative reform definition. Butif growth rates since 2000 are considered, the relationship seems to reverse,since the CIS with reform attainment as measured by the EBRD still far

24 Background


behind Central Europe, has had much higher growth rates. One possibleexplanation is that what really matters for growth, as such, is the eventualattainment of a certain threshold level of market reforms, which implies thatgradualism in the sense of a late start is not an impediment for eventualgrowth recovery, though of course the output lost before the start is higher.The threshold hypothesis and other explanations are explored further inChapter 2, but here it should be noted that even if long–run output growthis not worse under gradualism, a late start can result in the undesirable out-come of state capture; indeed a central thesis of this book is that this is pre-cisely what happened in the 1990s.

A different argument on the gradualism side is that a too early or too rapidstart causes a lot of decline in output, job losses and other social costs. Thetheory on this is easy enough to imagine and again with the right assump-tions about the process of reallocation and unemployment–reemploymentdynamics can easily go both ways. The realities, as Chapter 3 will show, sug-gest early starters in fact suffer less of these costs than late starters.

A succinct assessment of this debate’s relevance using the case of Bulgariais found in Bristow (1996) and is worth quoting at length:

what has happened in the past five years adds up to serious macroeco-nomic trauma … was it all necessary? … Some of the real depressioncould have been avoided if trade liberalization had not been instituted sosuddenly and if monetary policy had given a lower priority to the controlof inflation. Inflation would have been less explosive … if a more gradualapproach … to the liberalization of prices. However, if macroeconomicreform had been more gradual and if monetary policy had been con-cerned more with the fall in output … restructuring would have beendelayed even more … There is thus no real trade-off between macroeco-nomic and microeconomic objectives. Those who claim that ‘shock ther-apy’ is all shock and no therapy fail to recognize that the therapeuticresults do not inevitably follow the shock. The unpleasant macroeco-nomic medicine is designed to assist economic reform: the latter is notsomething which happens automatically if … Bulgaria’s problem is not somuch that there has been a dramatic decline in living standards, but thatthe suffering will have been in vain if reforms at the more microeconomiclevel do not take place.

On balance, the case for gradualism has been weakened in recent years byseveral counter-arguments. First, the mathematical models of gradualistshave been used to derive the opposite conclusion: under certain conditionsbig-bang is theoretically the best strategy. Second, the evidence on growth atleast until 2000 is more consistent with the rapid-reform view. Third, the evi-dence on social costs shows they are smallest for the early and rapid reformers.Fourth, the evidence on rent-seeking phenomena suggest a late start and

gradual reforms increases the probability of state capture. This may notaffect medium-term growth prospects but does affect long-term economicand political stability as well as democracy.

Two important arguments remain in support of gradualism. The surge ofgrowth in the CIS may with time result in an overall recovery that is greaterthan in Central Europe with its much reduced growth rates, and this mayalso lead to a reversal of the dominant oligarchic power in the CIS. Time willtell, but it is ironic how the tables have turned: in the mid-1990s proponentsof gradualism criticized the high costs of rapid reform, while those on theother side of the debate asked for time to pass to see the recovery effects; nowgradualists may be needing time to pass before the lagging reformers returnto the right path and achieve equivalent results. The second major argumentin support of gradualism remains the same as it has been from the start, themuch greater and smoother success of China. As noted, this will be dealtwith later in the book.

The large output fall

That output would fall in the initial years of transition was widely predicted.Kornai (1994) explicitly titled his article ‘The Transformational Recession’and argued the key reason was the need to squeeze out inefficient produc-tion in the soft-budget state sector. Bruno (1993) discussed why the processof macro-stabilization after a growth crisis seen in previous emerging-marketcrises – especially in Latin America – would repeat itself in the early years oftransition. But that output would fall so much was not predicted. Thus,Bruno admitted that even if the fall was exaggerated by soviet accounting’soverstatement of earlier GDP, ‘the extent of the post-1989 collapse has noprecedent elsewhere’. The unexpectedly large decline continued to be animportant element of gradualist criticism despite the clear evidence by thelate nineties that those economies which reformed earliest and fastestdeclined least.12 Consider some of the reasons for this large output fall,which are most usefully summarized in Campos and Coricelli (2002).Chapter 2 will consider the possible overstatement noted by Bruno, but theanalysis there will still show that, at a minimum, output fell at least 20 per centin the case of Poland, about comparable to the magnitude of US decline inGDP after the Second World War: in all other countries it was higher.

The major explanations put forth in the literature are:

● a Keynesian decline in consumer demand associated with stabilizationmeasures;

● trade disruption with the end of the Council for Mutual EconomicAssistance (CMEA);

● hard-budget effects remove incentives for production of negative value-added goods and allow the reallocation of resources through creativedestruction;

26 Background


● excessive monetary tightness leading to a credit crunch; and● disruption of soviet period supply links (sometimes labelled as

‘disorganization’).

Campos and Coricelli (2002) are no doubt right to say pure Keynesian effectswere far less important than supply-side phenomena, though their claimthat in emerging markets stabilization programmes were not associated withsignificant output decline is wrong. Bruno (1993) was cited earlier to suggestthat a more than minor decline did indeed typically follow stabilization; theoutput reductions in the 1990s crises of East Asia and Latin America werealso substantial (ranging from 10–15%) even if quickly reversed. The secondand third points above are not unrelated to the first: hard budgets comingfrom proper pricing of unwanted goods (including the hidden distortions ofCMEA barter trade), plus cuts in subsidies from the budget, in effect reducedemand for the output of inefficient firms (not just consumer demand) andachieve the desired reductions of inefficient output as needed for resourcereallocation. The supply-side phenomena that were most important are thenew hard-budget policies and the disorganization effects as soviet supplylinks are only slowly changed to market-based links. The disorganizationmodel of Blanchard and Kremer (1997) has tested well in empirical studies,such as Commander and Coricelli (1995), and in the review of labour marketflows in the transition by Haltiwanger et al. (2003). Notwithstanding theview that earlier reformers recovered earlier, and that the laggards experi-enced a much longer period of reorganization, a reasonable conclusion isstill that the process of reorganizing new supply chains and markets tooklonger than may have been expected. In this regard Bruno (1993) was under-statedly prescient: ‘while there is a lot to learn from the reforms of distortedsubsystems in other economies, it makes a world of difference if almost thewhole economy is centrally controlled, non-market and publicly owned …and where a market economy has to be created ab initio’.

The credit-crunch hypothesis first gained prominence in Central Europe,including an intense debate on this for Poland, where critics of the Balcerowiczprogramme argued the degree and duration of monetary tightness wasexcessive and resulted in a bigger recession than needed for the hard-budgeteffects (see Kolodko, 2000a, Calvo and Coricelli, 1993). With Poland expe-riencing the least amount of decline and having the earliest recovery thisseems in retrospect like a marginal debate, valid as the argument may be.Indeed, to argue that Poland was ready for a loosening of monetary policywithin 18–24 months of the 1989 big-bang implies the conviction thathard-budget constraints were solidly in place. The issue of when stimulativepolicy, monetary or fiscal, can be put in place in a transition economy is infact an understudied question. The dilemma is simple: on the one handstimulation if it works can go a long way to minimize output decline andlead to recovery; on the other hand, if stimulation occurs too early – or the

worst case if it occurs by subsidies broadly-writ (directed credits toenterprises, tax breaks, tax arrears) – it will undermine the effort to impose ahard-budget mentality on producers, and inefficient negative value-addedproduction will continue. There is no consensus on the reasons for theexcessive output decline, nor indeed for how excessive it was. This last pointis examined again in Chapter 2 where it is shown that the official estimatesmay exaggerate the magnitude, but that there is no doubt it was historicallyvery high. On the reasons for this it is at a minimum clear – and againChapter 2 gives the details – that the magnitude of decline was greatest forcountries with slower progress in transforming their economic systems.

How to privatize

With the possible exception of Belarus under Lukashenka, and Turkmenistanunder Niyazov, there was a universal consensus that transition needed asubstantial increase in the share of the private sector, hence a large transferof ownership of state assets into private hands. The very first conclusionto draw about this debate is that the emphasis was too much on the processof state asset disposition, or ‘denationalizing’ ownership, and too little onestablishing conditions to stimulate de novo enterprise-formation. Indeed,the term privatization has become nearly synonymous with the process ofstate-asset disposition, though it would be better to think of the neededchange more broadly, as private-sector development. It is too late to use theterm privatization in the broader sense, and too awkward to employ thephrase ‘state-asset disposition’ to refer to the transfer of assets to privateownership, but readers will be able to infer which use of ‘privatization’ isintended. The voluminous new empirical literature on the consequences ofprivatization has come to recognize the distinction, and also to provide ten-tative conclusions on three key questions: What is the optimal method forstate-asset disposition? Has privatization led to the expected efficiencyimprovements? Where high concentration of ownership occurred has thisbeen good or bad?

The choice of technical method for disposition of state assets dominated thediscussions on privatization for most of the decade, with various positionsbeing taken on the advantages and disadvantages of mass privatization bydistribution of vouchers, direct sales to outsiders, and preferential salesto insiders, that is management and workers.13 The dilemmas addressedincluded fairness, speed of the transfer, usefulness of prior restructuring,prospects of early efficiency improvements, and, related to the last, ‘agency’problems – that is, avoiding the situation where new managers act not in theinterest of profit-maximizing owners, but in their own interest. These issueswere correctly considered to be of limited importance for small state enter-prises (retail outlets); hence the discussion was focused on large scale entities.

Each of these methods had some advantage over the others: mass voucherprivatization was thought to be the fairest, and could be implemented

28 Background


relatively quickly. But in practice, as described in Havrylyshyn andMcGettigan (2000) and Svejnar (2002), very imperfect secondary markets forvouchers developed leading to low-cost vacuuming up by poorly regulatedinstitutions, or by a few embryonic capitalists with insider knowledge. In theformer case, fraud pyramid schemes undermined public support as in theCzech Republic (Harvard Fund), and in even greater degree Albania andRomania. In the latter cases, especially relevant to Russia, Ukraine, and otherCIS states, this allowed insiders to obtain assets at far below potential value,and has been seen by the populace and outside observers as legalized theft(Freeland, 2000b). In addition, as Svejnar (2002) notes for the Czech Republic,Slovakia and Lithuania, where in some cases large amounts of shares wereretained by individuals, this meant a wide dispersion of ownership and aconsequent poor-governance situation allowing ‘managers or majority share-holders to “tunnel out” assets at the expense of minority shareholders’.

Insider privatization was thought to allow very speedy transfer, andimmediately provide a group of owners and managers with a strong stake inimproved efficiency. In practice, this method too was often abused, asmanagers in effect controlled the workers’ shares given the inertia of therelationship in the soviet period, and thus two problems arose. First, a poten-tial for excessive concentration of control if not actual ownership (as describedin Barnes, 2003a, in a comparison of Hungary, the Czech Republic andRussia; and World Bank, 2004, for Russia). Second, agency problems as con-trolling insider managers were able to tunnel-out assets at the expense ofworker-owners and any outside shareholders.

Sales to outsiders was considered to be very slow, and of course unfair tothe population, but it had the advantage of avoiding the corporate governanceand agency problems that both of the other methods suffered from. While inpractice almost every country used some combination of the three methods,there are discernible tendencies to concentrate on one type; Svejnar (2002)even sees ‘remarkable differences across transition countries in the strategyof privatizing’. Havrylyshyn and McGettigan (2000, table 3) does suggestthat outsider sales were the major tendency in Estonia and Hungary, andthese cases are generally judged to have had far better outcomes than others.As Svejnar concludes, this ‘provided much-needed managerial skills, externalinvestment funds, generated government revenue, and effective corporategovernance, and turned out to be relatively fast when carried out by deter-mined governments’. This seems to point to outside sales being more effec-tive in practice than either voucher or insider buy-outs, both of which canlose their advantages if the process is captured and abused by insiders orfraudulent financial operators.

But if one considers the result of outsider sales in Russia (Barnes, 2003a,2003b; Frye, 2003; Stiglitz, 1999), starting with the loans-for-shares arrange-ments of the mid-1990s (and the later outsider sales in Ukraine: Aslund,2002; World Bank, 2000), the conclusion is much weakened. The governments

may have been insufficiently ‘determined’ in those cases, or more likelydetermined but already captured by the wealthy new potential purchasers.One cannot characterize Ukraine and Russia as cases of rapid mass privatiza-tion via vouchers; yes, there was a lot of that but it was subverted by thevacuuming process, and in any event as World Bank (1996) recognized,‘Russia’s mass privatization program of 1992–94, although it used vouchers,was basically a case of management-employee buy-out because of its prefer-ential treatment of managers and workers’. Furthermore, as Barnes and oth-ers describe, there were later a lot of direct sales, in particular to banks in theloans-for-shares exercise (some readers will doubtless replace the word ‘exer-cise’ with ‘scandal’).

In the end, the choice of method has turned out to be far less importantthan two other, related, characteristics of the process: the transparency ofinformation and process; and the role of political insiders who formally orinformally influenced the mechanism of implementation, and at the sametime were in a position to be the new owners. If one uses insider in thebroader political sense, the larger their role, the more likely transparency didnot prevail, and as a consequence any one of the technical methods could besubverted to the interests of these insiders. Conversely, the less insider influ-ence, the more successful privatization regardless of method. Thus we seenow all of the Central European and Baltic countries with a fundamentallygood outcome, despite the fact that Poland and Slovenia delayed the processfor a long time then used some combination of sales and vouchers, or thatLatvia and Lithuania relied relatively more on vouchers than did Estonia.

The efficiency effects of privatization can today be assessed on the basis of alarge number of studies both at the country and firm level; the conclusionshere are based on three review articles which summarize well over a hundredempirical studies covering a large proportion of the transition countries andthousands of individual firms: Havrylyshyn and McGettigan (2000); Nellisin Havrylyshyn and Nsouli (2001); Djankov and Murrell (2002); andHaltiwanger, Lehmann and Terrell (2003) who focus on job-creation effects.The vast majority of studies find that private firms have greater productivity,faster growth and generally create more jobs. The effects appear to be muchstronger in transition economies than in the earlier experience of othercountries, which is reviewed by Megginson and Netter (2001). Some doubtsremain, as there are important studies which do not support this conclusion,but these, like Earle and Estrin (1997) for Russia, tend to be earlier ones whenthe period of time was perhaps too short for a good assessment, especiallyoutside Central Europe. Also, it is not always clear as Svejnar (2002) notesthat major sampling and data problems, for example ‘cherry-picking’, havebeen adequately addressed. Some individual studies do account for cherry-picking and some but not all find positive effects. Thus, despite the uncer-tainties, it seems difficult to ignore the sheer magnitude of the tendency inthe econometric studies, which as the extensive review by Djankov and

30 Background


Murrell (2002) puts it: ‘privatization is strongly associated with moreenterprise restructuring’.

The studies also show a strong consensus on the rank-ordering of efficiencyeffects according to the type of firm. The highest are de novo firms, small andmedium-scale firms are next, large privatized ones follow, and the least effi-ciency improvement is found among state-owned firms. This carries an impor-tant implication: new and small firms are the most likely source of growth inthe transition process, much as they have been in development experienceearlier. Such a finding vindicates those who advocated at the outset puttingmore emphasis on new development of the private sector and not so much ondisposing state assets; see for example Krueger and Winiecki’s comments inClague and Reuser (1992). The reasons for the key role of de novo firms areelaborated in Havrylyshyn and McGettigan (2000). With no vested interests toprotect, these firms are more likely to be highly energetic and innovativesupporters of more liberal open markets as well as even-handed rule of law.Initially, they will operate in low capital-intensity areas thus creating morejobs, and they will not count on government support or privileges simplybecause they cannot get them. Given that the new, small sector is still verysmall in many of the transition economies outside Central Europe and theBaltics, the lesson here still has tremendous relevance for the future.

One other important finding deserves attention, even though it is basedon a relatively small number of studies: the interplay of ownership per se,and the market environment facing firms. A key article by Zinnes, Eilat andSachs (2001) uses cross-country analysis and index measures of institutionssuch as degree of competition, openness to entry, the quality of rule-of-law,to show that it is the combination of markets and competition which leadsto efficiency improvements, while private ownership alone has virtually noeffect.14 Recall Stiglitz’s phrase ‘There must be both competition and privateproperty.’ The issue of general institutional improvements has in fact cometo the fore since the mid-1990s as an under-emphasized dimension intransition policy, and it is furthermore linked to the problem of excessiveconcentration of ownership discussed next.

While it makes sense to focus on efficiency improvements as an empiricalmeasure of how well or badly privatization has gone, there is another outcomethat for the long run may be equally important: the degree of concentration ofownership, or the ability of a small number of new capitalists to influence statepolicy. Those who criticize strongly the results of privatization no longer gen-erally deny the measured efficiency gains, rather they focus on the nature ofthe economic and political organization that is generated. The positionsof major critics of transition so far was outlined in the first section of thischapter, where I argued that their full-frontal attacks on the WashingtonConsensus are mistaken or misplaced. But there is considerable merit to thepart of their arguments which contend insider-dominated privatization hasbeen neither good for the economy nor the polity, it has been unfair and

undesirable, and does not augur well for future prospects. A large number ofempirical studies, some qualitative, some quantitative, provide evidence onhow insider-oriented processes of state-asset disposition tended to result inownership concentration, creation of oligarch groups, and eventually statecapture by these vested interests.

Thus Barnes (2003a) compares Hungary, the Czech Republic and Russia,explaining in great detail how in the latter the insider mechanisms worked,how they were explicitly advocated by many reformers inside and outsidethe country to ensure co-optation of the Soviet-period directors to achieve aspeedy transfer, how the new banks played a role in this and in fact oftenbecame the majority owners, and how in the end this led to the concentra-tion of ownership in the hands of about two-dozen clearly identifiableindividuals heading enormous business groups. Barnes includes a useful dis-cussion of the ways in which these ‘oligarchs’ currently influence policyensuring it is favourable to them and protects them from potential competi-tion, internal or external. Acquisition of upstream or downstream compa-nies – steel companies acquiring coal – mines or export ports – not onlydefends against competition, but often allows regional power concentrationwhich in turn provides political leverage through bribes, campaign financ-ing, threats of closure, job cuts. A revolving door of personnel from top posi-tions of industry to government and back permits economic interests toinfluence or even draft laws and decrees pertinent to their business. Morerecently a World Bank (2004) study on Russia not only gives a list of ‘oli-garchs’ remarkably similar to that in Barnes (2003b), but argues in greatdetail and with considerable supporting empirical evidence ‘that concentra-tion of ownership in the hands of a few major players can result in collusion,create barriers to entry, and eliminate healthy competition’. The forcefulnessof the conclusions and the explicit naming of names by an internationalinstitution like the World Bank is unusual enough to also speak volumes.

The possibility of rent-seeking interests acting against the greater good ofthe society is not a new concept in economics. As a general proposition, itwas thoroughly explored in theory and practice by Krueger (1974). It foundits most elegant expression for transition economies in Hellmann (1998),who wrote about how the ‘winners’ – the new capitalist class which in factarose most often from some part of the former communist elite – dominatedthe process of taking over ownership of state assets and used this economicpower as a lever to capture the mechanisms of the state and polity. Thepotential for this was recognized earlier with more or less clarity in thewarnings about asset-stripping in the late 1980s (Clague and Reuser, 1992),or concerns that newly established rent-seekers opposed to full liberalizationwould control policy and ‘freeze’ the transition (Havrylyshyn, 1995). This,more than any other aspect of privatization should in my view become themost important issue for discussion looking forward.

The oligarchic outcome of insider-privatization in Russia, Ukraine andmany other countries has resulted in a high degree of state capture; this

32 Background


process is described in more detail in Chapter 6. Such an outcome has alsobeen used by critics of rapid reform to buttress their argument, I believe incor-rectly. Reddaway and Glinski’s (2001) central thesis may be summarized assaying that the effort to impose ‘shock-therapy’ (their term) was bound to failas preexisting nomenklatura interests would subvert it in their favour. In thesame vein, Murrell (1996, p. 42) argues that having failed in 1992 with a pre-cipitate stabilization and liberalization, Russian reformers ‘… gambled thatprivatization is a sufficient condition for all other reforms’. The implicationof a causality running from big-bang to vested interests pursuing rents andeventual ownership concentration through insider-privatization is also clearin the argument of Stiglitz: ‘corruption and rent-seeking itself may have beenincreased by the manner in which reforms were conducted’.

The problem with such logic is two-fold. Conceptually, it must deal withthe established rent-seeking literature’s consensus that it is precisely too-slowrather than too-fast reforms that increase opportunities for vested intereststo act, both because distortions remain allowing high rents, and becausemore time is given to them to lobby against reducing those distortions.Empirically it must deal with the fact that, even if Russia is considered a caseof rapid-reforms, only one other CIS state where state capture took place canbe so labelled – Kyrgyz Republic – while all the others pursued a gradualstrategy. The vast majority of the rapid reformers in Central Europe and theBaltics, which did pursue rapid reforms, avoided the worst of the insiderprivatization, oligarch-economy situation. Indeed for Russia it can easily beargued that the big-bang of 1992 was at best very short-lived and soonaborted, and it was its reversion to gradual and incomplete reforms that ledcausally to insider privatization.

To summarize briefly on privatization, consider the three questions put atthe outset. First, there is no one technical method of privatization that hasworked better than any other, because the dominant determinant of out-comes has been the degree of non-transparent political insider influence inthe process. Second, all forms of privatizing appear to have some beneficialeffect in efficiency, but in a clear rank-order starting at highest: new enter-prises, small and medium enterprises, large enterprises, and still state-ownedones. An important related conclusion is that private ownership alone has alimited effect, but adding a favourable institutional and competitive envi-ronment gives much greater results. Third, a high degree of ownership con-centration may not in the short run preclude improved efficiency, but in thelong run it is problematic if it leads to state capture by a handful of powerfulbusiness groups, or oligarchs.

Development of market-enhancing institutions

There is so much agreement that institutions matter in developing an effec-tive market economy15 that the debate on this, if not properly focused, candegenerate into sterile straw-man accusations. Reference to good marketinstitutions, government regulations and a strong legal system is to be found

in virtually all early writings on how to undertake transition, hence debatingwhether the Washington Consensus road-map did or did not excludeinstitutions is not very useful: conceptually it did so very clearly. Argumentsthat institutional development should have preceded liberalization andprivate-sector development are only useful if very narrowly focused on precisesequencing of specific elements of reform and specific types of institutions.The institutionalist approach, as North (1993 and 1995) and Murrell (1996)note, is based on the historical lesson that it took decades if not centuries forproperty rights and related institutions to develop. If this argument is puttoo broadly it is unrealistic in practice as it implies waiting until institutionsare in place. Thus, the broad criticism of Stiglitz (1999) is not helpful, whilenarrower analyses that demonstrate which specific institutions creating acompetitive environment to ensure transfer of ownership has beneficialeffects (Zinnes, Eilat and Sachs, 2001) can provide specific and implementablerecommendations.

Also useful are the criticisms, whether right or wrong, that suggest toolittle weight was given to institutional development in the early period, asin Moers (1999) and IMF (2000). The lessons of recent history on the roleof government and the development of institutions are explored verythoroughly in the World Bank’s 2002 World Development Report, not only fortransition economies but other developing countries.

The somewhat surreal agreement by all observers that institutions matterin the transition hides the important differences, which can be summarizedin two categories:

● What are the most important market-enhancing institutions needed; thatis, is there a minimum critical mass?

● Should they be developed before, during or after the main steps ofstabilization and liberalization?

As there are many relevant institutions, it is not surprising there would bedisagreement on relative importance and sequencing. The list of institutionsor institution groups given in the World Bank (2002) study, and the analysisthere, does suggest there are some that should be very early, some that canbe developing simultaneously with introduction of specific reforms, andothers which can be allowed to evolve and respond to the market’s needsover a much longer period. The first category would surely include preexist-ing elements such as the state’s ability to enforce basic law and order, butalso new elements such as a market-oriented laws and government agencies:a Central Bank, a Finance Ministry that enforces budget discipline, regula-tory agencies for enforcing codes of commercial behaviour, an anti-monopolyregulator.

The second category is exemplified by the link between, on the one hand,privatization and freedom of private-sector activity, and on the other the

34 Background


legal basis for secure property rights plus ensured competitive behaviour.Without a legal basis of property rights, new owners will be very cautious aboutengaging in new ventures, expanding output and employment, and so on.In some countries this was established very early, with, in cases like Polandand neighbours, a smooth reintroduction of pre-communist commercialcodes by 1991–92, while in Russia, Ukraine and most other CIS countriesnew laws allowing free enterprise coexisted for some years with Soviet-period laws deeming it illegal until about the mid-1990s. Ensured competi-tion and open entry are needed to avoid monopolistic behaviour which canlimit the benefits of the transition from socialist to private ownership, asmany have by now shown. This was much more of a problem, partly becauseof delays in introduction of laws, and partly because of ineffective applica-tion of laws. This last merits a separate discussion below.

Other links in the second category include introduction of a commercialadjudication infrastructure for disputes, bankruptcy and so on. That thisneed not be fully in place before private activity begins, is illustrated by thecase of Lithuanian judges in new bankruptcy courts who had no experienceand no fully developed regulations on how to apportion creditor rights.They proceeded by a combination of trial and error, learning about proce-dures used in Western Europe, participating in training opportunities offeredby external technical assistance, and recommending to their governmentadjustment to the laws on the basis of early experience. An analogousprocess occurred in all countries with tax reforms, with gradual introductionof new forms of tax (value-added tax, property tax, lump-sum taxes at firstfor small entities like bazaar-merchants), and new regulations concerningtiming of payments, declarations of income, refund procedures for exportedgoods, penalties for late payments, and so on. Very few countries even inCentral Europe were able to put in place a complete set of tax laws and reg-ulations before at least the mid-1990s, and most continue with refinements.That all have seen significant economic recovery leads to the conclusion thatthe process of developing market institutions can be symbiotic with expansionof private market activity and does not need to precede it.

There may not be clear examples of institutions that are fully postponableuntil after the basic reform steps are done; a more useful distinction isbetween having a basic and simplified legal-regulatory framework in place asthe new market regime begins operation, and refinements of these regulationsover time. It is therefore an ad hoc judgment to be made in each situationwhat the basic minimum may be, and how soon the refinements take place.In retrospect, it is clear that most of these economies proceeded along such apath, albeit at different speeds. The Central Europe group moved fastest usingtwo vehicles: in the early 1990s reintroduction of pre-communist laws, inparticular Commercial Codes or copying from Western European exam-ples. This was the case in Poland and less so in the Czech Republic, Slovakia,Latvia and Lithuania, from the mid-1990s implementing the chapters of the

Acquis Communautaire in their efforts to accede to EU membership. Thecountries of the CIS – with some exceptions such as Armenia’s land laws –moved much more slowly, intermittently introducing new autochthonouslaws and abolishing Soviet ones.

But this speaks only about the laws on paper, not about the effectivenessof their implementation. It is a widely held view of Kremlinologists that theUSSR in particular, but also the satellites, were states in which informallegalities were far more important than the paper laws. Thus, republics of theUSSR were constitutionally free to secede from the Union, but that none didso before 1991 and all did after that cannot be interpreted as a sudden massconversion to secessionist desires. Similarly, private economic activity waslargely banned, but as Handelmann (1994) elucidates, the existence of under-ground economic activity and a ‘soviet-mafia’ well-known to the authoritiesyet largely untouched speaks volumes about what the real law was. Untilrecently, economists have paid limited attention to the informal institutionswhich political scientists have long studied; but in the transition debates ithas come to be recognized the issue revolves around the effectiveness of imple-mentation, not the paper laws.

Recognizing this analytical work on the level of institutional developmentin transition rarely measures this by the existence of legal text, but rathertends to rely on subjective perceptions of the effectiveness of different cate-gories of institutions. Early compilations of such synthetic indicators of mar-ket institutions come from Freedom House and the Heritage Foundation,and Transparency International with its narrower corruption index. Morerecently, a comprehensive compilation of various sources and some newindicators have been put together by the World Bank (Kaufmann, Kraay andZoido-Lobaton, 1999) and continue to be updated in annual surveys of thebusiness or institutional climate. Weder (2001) uses such data to assess thedegree of institutional development in the region, and Table 1.1 summarizesthe results. Clearly, only Central Europe and the Baltics are beginningto approach the levels of advanced market economies, but many of thecountries in South-East Europe and the CIS were by the late 1990s alreadycomparable to developing market economies in the middle and lower rangeof institutional development.

What does this tell us about the debate on the role of institutionaldevelopment: were there damaging delays in putting in place effective insti-tutions? And were there some particular institutional elements that werecritical? The values of Table 1.1 are good evidence that outside of CentralEurope and the Baltics, institutions still lag far behind desirable levels, con-firming the concerns of critics like Murrell, Stiglitz and many political scien-tists as discussed above. But in a cross-country comparison to be elaboratedin Chapter 3, it becomes clear that the pace of institutional development isbroadly related to the pace of economic reform which is not consistent withthe conclusion that rapid reform went too far ahead of institutional changes.

36 Background


An alternative stylized fact is that where there was an ability to move fast oneconomic reforms, there was a parallel capacity to move quickly on supportinginstitutions.

When one turns to the issue of isolating critical institutions that may havedone the greatest damage if missing, or the biggest contribution if in place,the answer seems to be: a minimal degree of property-rights security, and aliberal free-entry environment equal for large and small enterprises. This canbe seen referring back to the discussion on private-sector development. Theevidence points to several stylized facts. In most of the Central Europe andBaltic (CEB) countries, early establishment of property-rights regulationsand, more importantly, the perception that this was applicable to small newunits, led to a boom in entrepreneurship and contributed importantly to theearly recovery. In contrast, the uncertain environment in other countriesdrove any embryonic new capitalists underground,16 and discouraged themfrom becoming so large as to show their heads above the surface. Furthermore,the weak property rights plus great administrative barriers to new entrants,at a minimum facilitated the process of development of small groups of oli-garchs who became owners of a large proportion of privatized state assetsthrough insider-privatization privileges.

This last fact is one of the most important and valid points made by criticsof the reform process so far, though as argued above they often underminetheir own argument by linking ‘bad’ privatization with rapid reform, whilein fact the broad picture is that rapid-reform countries most often avoided

Table 1.1 Quality of institutions in transition economies 1997–8

Average value of index(a) Developing countries in (Range �20 to �20) same range

Central Europe(b) �6.0 Chile, Korea, South AfricaBaltics �4.0 Uruguay, UAESouth-East Europe �2.8 India, Lebanon, PakistanCIS Moderate Reforms �6.1 Peru, Burkina Faso, GuatemalaCIS Limited Reforms �10.3 Kenya, Haiti, LaosAverage IndustrialCountries �12.6

Notes: (a) The Weder index is a composite of five governance indicators from the World Bank database noted in text. (b) The country groupings are elaborated in Chapter 2 and reflect approxi-mately the degree of progress to market economy status. Central Europe includes: Hungary,Poland, Czech Republic, Slovakia, Slovenia, Croatia, Baltics, Estonia, Latvia, Lithuania. South EastEurope: Bulgaria, Romania, Macedonia, Albania – no values are available for Bosnia–Herzegovinaor Serbia–Montenegro here, but these countries are included for other data tables. CIS ModerateReforms: all CIS except Belarus, Uzbekistan, and Turkmenistan which fall in the group CIS LimitedReforms.

Source: Author’s calculations based on data in Weder (2001).

‘bad’ privatization. Nevertheless the issue deserves attention as one of themost important problems looking forward.

Main issues looking forward

At the beginning of this chapter I listed eight debates or contentious issuesof the past decade; two of them should definitely remain in the past as theyhave become sufficiently resolved or at least made less relevant. The speeddebate has become far less relevant, and the evidence strongly suggests thatunless Chinese initial conditions of large surplus labour prevail, early andrapid reform has in practice generally given better results with the possibleexception of excess speed of privatization in some cases. The arguments thatlarge output falls were due to excessively tight stabilization and too-rapidliberalization are also shown to be questionable, in particular if some correc-tion for Soviet-era ‘GDP’ values is done. The other six issues remain of someimportance. Three of them are more historically important but not critical orcontentious for future considerations: the large differences across countriesin transition progress and performance; the costs of transition, both economicand social; and China’s approach as a model. The first two are consideredrespectively in Chapter 3, Chapter 4; the case of China is beyond the scopeof this book, though many relevant comparisons are made throughout.

The remaining three, plus two new issues, constitute the most interestingfocus for forward-looking discussion. The three carry-over issues are: bestapproaches to privatization; development of market institutions; and therelationship between markets and democracy. But for the future somewhatdifferent aspects of these problems are most relevant. The first new issueconcerns the influence of ‘oligarchic’ owners on further liberalization; thatis, will they support it for the sake of securing property rights, or will theyoppose it for the benefits of rent-seeking gains. Looking back, this is connectedclosely both to the way privatization was done allowing in some countriesthe development of such strong vested interests, and to the underdevelop-ment of institutions which facilitated a non-transparent transfer of assets.Looking forward it is critically related to further progress on institutionaldevelopment, completion of economic liberalization, and continued democra-tization. It is also related to the second new issue: the prospects for continu-ation of the surge in growth in the relevant CIS economies. To sum up, thehuge transition problems for the future are:

● correcting inadequacies or negative effects of privatization;● institutional development;● the relationship between markets and democracy;● does state capture lead to a freezing of transition? and● can the surge of growth in the CIS continue and lead to further progress

in reform?

38 Background


These new issues are of limited relevance to countries that have alreadybecome members of the EU as of May 2004 – the Czech Republic, Estonia,Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia – and probablyalso for some of those in South-East Europe that have not, but are likely todo so in a few years – Bulgaria, Croatia and Romania. The other South-EastEurope countries are in a marginal zone, while the 12 CIS countries are clearlythe ones for which these issues are most relevant. But a historical under-standing of why many countries no longer face such problems will be neces-sary for a good analysis of the future prospects for those that do. Therefore, allthe countries will be considered in the backward-looking analysis.

For the future, the challenges for the first group of countries are related toEU membership, macro and exchange-rate policies upon accession to theEU, adoption of the euro and its timing, and the like. For the others, the fivenew issues are the important ones. For them, I propose the most useful andcritical aspects of discussions or debates (as they have already started) can begrouped under two schools of thought. The first argues that once a mini-mum of stabilization, market liberalization and privatization is achieved, fur-ther progress in transition is inevitable, (indeed it will also help eventually infurthering democratic processes). The second takes a conditional view of thisinevitability, arguing that where the reform process allows vested intereststo build up quickly and benefit from rent-seeking opportunities of partialliberalization, they acquire a concentration of state assets in opaque privati-zations to become what is popularly known as the oligarchy, and finally cap-ture of the state. Their interests are not to liberalize or democratize further,rather exactly the opposite. The transition is frozen (TF) into a capitalist butnot competitive economy and an autocratic polity perhaps in superficialdisguise of a voting-democracy.

The transition inevitable (TI) paradigm does not simply argue that a criticalmass of private ownership and market-based decision-making makes rever-sal impossible and progress inevitable; it has a more sophisticated logic cen-tring on the value of property rights. Once a major part of assets is in privatenot state hands, the old bureaucracy and nomenklatura no longer has apower base to oppose reforms, but, more important, the new capitalists willwant to have security of property rights and create a demand for rule of law,transparency, law and order and so an. Thus, Shleifer (1997) summarizes:‘Russia’s experience shows how privatization, combined with equity incen-tives for enterprise insiders, transfers control rights from the bureaucratsand stimulates political and economic pressures to protect private propertyrights.’ Similarly, Aslund (1997) argued that ‘Russian capitalists want to beindependent of bureaucrats and safeguarded by a system of law.’ A veryrecent exposition of this optimism is in Aron (2003) in the New York Timeseditorial opinion addressing the jailing of Khodorkovsky, head of Yukos:‘oligarchs can help advance the cause of Russian democracy’. This view ismore broadly applied beyond Russia by others, starting perhaps with the

first mid-stream review of transition by the World Bank (1996 WorldDevelopment Report),17 and includes important writings on how privatiza-tion should be done such as Boycko, Shleifer and Vishny (1995), as well ascurrent recommendations for countries suffering from oligarchic concen-tration of ownership: Boone and Rodionov (2001), Cottrell and Ostrovsky(Financial Times, 16 April 2001).

Buiter (2000) (whom I would put in the TF school) points out that the the-oretical basis for this is in Coase’s proposition that efficiency only requiresthat property rights are assigned unambiguously, which he restates in morepopular terms as ‘yesterday’s thief is the staunchest defender of the sanctityand inviolability of property rights’ (p. 606).

The counter-argument of the TF school is simple: capitalists will favourrule of law not out of benevolence or ideology but only if it is in theirinterest. The essential difference with TI is the empirical proposition that insome cases – Russia and other CIS countries among them – it is not in theinterest of the new capitalist class holding concentrated ownership of assets.Thus, for example, Polischuk and Savvateev (2004) argue ‘inequality of own-ership … could make wealthier agents favor less than full protection ofproperty rights … [and] property rights will not emerge from the grassroots’.An important corollary is that while large oligarchic owners do not necessarilygenerate an endogenous demand for property rights and good rule of law,small owners, and especially new entrepreneurs, do. The idea of a frozentransition has many antecedents before this formal model of PS, and corol-lary elaborations. An interesting aspect of these writings is that many if notmost of these sceptics of coopting existing vested interests by allowing themprivileges on privatization, were in fact otherwise ardent proponents of theWashington Consensus. I will make only a few references.

Interestingly, the earliest hints of these different approaches are seen inthe views of two well-known proponents of the Washington Consensus. In a1991 conference gathering of eminent Western economists (proceedingspublished in Clague and Reuser, 1992), Fischer while recognizing the roleof new enterprises emphatically stated ‘privatization of state assets is anessential step in the creation of the private sector’. In contrast, Kruegeremphasized that,

experience from developing countries is, by and large, that growth hastaken place primarily through the emergence of new activities, not theadaptation of older ones … [and] focus upon privatization of existingassets … searching for the least unfair process … diverts attention fromthe more important problem of creating new earnings streams.

Fischer was certainly not saying that competitive rules of the game couldwait, nor was Krueger arguing not to bother with privatization, but therelative emphasis is critical.

40 Background


Havrylyshyn (1995) formulated an early version of the TF argumentnoting that already ‘markets have been created in CIS countries with privateownership and profit opportunities … opportunities restricted … to themore privileged few and not opportunities for competitive capitalism …various interests will do everything to avoid further reforms and keep thetransition process frozen’. A pioneering contribution in the TF school is inHellman (1998), who confirms the fears that privatization of the typeadvocated by Shleifer et al. (which coopts existing politically and financiallypowerful individuals by giving them a quick and low-cost insider access toprivatizing assets), did indeed result in a concentration of assets that wouldbe inimical to further liberalization. He underlined a paradox in the matter:from the mid-1990s the strongest and most effective opposition to furtherreform in many CIS countries came not from the losers, the large mass of thepopulation which felt the pain of unemployment and lower living standards,but from the ‘winners’, the new capitalists benefiting from a non-transparentand generally inequitable transfer of state assets.18 A key conclusion ofHelman’s article is that these ‘oligarchs’, as they are now commonly called,have the ability to capture the state and ensure its policies are favourable tothem, rather than to open competitive markets.

Buiter (2000) expresses concern that this predation inhibits secureproperty rights and thus depresses capital formation and growth. Yavlinsky(2003), a closer observer than any of the others, makes a poignant plea forbattling the oligarchy, a plea firmly planted in simple economic theory ofcompetition: ‘the larger and more influential the group, the greater are itsopportunities to deviate from the universality principle of the businessclimate and fair competition’.

For those who are not convinced until a model is formulated mathemati-cally, Polischuk and Savvatev (2004) provide such a formulation. It hingeson the simple trade-off calculation made by an oligarch: is the benefit ofnon-transparent rent-seeking greater than the cost of some uncertaintyabout the oligarch’s own property rights security? Since the former is large(as earlier developing-country literature showed from the time of Krueger’s,1972, seminal piece on rent-seeking) and the latter is easily bought infor-mally, it is not surprising that Polischuck and Savvater can conclude ‘somewealthier agents would prefer a hybrid equilibrium (with informal proce-dures and rent-seeking opportunities) to the market one, and thus wouldresist secured property rights’. In simple words, that’s because secured prop-erty rights allow capitalism for all, while oligarchs currently in control muchprefer capitalism for the few.

A large number of empirical studies provide evidence on how these processesof privatization resulted in concentration in some countries but not others.Barnes (2003a) and the World Bank (2004) have been cited earlier on this.

A related large literature on the effects of privatization reaches conclu-sions that are consistent with the TF’s logic: concentrated ownership is not

favourable to competitive efficiency or further liberalization. The consensusfindings generally show that any privatization is better than none, but thatby far the best privatization is one which allows small- and medium-sizedenterprises (SMEs) to thrive, and new entrants to enter. Several studies showthat privatization alone without a competitive environment has little or noeffect, but with such an environment it is very much more effective.

Another relevant strand of recent literature is on voting patterns. Thegeneral finding in studies that cover both CEB and CIS is that small entre-preneurs, and new entrepreneurs, are particularly strong supporters forrule of law, competitive markets and democracy (see on this Fidrmuc, 2000;Raiser, di Tommasso and Weeks, 2001; Frye, 2003; and Jackson, Klich andPoznanska, 2003).

The future importance of this debate should not be underestimated, orovershadowed by the current reality of a surge in economic growth in theseeconomies to levels of 5–8 per cent annually or more, approaching rates seenin the period of the East-Asian boom in the 1970s. The direction in whichsuch polities go, freezing oligarch power, or yielding it for the sake of secur-ing property rights for assets accumulated so far, has implications for growthprospects, further liberalization of the economy and further evolution of thefragile democracy status achieved. Since the TI and TF issues are very muchpolitical economy ones, it is of no small interest to refer to the politicalscience literature and debates on post-communist democratization, which issummarized in the next section.

The transitology paradigm in political science literature19

There is an analogy between the TI v. TF debate in economics and thetransitology debate amongst political scientists. This section will first sum-marize the main elements of the debate in the political science literature,and then draw on this to clarify different views on the link between marketreforms and democratic developments in transition economies.

The transitology paradigm predates the post-communist ‘transition’, andthe similarity of terminology is coincidental. Transitology is one of the the-ories or views about how change from autocratic to democratic regimes takesplace, and is distinguished from other models by its linearity: that is the viewthat once the regime begins to move away from authoritarianism it movesineluctably towards democracy, albeit with bumps on the road. In this itbears a similarity to the TI argument in economics. The broader literature onchange from authoritarian to democratic regimes became prominent in the1970s and 1980s with writings on Latin American democratization, a cen-trepiece of which is a four-volume study by O’Donnell and Schmitter (1986).But this transitology paradigm is subject to considerable attack in the litera-ture, Carothers (2002) arguing the linearity is an illusion as one goes around

42 Background


the globe, and McFaul (2002) and Motyl (2004) making a similar point forthe post-communist states and perhaps going even further to say that, ifanything, one sees by 2000 a trend to polarization rather than a convergenceto democracy. Consider first some of the general debate on the transitologyparadigm.

The transitology paradigm in the study of democratization, both in thepost-communist world and more broadly, has been difficult to verify empir-ically, and is subject to much criticism, but remains an important tenet ofpolitical science studies of democratization. A seminal exposition in politicalscience is provided by O’Donnell and Schmitter.20 One of the most recentcritiques of transitology is by Carothers,21 who lists five key assumptionsunderlying this paradigm:

1 Any country moving away from authoritarianism is moving towardsdemocracy.

2 Democratization takes place in a set sequence of stages (this will befurther elaborated below).

3 Elections are the key to democratization and further democraticconsolidation.

4 The structural conditions of democratizing countries, such as the levelsof economic development or ethnic composition, as well as historicalexperiences, are not crucial influences on the course and outcome ofdemocratization.

5 The ‘third wave’ of democratization takes place in functioning states.

Perhaps the most contentious issue is the assumption of linearity in theprocess of democratization, democracy proceeding in three stages. The firstis democratic opening, liberalization under the authoritarian regime, usuallymanifested in rifts between reformers and hardliners within the regime. Thisis followed by the collapse of the authoritarian regime and its rapid replace-ment by a new, democratic system. Democratic institutions are introducedand codified in a democratic constitution. A prolonged stage of consolida-tion follows, during which ‘democratic forms are transformed into democra-tic substance’, so that democracy becomes the only game in town, acceptedby all major political actors (Carothers, p. 7).

One of the key objections to the transitology paradigm is that it is toodeterministic in assuming that in each case democratization proceedsthrough a similar set of stages. Fish (1999), for example, argues that transi-tologists focus too much on studying the impediments to the ‘completion’of democracy and not enough on the exploration and theorization of thedifferences in democratization among countries undergoing that process.22

Alternative theories of democratization have not always arisen in responseto the transitology paradigm, but sometimes in parallel to it. The most notableis the structuralist theoretical approach, focusing on economic, social and

cultural preconditions of democracy.23 The common assumption is thatcountries at a higher level of economic development, characterized byethnic homogeneity, or certain cultural prerequisites, have a greater chanceof experiencing democratic breakthrough and sustaining a democratic politicalsystem. Yet, Fish’s reexamination of these postulates yields a less thanconvincing result, finding weak correlation between levels of economicdevelopment and ethnic homogeneity on one side, and democratization onthe other. Beyond structuralist explanations, other scholars emphasize theimportance of factors such as the extent of institutional pluralism withinmature authoritarian systems.24 Still others emphasize the importance offunctioning party systems as the key institution leading to democratic con-solidation.25 Yet another approach is the argument that uncertain balance ofpower which requires all to compromise is more likely to lead to democraticinstitutions, but in post-communist experience McFaul (2002) shows that, tothe contrary, uncertain balance has tended to undermine democratic forma-tions and allow authoritarianism to renew itself, while successful democrati-zation has been based more frequently on a historical moment of unequalpower as in Poland and Czechoslovakia.

Thus, one sees that the transitology paradigm is strongly challenged,although no one single alternative has replaced it since the challenges donot form a compelling unified paradigm with forceful empirical support.In that regard, the economic literature may be less complex with at mostthree paradigms. For countries that are clearly in the sphere of the EU (mem-bership or near-certain prospects), the paradigm is simple: any remainingdistance on the road to market economies will be automatically closed bythe legal dictates and moral suasion of the EU. The others are on the Occam’srazor of oligarch’s preferences between secure property rights and continuedhigh ‘rents’ from an unequal and non-transparent political process: theTI–TF debate has to do with the probabilities of falling on one or the otherside of the razor edge.

There is in fact a very analogous proposition about democratizationprospects for post-communist countries in Motyl (2004), who puts CentralEuropean states in the advanced democracy group, Belarus and much ofCentral Asia in the least advanced, and all the others in a middle categorywith a razor-edge possibility of going either way. He suggest that recentlythe dynamic of this middle group already shows several about to join theadvance category (Bulgaria, Croatia, Romania), with others falling in theother direction (Kyrgystan, Russia, Ukraine). He does not, however, attributethis to EU prospects, but vaguer historical conditions.

What can one draw from this analysis to understand the linkage betweenmarket liberalization and political liberalization? A good starting point is theimportant prediction of Przeworski (1991) that market reforms could only beeffectively implemented in a less than fully democratic regime. The mostrecent political science writings argue that this does not seem to be upheld.

44 Background


As McFaul (2002, p. 221), puts it: ‘Many had predicted … that reorganizationof economic institutions would undermine democratic transition … tothe contrary those countries that have moved the fastest on economic trans-formation also showed the greatest success in consolidating democraticinstitutions.’

One broad-brush indicator of this understandable to economists andothers is the striking conclusion in EBRD (2003) of a strong positive relationbetween the degree of democracy and progress in market reforms, ratherthan a negative one (see Figure 2.2 in Chapter 2). There is room for discus-sion on the direction of causality of the relation, that is whether marketreforms led to democratization, or whether the preexistence of democracyand a more civil society enabled rapid reforms, or even a circular causation.Chapter 2 devotes several pages to this relationship using updated informa-tion on democracy indicators. It may be that third factors play a role, andthat in some cases the causality is in one direction and in others the oppo-site. Balcerowicz (1993) argued forcefully that contrary to Przeworski’s fearsearly democratization facilitated economic reforms, providing a historicalopportunity to give the population its first desire: a move away from com-munism. But the Balcerowicz approach did not ignore the concerns ofPrzeworski, and precisely because there was a risk that the adjustment costsof economic reforms could turn popular opinion in a democracy againstreforms, he argued for a big-bang approach to put in place as much of themarket as possible before the honeymoon ended. He contends that inCentral Europe one witnessed a reversal of the classical sequence ‘capitalismfirst, democracy later’.

But recent Freedom House indicators of democracy showing deteriorationfor many CIS countries with moderate transition progress (CISM) countries,27

after initial improvement suggests causation is not in all cases from democ-racy to comprehensive reforms. There, slow movements to democracy andslow reforms were the norm, and rather than reinforce each other towardsfurther progress the partial nature of each has resulted in fragile achieve-ments in both and recently already a reversal on the democracy dimension.

Perhaps the lesson to draw from this is that while doubtless there is atension between democracy and market liberalization in a transition fromnon-liberal polities, to characterize it as a conflictual either–or situation iswrong. It is more of a delicate balance and set of trade-offs. Any categoricalinterpretations are mistaken, because they do not distinguish enough theends and the means.

On the side of democracy, it is a common and faulty perception thatensuring a reasonably open voting process is a sufficient means to the end ofa liberal democracy. As far back as John Stuart Mill it was recognized thatdemocratic voting and liberality were not identical, and recently Zakaria(2003) has raised the same concerns arguing that even in mature democra-cies of Europe and North America, the mechanisms of vote-maximizing are

degenerating to allow ‘capture’ of governments by political professionalswith no ideological commitments and no automatic concern for liberalism.He argues that the degree of capture is of course far greater in less-developeddemocracies, and makes the same point as Carothers (2002) that one can eas-ily be deluded into thinking that ‘minor’ voter manipulation by entrenchedpolitical elites in post-communist or emerging market polities is simply areflection of these countries not having completed their march to liberaldemocracy. Carothers, McFaul and Motyl give compelling arguments thatthese are sustainable political equilibria.

In much the same way the recent debate in transition economies has anoptimistic side which argues that imperfections in a competitive market willbe corrected endogenously. This argument in effect says that the means,private ownership, will lead to the desired end point: a competitive andopen economy. Critics of this position argue that while the powerful neweconomic interests may want to see secure property rights for themselves,this is not automatic (just as vote-maximizers do not automatically upholdliberality), and indeed in current circumstances on the ground they are morelikely to conclude that maximizing profits for the foreseeable future meansretaining, and not giving up, their influence. In a word, analogously to thepolitical science views about sustainable regimes with democratic voting butauthoritarian results, the ‘transition-frozen’ arguments say that state captureby powerful economic interests is not a way-station to a liberal market but asustainable economic equilibrium based on well-known profit-maximizingprinciples.

46 Background

2Measuring Progress in Transition

47

How to measure progress

There is a broad consensus amongst experts that Central Europe and theBaltics have progressed much further towards a market economy than havecountries of the Commonwealth of Independent States (CIS); and that of thelatter, Belarus, Turkmenistan and, arguably, Uzbekistan have not changedmuch. Such a view is also widely held by the citizens of these countries.While such broad judgments are probably correct, for analytical purposes itis surely better to have a more rigorous, objective measure of transitionprogress. Unfortunately, there can be no single measure of a country’s progressalong this path since transformation from a centrally-planned economywith state ownership to a market economy with private ownership involvesmore than changes in economic arrangements, and encompasses political,social and – as is increasingly recognized – institutional changes.

Many different metrics have been used in the literature coveringeconomic, political and social indicators. Economists tend to focus on mea-sures of positive economic results or performance such as growth, inflation,resource-reallocation measures, job-creation, productivity improvements, aswell as possible negatives or social costs such as unemployment incomeinequality, poverty ratios, mortality rates and so on. All the above indicatorsare results of the policy process and, while subject to measurement anddefinitional errors, are relatively objective metrics. There also exist indicatorsone may call ‘policy inputs’ which attempt to measure the distance a countryhas travelled along the transformation path using some estimate of policyreform undertaken. These tend to be somewhat more synthetic data com-pared to the results variables and need to be interpreted with even greatercaution, though I will show below that one such indicator serves extremelywell as a summary of progress in transition. This is the well-known set of‘Transition Progress Indicators’ (TPI) produced annually by the EuropeanBank for Reconstruction and Development (EBRD) since 1994. Similarindicators were compiled by the World Bank (1996) for an earlier period.

In addition, there now exist a wide array of institutional or business climateindices covering corruption, regulatory complexity, tax burden, security ofproperty rights and the like, put together by many institutions including theWorld Bank–IFC, the BEEPS data bank, Freedom House, Heritage Foundation,and Transparency International.

In the political science and history literature, the focus is of course onassessment of progress towards establishment of democratic institutions,often including a variety of quantitative and relatively ‘objective’ data suchas voter participation, turnover of governments, number of parties, andslightly more synthetic ones such as degree of democracy, civil liberties,media freedom and reliability of rule of law. This type of measure comesfrom the same approaches and sources as the institutional indicators men-tioned above and is again somewhat more subjective. This is because theprinciples or concepts these indicators attempt to measure cannot be easilyobserved directly and are therefore constructed by combining perceptions ofthe public and businesses, with the assessment of experts.

This book is primarily an economic analysis, therefore the central focuswill be on economic variables with two important exceptions. First, severalindicators of institutional development that affect economic agents will benoted and discussed frequently throughout the volume – for examplecorruption, state capture by vested interests, rule of law. Secondly, this chap-ter will analyse in some detail the relationship between economic indicatorsof progress and political indicators of democratization.

The EBRD transition indicator: how good a measure of progress?

The evolution of the EBRD transition indicator by country and region

It has long been a widespread practice in the economic developmentliterature to use growth of per capita income as a summary indicator ofprogress, subject only to qualifying arguments that income distribution andpoverty are not made permanently worse by growth. This tradition datesback to the view of a pioneer in the development field, Arthur Lewis, whoargued that while widespread reduction of poverty could be better defined asthe final goal of development, it was unimaginable in the long run that thiscould be achieved without substantial growth. Recently, the debate has beenrenewed (see for example World Bank, 2000) by critics who do not denyLewis’ point but suggest policy efforts should go beyond promoting growthto ensure growth results in poverty reduction within a reasonable timeperiod. A different strand of thought came even earlier, with the UnitedNations Development Programme (UNDP) effort to construct a broadermeasure of success, the Human Development Index (HDI) which combinesgrowth, distribution and several other measures of well-being.

48 Background

Measuring Progress in Transition 49

In the transition literature, starting about 1995, growth was used as aproxy for success or progress by many writers including this one.1 In thisbook, I propose to focus on the policy inputs approach of the EBRD as abetter measure of progress on the transition path. The reasoning is as fol-lows. Transforming these economies into market-based ones has the purposeof leading to better performance and improved material standards for thepopulace, and this is surely what people expect. But despite the econometricevidence showing a broad correlation between degree of reforms and growth(discussed later in this chapter), the result is not always immediate, it entailssome social costs at least in the short run, and affects popular perceptions inways that may make the relation an endogenous one, as predicted byPrzeworski (1991) and recently confirmed in Kim and Pirtilla (2004) – that is,if reforms cause too much pain, electors will vote in a new government thatpromises slow reform.

For the most part, in transition countries we are not yet observing thelong-term equilibrium path of growth as in developing countries, but onlyan adjustment to a new equilibrium path. This last point is nicely illustratedby the near-universal finding in econometric studies of transition growththat investment ratios are not significant, in sharp contrast to traditionalgrowth study results. It will be useful, therefore, to distinguish the effects oftransition on growth and distribution of income from some measure of thechange in the economic system on its presumed path to a market economy.This is in principle precisely the purpose of the EBRD Transition Indicators.

The remainder of this chapter will discuss the degree of such progressachieved by different countries over the period 1989–2003, and show thateven if one prefers economic performance indicators, human developmentsocial indicators, or democratization indicators, it turns out that the transi-tion progress indicator (TPI) is highly correlated with these. It is incidentallynegatively correlated with social costs, a finding elaborated in Chapter 3. Thislast point means that further progress towards the market did not result inlarger social costs as gradualists argued, in fact the contrary. As to countrycoverage, the sample consists of those EBRD members for whom the index isconstructed, a maximum of 27. None of the Asian countries is covered, butthe reason goes beyond non-availability of a comparable index; it is arguedhere that the Asian experience had such an incomparably different startingpoint that it deserves a separate analysis. This is a very contentious point inthe literature, with many arguing that the more gradual and less disruptivetransition of China should have been followed in the European region as well.2

Turning now to the EBRD indicator, we see in Figure 2.1 a distribution ofcountries on a scale of 1.0 (no market reform) to 4.3 (full market reform) in2003. The value for each country is a simple average of eight separateindicators: large-scale privatization, small-scale privatization, enterprisegovernance, price liberalization, trade liberalization, banking liberalizationand other financial sector reform.3 The EBRD often cautions against such

averaging, but then frequently does exactly that for comparative analysis; asthere is no case for weighted averaging, simple averages are not unreason-able. The picture suggests four or five groups of countries: Central Europe(CE), the Baltics (I will on occasion refer to the two as a group, CEB); South-East Europe (SEE); CIS countries with moderate transition progress (CISM);and CIS countries with limited progress (CISL).4

The nine CEB countries are the most advanced in transition to marketeconomies, with values in the range 3.4–3.9. Though the Baltics are wellwithin the group, Estonia (3.8) being second only to Hungary (3.9), andLatvia (3.6) higher than two Central European countries, it will for manypurposes be useful to keep the Baltics separate given their unique experienceof having been part of the USSR. Note that in the CEB group, all but Croatiahave acceded to EU membership as of 1 May 2004. The next group, SEE,comprises six countries ranging in value from 2.5 to 3.4, clearly below theCEB group. The CISM group of nine countries ranges from 2.4 to 3.1,somewhat overlapping the preceding group, raising the question why someCIS are not associated with the higher group. Geographical convenience is aminor reason; a second is that the SEE laggards, Bosnia & Herzegovina,Serbia-Montenegro, have had a very short period since civil conflicts quietedand seem headed upwards. But the main reason is that most of the SEE coun-tries are clearly on track for a second or third wave of EU accession, while as

50 Background

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5H

UN

CZ

EC

HP

OL

SV

KC

RO

SV

N

ES

TLV

ALI

TH

BU

LR

OM

MA

CA

LB BiH

S&

M

KY

RA

RM

GE

OR

US

KA

ZU

KR

MO

LA

ZE

TA

J

UZ

BB

EL

TU

RK

Countries by group

TP

I val

ue

Figure 2.1 EBRD transition progress indicator (TPI), 2004


of the end of 2004 none of those in the CIS were.5 Finally, three countriescomprise the group least advanced in transition, CISL, with values 1.3 forTurkmenistan, 1.9 for Belarus and 2.2 for Uzbekistan.

It is useful to compare these results with some other potential measures ofpolicy progress. Consider first the ‘Competitiveness Index’ constructedannually for a large number of economies around the world by the WorldEconomic Forum in Switzerland. Zinnes, Eilat and Sachs (2001) applied sucha methodology comprising business and expert perceptions and assessmentsin a synthetic index to a similar group of 25 transition countries for 1998.The CE group, in order of country as in Figure 2.1, had rankings of: 1, 2, 3, 5,8, 9; the Baltics, 4, 6, 7; virtually a precise correspondence. For the SEEcountries, their rankings were 10, 12, 20 (Macedonia) and 22 (Albania); andfor CISM 11, 13, 14, 15, 16, 17, 18, 19 (Azerbaijan), 24 (Tajikistan). The coun-try names in brackets show placement slightly out of the Figure 2.1 order; forexample Macedonia and Albania are lower-ranked than Azerbaijan, butoverall the groupings and the overlap between SEE and CISM are broadlymaintained. Finally, their rankings put the CISL countries at the end (21, 23,24) save for Tajikistan which since 1998 has moved forward while the othershave even slipped. In summary, this very different approach to rankingeconomic competitiveness of countries gives virtually the same results.

The role of institutional development as discussed in Chapter 1, is consid-ered by all analysts to be as important in the long run as liberalizationpolicies, indeed some would assign it an even greater weight. It is thereforeessential to verify that the pattern of TPI discussed above is consistent withthe pattern for institutional indicators. This is in fact the case as a simplecomparison of Table 1.1 and Figure 2.1 shows: the ranking of our countrygroupings is identical. This holds for individual country rankings as well; thecorrelation coefficient between the TPI and the Quality of Institutions is�0.896. In a later discussion of time trends for TPI, it is also seen clearly thatEBRD’s own measure of institutions (they label this as ‘second phase’reforms) results in the same country-group ordering as Figure 2.1.

The Appendix to this chapter gives a more detailed picture with TPI valuesfor each country in a rank order, as well as time paths for individual countries,but the five groups noted here will be the basis for comparative analysisthroughout most of this volume. On the one hand this has the disadvantageof missing out on some of the variations within the group, but where impor-tant this will be noted selectively, using as reference the Appendix countryvalues for the same variables. On the other hand, an obvious advantage ofsuch a grouping is the ability to analyse the main tendencies in a simplerframework; given the inevitable margin of argument about a quasi-subjectiveindicator, it is all the more sensible to group broadly similar countriestogether. The first such comparative analysis by group is to consider thecorrelation between the EBRD indicator of policy change and some commonmeasures of economic performance results.

Economic performance indicators

As noted, it is useful to think of the TPI index as a policy input, withresults being measured by standard economic performance variables. Threemeasures of performance are considered: recovery of GDP by 2003; inflationcontrol; and foreign direct investment (FDI) per capita. There are many othervariables one could use in addition (or instead), but this choice is probablymost reflective of the countries’ key goals in the first decade of transition.There is a wide consensus that the high inflation of the early 1990s in virtu-ally all countries had to be controlled quickly and financial stability achievedbefore other reforms could have a favourable impact on growth. Economicgrowth is not only a measure widely used to assess performance, there is alsolittle doubt that the expectations of the populace were primarily focused ona better living standard even more than on democratic freedoms alone.6 Asto FDI, it is not only an important contributor to new growth potential,especially in economies that had been quite closed, but more importantly itis likely to reflect the assessment of foreigners about the climate for invest-ment in general and future growth prospects. It is a better indicator thantotal capital inflows, because the latter includes sovereign borrowing, whichis more short-term and can be due to instability rather than stability. Arguably,total investment including domestic investment might be an even betterindicator, but the problems of informal activity and the embryonic nature offinancial intermediation make this variable problematic for now.7

A few qualifiers are in order concerning the growth or recovery variable inparticular. First, it uses official GDP estimates (EBRD reports are the source forall three), and excludes informal economic activities. The Appendix addressesthis problem in some detail reaching the conclusion that the latest and mostcomprehensive estimates seem to imply either that the biases are minor, ortheir direction does not affect the overall comparison.8 The reasoning is as fol-lows: for many countries the share of informal activity is either small or con-stant over time, hence bias is very small; for most others it is a clear trendwith a rising share early, and then a decline later, as there are no cases of sig-nificant volatility in this share the growth rate estimates are not biased. It isalso noteworthy that the share of informal activity for transition countriesmay be slightly lower than for developing countries (38% v. 41% as given inSchneider and Klinglmair, 2004). Interestingly, the sharp criticisms of transi-tion growth econometrics being biased due to exclusion of activity do notseem to be echoed in the area of developing-country growth empirics.

A second problem with official growth data is the non-comparability ofSoviet accounting with the UN standards used in all transition economies.Starting with Lipton and Sachs (1994), and more thoroughly in Aslund(2001), many have argued that the output fall was exaggerated. Aslundmakes three key points: the transition started in 1989 for Central Europe, butonly about 1991–92 farther east; central planning motivated managers tooverstate output; and overpricing for unsaleable unwanted goods meant

52 Background


real output value was much lower, with this effect being strongest in the sec-tors that saw the greatest output cut-backs in the 1990s. Note that this doesnot include any valuation for time in queues and lack of choice of goods, buteven so the adjustments proposed by Aslund are enormous, for example theofficial output decline in the Baltics virtually disappears. In the data forrecovery of Table 2.1, the arbitrary adjustments account for only 50 per centof Aslund’s adjustments. A fuller analysis of such adjustments is undertakenin Chapter 3 which presents various estimates of lost output during thetransitional recession.

With these caveats in mind, consider first the evidence of official GDPrecovery in Table 2.1. CE is clearly above the others, with the Balticsdistinctly lower, about equal to the SEE group. The CISM group is lower still,however the CISL value is second only to CE. The correlation betweentransition progress and recovery appears weak.

With adjustment for Soviet accounting problems, the Baltics fall intoplace, though CISL remain distinctly better performers than CISM or SEEcountries. One possible explanation is that where there is no pain there willbe no gain; that is, with so little reform even the minimal output fall that isnecessary for efficient restructuring has not yet occurred. These countrieshave continued to support output of loose monetary policy. The unsustain-ability of this is comparable to what happened in the mid-1990s in Bulgariaand Romania (Havrylyshyn et al., 1998); by 1994 both saw some stabilization,and modest growth, but this was followed by a crash of output and a resurgenceof inflation in 1996.

It is notable that inflation levels in CISL (Table 2.2) are much higher.Another explanation is that in all three cases the official numbers exaggerategrowth significantly. For Belarus, growth has been driven by exports to

Table 2.1 Recovery: estimated index of GDP, 2003(1989 � 100 with partial adjustment for Soviet accounting)

Unadjusted 2003 value Adjustment

Central Europe 115 139Baltics 90 112South-East Europe 82 (96)** 86CISM 67 84CISL 104 n.a.

Adjustment: adjusted by 50% of Aslund (2001) estimates as shown inhis table 3.1, except CISL, explained in the text.

** The first number is for all six countries; the brackets value excludesBosnia-Herzegovina and Serbia, whose crises and hence transition startcame much later.

Source: Unadjusted value: EBRD Transition Report, 2004.

Russia in barter exchange for energy at terms favourable to Belarus (see IMF,1999); the decline of growth in recent years as barter was reduced is consistentwith this hypothesis. In Uzbekistan growth has also fallen sharply, while inTurkmenistan exports of natural gas drive growth, but the export figures areon an accrual basis, while actual payments for gas deliveries are far belowcommitments, so the real value of output is much lower. Taking into accountthese arguments recovery in CISL may be much lower and more in line withtheir TPI ranking. The inflation and FDI values show a much closer andcleaner correlation with transition progress. In Table 2.2, one sees medianinflation at very low levels for CEB with the Baltics even lower, then SEE,CISM and CISL follow in order. Finally, Table 2.3 shows that FDI per capita isby far the highest in CE, only slightly lower in the Baltics, followed farbehind by SEE, CISM and CISL. If one excludes three energy exporters:Azerbaijan, Kazakhstan and Russia, the CIS value is far lower at 136.

It is not in the end surprising that the index of transition or reform progressis positively correlated with output recovery, as many early econometric stud-ies have found that the degree of reform progress is one of the most impor-tant explanatory factors for growth in transition. But it is important to verifythat whether one measures progress as the policy input on economic reforms,or as resulting economic performance, the ordering of countries is broadlythe same and comprises the above five very distinct groups.

54 Background

Table 2.2 Inflation performance (CPI increase in 2003, per cent)

Range mid-point Low case High case

Central Europe 2.5 Czech Rep. (0.2) Hungary (4.7)Baltics 0.9 Lithuania (�12) Latvia (3.0)South-East Europe 8.8 Macedonia (1.1) Romania (15.4)CISM 9.7 Kyrgyz (3.1) Tajikistan (16.3)CISL 13.3 Turkmenistan (6.5) Belarus (28.5)

Source: EBRD Transition Report, 2004, table A2.3.

Table 2.3 Cumulative FDI per capita, 1989–2003 ($US; by subregion)

Group average Low countries High countries

Central Europe 2,305 Poland (1,355) Czech Rep. (3,710)Baltics 1,641 Lithuania (1,070) Estonia (2,400)Southeastern Europe 465 BiH(282) Bulgaria (795)CISM 339(136)* Russia (31) Kazakhstan (1.094)CISL 168 Uzbekistan (35) Turkmenistan (269)

* Value in brackets excludes the energy exporters: Azerbaijan, Kazakhstan, Russia.

Source: EBRD Transition Report, 2004, table A2.8.


Progress in market reforms and democratization

In as much as democratization is also an important goal of the transition,and for the EBRD in particular – unlike the other International FinancialOrganizations – promoting democracy is explicitly part of its mandate, someattention to the relationship of economic progress and political progress isrequired. The EBRD itself has addressed this, first in the Transition Report2000, and then more thoroughly in the 2003 volume. This section discussesthe findings of the EBRD as well as those of some independent analyses byacademics and other institutions. The central hypothesis here is whether therelation between economic liberalization and democratization is a positiveone. For transition countries, Przeworski (1991) argued strongly that marketliberalization could only be successfully applied by suspending democracytemporarily. The EBRD analysis purports to show this has not been the case.

Before turning to the evidence, consider some problems with defining‘democratization’. This is a big and difficult question and a permanent intel-lectual debate; here, I will draw on the summary of the political scientists’debate on transitology provided in the previous chapter. Zakaria (2003)which some consider a popular treatise but is often cited in the academicliterature, points to the key issues of relevance for this volume. Holding elec-tions is not a sufficient indicator of democracy because electoral democracycan exist without secure or complete civil liberty; over time electoral rightscan increase without civil liberties improving (a concern for the assessmentof ‘democratizing’ societies); or electoral rights are not reduced even as civilliberties deteriorate (Zakaria suggests this may be happening even in democ-ratic societies such as the United States).

To its credit, the EBRD analysis addresses these definitional difficultiesvery directly, and begins by showing there is no statistical correlationbetween the transition indicator and frequency of elections. The point isthat ‘elections alone are not enough’ (EBRD, 2003, p. 23) because withoutcivil liberties comprising freedom of expression, openness and credible ruleof law, the elections can be biased and influenced even without explicit ille-gal actions such as ballot-stuffing. The recent drop in Freedom House ratingsfor democratic freedoms in many CIS countries9 exemplifies the problem.

EBRD (2003) then compiles a broader, albeit more subjective, measure ofconstitutional liberalism for 1996 and finds a strong correlation with theaverage TPI for the period 1997–2003. An update of their compilation andthe consequent relationship with TPI is shown in Figure 2.2 for 2004. Thecorrelation may be even stronger because since 1996 several lagging reform-ers fell in the constitutional liberalism rankings (Belarus, Moldova, Ukraine),while some advanced reformers have risen (Croatia, Slovakia). For 2004 thevalue of the regression coefficient R2 � �0.723.

This overall measure of democracy might be questioned as it includes atleast one component that is not in principle inconsistent with the enlight-ened autocracy Przeworski thought necessary for economic reforms: rule of

56

Figure 2.2 Constitutional liberalism and progress in transition

YUGYUG

MDAMDABIHBIH

TKMTKM

BLRBLR

TJKTJK AZEAZE

KAZKAZ

UKRUKRRUSRUS

ALBALB

GEOGEOARMARM

KGZKGZMKDMKD

ROMROMBGR, HRVBGR, HRV

LTV, LTULTV, LTU

SVNSVN

CZECZE

ESTESTHUNHUN

POL, SVKPOL, SVK

y = 0.1988x − 0.1302

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.01.0 1.5 2.0 2.5 3.0 3.5 4.0

Transition progress indicator, 2004

Con

stitu

tiona

l lib

eral

ism

, 200

4

Country nameAlbania

ArmeniaAzerbaijanBelarusBosnia & Herzegovina

BulgariaCroatia

Czech REstonia

GeorgiaHungaryKazakhstanKyrgyzstan

LatviaLithuania

MacedoniaMoldovaPoland

RomaniaRussiaSerbia/M

SlovakiaSloveniaTajikistanUkraineUzbekistan

ALB

ARMAZEBLRBIH

BGRHRV

CZEEST

GEOHUNKAZKGZ

LAVLTU

MKDMDAPOL

ROMRUSYUG

SVKSVNTJKUKRUZB

Country code

R2 = 0.7231

UZBUZB


law. If one takes just ‘the media freedom’ component as perhaps a betterdirect measure of civil liberty, the average for each of the five groups desig-nated above clearly follows the rank order of TPI. The results are on a scale of0–100: CE � 68; Baltics � 71; SEE � 49; CISM � 39; CISL � 20,10 confirmingthe proposition of this chapter that the TPI is not only a good proxy forprogress on the economic dimension, but also for the political-democraticdimension.

This is also consistent with the recent conclusions of several politicalscientists noted in Chapter 1 that contrary to Przeworski’s concerns that eco-nomic liberalization would undermine democracy, democracy has in factprogressed farthest in countries that have advanced fastest on market liber-alization. Let us consider why his prediction may not have materialized.There is a vast literature by political scientists and historians interested in thetransformation, and I give here only a selective illustration of the differingviews expressed. At one end of the spectrum, Reddaway and Glinski (2001)writing on Russia argue forcefully that because economic reforms were intro-duced in a ‘shock-therapy’ fashion, the embryonic democratization in theearly years was shunted aside to allow rapid reform; this is very much in linewith the Przeworski prediction. However, this says nothing of other countries,in particular Poland, the Czech Republic, the Baltics and other transitionleaders, which also undertook rapid big-bang11 reforms and did very well oneconomic performance as well as democratization. Nor does it adequatelyaddress the common criticism that it was the incompleteness of reformsrather than their introduction that led to the failures. But Reddaway andGlinski do make a compelling case about the consequences of insider priva-tization for both economic liberalization and democratic liberalism; thistouches on a central hypothesis of this volume elaborated in Part II.

Other political analysts, however, come to a very different view withMcFaul (2002, p. 221) perhaps at the opposite end of the spectrum contend-ing that, contrary to predictions of inevitable discordance between marketreforms and democracy, ‘those countries that moved the fastest on eco-nomic transformation have also achieved the greatest success in consolidat-ing democratic institutions’. One of the earliest precursors of this view mayhave been Brzezinski (1993) who spoke of ‘the primacy of political reform asthe basis for effective economic reform’, and the expectation that the twowould in fact go hand in hand. He also foresaw precisely the wide dispersionof outcomes that one sees in both market and democratic reforms. A similarsentiment is found in Carothers (2002) and Motyl (2004), who criticizestrongly the general ‘transitology’ paradigm that countries will keep movingforward towards full democracy once a ‘breakthrough’ from autocracyoccurs. They show instead that despite early democratization, some coun-tries have clearly fallen back into a solid autocratic path, some are in anuncertain middle range or ‘gray zone’, and only about a dozen are clearlywell on the way to meaningful liberal democracies. Combining the lists of

these authors, we find in the first group the CISM countries plus some inCISL, in the second most CISM and SEE countries, and in the third mostlyCEB countries plus tentatively some from SEE like Bulgaria or Romania.

That a strong correlation between economic and political liberalizationexists after such a short period of transition seems clear, but, as Bunce (1999)emphasizes, the deeper relationship and the direction of causation is forpolitical analysts the most interesting issue. On the one hand, one mayargue that early and successful reforms bring early results, and make it easierto advance on the democratic front (that is, economic reform causes democ-racy). This is broadly in the spirit of an older view amongst political scien-tists that higher income makes democracy more likely (Lipset, 1959). On theother hand, there is evidence of causation running in the other direction.Kitschelt (1995), in a multivariate analysis, shows that prior history ofdemocratization (the interwar period for our CEB group) is an importantdeterminant of democracy, and Shleifer (1997) argues that civil liberties ofa democracy translate into both a demand for and a supply of economic lib-eralism, especially in the dimensions of property rights security and rule oflaw. But both of these views are broadly consistent with the reality of CentralEurope at one end of the spectrum, and most of the CIS at the other end. Inaddition, as EBRD (2003) notes, ‘there can be little doubt that the prospect ofEU membership has been a major spur for structural reform in the accessioncountries’. Does this mean that without the prospect of EU membershipthe advanced reformers may not have advanced as far, or have advanced onthe economic but not democratic dimension? This issue plays a critical rolein the main thesis of the present volume and will be addressed morethoroughly in Part II, especially Chapter 7.

To conclude this section, it suffices to say that the EBRD TransitionIndicator, despite its shortcomings as a somewhat synthetic index and notsubject to a rational weighting scheme for its different components, turnsout to be a very good proxy measure of progress on the path to a liberal mar-ket economy, as well as progress towards a liberal democracy and effectivecivil institutions. The next chapter will analyse the empirical evidence onthe economic and social costs generated by the transition, in effect testing akey hypothesis of the gradualist school that social costs are greater under big-bang reforms. The antecedent of such a hypothesis is the Kuznets U-curvehypothesis in the development literature, that economic growth at first leadsto worsening inequality (see for example the WIDER study by Wan, 2002).But first it may be of interest to look at the evolution of TPI overtime for thefive country groups.

Patterns of reform progress over time

It is not surprising to find that the pace of market reforms was not a steadyone over time nor across regions; indeed, it also differed by type of reform.

58 Background


A broad overview is seen in Table 2.4 showing for all the 27 transitioncountries the level of progress for what the EBRD defines as ‘initial phase’reforms, and ‘second-phase’ reforms sometimes described in the literature assecond-generation reforms. The scaling is the same as for TPI in Figure 2.1.The first phase includes price liberalization, foreign trade liberalization andsmall-scale privatization, and might be more usefully labelled as ‘liberaliza-tion’, while the second phase comprises the EBRD’s remaining indicators:this last can be thought of as ‘institutional’ reforms, and indeed in its analy-sis the EBRD discusses them in that way. EBRD notes for the average of allcountries an early surge of liberalizing policy changes from 1989 to about1994, a somewhat slower but still steady pace to 1996, then a virtual flatten-ing at a level of about 3.2 to 3.3. This is still well-below the top value of 4.3representing a fully functioning market economy, but importantly the rea-son for this flattening turns out to be the difference between the advancedcountries and the slower reformers as seen in Table 2.4.

Table 2.4 breaks this down by the five country groups. It is evident that inall regions institutional reforms lagged behind liberalization and in generalwith less of a surge as seen for liberalization in the early 1990s. This patternis consistent with the common view that institutional reforms cannot beimplemented so quickly, but it may also be consistent with the criticismaimed at the Washington Consensus that they were not given enoughimportance early on. This debate cannot be easily resolved, as there is nobenchmark to define the phrase of Vaclav Klaus ‘as fast as possible’. The evi-dence is therefore easily used by both schools of thought in support of their

Table 2.4 Transition indicator values by year and type

1994 1999 2003

Central Europe Liberalization (LIB) 3.7 4.2 4.27Institutions (INST) 2.7 3.1 3.3

Baltics LIB 3.7 4.1 4.27INST 2.3 2.9 3.3

S.E. Europe LIB 3.0 3.9 4.1INST 1.7 2.2 3.0

CISM LIB 2.2 3.7 3.9INST 1.4 2.1 2.3

CISL LIB 1.9 2.0 2.2INST 1.4 1.6 1.5

Source: Averages calculated from EBRD Transition Report, 2000 and 2003,country tables. Liberalization is the average of the following indicators: priceliberalization, foreign exchange and trade liberalization and small scaleprivatization. Institutional reforms comprise: largescale priviatization,governance, competition policy, infrastructural policy reforms, financialsector reforms.

arguments. However, much more can be inferred by comparing the timepatterns across the five country groups.

First consider what that comparison shows in Table 2.4, with the firstphase ‘liberalization indicators’ averaged under the label LIB, and the second-phase ‘institutional indicators’ averaged under INST. By 1994 the CentralEuropean and Baltic countries had reached the very high LIB level of 3.7,broadly comparable to many mixed market economies with some state-ownership, and price regulation of a few basic goods (housing, some staples).The value of TPI � 4.0, for comparison, is described for the three liberalizationdimensions by the EBRD as follows:

● Comprehensive price liberalization except for a small number ofadministered prices.

● Removal of all quantitative and administrative restrictions on importsand exports.

● Complete privatization of small companies with tradable ownershiprights.

By 1999 this had increased substantially, and by 2003 had effectivelyreached the maximum rating of the EBRD index. Six of the CEB countrieswere in fact rated at 4.3, and three (Croatia, Estonia, Slovenia) at 4.2, hencethe mechanical result of a 4.27 average in these groups. For most observersthat should be a distinction without a difference: on basic liberalization ofeconomic activity, all nine countries must now be considered as fully func-tioning market economies.12 It is particularly notable that the Baltics, start-ing later than the others, already caught up to them by 1994 and kept pacein the final liberalization drive.

South-East Europe lagged somewhat behind, though not nearly as muchas the CIS group which in 1994 was far below the level of SEE (no 1994data available for conflicted Bosnia-Herzegovina and Serbia-Montenegro).Indeed, at the mid-1990s point, the gap between the moderate CIS reformersand those with limited reforms is not visible. That quickly changes, however,and by 1999 the CISM group’s progress relative to the CISL is evident, as itsurges ahead in liberalizing measures to achieve about the same position thatthe CEB countries had five years earlier. I argue below that it is not just acoincidence that the CISM output recovery began after reaching about thesame level of TPI as had been reached by the Central European and Balticcountries when their recovery began. What is striking is that by 2003,though the CISM countries had not yet caught up to the liberalization levelsof Central Europe, they, and even more so SEE, were very close to the 4.0mark for the liberalization indicators, reflecting reasonable well-functioningmarket mechanisms if not institutions. The nearly stagnant process of liber-alization for CISL countries is particularly evident in the Table 2.4, and rein-forces the distinction drawn in Figure 2.1. Had the latter been compiled using

60 Background


only the LIB indicators, the gap between these two groups would be fargreater. The values for CISM range from 3.6 (Tajikistan) to 4.2 (Georgia, Kyrgyz),while for CIS they are 1.7 (Turkmenistan), 2.4 (Belarus) and 2.5 (Uzbekistan)(see Appendix figure for LIB). Indeed, even the case of Tajikistan is no longermarginal: its LIB value in 2003 was 3.6, which is well-above the 2.5 ofUzbekistan; while in 1994 Tajikistan was second last with only Turkmenistanlower (1.6 and 1.2 respectively); it has progressed steadily while the threelagging cases of CISL did not.

Consider another aspect of institutional change. Table 2.4 reinforces theEBRD (2003) story that in all countries institutional reforms lagged consid-erably behind liberalizing measures and still do so even in the more advancedcases. While in Central Europe the index is a respectable 3.4 – with the high-est value in Hungary at 3.7 – this is much less than the average of 3.7 forliberalization in Central Europe nearly 10 years earlier. The descriptors usedby the EBRD for the value of 4.0 in the institutional indicators are:

● Competition policy: significant enforcement actions to reduce abuse ofpower and to promote a competitive environment.

● Banking reform: significant movement of banking laws towards BISstandards.

● Securities markets: laws and regulations approaching IOSCO standards.

Clearly, the attainment of institutional development is still far from the abovedescriptions; hence, unlike with market liberalization; one cannot even con-sider these countries as being fully comparable to other market economies.

It is notable that the Baltics are at the same level and South-East Europe isnot however that far behind, certainly much less so than the CISM wherethe best that can be said is that they have been moving forward measurablysince 1994, unlike the CISL.

Finally, what do the values of Table 2.4 imply for the debates about therelative importance and sequencing of liberalization and institutions? First,note that the degree of institutional reform reached before growth startedwas not that high in Central Europe, the Baltics, or for that matter in theCISM countries. Indeed, even by 2003 the level of development of institu-tions still had a long way to go in the advanced countries, yet no major dam-age to the performance of the economies appears to have occurred. True,growth slowed a lot in Central Europe, but not in the Baltics – the reasons arediscussed below.

Johnson and Subramanian (2005) propose that, generally, good economicpolicy alone can give growth a start, but sustained growth requires improvedinstitutions. The case of Central Europe appears to fit this well. In the firstrecovery phase 1993–98, GDP grew after the surge of liberalization but thenslowed as institutional reform lagged. As these picked up slowly, strongergrowth returned after 2000.

Second, there is not a single instance of slow liberalizers moving aheadmore rapidly (or even at the same pace) with institutional reforms as a truegradualist strategy implies. To the contrary, those that liberalized fastest alsomoved fastest on institutional reforms, albeit with a lag.13

The EBRD makes the point that partial liberalization creates ‘winners …who block further progress in reforms’ (EBRD, Transition Report, 2000, p. 30).Perhaps, the most important inference is that the slow reformers were notslow reformers for the reasons often given by internal proponents of gradualliberalization – ‘the economy is not ready for market operations’ – else theywould have speeded up institutional reform to prepare the economy forsubsequent market liberalization; significantly, nowhere is this pattern visible.This may help one understand the main problem with gradualist arguments.Recall they were motivated by the notion that liberalizing too fast ahead ofinstitutional developments would be less effective and create more disloca-tion and pain. Many countries did move slowly but none did what theoreti-cal gradualists recommend, moving faster on the institutional side. Why?Self-interest of the political leadership. Part II goes into this more deeply.

Main determinants of growth in transition14

Many studies of growth and recovery in the transition had already beendone by the late 1990s, less than a full decade after the process began. Therewere already over a dozen econometric cross-country studies covering mostof these countries, surveyed by Havrylyshyn (2001) and Campos andCoricelli (2002). While these studies fell short of definitive answers on all ofthe key debates of the 1990s, they showed a surprising degree of consensus:the standard factor input variables are not important; prior financial stabi-lization is virtually a sine qua non; liberalization and structural reforms arekey explanatory variables; unfavourable initial conditions can reduce growthprospects but this effect declines as time passes; and good institutions domatter but complement rather than substitute for liberalizing policies in thesense that they need not precede liberalization, but must soon begin to catchup. Consider each of these conclusions.

In the 1990s there was a renewed interest by economists in explaininggrowth by going beyond the role of factor inputs – land, natural resources,labour, physical and human capital – which were central to earlier Solow-type models. Factor inputs continue to play a large role, but other explana-tory variables have been added as exemplified by the work of Barro andSal-i-Martin (1995). But transition economies’ growth in the 1990s is notanalogous to the long-term equilibrium growth path that is usually mod-elled in growth studies. As Havrylyshyn et al. (1998) noted, the dynamics inthis period are not a matter of moving the economy to a higher production-possibility frontier (PPF) through expansion of factor inputs or even techno-logical change. Rather, it is a matter of correcting the inefficiencies of the

62 Background


communist period including moving from inside the PPF to the PPF, andshifting allocation along the PPF to an international comparative advantageposition. It is in a word the application to the production function of theKornai–Blanchard (KB) paradigm changes noted in Chapter 1. Therefore it isnot surprising that all efforts to include capital, usually proxied by theinvestment/GDP ratio, show insignificant and often negative results.

A slight clarification may be needed. The above results do not imply thatinvestment is not needed in the transition process of reallocating resources.To the contrary, at the micro-level of enterprises a lot of new (but oftensmall) investment is taking place. But, in the aggregate, the amount of newinvestments in early phases of the transition may not – and need not –exceed replacement levels for the preexisting capital stock. Indeed, Camposand Coricelli (2002) list as one of seven stylized facts of growth in thetransition that in the aggregate ‘capital shrank’, if old industries are ineffi-cient, a shift from them to more efficient ones or – as was also often the case –a shift within firms to more profitable product lines, can take place in anenvironment of negative or low net investment, as long as small gross newinvestments are going into more efficient production.

The primacy of financial stabilization as a prerequisite for growth recoveryis not a surprising result, nor indeed was it a controversial issue; as noted inChapter 1, most critics of rapid reform or the Washington Consensus agreedon the need for stabilization. A couple of aspects of this do deserve attention,however. Some observers argued for the use of exchange rate anchors as thecentre piece of any stabilization strategy. The econometric evidence does notgive a clear-cut answer on their effectiveness, because in fact several categoriesof cases emerged historically. Some did indeed achieve successful stabiliza-tion while using an anchor (currency-board countries like Estonia, Lithuaniaand, effectively too, Latvia, then later Bulgaria), but a large number ofCentral European and later most CIS and SEE countries achieved stabiliza-tion without this anchor, though some of these had an approximation in theform of crawling/adjustable pegs (Poland), and some maintained a de factoproximity to a peg (Croatia). Russia is still debated (as on everything else!!),but it arguably had a peg of sorts until 1998, with demonstrably limitedsuccess in stabilization.

Another unresolved detail in the econometric literature is whether budgettightness or inflation control are the determinant variables, or both. Theattempts to sort this out are mixed: there is a consensus that lower inflationstimulates growth, but separate effects of inflation v. budget deficits are noteasily established. This may be due to two factors. First, almost all these modelsare ad hoc and not derived from structural equations, including for examplesimultaneous determination of inflation and growth. In cases where infla-tion is separately determined, deficits do show positive and significant cor-relation with both inflation and growth. Second, fiscal deficits may havebeen too narrowly measured, excluding off-budget transfers, central-bank

directed lending and so on. Since stabilization cannot be narrowly defined,a typically good proxy for the entire strategy may indeed be inflationreduction; hence the results one observes: inflation control is highly significantin growth regressions.

Liberalization of markets and related structural reforms also show up as one ofthe main determinants of growth during the transition, though this is truefor the aggregated synthetic measures, such as the EBRD transition index,less so for individual components. Thus, price liberalization alone is signifi-cant in only a few studies; privatization also comes out insignificant in mostbut occasionally significant in a few specifications. This suggests it is thecombined effect of several policies that matters in creating new opportuni-ties for private-sector activity, not a surprising or controversial result. It maybe more controversial and important if one brings into the picture marketinstitutions. Thus, as noted already in Chapter 1, quantitative analysis of theeffects of privatization have come to a clear consensus, that transfer of own-ership alone may at best have some small positive effects, but significantbenefits come only with the complementary development of competitivemarket institutions. What this means precisely is not easy to define, becauseinvariably all of the studies use some broad synthetic index of institutions:competitive environment, security of property rights, rule of law, govern-ment corruption. But it does strongly confirm the view that some mini-mum degree of institutional development is needed alongside private-sectordevelopment.15

It is important to note that neither market liberalization, privatization norinstitutions alone have an overwhelming explanatory power, but rather allof them matter as they act in a complementary fashion. This last economet-ric result may teach a humble lesson both to big-bang reformers and gradu-alists. Rapid-reform advocates have by now understood it was not enough torecognize conceptually the role of institutions – the fact that they developedmuch more slowly in some countries than others may reflect insufficientweight given to them in policy recommendations. For both gradualists andinstitutionalists, this result says it was indeed necessary to move on severalfronts at once, and there surely would have been little mileage in pushingfirst for institutional development while delaying the liberalization andprivatization elements.

Only a handful of econometric studies of growth have grappled with thedebate on gradualism v. big-bang, but have not been as conclusive as simplerqualitative analysis, because each has defined speed in a different way.Heybey and Murrell (1999) find speed does not matter, rather the cumulatedlevel of reforms does; Berg et al. (1999) define it as early attainment of acumulative level and find it matters, in the same way the simpler analysis inthe early part of this chapter shows that early reformers have performedmuch better. That a cumulated level of reforms eventually also brings growthmay be right, and the evidence of the CISM growth surge discussed below is

64 Background


consistent with such an interpretation. But that does not gainsay the interimperiod loss of output that slower reformers suffer.

Concerning the importance of initial conditions relative to reform policiessuch as stabilization, liberalization and institutional development, there isno clear-cut result in the econometric studies or in qualitative analyses. Inthe econometrics, these have been measured variously as degree of overindus-trialization, share of defence industry, years under communism (a proxyfor market memory or ‘mental’ distance from capitalism), distance fromEuropean markets, war or civil conflict, and so on. Because the possible num-ber of measures of initial conditions is so large, the results vary according tochoice of variable, choice of period, econometric specification. One of thestrongest findings showing a strong role of initial conditions is de Melo et al.(1997). Later, Havrylyshyn et al. (1998), using the same measures withadditional years of data, point out that even if this was true in early years,the statistical significance of initial conditions declines over time. In thesame spirit, Zinnes et al. (2001) distinguish immutable conditions (geogra-phy, history) from changeable ones (degree of industrialization, share ofdefence), and also find the latter matter little after a short period of time.

Perhaps the strongest argument against the relevance of initial conditionshas not been tested in the literature, namely that some of them may haveeither negative or positive effects on growth. Thus, for example, the highshare of defence industry in some countries (high in Ukraine and Russia,very low in the Baltic Republics; high in Slovakia, lower in Czech Republic)can be both a drag on the reallocation to new industries, but also, given thatit contained the highest level of human capital and technology, an opportu-nity for generating a lot of new growth under the proper incentives. This isanalogous to the common arguments about natural resources, which inprinciple under good policies should be a benefit to the country, but in prac-tice because it may lead to complacency and bad policies often turns out his-torically to be a negative influence on growth. That defence industries wereoften strong lobbies for slow adjustment is hardly debatable.

It has become a nearly unanimous view that institutions are importantfor sustained growth, though as an excellent review by Johnson andSubramanian (2005) warns, this sometimes verges on faddism, with a largeunanswered question: is there any way one can effectively promote goodinstitutions? The works covered by their review as well as many earlier onesaddress the role of institutions in non-transition economies, starting withthe pioneering contributions of North (1993; 1995) and ending with themost recent revival of his ideas for developing countries. Here I note only afew points about institutions and growth most pertinent to transition. Onlya handful of the writings analysing growth in transition include institutionalquality as a variable. Moers (1999), Havrylyshyn and van Roeden (2003)conclude that institutions do contribute significantly to growth in transi-tion, but especially in the later phase of sustained growth, while with

liberalization, stabilization and initial conditions are more important in theearly recovery. Beck and Thorsten (2005) show econometric results thatattribute almost all the explanatory power to institutional quality alone, asuspicious result for the short term, though consistent with the ‘deep expla-nation’ school of thought, as Johnson and Subramanian (2005) labelled it,that is that, in the long run, since good institutions lead to good policies,they alone fully explain growth.

Becks and Thorsten (2005) also attempt to model factors behind institu-tional development, and find that greater inertia of socialist ideology plusreliance on natural resource exports are statistically significant. Simply put,they argue these two factors lead to strong rent-seeking behaviour and evo-lution of vested interests which oppose institutional development. Chapter 5elaborates on the role of ideology or commitment, and is generally support-ive of their view. But the importance of natural resource exports is lessconvincing; a striking counter-example is Ukraine, an energy-dependentcountry where oligarch evolution was second to none. Chapter 6 tells themore complete story on rent-seeking in post-communist transition.

Such endogeneity amongst initial conditions, policies, institutions andgrowth is the strongest argument for the importance of initial conditions,albeit a fragile one. It cannot be disputed that policy choices are not made inan abstract text-book vacuum, but must be influenced by the economic andpolitical circumstances facing governments. It is entirely legitimate todescribe the process as one in which actual policy choices made (say the TPIvalues of Latvia v. Ukraine in 1994) were influenced by the different initialconditions. But this logic in its extreme leads to nothing more than histori-cal determinism, emasculating the role of any policy choice, and must there-fore be argued with great caution. That this extreme version does show upoften is manifested by the immense popularity of the view that CentralEurope, being close to Western Europe and having a shorter period undercommunism, was bound to do better. If one buys into these simple explana-tions, one must also conclude there are millions of wasted pages of print dis-cussing transition policies and recommendations in the past decade; such anargument basically says what happened would have happened regardless ofpolicy advice.

I take the view here that there was a relevant policy choice despite theimportance of these historical influences. This is exemplified in the early 1990sby the very slow reforms in Romania, the aborted efforts to move quickly inAlbania, Bulgaria, Kyrgyz Republic, Moldova and especially Russia, and thevery early efforts of Armenia and Georgia which were stalled by civil con-flicts. Some of these were reversed by policy (Bulgaria, Moldova, Russia),some frustrated by the land-bound geographic isolation (Kyrgyz Republic)and some by inattention to financial time-bombs (Albania and less dramati-cally the Czech Republic). These examples show that choices could be andwere made contrary to what the historical forces dictated. This view is also

66 Background


informed by the reality of how choices are made; that is, not objectivelyaccording to the most compelling intellectual argument, but influenced byvested interests.

Anyone closely involved with policy-making in the early euphoric periodof post-communism will find the following examples familiar. In the formerUSSR republics, many politicians and policy-makers had been members ofthe political or technical nomenklatura, and it was not difficult even for out-siders to understand quickly who was in favour of a shift to the market andwho was, for various reasons, opposed. The opponents never stated publicly,or even in large closed meetings, that they were against the market; insteadthey played for time to find good arguments against, or at least for not goingall the way to a private-market economy. Intellectual debates on how best todo it, the many methods of privatization, the burden of initial conditions,the pain of too-rapid reforms provided the opponents to reform with the‘scientific’ rationale they needed. Thus, one soon heard from them that onemust go slow lest the people starve as prices become unaffordable, that one canprivatize bakeries but require new owners to produce only bread lest there beshortages, that one must put in place all the legislation and agencies thatregulate competition before privatizing. A personal favourite of mine is theuse of market-memory arguments, especially by Kolkhoz directors arguingagainst too rapid privatization of the land: ‘It’s a good idea of course, but itmust wait until the people are ready for the market.’ Invariably when suchwords were uttered, it was in fact the speaker who was not ready for themarket.

Let me be clear that the above discussion is not part of the logical criticismof gradualist arguments: one must not hold intellectuals responsible for themisuse of their arguments by politicians and policy-makers in government.Thus, intellectuals who recommended gradual reforms cannot be at faultthat their arguments were used to buy time by politicians who were againstany reforms. But at the same time, one should also not fault proponents ofbig-bang reforms whose arguments were misused by politicians who had nointention of doing the full package but were playing off various interests andthe public for popularity, as in Russia early on. It is, however, the intellectualresponsibility of both sides when doing an ex post analysis to recognize thepotential abuse of ideas by practitioners, and to factor this reality of imple-mentation of ideas into the historical assessment.

Understanding the surge in CIS growth rates after 1998

The econometric studies of growth generally cover only the 1990s given thetime lag between data availability and publication of research papers.Unfortunately they miss a key turning point in the recovery, that is the surgein growth rates after 1999 for CISM countries, at which time interestinglythe CISL experienced a slowdown. The broad outlines of this trend are seenin Table 2.5. The simplest and most popular explanation has been the sharp

increase in oil and gas prices which directly benefited Azerbaijan, Kazakhstan,Russia and Turkmenistan and was thought to benefit indirectly others in theregion through the spillover effect of increased imports. The problem withthe spillover argument is that it was not enough to explain the equally highgrowth rates for major energy importers such as Ukraine – surely the terms-of-trade loss should have kept their rates lower. Furthermore, the importeffect was declining over time, as the diversification of trade away fromintra-CIS trade continued and for many in the region the share of exports toRussia had fallen from well over 50 per cent in the 1990s to a third or less by2002 (see Elborgh-Woytek, 2003).

Owen and Robinson (2003) demonstrate that even for Russia oil was notthe whole story – at least as important was the beneficial side-effect of the1998 financial crisis of a real exchange rate adjustment, initially nearly 50per cent devaluation, followed by a gradual appreciation to 75–80 per cent ofpre-crisis value until 2003. The rouble then rose sharply to regain its real1998 value by mid-2005. Most of the other CISM currencies eventually fol-lowed the rouble devaluation, and hence also benefited from this effect ongrowth of export and import-substituting domestic production. Berengautet al. (2002) provides a good analysis of the various possible factors behindthis growth surge, and include, besides the above two, the simple possibilitythat Ukraine (and by extension others) had hit such a low point in thedecline, that the rebound was bound to be strong.

It is useful to recollect the very high growth rates (5–10%) in the mid-1990s when war and internal conflicts subsided in countries such as Albania,Armenia and Georgia. Tajikistan, since 2000, may be a similar case. ButBerengaut et al. (2002) also include a policy variable in their explanation forUkraine: a distinct hardening of the budget constraint especially as it relatesto implicit energy rents and subsidies under the more reform-minded PrimeMinister Yuschenko and his Energy Minister Tymoshenko. Owen (2004) andAslund (2004, Moscow Times) and others describe a similar hardening inRussia under Putin, with regional budgets subordinated to the federal budget,tax collections greatly increased, and oil revenues wisely used to pay offsubstantial portions of the external debt, which fell from over 60 per cent ofGDP in 1999 to about 25 per cent in 2003.

Does this surge in growth conform with the econometric consensusdescribed above? In one way it is contrary to expectations: the CentralEuropean countries, with much higher values of the TPI, have now seengrowth decline to an average far below that of the CISM, as seen in Table 2.4,though the average for the Baltics remains high and comparable after a sharpdip in 1999 reflecting the Russian crisis.

However, this is too static an interpretation of the relation between levelof market progress attained and growth. For the Central European and Balticcountries which by this time had TPI values in the range 3.4 to 3.8, or veryclose to a well-functioning market economy, and indices of output that even

68 Background


in official terms were beyond full recovery, the short-term factors whichexplain recovery begin to be replaced by conventional explanations. This isnot the place for a detailed analysis of that sort, but note that the muchfaster growing Baltics have kept their public deficits well under 3 per cent,while those of Central Europe have exploded way beyond 5 per cent.

For the CIS countries, the most relevant question to ask with the back-ground of earlier econometric studies may be whether they had by 1999reached the same cumulative level of TPI values that one saw for CentralEurope and the Baltics at the time of their first recovery; that is, an averagein the year preceding first positive growth of TPI � 2.55 (2.50 for CentralEurope and 2.65 for Baltics). Table 2.5 shows for each CISM country theapproximate year in which this value was reached, in brackets the TPI value,and then the year of first positive growth. With a few exceptions,16 thepicture one sees is that when the first positive growth was seen in CISMcountries, they had reached something close to the same magnitude of TPIvalues as had been the case in the Central European recoveries. In most

Table 2.5 Growth of GDP since 1998

1999 2000 2001 2002 2003

Central 2.4 3.7 3.5 3.3 3.2Europe

Baltics 0.1 6.0 7.0 6.3 5.7CISM 4.2 8.8 6.4 5.5 6.7

Source: EBRD Transition Report, 2003, table A3.1. For 2002 the averageexcludes Kyrgyz where a goldmine incident caused growth to fall fromabout 5–6% trend to �0.5%.

Table 2.6 Year that the TPI growth threshold (2.55) was reachedfor CISM countries

Year of firstYear reached TI value � growth

Armenia 1997 (2.45) 1994Georgia 1996 (2.5) 1995Kazakhstan 1996 (2.6) 1996Russia 1995 (2.5) 1996Kyrgyz 1995 (2.5) 1996Moldova 1995 (2.5) 1997Ukraine 2000 (2.54) 2000Azerbaijan 2001 (2.45) 1996Tajikistan 2003 (2.39) 1997

Source: EBRD Transition Report, various years.

CISM cases this was before 1998, but the beginning of the recovery trend washit by the financial crisis in Russia, though the delay was not long with therecovery even picking up additional momentum from the post-1998 factorsmentioned above. Therefore, one can conclude that the recent surge ingrowth, while attributable to several factors, is at least in some part a reflec-tion of having finally reached a sufficient degree of progress towards amarket economy to stimulate local economic activity.

The prospects for this being sustained are better left to a later discussion inChapter 8, in particular for the countries where the role of vested interests or‘oligarchs’ has become so important that their influence on policy must beconsidered when asking whether reforms will continue as they did inCentral Europe, or perhaps be stalled if not actually reversed.

Appendix: unrecorded economic activity and measures of GDP

Economic activities not captured by official GDP figures have many labels such asshadow, underground, unofficial or unrecorded. I use the last and Feige and Urban’shelpful shorthand YU. Feige and Urban (2004) is perhaps the most up to date andcomprehensive review for transition countries, containing an excellent assessment ofboth the measurement issues and the analysis of factors driving activity underground.Their analysis by no means originates with transition country concerns but dates backat least to the early 1980s and previous work of Feige and Tanzi, and for transitionstarting with the comprehensive estimates by Kaufmann and Kaliberda (1996). ThisAppendix addresses only the measurement, focusing on the question of possible biasimparted to official GDP data.

The magnitude of unrecorded activities in the post-communist period is generallyconsidered to be a problem because it has increased sharply as a proportion of totaleconomic activity (TEA in Feige-Urban), raising issues of GDP measurement, losttaxation and policy needs. How big is YU compared to other economies and howmuch has it increased?

A recent comparison for 110 countries by Schneider and Klinglmair (2004) showsthe average ratios in 1999–2000 by groups are: OECD countries 18 per cent; develop-ing countries 41 per cent; and transition countries 38 per cent. I draw from these twooperative conclusions. If there is a problem using official GDP for analysis in transi-tion countries, the problem is probably as large for all developing countries – yet, littlehas been written about the need to redo the vast numbers of econometric studies ofgrowth for this group. Nevertheless, in what follows I take seriously the criticismsabout official GDP studies and try to ask how in the transition process there may bespecial biases. A second conclusion is that, with OECD country values around 18 percent (low of 8.6% for Switzerland and high of 28.7% for Greece), a working cut-offpoint for ‘low’ values will be set here at 20 per cent.

How much of an increase there was in the post-communist period is shown byAlexeev and Pyll (2003), who estimate a value back to 1979 for the FSU by republic.They show a (weighted) average for the USSR then of 34 per cent, dropping to 25 percent in 1989. For comparison with post-independence years and other studies, anunweighted average is more useful. Using 1989, as this is the earliest available inCentral Europe, the averages are: five Central European countries 19 per cent (22% if

70 Background


the outlier Czechoslovakia is excluded); Baltics 22 per cent; four western CIS countries25 per cent; four Caucasus-Central Asia countries 33 per cent. The comparable valuesin 1999–2000 given by Schneider and Klinglmair (2004) are: CE 29 per cent; Baltics35 per cent; western CIS 47 per cent and Caucasus Central Asia 48 per cent. As someestimates even for the recent period after decline are in CIS higher than this, it is evi-dent that the post-communist period did indeed see a considerable rise in the averages –which means it is important to analyse how the changes may bias GDP. Anotherimportant conclusion is that in the Soviet period, Central Europe, the Baltics and thewestern republics of the USSR were not very different with regards to unrecorded activ-ities. This has great significance in the discussion of Part II of how the new oligarchsmay have had roots in the soviet ‘mafia’, suggesting this force was not much less inCentral Europe than in the USSR.

A large unrecorded economy has the immediate bias that official GDP underesti-mates the total, and underestimates income and consumption. If this share is grow-ing, as it was, the measures of increased poverty and inequality need to be consideredcautiously. I return to this in Chapter 3. In this chapter, the concern is with growthrates and indices of recovery after the transition recession of the early 1990s. There arefour categories to address the increasing bias problem. First, for countries where theshares are low, say about 20 per cent or less, as in OECD countries, there is little rea-son to undergo substantial effort to adjust or correct the calculations; second, forcountries where the share is high, but fairly constant over the period, again too littlebias exists in official figures to justify corrections; third, countries may show a cleartrend up, or down, or a rise followed by a decline – this last is in fact the most com-mon situation, and it implies a strong bias but one whose timing and direction is evi-dent and correctable at least qualitatively; fourth is the worst case, where the share isvolatile.

Feige and Urban (2004) provide a thorough summary for 25 countries for 1990–2001,using several different measures. The authors conclude on a note of concern that theoverall magnitudes, the sometimes large differences among methods, and the occa-sional ‘nonsense’ result of a negative share make estimates of YU very unreliable. On aconstructive note with which the present author very much agrees, the vast amountof new information this study provides should be used to better understand thedynamics of unrecorded activities and thereby the transition process itself. What theydo not discuss is the bias effect on official GDP growth estimates and hence the relia-bility of growth analyses. I will, with some caution, attempt to do by inference fromtheir data, at the risk of coming to conclusions with which they might not agree.

It is striking that for most countries different methods of estimation may show largeabsolute difference in some countries – over 20 percentage points in Azerbaijan – butvery rarely do they show a different trend. This leads one to conclude that methodvariation is not a serious problem for growth-rate bias. Considering the four categoriesof outcome above, the most important and comforting inference is that onlyTurkmenistan shows considerable volatility over the decade, with about 2 1/2 cycles ofups and downs over a range of 0–40 per cent. The data problem for Turkmenistan isbroader than this and, as noted in the body of the chapter, growth rates are probablyoverestimated. In the first category of low ratios we find five of the six Central Europecountries of Figure 2.1. The levels are in the range 10–25 and all trend slowly down-wards over the decade, making the problem very limited by 2000. If there is a slightbias, it is to underestimate GDP growth in early years. The exception is not a majorone: Croatia has values between 20–30 per cent, but quite constant in the trend,putting it into the second category and implying that growth rates are not much

biased. Only one other country shows a similar constancy in the trend line,Uzbekistan in the range around 10 or around 30 depending on method; again, the dif-ferences in method do not have significance for robustness of growth-rate estimates.Recall that in the main text discussion it was suggested that rates are overestimated,but for reasons unrelated to unrecorded activity share.

Most of the countries are in the third category: medium to high ratios and a risingtrend to about the mid-1990s and then a decline. There are some systematic differ-ences among the groups of Figure 2.1, but these have little implication for growth-ratebias; the rise and decline trend for all these groups is the fact of greatest relevance tobias. The Baltics differ slightly from the other CEB cases with somewhat higher ratiosin the early 1990s, between 35–45 per cent; clearly too high to consider as in the OECDrange, and insignificant. But they experience a very sharp decline, especially after1995, reaching CEB averages of 10–15 per cent by the end of the decade. SEE countrieshave ratios in early years as low as about 25 per cent (Romania), up to 60 per cent(Albania). The CISM group have all much higher ratios, typically 50 per cent or more,but declining to 20–40 per cent. That the CISL cases are lower is a reflection not ofmore advances, but the contrary: the freedom of private activity is still more circum-scribed in these least-reformed economies so that the risk of penalties is great enoughto discourage such activity. On this see inter alia Feige (1997), Feige and Urban (2004)and Kaufmann and Kaliberda (1996).

In summary, only one country suffers from volatility of these ratios, but it has otherserious data problems which make all interpretation for it uncertain. Another fivecountries in Central Europe have ratios too low, especially after 1995, to make muchdifference. The rest have medium to high ratios but a clear pattern of rise then decline.This means that official growth rates in the early years are underestimated comparedto total economic activity (TEA) – in practice less negative than official data – and inlater years somewhat overestimated, but as the share declines this bias narrows.Qualitatively, this applies to CE countries as well, since they too saw a rise and declinetrend, but the effect is much smaller given the low ratios. Is this bias enough to reversethe common view that all economies suffered a decline?

Feige and Urban (2004) compare period growth rates using official data and theirTEA estimates, and the answer is clearly no, the transition recession remains a fact.Is the bias differentiated by countries enough to reverse the common view that thedecline is greater as one goes East? The CE group in their estimate of TEA had less of arecovery than official data suggest, but still an overall positive growth over the period,hence an index by 2001 over 100 per cent compared to 1989. The CISL all showmore of a decline, which is in keeping with the arguments made earlier that theirdecline was underestimated and recovery overestimated. For the Baltics period growthis a greater negative, because the recovery saw an apparent absorption of unrecordedactivity. There is a mixed message in this: given the total activity in 1989 includingunrecorded, the total 12 years later was still lower, but most unrecorded activity hadbeen enticed into the open by the reforms including, presumably, perception of fair-play and a secure rule of law for small firms. The reader is also reminded of the factthat other reasons for underestimates of the absolute achievements exist unrelated tounrecorded activity – these are reviewed in Chapter 3 and while rough adjustments areshown in the second column of Table 2.1, the CISM countries show much variation bycountry, but for the group as a whole the net decline since 1989 is not nearly as largefor TEA as for official data; however, it is still larger than in the other groups.

Overall, then, the broad-brush picture of Table 2.1 is only moderately modifiedby taking into account the ‘best-practice’ information currently available aboutunrecorded economic activities.

72 Background

73Table A2.1 Transitional indicator averages (by group)

1989.0 1990.0 1991.0 1992.0 1993.0 1994.0 1995.0 1996.0 1997.0 1998.0 1999.0 2000.0 2001.0 2002.0 2003.0

CEHungary 1.5 2.0 2.3 2.5 2.9 3.1 3.5 3.6 3.8 3.8 3.8 3.9 3.9 3.9 3.9Poland 1.3 2.0 2.4 2.5 3.0 3.1 3.4 3.5 3.5 3.6 3.6 3.7 3.7 3.7 3.7Czech R. 1.0 1.0 2.1 2.6 3.1 3.3 3.4 3.5 3.6 3.6 3.7 3.7 3.7 3.7 3.7Slovakia 1.0 1.0 2.1 2.6 3.0 3.2 3.3 3.4 3.4 3.4 3.5 3.5 3.5 3.6 3.6Slovenia 1.6 1.7 1.9 2.0 2.8 2.9 3.1 3.2 3.2 3.3 3.3 3.4 3.4 3.4 3.4Croatia 1.6 1.7 1.8 1.9 2.1 2.6 2.8 3.1 3.2 3.2 3.2 3.3 3.3 3.4 3.4

BalticsEstonia 1.0 1.0 1.3 1.8 2.6 3.0 3.2 3.3 3.5 3.5 3.6 3.6 3.7 3.7 3.7Latvia 1.0 1.0 1.1 2.0 2.1 2.5 2.9 3.2 3.2 3.2 3.2 3.3 3.3 3.5 3.6Lithuania 1.0 1.0 1.1 1.6 2.5 2.8 3.0 3.1 3.2 3.2 3.3 3.3 3.5 3.6 3.6

SEEBulgaria 1.0 1.0 1.6 1.8 2.0 2.3 2.5 2.5 2.9 2.9 3.0 3.2 3.2 3.3 3.3Romania 1.0 1.0 1.2 1.6 1.9 2.3 2.6 2.6 2.9 2.9 3.0 3.0 3.0 3.0 3.0FYR Macedonia 1.6 1.7 1.8 1.8 1.9 2.4 2.6 2.8 2.8 2.8 2.8 3.0 3.0 3.0 3.0Albania 1.0 1.0 1.1 1.6 1.9 2.1 2.5 2.6 2.6 2.6 2.6 2.7 2.8 2.8 2.8Bosnia-Herzegovina 1.0 1.0 n.a. n.a. n.a. n.a. 1.1 1.4 1.6 2.1 2.1 2.2 2.3 2.3 2.4Serbia & Montenegro 1.0 1.0 n.a. n.a. n.a. n.a. 1.5 1.5 1.5 1.4 1.4 1.4 1.9 2.4 2.5

CISMArmenia 1.0 1.0 1.0 1.5 1.5 1.5 2.2 2.5 2.6 2.7 2.7 2.7 2.9 3.0 3.0Georgia 1.0 1.0 1.0 1.2 1.4 1.4 2.1 2.6 2.8 2.9 2.9 3.0 3.0 3.0 3.0Kazakhstan 1.0 1.0 1.0 1.3 1.5 1.8 2.3 2.8 2.9 2.9 2.8 2.9 2.9 2.9 3.0Russia 1.0 1.0 1.1 1.9 2.2 2.4 2.7 3.0 3.0 2.6 2.5 2.6 2.8 2.9 2.9Kyrgyzstan 1.0 1.0 1.0 1.5 1.8 2.5 2.9 2.9 3.0 3.0 3.0 3.0 3.0 3.0 3.0Moldova 1.0 1.0 1.0 1.4 1.7 2.1 2.7 2.7 2.7 2.8 2.8 2.8 2.8 2.8 2.8Ukraine 1.0 1.0 1.0 1.2 1.3 1.5 2.3 2.5 2.6 2.6 2.6 2.7 2.7 2.8 2.8Azerbaijan 1.0 1.0 1.0 1.1 1.3 1.3 1.8 1.9 2.3 2.5 2.5 2.5 2.2 2.7 2.7Tajikistan 1.0 1.0 1.0 1.3 1.3 1.3 1.8 1.8 1.9 2.1 2.2 2.3 2.3 2.4 2.4

CISLUzbekistan 1.0 1.0 1.0 1.1 1.4 2.0 2.4 2.4 2.3 2.2 2.1 2.1 2.0 2.2 2.2Belarus 1.0 1.0 1.0 1.1 1.6 1.6 2.1 1.9 1.8 1.6 1.5 1.6 1.8 1.8 1.9Turkmenistan 1.0 1.0 1.0 1.0 1.0 1.1 1.3 1.3 1.6 1.5 1.5 1.4 1.3 1.3 1.3

Source: Author calculations based on EBRD Transition Report, various years.

74

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

1989

.0

1991

.0

1993

.0

1995

.0

1997

.0

1999

.0

2001

.0

2003

.0

TP

I ave

rage

HungaryPolandCzech R.SlovakiaSloveniaCroatia

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Ave

rage

LIB

inde

x

HungaryPolandCzech R.SlovakiaSloveniaCroatia

0.00.51.01.52.02.53.03.54.04.55.0

1989

.0

1990

.0

1991

.0

1992

.0

1993

.0

1994

.0

1995

.0

1996

.0

1997

.0

1998

.0

1999

.0

2000

.0

2001

.0

2002

.0

2003

.0

TP

I ave

rage

EstoniaLatviaLithuania

Figure A2.1 CE transition progress indicators (avg. per year)

Figure A2.2 CE average liberalization index per year

Figure A2.3 Baltics transition progress indicators (avg per year)

75

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Ave

rage

LIB

inde

x


00.5

11.5

22.5

33.5

44.5

5

1989

1991

1993

1995

1997

1999

2001

2003

TP

I ave

rage

BulgariaRomaniaFYR MacedoniaAlbaniaBosnia-HerzegovinaSerbia & Montenegro

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1991

1993

1995

1997

1999

2001

2003

Ave

rage

LIB

inde

x BulgariaRomaniaFYR MacedoniaAlbaniaBosnia-HerzegovinaSerbia and Montenegro

Figure A2.4 Baltics average liberalization index per year

Figure A2.6 SEE average liberalization index per year

Figure A2.5 SEE transition progress indicators (avg. per year)

76

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1989

1991

1993

1995

1997

1999

2001

2003

TP

I ave

rage

ArmeniaGeorgiaKazakhstanRussiaKyrgyzstanMoldovaUkraineAzerbaijanTajikistan

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Ave

rage

LIB

ind

ex

ArmeniaGeorgiaKazakhstanRussiaKyrgyzstanMoldovaUkraineAzerbaijanTajikistan

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1989

1991

1993

1995

1997

1999

2001

2003

TIP

ave

rage

UzbekistanBelarusTurkmenistan

Figure A2.7 CISM transition progress indicators (avg. per year)

Figure A2.8 CISM average liberalization index per year

Figure A2.9 CISL transition progress indicators (avg. per year)

77

00.5

11.5

22.5

33.5

44.5

5

1991

1993

1995

1997

1999

2001

2003

Ave

rage

LIB

inde

x

Uzbekistan

Belarus

Turkmenistan

Figure A2.10 CISL average liberalization index per year

3Economic and Social Costs of Transition: A Hard Look at Soft Facts

78

Introduction

Experience in developing countries of adjustment to macroeconomicimbalances and price distortions showed that some short-run costs wereinevitable before improvements came. This accords with the economic the-ory of market equilibrium, that before investors, producers and consumerscan react optimally, they need to have equilibrium price signals, but thesecome only after some period of trial and error. During this period there maybe severe welfare costs as restructuring and reduced government supportresult in job losses, cuts in social services and a widening of income distrib-ution differences. This was expected, but what was not predictable in advancewas the magnitude of these costs, their burden on different groups, and theirduration. This chapter will present the best available information on suchcosts over the period 1989 to about 2003–04. The main findings are three.First, these costs occurred in all countries and their magnitude ranged fromsignificant to enormous. Second, the evidence also shows that many earlyassessments overstated the costs, partially because they were done too soon,at the mid-1990s bottom of the transition recession, and partly because ofdata biases. Third, the evidence shows these costs varied considerably acrosscountries, with the faster reformers experiencing much lower negativeeffects, and over time even with lagging reformers having generally bottomedout about 1995 and reversed direction.

The chapter is structured as follows. The next section will discuss con-ceptually the nature of transition costs, how to assess them relative to thebenefits, how to measure them, and the caveats of statistical comparisons.We then give some numerical estimates of the aggregate magnitude of lostoutput and employment, followed by a focus on alternative social indicatorsincluding the UNDP’s Human Development Index (HDI), poverty, incomeinequality, health and education. A final section will draw some keyconclusions about the patterns of the transition costs over time and acrosscountries.

Economic and Social Costs of Transition 79

Were negative effects inevitable?

An assessment of transition costs is best made in a Pareto-optimality context.There are three possible outcomes: first, only positive effects occur and apure Pareto optimal state obtains with no-one made worse off; second, someindividuals suffer costs but are adequately compensated; third some individ-uals suffer costs but are not fully compensated.1 The simplified framework ofChapter 2 describing what transition from plan to market entails leads to acorollary about negative transition effects of at least three types: directeffects, as jobs are lost in the process of reallocating resources from old inef-ficient firms and absorption by new efficient firms is delayed or partial; indi-rect effects on social programmes as government budgets are unable tomaintain an adequate level of compensation; and distributional effects as amarket-economy regime results in a widening of income distribution.2

With the possible exception of China where surplus labour in agricultureallowed new job-creation to precede destruction of inefficient firms, it islikely that all transition countries experienced these three effects to somedegree and therefore experienced either a compensated Pareto-optimal resultor an uncompensated one. The assessment I make in this book is still farfrom definitive as it may even now be too soon to make a judgment; butwith that caveat it is possible to indicate where in this spectrum each of thecountry groups fell, and to note deviations of individual countries where rel-evant. In making such an assessment, some judgment on a time-horizon forcompensation is necessary: a displaced worker who in a few years attains astandard of living superior to or with better prospects than in the socialistperiod might consider the temporary unemployment as compensated. Themost practical question is: for any social indicators showing a deterioration, hasthis recovered to original levels, or at least turned around? Given that in all27 countries in our groups, even the lagging ones, economic recovery is wellon the way and indeed in many CIS countries is very strong since 2000, neg-ative answers to these two questions would lead one to conclude that if after10–15 years recovery to or beyond original levels is not attained, this issurely not a compensated Pareto optimal state. If the answer is ‘yes’ the judg-ment might be that a compensated Pareto optimal state, while not attainedyet, is close.

How to measure the economic and social costs? It is already possible toundertake an exhaustive inventory of economic and social indicators andthe changes over time. Indeed the existing literature on this is already verylarge, and the availability of data continues to grow with many recent andcontinuing micro surveys on incomes, jobs, unemployment, health andeducation, even perceptions and attitudes. That literature reflects a vigorousdebate between those who view the costs as excessive due to reforms thatwere too rapid, and those who argue that the costs were in fact lower wherereforms were more rapid. I will provide here a selective rather than exhaustive

set of information which captures the most important indicators of well-being, and attempt do so with balanced consideration of the two sides in thisdebate. As I associate myself with the rapid-reform school, it is appropriate toavoid methodology and data sources that bias results in that direction. Inparticular let me note that most of the data on social indicators is taken fromdetailed studies by proponents of the ‘too-rapid’ school and UN sources suchas the UNDP Human Development Report and UNICEF Transmonee database;these sources are not generally suspected of Washington Consensus bias.

The variables used here are: a measure of the output loss using official andadjusted GDP; some measures of unemployment; the percentage of popula-tion below the poverty line; Gini coefficients measuring equality or inequal-ity of income distribution; the Human Development Index of the UNDPwhich has stood for more than a decade as the key alternative to GDP percapita for measuring well-being; life expectancy, considered by most analystsas perhaps the best single proxy for health status; and finally some measuresof educational attainment.

There are many data problems for three main reasons: comparisonbetween Soviet-period definitions and market-period ones (the GDP losscalculation is most affected); a limited data availability for most recent years(income distribution); and accuracy of measurement (actual consumption;actual unemployment). Where these are particularly severe, I present a rangeof estimates. Specific data problems are described in more detail in thediscussion for each of the variables.

One last prefatory comment needs to be made on the problem of bench-marking. Part of this is a data problem as for example with the Gini-coefficientmeasure of income inequality. Values for the Socialist period abound, butusing them as a reliable benchmark is not entirely objective,3 because itignores the importance of an institutional reality of that period nicely cap-tured in Feige (1997): ‘although incomes were distributed equally, differentialaccess to [goods and services] and selective opportunities for illegal wealthaccumulation created a highly unequal distribution’. This reality must at leastbe recognized in any assessment of changes after the transition to market. Butthe benchmark problem here is more than data definitions: a shift to privateownership and market determination was expected to lead to some wideningof income distribution and that this would be arguably a good thing, creatingincentive to greater effort both by labourers and entrepreneurs. The properbenchmark comparison is then not with the Gini levels of 1989, but perhapsthose of market economies considered having ‘reasonable’ patterns of distrib-ution. Another possible benchmark is China given its huge success in achiev-ing high growth; the results may be surprising to some.

To avoid misunderstanding, it should be clear from the outset that theabove remarks on a benchmark for Gini do not apply to poverty measures.These are subject to data problems in both periods, but there is no analogousrational argument of a ‘good’ rise in poverty for motivation purposes. If

80 Background


poverty ratios increase, this is prima facie evidence of a deterioration thatrequires compensation.

Aggregate economic costs: lost output and unemployment output losses

A striking feature of the transition in the non-Asian region has been asubstantial and extended decline in output. The extent of this output lossusing official data is shown in Table 3.1, showing in the first row for eachcountry group the index of aggregate GDP with 1989 as the base year. It isoften characterized as ‘higher and more persistent than during the GreatDepression’ (Grun and Klasen, 2001, p. 360). Many works have analysed theextent of this decline and consequences for welfare, and it is worth citingtwo important studies. UNDP (1998) analysed in great depth the evolutionof various social indicators through 1995 and labelled them ‘the most acutepoverty and welfare reversals in the world’. Milanovic’s (1998) analysis ofincome distribution, while noting ‘the economic landscape is changing for

Table 3.1 Official and adjusted index of output (1989 � 100)

Output index

Low point 2002

Central EuropeOfficial 80 120(Adjusted) (94) (148)Average 87 134

BalticsOfficial 52 82(Adjusted) (80) (128)Average 66 105

SEEOfficial 61 81(Adjusted) (80) (85)Average 71 83

CISMOfficial 40 55(Adjusted) (70) (96)Average 55 76

CISLOfficial 60 96(Adjusted) (70) (112)Average 65 104

Source: Official figures from the EBRD Annual Transition Reports;adjusted based on Aslund (2001), as explained in the text.

the better’ nevertheless concludes that ‘massive dislocations … have hadhuge social costs’. For the worst-hit region, CISM, the index reached a low of45 in the mid-1990s, and as late as 2003 had recovered only to 67. ForGeorgia these figures are correspondingly 27 in 1994 and 41 in 2003. Thelong duration of the decline is further exemplified by the fact that it contin-ued as late as 1999 in Moldova and Ukraine, that is a full decade of negativegrowth. With numbers like this, it is small wonder that many observersconcluded that transition has created a human disaster.

While the fact of a large output decline is virtually undisputed, theseorders of magnitude are overstated by varying degrees in different countries.Furthermore, the temptation to focus on the worst cases is misleadinginasmuch as the best cases – especially in Central Europe – can easily be char-acterized as having suffered relatively minor losses over the period. As anillustration, consider that Poland’s decline lasted only two years, and evenwithout any adjustment of official GDP the low point was 80; by 1996 theindex had surpassed 100, and in 2003 was 135. For the Central Europe groupin this study, if one excludes Croatia, the bottom was reached in 1992–93 atabout 75–80, 100 was reached variously between 1997–99, and the averageindex in 2003 was 115. The correlation between limited decline and greaterrecovery on the one hand, and the early start on reforms on the other, hasalready been discussed in the previous chapter. The corollary point to bemade here is that characterizing output decline in transition as a disasterwithout reference to the far superior performance in Central Europe, andeventually also in the Baltics, is somewhat misleading.

An additional source of overstatement comes from using official GDP andthe same benchmark date of 1989 for all countries. This point has been madeby many analysts as noted in Chapter 2, but is most comprehensively dealtwith in Aslund (2001). There are several corrections made, and his workprovides estimates of how each of these adjustments change the index ofoutput for 20 countries in 1995. His adjustments may go too far, and to min-imize bias in my argument they will only be used for a broad-brush estimateof what might be the absolute minimum output losses for our five countrygroups. First consider a brief outline of the corrections. The simplest andeasiest is the start date: Aslund argues that 1989 can be considered the lastyear of socialism in the four Visegrad countries, but that for almost all therest – former Yugoslavia, former Soviet Union, Bulgaria, Romania – transitioncould not have begun earlier than 1992, making 1991 a more reasonablebenchmark.4 This seems eminently sensible and indeed a bit conservative:one could argue that attributing output loss to transition reforms shouldstart with the year in which significant reforms began, not the year in whichthey could have begun. This would mean the start year in much of the CISother than Russia and Kyrgyz Republic would be about 1994–95, and in theCISL group arguably still in the future. But I have not done this, rather I takeAslund’s suggestions of 1991 for some countries. The main point is that any

82 Background


loss before the start year should be attributed to socialism and not to transi-tion. Aslund’s change in start year alone accounts for 15–25 points of the CIS55 per cent decline. A second correction, perhaps the most important, is todiscount the soviet period GDP for ‘value-subtracting activities’, that is theenormous inefficiency of producing unusable or very poor quality productsall valued as if they were saleable on open markets.5 Informal activities andinternational trade mispricing make for further corrections, but of less sig-nificance generally than the previous ones. It is notable that Aslund’s cor-rections do not attempt to discount the soviet period GDP for the cost toindividuals of queuing and attendant frustration of not always obtaining thedesired goods. This is another reason his corrections are not greatly exces-sive. Value-subtraction adjustments, according to Aslund, account for 20 percent or so of the loss under official GDP estimates.

Aslund’s total adjustment paints a very different picture of output trends.His index values for 1995 show Central Europe (excluding Croatia) alreadymore than recovered to 112, and the CIS is put at 86 (compared with officialindex values about 90 and 45 respectively). The adjustment for CIS is fargreater because of the shift in the benchmark year to 1992 and the higherdegree of ‘value-subtraction’ in the Soviet period compared to CentralEuropean satellites. To allow for a comparison with official data and a track-ing of output at the trough, and in the latest year, I have projected Aslund’s1995 estimate to the trough year where necessary and to 2002 using officialgrowth rates. Aslund does not cover all the 27 countries, hence group aver-ages using his sample are applied to the five groups here. The resultingcalculations provide the comparative estimate of Table 3.1 (in brackets in thesecond row for each country group), and show a substantially greater recov-ery for all regions. Central Europe and the Baltics are well-beyond initiallevels, SEE is close to recovery, and the CISM group had virtually recoveredby 2002; the high growth rates in 2003 and 2004 might easily bring themover initial levels.

While Aslund’s estimates are a reasonable but still conservative adjustmentfor this historical comparison, I propose to take an even more conservativeapproach and use an average of official and Aslund values for a calculationof cumulative output losses, in effect allowing for only one-half of theAslund adjustment. This average is in the third row of Table 3.1 within eachgroup. An alternative calculation not shown here, to adjust the GDP indexonly by shifting forward the start year for countries outside Central Europe,in fact gives values quite close to this average. Thus, even ignoring the effectsof differences between Soviet and market accounting, and correcting onlyfor the timing problem – that is attributing declines before regime change tosocialism and not to the transition – changes the story considerably.

What then is the picture that Table 3.1 paints? A painting may inadver-tently be a nice analogy, because the vision will be in the eye of the beholder,ranging over three variants. The official data show a dramatic decline of

60 per cent in CISM, 50 per cent in the Baltics, 40 per cent in SEE, but muchless in Central Europe. The Aslund-adjusted decline is much less dramatic,about 30 per cent in CIS, even less in Baltics, and only 5–6 per cent declinein Central Europe. The average of the two moderates the worst-case scenariounder official data but still implies a significant decline to the mid-1990soutside Central Europe of about one-third or less in the Baltics, SEE andCISL, and 45 per cent in CISM. This is an important and indisputableconclusion: with the exception of Central European countries, even the esti-mates after adjustments for accounting definitions (Aslund) reveal an outputdecline of at least 20–30 per cent in the first half of the 1990s. This may falla bit short of the decline in the 1930s Depression, but is large enough toqualify historically as a serious deterioration and cannot but have hadpainful consequences for a large proportion of the population.

The alternative picture of the economic cost of transition, providing anestimate of cumulative loss of output over the decade, again shows a rangebased on official data and the partial adjustment (third row) of Table 3.1. Theexercise is necessarily very approximate for two reasons: the loss must be rel-ative to a hypothetical counterfactual of output levels had transition notoccurred, a counterfactual which can be disputed; and official annual dataeven where available for the whole period suffer from severe inconsistenciesbetween recent and earlier publications.

The technical details of these assumptions and the calculation mechanicsare found in the Appendix to this chapter. Two counterfactuals are proposedhere, both relatively favourable to the ability of continued socialism tomaintain its output level. The first counterfactual, extremely favourable tosocialism, is that for another decade or so under socialism, output mighthave continued to grow modestly for a few years, then to decline slowly. Thisis approximated roughly by an output level that stays steady at the 1989 (or1991) level for a decade, and is implicitly assumed in most discussions ofoutput loss.6 That this can actually be done with continued subsidization,inflation – creating credits, and barter trade arrangements – is arguably sug-gested by the cases of Belarus and Uzbekistan. A less favourable assumptionis that the sustainability policies are less successful, and that after a year ortwo of mild positive growth, output crashes (as in fact it did before transitionstarted in virtually all countries) reaching the same level as did actual outputat about the same time in the mid-1990s. This is roughly approximated by asteady output level to the mid-1990s and a full crash to the actual low pointreached.7

Even the last counterfactual is still relatively favourable to socialism. Infact it is undisputed that output started to decline in almost all countrieswell-before any serious transition reforms began, and it would not beimplausible to posit as the least favourable counterfactual that continuedsocialism would have resulted in an output decline to the mid-1990s notmuch different from the actual one observed.8 The implication? Cumulative

84 Background


output loss would be close to zero or at least very low until the actual low pointwas reached. To retain a conservative bias in calculations, I do not employ thisleast favourable counterfactual but it is discussed more fully in the Appendix.

In assessing the values of Table 3.2, I wish to emphasize once again thateven the lowest values rely on making the most conservative assumptionsabout the advantages of rapid market reforms and disadvantages of contin-ued socialism. None of its values assumes continuation of socialism wouldhave resulted in a decline after 1989–91 similar to the actual one observedbefore, though the case for this could easily be made. The smallest cumula-tive value, 25 per cent of annual GDP minimum loss in Central Europeunder the average adjustment, is in fact more conservative than the fulladjustment of Aslund (about 10 per cent loss), and further takes no accountof eliminating the costs of queuing in the shortage economy. On the otherside, the highest values of maximum loss using official data, 650 per cent ofannual GDP, lean strongly in the direction most favourable to socialist con-tinuation in two ways: there’s no need to make adjustments for overvaluedsocialist output, and socialist continuation would have allowed sustaining aviable and only declining economy for a decade beyond 1989. For those whoaccept both these assumptions, the loss in one of the regions could be said toreach nearly seven years of output. As noted several times, however, both theassumptions are highly implausible.

The range of values in Table 3.2 is still very wide with a low of 25 per centof initial output lost in 14 years in Central Europe under the partial Aslundadjustment, to a high of 650 per cent or nearly seven years of output in theCISM group under official accounting and assumptions most favourable tothe viability of the socialist economy. But even the minimum values clearlyshow that for all but the Central European countries the losses were unusu-ally high by historical standards, and proponents of rapid reform cannotdeny the fact of a large aggregate cost. On the other side it seems equallyclear that terms like ‘dramatic disaster’ are misplaced. Furthermore, if the

Table 3.2 Summary estimates of cumulative ouput loss (as per cent of 1989 GDP)

Maximum Minimum I Minimum II(official data and (partial adj. and (full Asland adj. and most-favourable less-favourable less-favourable counterfactual) counterfactual) counterfactual

Central Europe 100 25 10Baltics 190 45 25SEE 360 90 60CISM 650 160 110CISL 400 120 100

Source: Appendix Table A3.1.

time horizon for Pareto-optimality judgments goes beyond one decade,there are already substantial gains in Central Europe, somewhat less but stillpositive in the Baltics and the SEE group. The CISM countries with recentgrowth rates 5–10 per cent per annum are probably approaching the gainsphase as well.

The last column in Table 3.2 shows the cumulative losses using the fullAslund adjustment, which might be indicative of applying the leastfavourable counterfactual on socialism. This would amount to saying thatsocialism was already unable to sustain output by 1989, and without transi-tion the decline would have been inevitable and perhaps not much less thanthe actual pace observed. Under this assumption much less output loss couldbe attributed to market reforms.

Readers will choose what they judge most reasonable in this picture: theauthor’s (conservative) choice is that Central Europe experienced cumulativelosses up to one-quarter of initial output, the Baltics one-third or so, SEEprobably less than one year, and CISL somewhat over one year. The strongestimpact came in the moderately reformist CISM with losses significantlymore than one year, but probably less than two years of initial output. Iadmit to being tempted by the assumption that continued socialism wouldhave resulted in nearly the same decline as actually occurred, in which casethe cumulative losses would be lower still.

The focus of this chapter is on the costs of transition rather than on itsbenefits. It may nevertheless be useful to give some indication of how muchof the aggregate lost output has been recouped or surpassed. The simplestapproach is to estimate how much cumulative GDP gain there has been fromthe time the index reached 100 per cent of the benchmark year value. Takingthe average values of Table 3.1 (partially adjusted index), full recovery camesometime after 1995 but certainly well-before 2000. Hence, on average therehas been about five years of pure ‘gain’, with index values by 2003 reaching135–150 per cent (partial adjustment), which implies a cumulative gain of50–60 per cent annual GDP. This gain exceeds the median estimates of loss forCentral Europe and is about equal to the loss in the Baltics. Ignoring the CISLgroup with questionable data, the other groups, SEE and CISM, are onlyrecently achieving full recovery, and net gains are minimal or still to come.However, the above characterization is the one least favourable to the case formarket transition. If the hypothetical counterfactual were that under contin-ued socialism output would have continued to deteriorate to reach nearly thesame bottom as did actual output, then pure gain would have started from thefirst year of positive output growth. It would thus be far greater for CEB coun-tries than the above rough estimates, and already significant for the others.

Trends in employment and unemployment

As with GDP, labour force data are subject to several problems of measure-ment and comparability. With appropriate caveats, it is nevertheless possible

86 Background


to give a reasonable picture of the trends and note some stylized facts, theprincipal one – on which there is broad consensus – being that all countriesexperienced a substantial increase in unemployment. This section describesthe main conclusions of earlier analysis on three issues: trends for unem-ployment rates by country group over time; labour-force participation andmobility; and some of the explanations for the fact of a very high and per-sistent unemployment. First, however, comes a brief discussion on thenature of the data problems.

There are two main sources of unemployment data which provide cover-age of all these countries for most of the period since the early 1990s: theUNICEF Transmonee database, and the EBRD Transition Reports.9 In princi-ple, they both use the same national data sources, but some anomalies arise.EBRD provides more recent data, and for a number of countries, especiallySEE, gives a longer time series. But while Transmonee indicates that for allcountries the concept of unemployment is ‘Labor-Force-Survey’, which gen-erally avoids the underestimation of ‘official registration’ data, the EBRDnotes that in many CIS countries surveys are unavailable and the data shownis therefore registration-based. The anomaly is that despite the different def-initions given, the actual values are very similar in Transmonee and EBRD, asseen in the group averages of Table 3.3; I infer from this that in fact boththese sources give data from labour-force surveys where available andregistration where not.

This of course makes comparability across countries difficult, though thebias is clear. Many CIS countries show values less than 5 per cent, quiteunlikely on the face of it; hence, their lower average is partly explained bythis data problem. Thus, one sees in these two sources very low 2002 valuesfor individual countries Tajikistan (2.4), Ukraine (3.8), Uzbekistan (0.4),Belarus (3.0), and the extreme case of Turkmenistan where employment is‘guaranteed’, hence the official rate of unemployment is precisely (0.0).EBRD wisely treats the last case as ‘unavailable’ for the period. Transmoneeshows a blank for all years, except 1999 where it is given at 4.9%, with unfor-tunately no explanation. These individual underestimates obviously implythe group averages for CISM and CISL are much too low.

In a handful of cases informal estimates of actual unemployment exist,and EBRD (2003) notes the much higher figures: Tajikistan about 30 per centand Turkmenistan about 19 per cent in 1998. Beyond this, comparabilityover time is also in question, as suggested by the cases of Georgia andKyrgyzstan: for both EBRD shows very low rates until about 1997 and afterthat the values jump to the teens, 12.3 per cent and 17.4 per cent by 2002.As a last example of the difficulties such soft data create for a researcher, con-sider that in the EBRD (2000) summary of unemployment, Azerbaijan, alongwith Armenia, Kazakhstan and Russia, is one of four CIS countries withplausible values in the teens, suggestive of the possibility the data is based onsurveys; but unhelpfully, there is no country-specific information on the

concept, only the general statement that in many CIS countries onlyregistration is available. Also an internal anomaly appears within EBRDreports since 2002, showing in country tables values for Azerbaijan through-out the 1990s about 1–2 per cent. Transmonee shows only one year 2001,12.8 per cent.

Nevertheless, for most countries the two sources are broadly consistent,the values are plausible or the bias is known, and therefore general patternscan be deduced from the averages in Table 3.3. Ignoring the implausiblevalues of the CISL group, it is clear all countries experienced a substantiallevel of unemployment during the transition. The higher values in theadvanced reformers reflect a combination of the data underestimate in CIS,plus the well-recognized fact observed by many that, until the late 1990s,slow reformers’ workers were indeed kept on in the payroll, but they did lit-tle and were paid little, that is ‘in these countries labor market adjustment

88 Background

Table 3.3 Unemployment rates in the transition (percentage oflabour force)

1996 1999 2001–2

Central EuropeT_M 9.3 11.6 12.3EBRD 9.5 11.6 12.2

BalticsT_M 15.3 13.5 14.2EBRD 15.3 13.6 12.0

SEET_M 17.6 18.4 18.6EBRD 10.5(28.8)* 15.1(32.9) 13.6(33.8)

CISMT_M 8.7 12.7 10.3EBRD 5.5 9.3 8.0

CISLT_M n.a. n.a. n.a.EBRD 0.7 1.4 1.9

* The values in brackets are the average for 3 outliers: Bosnia-Herzegovina,Macedonia and Serbia-Montenegro. The unbracketed values are forAlbania, Bulgaria and Romania.

Source: The first row for each group is based on UNICEF Transmoneedatabase (2004); the second row is from EBRD Transition Report (2000),table 5.1, and the 2003 report for latest values.

Explanations: Transmonee data is in principle labour-force survey-based,EBRD official registration. For SEE, EBRD data is shown for two subgroups:the first number includes Albania, Bulgaria and Romania; the second inbrackets Bosnia-Herzegovina (missing in 1996), Macedonia and Serbia-Montenegro. For CISL the average excludes Turkmenistan.


took the form of … wage arrears’ (Campos and Coricelli, 2002, p. 807).10

The increase over time for the CIS is also a reflection of data problems andthe gradual ending of informal employment, though in much of CentralEurope where recovery occurred early, unemployment generally fell onlyslightly then rose again in the late 1990s.

There are only a few exceptions to the persistence of high unemployment;in Hungary and Slovenia one observes a peak of 9–11 per cent about 1994,then a steady decline to less than 6 per cent by 2002. In the Baltics, Estoniahad lower values, but also a continued rise until 2000, while the other twopeaked at higher levels by 1995–96 with a steady decline thereafter. In SEEBulgaria has somewhat the same pattern, but much higher values peaking at20 per cent, declining to 15 per cent. Incidentally, there are two distinct sub-groups, with Albania, Bulgaria and Romania in the range of just below 10 percent to about 20 per cent, while Bosnia-Herzegovina, Macedonia and Serbia-Montenegro, no doubt reflecting the civil conflicts and greater politicalinstability, range between 20–40 per cent. In the CIS, Kazakhstan and Russiashow faint signs of a peaking in the late 1990s and a small reduction to 2002consistent with the surge in growth since 1999, though the weak data of theregion may hide other such cases or mislead on these two. Available obser-vations are too few and too uncertain to allow any meaningful conclusionfor the CISL group; recall however the informal estimate for Turkmenistanunemployment of 19.8 per cent.

Patterns of labour-force mobility in the transition varied a great deal acrosscountries in line with the variation in the nature of the adjustment, a pointelaborated in Haltiwanger, Lehmann and Terrell (2003), who note that in theearly period job-destruction dominates, and later job-creation rates catch up.Since CIS countries were slower to reform and force adjustment, the netmisemployment came later than in Central Europe and the Baltics. But twoother aspects were similar: there was a lot of mobility out of the labour force(Boeri, 2000), reflecting the traditional discouraged worker theory, and verylittle geographic mobility (Campos and Coricelli, 2002). The latter also note,however, that the mobility or turnover was very high in the sense of consid-erable flows from state to private activities and, as EBRD (2000) also illustrates,from declining sectors to growing ones. I turn now to some explanations oftrends and an assessment of the unemployment cost of the transition.

Why has there been such a high and persistent level of unemploymentafter more than a decade of transition? It is not enough to say this simplyreflects the sharp decline in output, for two reasons: first, the actual outputdecline has probably been far less than official GDP figures show, especiallyin the Central European region; and second, the trend for unemploymentafter the mid-1990s did not follow the output recovery trend, but continuedto stay high, or even rise despite positive output growth. There are two main(and complementary) explanations found in the literature: the excess employ-ment of the communist period was very high and is still being worked out;

and institutional labour-market rigidities have prevented a greater reabsorp-tion of the unemployed.

That excess employment of labour characterized the central-plan period is awidely accepted consensus, but there were no systematic estimates of the orderof magnitude of this excess. The nature of inefficiencies, as explored ex postin Easterly and Fischer (1995), resulted in overemployment relative to the out-put produced. The excess is not simply attributable to the socialist principleenshrined in law that all are guaranteed employment. In fact, the taut-planning nature of the regime resulted in a ‘constant excess demand for labor’,as managers trying to meet monthly targets and obtain wage subsidies hoardedlabour (Campos and Coricelli, 2002, p. 797). Rather, the inefficiencies camefrom misallocation of resources to labour-intensive technologies, poor incen-tives for employees, and a rigid system of centralized factor allocation thatmeant a low elasticity of substitution between labour and capital. Togetherthese elements led to a level of employment that by international standardswas highly inefficient, hence a labour to output ratio that was excessive.

But the order of magnitude of this excess has not been estimated for thisperiod. One possible ex post clue may come from those countries where theadjustment has been fastest, and unemployment performance best. Polandis not such a case even though recovery came earliest and the index of out-put has achieved the highest level, for there the problem of unemploymentseems far greater than in other Central European and even Baltic countries.Poland’s unemployment rose sharply to about 15 per cent by 1993 beforedeclining gradually, but the decline was reversed by a sharp rise since 2000,reaching a new high of about 18 per cent in 2002. In contrast, unemploy-ment stayed at much lower levels in the Czech Republic, Estonia, Hungaryand Slovenia, with peaks reaching 9–11 per cent, albeit at different points ofthe output cycle. While this might be considered a minimum indicator ofSoviet-period overemployment, that value may be higher because these fourcountries were in Soviet times generally considered to have the highestefficiency (and not coincidentally highest per capita income). Furthermore,the process of cleaning out inefficient labour was certainly not complete bythe mid-1990s when the peak was reached, but as Haltiwanger andVodopivec (2002) show for fast-restructuring Estonia, job reallocation wasvery high even later, at a rate of 20 per cent, similar to that of the USA.Nevertheless, these better performers do set a rough benchmark for the oth-ers, making it difficult to deny the conclusion that the level of unemploy-ment has been probably higher or of longer duration than the excesses of theSoviet period alone would imply. Unlike many other indicators, one cannothere adduce the explanation that slow reforms are to blame, since the veryhigh rates are seen in many if not all of the rapid reformers of Central Europeand the Baltics and not just the CIS.11

It is useful to look for additional explanations, at least in cases ofunemployment rates persistently above 10–15 per cent. Some of these are

90 Background


country-specific, as the very high rates noted for three of the conflict-torncountries of SEE where political and social instability suppresses entrepre-neurial initiative and output activity remains very low, while of course infor-mal activity no doubt thrives. Feige and Urban (2004) show that Macedoniahas not only a much higher share of informal sector than many others, butalso a rising rather than declining trend. The EBRD (2000) analysis of employ-ment trends argues that weak labour markets still obtain even in the moreadvanced reformers, reducing labour mobility and reallocation. One aspectof this is geographic mobility, which is very low, and which Campos andCoricelli (2002) attribute to distortions in the housing market such as urbanregistration formalities and a low stock of housing, both rigidities a carry-over inherited from the Soviet period, though more of a problem in CIS thanCentral Europe.

These explanations have some merit, but it is hard to complete the circleof logic and say ‘Eureka’, we now understand, for none of the analysis is ableto assess how much of the actual unemployment is due to each of the deter-minants. For example, one might argue that institutions or housing con-straints impeding geographic mobility are not yet truly binding factors, forthe restructuring is far from complete in most of the countries, even in manyof the more advanced reformers, and the output recovery is not strongenough to absorb unemployed at the new ratios of capital to labour.

We are left in the end with a picture of substantial labour market shifts, con-siderable reallocation, but nevertheless rates of unemployment and discour-aged employees which are very high and a good indication of the individualcost to the population brought about by the transition. Better performance onmarket reforms, growth and stabilization has eased this somewhat in a hand-ful of countries, but the problem persists for most others.

Impact on individuals as measured by social indicators

This section will review in some detail a number of concrete social indicatorscommonly used in the analysis of economic development and well-being.A key source for such data is the United Nations Development Programme’sannual Human Development Report (UNDP-HDR); in addition a large numberof analytical studies of the UNDP, UNICEF and other agencies have producedstudies of the social costs which provide considerable data. The followingspecific indicators are analysed over the period 1990–2001: the UNDP over-all index of well-being (the Human Development Index, HDI); two incomedistribution indicators: the Gini coefficient and the per cent of population inpoverty; life-expectancy proxying for health in general; and education enrol-ment ratios representing education and human-capital developments.

Numerous studies exist assessing the worsening inequality, increasedpoverty and declining life-expectancy in the region, with many concluding

like Grun and Klasen (2001) that: ‘Due to low inequality and moderateincome levels, socialist countries enjoyed relatively high levels of economicwell-being. In the transition, rising inequality and falling incomes have ledto a dramatic absolute decline in well-being.’ An important work on thisissue, UNDP, Poverty in Transition (1998) succinctly summarizes these eventsas ‘the most acute poverty and welfare reversals in the world’. Such stronglynegative assessments may be exaggerated for three reasons: the data from thesocialist period overstate well-being and understate inequality of distribu-tion (primarily by ignoring the fact that distribution of income was moreequal than distribution of goods); second, not all countries suffered to thesame degree; and third, the time-lag of academic research resulted in cover-age largely for the period of decline and did not adequately reflect possiblerecovery of well-being measures since the mid-1990s.

Simply put, it may have been premature to assess costs of a process at itsexpected mid-point. Thus, the UNDP (1998) study compared 1990 and 1995values; this is easiest to correct, as data up to 2001 are now available in UNDPsources to update the findings. The variation by country or country group isalso easily observed and will figure prominently in this section. The mostdifficult correction to make is for the problem of data comparability betweenthe Soviet and post-Soviet periods. I will not attempt to adjust for the (real)possibility that the communist regimes exaggerated data in a favourabledirection, and use data as available, but I will refer to some evidence of suchexaggeration.

Students of economic methodology will find a lot of evidence ofMcCloskey’s (1987) ‘economic rhetoric’ in this literature, which includes notonly the very negative assessments cited, but also rather more positive ones,as for example the EBRD Transition Report analysis on social costs in 1997,1999 and 2000. The latter note inter alia a general tendency to recovery inthese indicators, and further demonstrate that the degree of deteriorationthat did occur was highest where economic reform was less advanced.Indeed, the EBRD goes as far as concluding that, in Central Europe, indica-tors such as life-expectancy and education saw virtually no decline even inthe first half of the decade. In the same conservative and balanced spirit asin the preceding data analysis, I rely as noted largely on data of the UNDP-HDR, presumably not subject to the ‘Washington Consensus’ bias. Indeed,the very purpose of the Human Development Report when it was first initiatedmore than a decade ago was to provide an alternative to the GDP per capitaand growth criteria most commonly used to evaluate development progress.This alternative is now exemplified by the HDI statistic the UNDP publishesannually. Though it may be useful to note that, just as economic perfor-mance statistics, social indicators are not problem-free and some caveatsneed to be laid out, the evolution of the HDI for the transition countries willbe the basis for the first important quantitative assessment of social costs intransition.

92 Background


Overall well-being: the human development index

The HDI is, like the TPI of the EBRD used in preceding chapter, a syntheticconstruct with some implicit subjectivity on two counts: the choice ofcomponents, and the weighting scheme. Both of the above indices use equalweighting of components for lack of a good rationale justifying any otherweighting. I do not propose any modification of either the TPI values or theHDI ones, and take them as they are reported by the respective institutions.Among the components of HDI, those most subject to difficulties areinequality indicators. Even in advanced economies there are many problemsrelated to survey coverage, definition of the poverty line, and comparabilityacross countries. For transition countries these are exacerbated by the natureof the communist regime with distorted prices, considerable in-kind supportby enterprises, the fact that distribution of income and distribution of goodswere different and separable concepts. Data on life-expectancy, mortalityand school enrolment may be most free of problems, but even they sufferfrom incompleteness of data and deteriorating statistical services in the earlytransition period.12

With these caveats in mind, consider the evidence on general well-beingor human development, as measured in UNDP’s HDI. It contains severalelements, one of which is GDP per capita, while the others are more directmeasures of well-being related to health, education, roles in society, genderand minority empowerment, and so on. Thus, it reflects not just social con-ditions but a combination of social and economic ones, and is meant to be abroader measure of well-being. Table 3.4 summarizes the tendencies for thefive country groupings.

Table 3.4 Average human development indicator values

1990 1995 2001 Trend

Central Europe .815 .821 .845 5 rising, 1 declinefull recovery�

Baltics .812 .779 .822 3 full recovery�

SEE (3 of 6) .752 .749 .772 1 rising, 2 decl., fullrecovery�

CISM .772 .721 .737 1 rising, 7 partrecovery, 1 cont’ddecline

CISL .767 .743 .760 3 decline, fullrecovery

(Maximum 2001, Norway � .944)

Note: ‘Full recovery’ means that the 2001 HDI value regains its 1990 level; a ‘�’denotes it exceeds the 1990 level.

Source: UNDP, Human Development Report, various years.

The EBRD conclusion that Central Europe suffered very little deteriorationis strongly confirmed, as the average HDI shows no decline even in the firsthalf of the decade. Within the group only Croatia shows a slight deteriora-tion 1990–95 (from 0.801 to 0.794) which may not be surprising given thepost-Yugoslav conflict. The EBRD conclusion on the negative correlationwith reforms is also confirmed, as country groups with less reform progressexperienced stronger deterioration in overall well-being. A strong deteriora-tion is undeniable in the Baltics; however, by 2001 all three had surpassedthe initial HDI level – a finding incidentally more in line with the adjustedoutput than the official output indicator of Table 3.1. The same appears to betrue for South-East Europe, though the data here cover only three of sixcountries. Even in the CISM group we observe a clear recovery since 1995,albeit only partial. The exceptions to this trend are Moldova which hassuffered a continued decline, while Kyrgyz Republic is shown to have beenrising continuously, a result inconsistent with the data of UNDP on povertyratios, but one that can be ignored without affecting the broad conclusion.

The view of many critics that some countries like Belarus and Uzbekistanhave avoided the worst of the deterioration (Grun and Klasen, 2001, p. 380)seems at first sight to be confirmed. The decline of HDI to 1995 was clearlymuch less for the CISL slow-reformers than for the CISM group. One should,however, leave to future analysts to determine whether the correct inferenceis that the very gradual reformers of CISL have successfully minimized thesocial costs, or whether their lack of adjustment and now-slower growthhave merely postponed those costs. Some evidence like life-expectancydeclines in Belarus and recent analysis of extensive poverty in Turkmensistan(World Bank, 2000) – discussed below – belie this rosy conclusion.

The evolution of HDI clearly supports the view that countries which movedfastest and farthest on economic reforms experienced far lower social coststhan slower reformers. This is contrary to the expectations of gradualists thatrapid reforms would lead to substantial deterioration in social well-being. Theevidence is consistent with a compromise view that partial and too slowreforms are worse than rapid, or no reforms: ‘slow transition might mean theprocess has been captured and delayed by old elites who are able to slow themove to a market economy and enrich themselves at the expense of the pop-ulation’ (Grun and Klasen, 2001, p. 379). The lesser impact in CISL than CISMcountries is consistent with the latter view. But the results for SEE countrieswhich were also partial-reformers raise a question: why did they not suffer thesame sharp deterioration as the CISM group? One possible answer is related tothe effect of a strong desire for integration with Europe or, more explicitly, EUmembership, which mitigates somewhat state capture and keeps reforms mov-ing forward even if slowly. As discussed in Chapter 7, measures of state captureshow lower values in SEE states. Finally, it is notable that the HDI values showa recovery trend from 1995 for all the groups, and individual country values(where available) show only one exception to this, Moldova.

94 Background


But the overall patterns do not tell us how significant, costly or painfulwere the declines that did occur.13 As the HDI measure is a synthetic one, aper centage decline is not very meaningful; I suggest a useful benchmark iscomparison to other countries in the HDI tables for the same year. Thus, theBaltics fell from a level in 1990 comparable to Malta and Argentina, to onein 1995 comparable to Mexico and Trinidad-Tobago. All these values are stillin the range of the UNDP Report’s category of ‘High Human Development’.For the SEE the small decline was the equivalent of falling from the level ofPanama to that of Saudi Arabia and Thailand, with the initial and subse-quent values all in the top half of the Medium Development range. For theCISM group the more substantial decline can be envisioned as a decline fromthe bottom of the High Development range (about the same as Mexico,Trinidad-Tobago) far down the Medium Development range (about the sameas Turkey and Ecuador). The recovery to 2001 has not lifted them much inthe relative ranking, as most other developing countries have experiencedcontinual improvements in HDI. Individual countries do not deviate muchfrom this picture; only Tajikistan’s HDI falls below 0.700 in 1995 to a valueof 0.665, which still put it solidly in the middle of the Medium Developmentrange, at about the same level as Algeria and Syria, and incidentally onlyslightly below China’s 0.679 value.

The above comparisons suggest that even the worst declines in well-being to1995 were probably short of disastrous, and kept these countries’ life-standardsin a broad sense well-above the levels of the least developed, that is theUNDP’s category: Low Human Development with HDI values below 0.500. Onthe other side of the ledger, it must be emphasized that HDI declining thismuch while in most other countries it was continually rising, is, if not disas-trous, highly disappointing; China’s HDI of in 1990 began well-below that ofthe CIS (0.624 v. 0.772) and had by 2001 nearly caught up (0.721 v. 0.737).The averages alone, which of course do not capture any internal dispersionin these countries, again show, as with output loss data, that there was unques-tionably a significant cost to transition. Some of the evidence on the disper-sion and specific dimensions of social costs is discussed next.

Change in income distribution: the Gini coefficient

While the overall average HDI values show significant deterioration only forthe CIS countries, the distributional impact was doubtless more severe, andthe compensated Pareto-optimal criterion set as a benchmark for this assess-ment dictates a closer look at statistics for income distribution and poverty.These measures are fraught with measurement difficulties even in marketeconomies where institutional arrangements are relatively stable over time,and these difficulties are compounded by the uncertain comparability ofdata in the socialist regime before 1989 and the market transformationperiod in the 1990s. The most recent review of inequality trends, Grun andKlasen (2001), addresses this issue and, relying on earlier detailed work of

Atkinson and Micklewright (1992), Milanovic (1998) and Flemming andMicklewright (2000) conclude ‘no strong case can be made that … data fromsocialist countries systematically understate true inequality’.

This may be too sanguine, because the reason a strong case cannot bemade of how much privileged access to goods and in-kind services14 down-ward biases a Gini coefficient is that no hard data is available. An inability tomake the case is not proof that there is no case to be made. The same can besaid for the costs of queuing – even a simple arithmetic of hours spent inqueues is not possible, and least of all is it possible to put a value on the costsof frustration at not getting the goods despite queuing. The last is not aminor nuance or splitting of hairs – in this chapter we are in the realm ofwelfare broadly defined à la HDI, where political empowerment is – as itshould be – an element of the well-being index; the frustration of queuing issurely in the same dimension of human satisfaction.

None of the authors cited is able to assess these critical characteristics ofincome–consumption links of the communist regime. Their efforts are limitedto comparisons of survey design and coverage, the importance of in-kindearnings, the role of rationing of goods, but do not attempt to value queuingtime. In fact, even on these matters, the cited works – while not able to makea strong case of bias – do admit there are biases, and generally in the directionof understating inequality in the socialist period. But without hard evidencethey conclude effects are small. Atkinson and Micklewright (1992), Flemmingand Micklewright (2000) note that Soviet data were ‘somewhat less reliable’;in-kind earnings were ‘not significantly’ more important in socialist coun-tries; and were ‘not generally’ more unequally distributed. If one focuses onthe italicized words, their sanguine conclusion can be drawn about compa-rability; if one focuses on the non-italicized ones, much more caution iswarranted. Indeed, the best interpretation of these technical assessments isthat there remains a suspicion of biases in the direction of understatinginequality in socialist countries. Milanovic (1998) puts it nicely (p. 324) ‘Wecannot do much [or anything] to remedy them.’ Indeed, he goes slightly fur-ther and expresses the belief that for Eastern Europe (our Central Europeand SEE) any effects are of a small magnitude and make comparisons overtime meaningful, while for the Soviet Union comparisons are ‘much lessreliable … [and] were biased … to show higher average incomes and lowerinequality’. The reason he cites for this conclusion is the selective oversam-pling of two-employee parent households, which does not even account forfurther possible biases of the goods-access sort.

What to make of all this? I would argue that while a strong quantitativecase for systematic downward bias in socialist measures of inequality cannotbe made, a strong case can be made that any biases were generally in thatdirection, and particularly so for the Soviet Union. Milanovic is right to saythey cannot be corrected in the data, but adds that ‘these caveats …should … provide some caution when it comes to interpretation of the

96 Background


results’. Here I will in fact use these data as they are, with the caveat that forCentral and South-East Europe the bias is likely small enough to ignore, butfor the countries formerly within the Soviet Union it is much larger, hencethe deterioration in any inequality measure is ‘almost certain to appearlarger than the actual change’ (Milanovic 1998, p. 324). Most important,some spotty evidence on very high poverty ratios in the Soviet periodsuggests the data on Gini values is distinctly biased.

Consider first the most general measure of income distribution, the Ginicoefficient, whose values theoretically range from ‘0.0’ with all incomesexactly equal, to nearly ‘1.0’ with a very few individuals enjoying most of thesociety’s income. Historically, typical Gini values were about 0.20–0.25 insocialist economies, 0.30–0.40 in advanced, industrial economies, and0.35–0.45 or more in developing countries.15

Turning to the available data for transition economies, it is not surprisingthat a good single data source cannot be readily identified. The UNICEFTransmonee database is the only source that gives Gini values for most of the countries from 1989 to 2001, and these are shown in Table 3.5: theTransmonee group average is the first row in each cell, the second is the

Table 3.5 Range of Gini estimates in the literature

Pre-transition Mid-transition Most recent1988–92 1993–96 1998–2001

Central 22 29 29Europe 20–29 25–36 26–39Transmonee average 19–28 (36) 22–35 24–31Transmonee rangeOthers range

Baltics 25 35 3624–25 28–39 32–4023–25 31–35 32–37

SEE n.a. 27 33(3 countries only) 16–26 23–29 28–40

16–23 25–33 28–41

CISM 27 42 4625–32 32–49 39–5223–30 (45) 34–50 29–52

CISL 25 n.a. 3323–26 n.a. 27–3423–28 n.a. 27–41

Source: UNICEF, Transmonee database (online) for the first two rows ineach cell; others are: UNDP, HDR various years; UNDP (1998); EBRDTransition Reports various years; Milanovic (1998), Commander, et al. (1999),Svejnar (2002), Popova (2004), Kolodko (1999), Notten (2004), Keane andPrasad (2002), Sukiassyn (2004), Berry and Karin (2003).

range of values across countries. Apart from the fact these are based on grossearnings – and therefore may overestimate inequality by excluding socialtransfers which we know took place16 – it turns out they are not always sim-ilar to the values found in the very large number of other studies and officialdata sources. Therefore, Table 3.5 shows, as the third row in each cell, therange of values found in a dozen such sources. These differences are too largeto make one comfortable relying on one source only simply because it is theonly one to have a time series. But all sources clearly coincide on the impor-tant conclusion, that the direction of change was towards an increase of theGini coefficient, that is a greater inequality of income distribution.

It must be emphasized that Table 3.5 is no more than a compilation fromabout a dozen studies culled from the literature, showing the range of Ginivalues for individual countries in the relevant groups, without adjustmentfor methodological differences: different years, different definitions, differentsurvey methods. The compilation is meant only to show that the variation isvery large, and conclusions about changes in the Gini coefficient must betreated with extreme caution. Grun and Klasen (2001) already noted thewide range of estimates for individual countries from just three differentsources. With a lot of new data, Table 3.5 shows an even wider range thanthey observed. This is partly due to cases of just one country in a grouphaving a much higher Gini coefficient, thus the significant outliers are iden-tified separately. Thus, in the pre-transition period Russia ranges from0.22 in one estimate for 1989 to 0.45 in another for 1992; in Central European estimate for Croatia shows a value of 0.36 much beyond the range for theothers in the group, 0.19 to 0.28. In the intermediate years, Russia with0.52 is again at the high-end of the group, but for all the transition countries;however, it is not that much an outlier since within the CISM several othersare in the upper forties range: Georgia 0.49, Ukraine 0.47, and Azerbaijan0.45. At the end of the period, Russia is in one study as high as 0.49, again anoutlier, with the rest of the group in the range 0.29–0.41. Similarly in the CISLgroup one sees a wide range with Turkmenistan at 0.45 and Belarus 0.27.

There are several reasons for this wide range of estimates: even forindividual countries, different surveys for approximately the same timeperiod show different results; adjustments for comparability by researchersmay lead to different estimates for the same country; the definitions vary,with Gini being calculated on earned income, net income, inclusive of in-kind transfers, adjusted for public transfers, or even in a few cases the basisis expenditures rather than income. I have decided not to attempt any rec-onciliation, partly to demonstrate an important point: the level of theGini in any year and hence the magnitude of change over the decade isextremely sensitive to assumptions made in survey designs, choice ofincome (or expenditure) basis, adjustments by researchers using the primarydata, and so on.

98 Background


Can any conclusions be drawn from such a wide range of estimates? I propose five, in approximate order of robustness:

● The wide variation strongly signals the need for caution in drawing anyconclusions.

● Despite the dispersion, it is evident that the Gini values rise in all countries.● The rise in available estimated values is distinctly greater in CIS countries

with Russia an evident outlier on the high side.● There appears to be a stabilization and perhaps even a decline in the values

from the mid-1990s to the latest period except in the Transmonee data.● There is a strong hint for at least some CIS countries that the first big

jump occurred in the first two or three years of the 1990s, that is beforetransition in any meaningful sense could have had much impact.

The first conclusion needs no further elaboration, and appears to be widelyaccepted in the literature. The other conclusions are more tentative andperhaps contentious, and merit some discussion.

On the basis of the available data used for Table 3.5, there is not a singlecountry case where the Gini does not rise from pre-transition values; thisappears to be a very robust conclusion.17 This should not be surprising andwas of course expected by all observers as an obvious effect of regime change.Socialist income precluded profit or rental income, and even wage structureswere intentionally and transparently compressed; the switch to private own-ership and market-determination of labour income necessarily widenedthe dispersion. I leave till later any normative consideration of whether therise in Gini was excessive or about as large as might have been expected.The precise amount of increase is difficult to specify given the wide range ofestimates but taking as a rough measure the difference between mid-pointsin the ranges of the three periods in Table 3.5, one can construct a roughapproximation as in Table 3.6. This shows a clear pattern across the region,with Central European countries experiencing the least increase, about3–4 points to the mid-1990s, compared to a rise in the same period of at least10 and as much as 15 points in the CISM group. For all the other groups therise is around 10 or less, though three cases of higher increase are seen:Estonia, Turkmenistan � 12–13; and Romania � 15, though the latter isonly 10–12 if one excludes the (arguably) implausible value of 0.16 in thepre-transition period.

The much greater increases in CIS countries may be systemically related tothe state capture phenomenon, but this is less applicable to SEE, and is cer-tainly not an adequate explanation for the Baltics. An additional factor is thegreater likelihood in the Soviet countries of underestimating inequality inthe socialist period as observed by Milanovic (1999). While it is not (yet?)possible to attempt any quantitative estimates of these two effects, it is surely

plausible to conclude that some part of the difference in CE’s rise of 3–4points and the 10–15 point rise elsewhere is attributable to an underestimateof the Gini coefficient in the early period. The greater entrenchment of com-munist elites and consequently greater abuse of privilege in the Soviet Unioncompared to Central Europe is, if not universally accepted, a common viewheld by Kremlinologists and political analysts of the transition.18

Did the rise in Gini peak at values reached in the mid-1990s? The availabledata seems to suggest it did for most countries, but is too weak to ascertain ifan actual decline occurred. Thus for example for Poland, Svejnar (2002) givesvalues for 1993 and 1998 at 0.30 and 0.33, while Kolodko (2004) shows apeak in 1996 at 0.34 declining in 1998 to 0.31. Two groups are clear excep-tions, Baltics and SEE with an apparent continual rise to latest years. The lastgroup is easier to understand as cases of stop and go transition. Two of thecountries, Bulgaria and Romania, experienced a renewed financial crisis inthe mid-1990s after a reform process that was at best hesitant, then undertooka fresh start in 1997–98. Macedonia continued (and continues?) to be buffetedby civil conflicts at its borders and internally. The explanation must be differ-ent for the Baltics where reforms were rapid and steadily progressing; the mostthat can be said is that the increases after the mid-1990s were much lowerthan the fist big jump in the early 1990s. The evidence of peaking is strongerfor CISM countries, and some Gini values in fact appear to fall somewhat fromtheir peaks of the mid-1990s in most countries. Even the outlier case of Russiashows a range of Gini values that is about the same for both the intermediateand latest period. The same tendency is observed for Ukraine in Berry andKarin (2003), who cites consistent estimates for the years 1999–2002 as fol-lows: 0.29, 0.29, 0.30, 0.31. There is also suggestive evidence from povertydata discussed below, showing that in Russia and Ukraine the peak of deterio-ration was about the time of the 1998 financial crisis, and that subsequently,as the strong recovery began in 2000, conditions improved measurably.

Some of the evidence suggests that in CIS countries most of the wideningof inequality occurred still in the socialist period or in the years before

100 Background

Table 3.6 Probable changes in Gini values

To the mid-90s Overall to about 2000

Central Europe 3–4 points 3–4 pointsBaltics 8–9 points 9–10 pointsSEE 8–9 points 12� pointsCISM 12–15 points 12–15 pointsCISL n.a. 9–10 points

Source: Calculated from Table 3.5 as follows: approximate differencebetween mid-point of ranges shown for each time period.


significant market reforms. For maximum possible consistency I will referhere only to sources which have data for more than one year, presumablyon a similar basis. Thus Popova (2004) gives a value for Russia in 1988 of0.27 and in 1992 a value of 0.45; Svejnar (2002) shows for 1991 a value of0.26 and in 1992 of 0.54. Commander et al. (1999) show a more gradual risefrom 1987 (0.26) to 1992 (0.29) with the first big jump occurring in 1993 to0.40. Some will conclude the jump was due to Russia’s 1992 ‘big-bang’, otherswill argue it was aborted soon and could not have had such a guide effect.Ukraine delayed any real start of stabilization and reforms until 1994–95 –as did many other CIS countries – hence any change in Gini before 1995 isclearly attributable to the socialist period or the inaction of governmentsand not to market reforms. For Ukraine, Svejnar (2002) shows Gini values in1988 and 1995 of 0.23 and 0.47 respectively, suggesting much of the widen-ing occurred before reforms.

Any conclusion about deterioration of inequality prior to significantreforms cannot be definitive given the limited and weak database.Nevertheless, it is not implausible that income distribution would widen inCIS countries before the transition started for two reasons: the sharp deterio-ration of the fiscal and inflationary situation began before 1992–93; CIScountries continued to operate a regime that was at least quasi-socialist, andcertainly did not approach market-like operations until the late 1990s,19

with an attendant sharp decline in output and hyperinflation. In addition,since the 1988 Law on Co-operatives legal opportunities for private owner-ship and profit income increased quickly. Both of these phenomena wouldresult in some early widening of income distribution.

At this point it may be useful to step back and ask the normative question:was the increase in Gini coefficients higher than appropriate for a regime switchfrom a planned socialist economy to a market economy with private ownership?There is no theoretically ‘correct’ benchmark for this, but to obtain somepurchase on the issue consider three possible benchmarks:

● the Gini values of Central European countries which, according to theHDI measure of well-being, went through the transition with virtually noworsening and certainly a net improvement after a decade of the process;

● the Gini values in other market economies, including in particular middle-income emerging markets, but also the high-income OECD countries;

● the Gini values in China, which like Central Europe saw a continuedimprovement in the HDI, and is frequently cited as a model of howtransition should have been done.

Comparative data for this purpose are shown in Table 3.7, which shows thewell-known fact that high-income countries of the OECD have much lowerGini coefficients than do middle-income developing countries, with rangesapproximately 0.25–0.40 and 0.30–0.60 respectively. Transition countries,

which as socialist economies earlier were even lower than this, now seem tofall in the middle between these two groups. Central European economieshave values in the mid-twenties comparable to the lowest of the OECD coun-tries (like Denmark, Sweden and other Scandinavian countries sometimescalled ‘welfare states’). The Baltics and probably SEE, with few exceptions,20

are in the range 0.30–0.35, hence comparable to the mid-range of OECDcountries and the low-range of middle-income emerging markets. The CIScountries show the widest range, with a few below 0.30, but several over0.40 and approaching 0.50.21 These last are then comparable to the middle-income and developing countries with much more unequal income distrib-ution. However, most except Russia still fall short of the very high cases withvalues of 0.50 and more, like Brazil, Colombia shown in the table, and Chile,Mexico, Venezuela and Paraguay. Similar high values are found in a numberof low-income African and Latin American countries as well.

From this one may conclude that increased income inequality as measuredby the Gini coefficient has certainly taken place after the transition, but com-paratively modestly for those countries in Central Europe, the Baltics and SEEwith more progress on market reforms. In the CIS cases of slower reforms, itapproaches levels at the lower end of the range for developing countries of

102 Background

Table 3.7 Range of Gini values by country groups

Lowest 2 countries Highest 2 countries

Transition countries Hungary, 0.24 Slovenia, 0.31Central Europe Poland, 0.25 Croatia, 0.36

Baltics Latvia, 0.32 Estonia, 0.37SEE Macedonia, 0.28 Romania, 0.41CISM Kyrgyz, 0.29 Armenia, 0.38

Kazakhstan, 0.31 Russia, 0.49CISL Uzbekistan, 0.29 Turkmenistan, 0.41

Market economies Denmark, 0.245 Portugal, 0.39High-income Sweden, 0.25 USA, 0.41Middle-income Indonesia, 0.30 Colombia, 0.57

South Korea, 0.32 Brazil, 0.60

Notes: Data ranges from 1996 to 2002. Ukraine (within CISM) estimates rangefrom 0.29 to 0.41 which could make it one of the lowest or one of the highest – anexcellent example illustrating non-robustness of Gini estimates. Memo item:China � 0.40.

Source: UNDP (1998) first column of table 4.5; all others UNDP, HDR, 2003. High-income countries: top 21 by HDR rank. Middle-income emerging markets: Korea,Uruguay, Chile, Mexico, Malaysia, Panama, Colombia, Brazil, Venezuela, Thailand,Peru, Philippines, Tunisia, Turkey, Jordan, Indonesia.


comparable income levels. One is tempted to say that with few exceptionsthe advanced transition countries are about where they should be on incomedistribution for a high-income market economy, and the lagging ones are alsoabout where they should be relative to middle-income market economies,although generally better than the most unequal distribution cases.

As to China, the UNDP’s most recent estimate for 1998 is 0.40, well in theupper range of the CIS countries. If one takes the high values of Table 3.5,China is exceeded only by Georgia, Armenia and Russia. Other studies showsomewhat lower values for China, about 0.32–0.35. Comparing with theranges in Table 3.5, it is clear that China’s income distribution by the end ofthe 1990s is certainly far less equal than that of Central Europe, slightly lessequal than in the Baltics, and comparable with the range observed for South-East Europe and most CIS countries excluding outliers. It is useful to look atthe changes in Gini values for China. Two recent studies, Gustafsson andShi (2001) and Benjamin, Brandt and Giles (2004), provide a thoroughanalysis over time, differentiating rural and urban Gini values. As they usedifferent methodologies, the results do not coincide precisely, but bothconclude that a substantial rise in the Gini coefficient took place. The first ofthese shows urban Gini values (income) rising nine points from 0.23 in 1998to 0.32 in 1995; the second for rural China shows a six-point rise from0.29 in 1987 to 0.35 in 1999. They attribute the change to falling agriculturalincomes as food prices declined, and to the rise of income opportunities innon-agricultural activities in rural areas. This reflects the well-known Chinesesuccess of liberalizing reforms in rural areas. What is most important in thiscomparison is that despite avoiding a net output decline, China experiencedincreased inequality of a magnitude not much different than the average fornon-Asian transition economies.

The explicit policy aims in China were to introduce market mechanismsfor their motivational and efficiency benefits, while receding only very grad-ually from various socialist instruments.22 Since China is generally consid-ered a success in its own right even by those who do not believe its approachcould have been used elsewhere – including the present author – it seemsreasonable to infer that an acceptable and natural change in the Gini coefficientmight be at least five and perhaps as much as 10 points. The China benchmarkleads to the conclusion that none of the Central European countriesexceeded this ‘reasonable’ increase (indeed they were generally at the lowerend), and very few of the others did so. Depending on which Gini measuresare correct, perhaps only Armenia, Russia, Turkmenistan and Ukraine didexceed the China benchmark by a few points, at least to the mid-1990s.

Changes in income distribution: poverty ratios

We turn next to direct measures of poverty, using in particular the poverty-line concept and the measure ‘per cent of population below poverty line’, orpoverty ratio. General caveats on data have been made earlier,23 but some

difficulties and inconsistencies particular to poverty measures should benoted. First, however, a point on normative interpretation is warranted inadvance. The reader will have understood from the preceding discussion onthe Gini coefficient that I argue some increase in its value from the socialistperiod is not only ‘normal’ (unfortunately an indefinable magnitude), to beexpected, and in fact a desirable part of the regime change to the extenthigher income opportunities not available under socialism are a motivatingfactor for both labour and capital to exert greater efforts and enhance pro-ductivity. The empirical evidence hints at the view that the extent ofincrease seen in Central Europe and perhaps in other countries (to about0.25–0.30 at least) is probably within this ‘desirable’ range. The same argu-ment is categorically not applicable in the case of indicators showing peoplebelow the poverty line; any long-term increase of this indicator immediatelyfails the compensated Pareto-optimality test set out at the beginning of thismonograph, and the word ‘deterioration’ is most appropriate. Simply put,governments of transition economies should at a minimum have ensuredthe necessary support and transfer mechanisms to avoid any worsening ofpoverty beyond a short term of, say, five years.

Unfortunately, enormous data and poverty-line definition problems mayexaggerate the extent of this deterioration, and furthermore underestima-tion of poverty in the socialist period analogous to that for Gini measuresadds to the bias of exaggeration. The most comprehensive coverage of esti-mates for the poverty ratio is found in EBRD annual Transition Reports from1998 which give values of the ‘international poverty line ratio’. But again, aswith the Transmonee source for Gini values, there are internal inconsisten-cies and there exist several other sources which give sometimes very differ-ent values either because they use the national poverty line (most commonin the transition literature – UNDP, 1998; Milanovic, 1998). Therefore, it willagain be best to show both the EBRD values and those in other sources in ourcompilation. Table 3.8 gives individual country values from four sources tohelp track evolution over time, while Table 3.9 shows ranges for countrygroupings to allow an easier visualization of the tendencies.

The first shortcoming of the EBRD compilation is that the year to whichthe data applies is not indicated except in the case of the 2003 Report, whereit ranges from about 1998–2000, with a few cases of 1996 and 2001. It is thusdifficult to assess the change in poverty over the transition, and for the CIScountries in particular where significant recovery only started in 1999–2000,difficult to see if there was a consequent decline in poverty. Also, the compi-lation switches sources and apparently definitions: from 1998–2000 thesource is UNDP and the poverty line is 4$/day at PPP; in 2001–02 it is WorldBank survey data at 4.3$/day; in 2003 it is the World Bank’s WorldDevelopment Indicators at 2$/day. Does this explain the dramatic declinesseen from the 2002 Report to that in 2003: Hungary 15.4 to 7.3; Poland 18.2to 2; Georgia 30.9 to 15.3; Kyrgyz 84.1 to 34.1? Maybe, but in that case the

104 Background

105

Table 3.8 Poverty ratio estimates: various sources

Pre-transition Mid-transition Most recent

Central EuropeHungary 1/2/2/1 3/15/25/4 -/7/-/-Poland 6/13/-2 19/18/24/20 -/2/-/-Czech R. 0/1/-/0 1/1/-/1 -/2/-/-Slovak R. 0/1/-/0 1/1/-/1 -/2/-/-Slovenia 0/1/-/0 1/1/-/1 -/2/-/-Croatia n.a. n.a. n.a.

BalticsEstonia 1/-/-/1 40/40/9/37 -/5-/-/-Latvia 1/-/-/1 25/35/-/22 -/8/-/-Lithuania 1/-/-/1 46/23/-/30 -/14/-/-

South-East Eur.Bulgaria 2/-/-/5 33/33/36/15 -/24/13/-Romania 6/-/-/6 39/45/22/59 -/21/-/-Macedonia n.a. n.a. n.a.Albania -/-/-/- -/-/29/- -/-/25/-Bosnia-H. n.a. n.a. n.a.Serbia-M. n.a. n.a. n.a.

CISMArmenia 14/-/-/- -/86/55/- -/49/54/-Georgia 13/-/-/- -/60/12/- -/12/-/-Kazakhstan 5/-/-/5 50/50/35/65 -/15/-/-Russia 2/-/-/12(10)(11) 45/50/31/50(-)(32) -/24/-/-(-)(15)Kyrgyz R. 12/-/-/12 84/84/51/88 -/34/64/-Moldova 4/-/-/4 65/85/23/66 -/64/-/-Ukraine 2/-/-/2 (12–30) 41/41/32/63 (55) -/46/-/-/ (45)Azerbaijan 34/-/-/- -/64/68/- -/9/50/-Tajikistan 51/-/-/- -/96/-/- -/51/-/-

CISLUzbekistan 24/-/-/24 47/31/-63 -/44/28/-Belarus 1/-/-/1 23/47/23/22 -/12/42/-Turkmenistan 12/-/-/12 57/58/-/61 -/44/-/-

Notes: Pre-transition is before 1992 for Central Europe, before 1993 for others;mid-transition is respectively 1993–96 and 1994–97; most recent period 1997 and later,1999 and later. EBRD Reports show the year of survey only in the 2003 reports, henceI have inferred years from earlier reports as follows. Central Europe, Reports of1998–2000: assumed to be pre-transition and only highest value shown here; reports of2000–02, assumed to be for mid-transition. For all other countries values in Reports1998–2002 are assumed to be for mid-transition. This is partly based on a judgment ofhow the order-of-magnitude of values in EBRD correspond to other sources.

Sources: The four values of poverty ratio estimate, separated by the forward slashes,are from the sources: UNDP (1998); EBRD TR; World Bank WDI; and Milanovic (1998)in that order. For Russia bracketed values is from Commander et al. (1999), Notten(2004). For Ukraine Berry and Karin (2003).

time-series data in this source can hardly be used to make an analysis overtime. The questions do not stop at this, because only the values shown in the2003 Report can be reconciled with the World Bank publication of WorldDevelopment Indicators (WDI); values in earlier reports cannot be reconciledwith published World Bank data.

The World Bank publication is also problematic, because it shows very fewvalues for the earlier part of the 1990s precluding an assessment of whathappened in the first part of the transition,24 when of course the expectationwould be a deterioration of poverty. Fortunately, this deterioration trend isevident in the EBRD data as well as the key 1998 study on poverty by theUNDP. These data conundrums suggest a better approach is, as with Gini, tomake a compilation with a range of estimates from different sources ignor-ing the methodological differences, and only refer to a single source when itgives useful comparative insights over time or across countries.

As with Gini coefficient estimates, the first striking fact about povertyratios in various sources is the very wide range. The main reason for the largedifferences is definition of the poverty line, whether it is national or inter-national, the threshold (for example for international it is sometimes 4$/dayPPP, sometimes 2$/day), the use of income or expenditure basis, and differentsurvey coverage. Just as an example of the large sensitivity, the mid-1990snational poverty rates shown in Milanovic (1998) for Kyrgyzstan and Ukraineon an income basis are an astonishing 88 per cent and 63 per cent – higherthan least developed African countries according to WDI data – whileEBRD in its first detailed analysis of social impacts (Transition Report, 1999,Annex 1.1), calculates that with expenditure base the rates fall to about half,46 per cent and 37 per cent respectively. Any effort to find consistent data,to reconcile different sources and definitions, or compile a time series forindividual countries for a consistent definition ineluctably leads to two con-clusions. First, it’s not possible to get enough consistent data for enoughcountries over the full period to allow a definitive assessment – or at least itwould be a huge undertaking. Second, it is clear that the high-end estimatesof the poverty ratio in the literature may be 2–3 times too high, because they

106 Background

Table 3.9 Range of poverty ratios by country group and period

Pre-transition Mid-transition Most recent

Central Europe 0–13 1–25 2–7Baltics 1 22–46 5–14S.E. Europe 2–6 15–45 13–24CISM 2–30 (51 Taj.) 12–96 12–64CISL 1–24 22–63 12–44

Source: Table 3.8, lowest and highest values in each group for each period.In CISM, Tajikistan is an outlier at 51%, and is thus shown in brackets here.


reflect simultaneously several upward biases: use of income rather thanexpenditure, a national rather than international poverty line, and an unre-alistically high poverty threshold (for example 4$ a day rather than the $2 or1$ commonly used in WDI for international comparisons).

The wide range is especially evident in the Former Soviet Union (FSU).Thus, in the middle period Estonia ranges from 9–40 per cent, Lithuania30–46 per cent, Russia 12–50 per cent, Ukraine 41–63 per cent and Georgia11–60 per cent. Outside the FSU, only for Bulgaria (15–36%) and Romania(22–59%) does one see similar wide dispersion, while in contrast themaximum variation in Central Europe is for Hungary at 7–25 per cent.

Can one infer anything about changes in poverty ratios since the start oftransition despite the wide range of estimates? The answer as with Gini valuesis clearly yes, and equally clearly that in the first five to seven years of tran-sition poverty increased substantially – though the high-end estimates maybe exaggerated. The biggest changes, as with the findings on Gini, are in theCISM group.

Many of the middle-period values in the CISM countries are so high thatthey are on the face of it doubtful: is 88 per cent for Kyrgyz Republic and66 per cent for Moldova truly plausible? These values would seem to bespeaka disaster of starvation or nearly so, and are neither consistent with the anec-dotal knowledge of the situation in CIS countries in the mid-1990s nor withthe observation made earlier related to the HDI values falling no farther thanto about the middle of the Medium Human Development range. Both theHDI and the Poverty in Transition report from which such values are taken areproducts of the UNDP, and to put it charitably they cannot both be right. Itis likely the estimates of the cited report are exaggerated by the use of apoverty line which is unrealistically high. The effect can be substantial,using the EBRD (2002) report with a poverty line of $4.3/day the CISM aver-age poverty ratio is calculated as 62 per cent. Using the (2003) report with apoverty line of $2.15/day, the CISM average falls by nearly half to 35 percent, a value generally below that found in the very low-income African andAsian countries. But even 34 per cent is a large increase from the valuesthought to prevail in the socialist period, that is well-below 10 per cent forthe western republics and moderately above 10 per cent for the others – withsome exceptions as seen in Table 3.8. However, here too great caution isneeded: as with the Gini coefficient, there may have been Soviet-periodbiases in the data showing less inequality and less poverty than in factexisted. This is clearly suggested in two detailed country studies.

A very detailed study of poverty in Ukraine, Berry and Karin (2003), pro-vides some further caution about interpreting these values noting the fact ofunofficial economy activity being about equal to official at this time. The studyalso provides a hint of the unrealistic levels of poverty-line definition: for1990 this is set at 100 karbovantsy (�100 Soviet roubles), which was a verytypical wage for a large majority of the population then. If projecting this

poverty line forward with inflation results in the 41 per cent poverty ratio inthe mid-1990s given by the UNDP (1998) Poverty Report or 63 per cent inMilanovic (1998), then either the poverty line is being set unreasonably highor the number of people in poverty in the socialist period was also very high.This last is a real possibility: for Ukraine Berry and Karin (2003) give povertyratio values for 1980, 1990 and 1992 respectively of 38 per cent, 12 per centand 30 per cent. These are for the 100 karbovantsy poverty line; for a lowerpoverty line defining the ‘ultra poor’ (83 k. which is not much lower and inmy mind still doubtful), the percentages for the above years are: 23 per cent,6 per cent and 17 per cent.

For Russia, Commander et al. (1999) also show initial values higher thanUNDP (1998), though much less so than Berry’s Ukraine numbers: 1990 �

5 per cent, 1991 � 10 per cent. As a further indication of how the Soviet-period official values may have been downward-biased, they refer to a studyof McAuley (1979) which suggests a poverty ratio in the late 1960s as high as35–40 per cent. These estimates strongly suggest that the initial pre-transition levels of poverty were rather higher than UNDP values show, anal-ogous to the evidence of higher Gini coefficients already noted. Note alsothe anomalously high values in the pre-transition period for Azerbaijan(34%) and Tajikistan (51%), compared to about 12–14 per cent for Armenia,Georgia and Kyrgyzstan, also very low-income republics. Given the nature ofthe regime, it is difficult to believe that the transfers to low-income republicswhich kept poverty lines there not much above 10%, were not applicable tothese two republics.

A comprehensive study of poverty in the Soviet Union by Matthews(1986) points to several sources of downward biases in official poverty ratios,even as individual regional studies in the 1970s showed far higher ratiosranging from 15–18 per cent in richer places like Moscow and Estonia,20–35 per cent in middle-income regions (Odessa, Novosibirsk) to 50–60 percent in low-income locations (Tuva). At a minimum, this casts great doubton the low-end estimates of single digits in the pre-transition period forUSSR successor states, including even the Baltics. If we take the Soviet-periodMatthews value of 18 per cent in Estonia as indicative for the Baltics andcompare the values in Table 3.8, we can conclude that poverty increased tothe mid-1990s by 10–20 percentage points, but by 2000 or so it fell sharplyto levels well-below the Soviet period. Berry and Karin (2003) makes anexcellent point that may explain the large differences: ‘many families were[never] too far above the poverty line’ and therefore it did not take much ofa decline in output and consumption to push a lot of people over the edgeinto ‘official’ poverty. Similarly, Commander et al. (1999) show that ‘Russiaentered the transition with a significant degree of inequality … significantlygreater than in … Central Europe.’

Nevertheless, this does not negate the conclusion that outside of CentralEurope there was a more than marginal increase in poverty, but rather a

108 Background


significant one.25 Even taking roughly the top values for the early period(10–12%) and the middle of the range in the intermediate period, thepoverty ratio jumps in FSU countries by about 20–30 points to values inmany countries of at least 40 per cent or more.26 This is by no means a smallnumber; poverty did increase dramatically, and it must have meant a greatdeal of suffering and pain for many individuals in the FSU countries, even inthe Baltics, for at least half a decade, probably more.

Consider briefly the other country groups. For much of Central Europe theincreases can indeed be considered negligible, and the levels not much above1–4 per cent, similar or even lower than some rich countries’ poverty ratios.Hungary and Poland were exceptions, with some estimates up to 15–20 percent respectively in the mid-1990s using the national poverty line, thoughthese have fallen a lot if one uses the WDI international line definition: from11 per cent and 15 per cent in 1993, to 7 per cent and 2 per cent in 1998–99.In the Baltics, initial trends were similar to that of other former republics,though the peak levels reached were rather lower, in the range 25–40 percent, but by the end of the period they all had values of 5–15 per cent, closerto Central Europe. Using an international line basis, the WDI shows for 1993values of 20–30 per cent, and for 1998 already 2–14 per cent.27

The extent of poverty decline in the SEE group is also very large, andindeed close to the CISM experience. In some cases it was probably exacer-bated by civil conflicts, but this was not a factor in Bulgaria and Romaniawhere the ratio reached 35–45 per cent, or even higher in some estimates.Thus a deterioration of 20–30 points as in CISM is a reasonable assessment.Finally, the changes for the least-reformed economies are, as often for thisgroup, open to interpretation. Those who want to argue these slower, grad-ual, reformers minimized social costs, will point out that national povertyline ratios rose less than in the CISM. Thus, UNDP (1998) notes in Belarusamong the western CIS, it rises only to 23 per cent, compared to 40–60 percent in the others; similarly, Uzbekistan had a value of ‘only’ 47 per cent, lowfor Central Asia. Those who argue the slow reforms did not help much tomitigate social costs (including this author) would counter with threepoints. First, other estimates (Milanovic, 1998) give values for Uzbekistanand Turkmenistan that are much higher and similar to the other CIS cases:63 per cent and 61 per cent. Similarly for Belarus, Demidkina in Heenan andLamontaigne (1999b) contends it may have been as high as 80 per cent, andactual unemployment not the 4 per cent official estimate but over 30 percent. Second, using an international line comparison as in the WDI 2004, formost recent years (2000–01) the average for CISM is 33 per cent, whileTurkmenistan (44%) and Uzbekistan (77%) are much higher, only Belarusremaining at very low levels (2%). Third, both Turkmenistan and Uzbekistansaw a continuing deterioration, from 1993 values of 26 per cent and 32 percent, while many of the CISM countries experienced a decline in povertyratio once the recovery began. I turn now to the recovery period trends.

Was there a general stabilization or even reversal of poverty ratios since themid-1990s? For Central Europe and the Baltics, some reversal is alreadyapparent by the late 1990s, as the last column of Table 3.8 shows, but thisconclusion is very tentative given the definitional change problem in theonly data compilation which gives very recent years, the EBRD. If oneignores the warning of the World Bank’s WDI that international povertyratios ‘cannot be compared with values reported in previous editions forindividual countries’,28 the apparent trend is a decline in the ratio from itspeaks for all but one country in Central Europe and Baltic group, also adecline for Bulgaria and Romania (no early years data for others). For theCISM, six of nine saw a decline (though marginal in case of Armenia: 55 percent to 49 per cent), while two saw an increase (Moldova and Ukraine); nodata given for Tajikistan.

Some country-specific studies of poverty also show evidence of a declineafter 2000. Notten (2004) gives values for Russia for 1994, 1998 and 2001 as12 per cent, 32 per cent and 15 per cent, respectively. The pattern here is asharp rise to the 1998 crisis and then a substantial improvement as the econ-omy recovered from 2000. Berry and Karin (2003) find a similar pattern forUkraine, but a less dramatic improvement from 2000, with an inferreddecline in the ratio from 55 per cent in the mid-1990s to 50 per cent or a bitless. On balance, a tentative conclusion can be made that the deteriorationin the poverty ratio has either flattened out, or even begun to reverse as out-put recovery translated into higher incomes, and the improved fiscal situa-tion perhaps also resulted in better social safety nets.

The worsening of poverty in the aggregate may have been mitigated by theshort duration and the large turnover seen in labour-force dynamics: do thepoverty percentages reflect a fixed group of people, or are there movementsin and out of poverty over short periods? If so, as EBRD Transition Report 1997argues, an additional relevant statistic is the share of population that ischronically or persistently poor, this being defined as falling under thepoverty line throughout a 2–3-year period of a longitudinal survey. A strik-ing result comes out: the share of chronically poor in the mid-1990s inHungary was 12 per cent, but in Russia it was no higher, and indeed in theearly 1990s Russia was lower at 8 per cent. One interpretation is that the sta-ble part of poverty in Russia (and perhaps in other CIS countries) is about thesame as in Central Europe, but the volatility of economic conditions (out-put, inflation, exchange rate movements) result in any one year in sharp upsand downs giving the higher poverty ratios typically observed. This resultdoes not reverse the basic conclusion that, in particular in CIS countries, alarge rise in poverty occurred with its attendant social costs: any reversalseen so far in that region is too little to be considered as compensation forabout a decade of suffering for a large percentage of the population. Onbalance, the CIS countries did not meet well the Pareto-compensatedcriterion.

110 Background


Life-expectancy: best proxy for health status

Among the most often cited, dramatic and undeniable statistics reflectingthe individual costs of transition is the figure of male life-expectancy inRussia ‘which declined by six years from 1989 to 1994’ (EBRD TR, 1999),specifically from 64 to 58 years, a level to be found only in the lower-middle-income countries. While this figure is an extreme, it serves well to focusattention on the fact that some of the transition countries did experiencesignificant declines in health conditions. It is widely recognized that life-expectancy is an excellent proxy for other health indicators, and is for thatreason the indicator the HDI formula uses it to represent health status. Inthis section I will use the male life-expectancy, as did the UNDP (1998) tran-sition poverty study, in order to focus even more narrowly on the specificproblem area.29 The quality of this statistic may be the highest for all thedata so far used, as it is a simple enough definition subject to little alterna-tive interpretation and generally thought to be reliably estimated in thesocialist period, though some doubts are raised by analysts. For example,Handelman (1994) suggests official data overstated life-expectancy in theUSSR, with informal sources giving a value of 63 years rather than 68 as inofficial data. While possible, this is less likely for these simpler data than forGDP or poverty indicator estimates.

If the six-year decline is the upper bound, what were the trends for the restof the region over the decade? As in our preceding analyses, the mainquestions asked of the statistics will be: how much of a change was there?how did this vary across the region, and in particular how was it related tothe pace of reforms? has there been a recovery from the mid-1990s? It willnot surprise the reader that the answers to the last two questions are verysimilar to the answers for all the other economic and social indicators.Table 3.10, based on the 2003 UNDP Human Development Report, summarizesthe changes in this indicator.

The Central European countries as a group and individually experiencedno decline between 1990 and 1995, indeed there was a substantial increaseto 70 years by 2001. In the early years of their transition, some of them did,however, experience minimal declines of 1/2 to 1 year to about 1993, so thephenomenon was marginally observable there as well. The early recovery ofoutput discussed in Chapter 2 and the early correction of budget difficultiesallowed the health expenditure to GDP ratio to recover and indeed rise froman average of 6 per cent in 1990–91 to 7.2 per cent in 1993–94 (UNDP, 1998).For the Baltics, the first half of the period brought a visible decline in malelife-expectancy of 2� years, but this was fully recovered by 2001. South-EastEurope for the most part avoided any decline; for Bosnia-Herzegovina andSerbia-Montenegro no data are available in the UNDP source, but anydeclines there would be likely be more attributable to war conditions.

It is only when we look at CIS countries that significant deterioration isevident, though the sharp decline seen in Russia was not common. Despite

the 3–6-year declines in some countries, the overall average fell much less,2 years to be precise, and half of this was recovered by 2001. The comparisonof CISM and CISL is interesting; while the output decline for CISL countrieswas apparently much less severe using official statistics, the decline in life-expectancy on average was the same as for the CISM, about 2 years. Indeed,for Belarus where the authorities and many outside observers proclaimthat their very gradual approach to the market avoided the pains seen else-where, the decline in male mortality was 3 years, second only to Russia’s andhigher than most other CISM countries which ventured farther on economicreforms. Furthermore, the recovery to 2001 is very modest in Belarus (64.0 to64.3) compared to a recovery of 1 year in Russia and Ukraine, 1.5 in Moldova,1.7 in Tajikistan and 3.1 in Kyrgyz Republic.

What kept the CISM average so far below the extreme Russian values?The declines were very minor in Armenia, Georgia and Kazakhstan, whileAzerbaijan actually experienced a modest increase throughout, rathersurprising for the Caucasus given the conflicts that took place and the hugereported decline in output. Tremendous resilience of consumption and self-care in such regions is one explanation (see Stillman and Thomas, 2004, forsuch an analysis of permanent income effects in Russia as well).30

The reasons for the decline in life-expectancy warrant a few observations.It has been noted that Central Europe’s early recovery and ability tomaintain public expenditures was a factor. Also, the inefficiency of healthdelivery reforms in western CIS countries (see for example the analysis ofclosure of health facilities in Jack, 2002) explains the poor results theredespite a very early recovery of health expenditure by the mid-1990s. Otherstudies have provided a thorough review of the linkages between the processof transition and health indicators such as life-expectancy: EBRD Transition

112 Background

Table 3.10 Male life-expectancy trends in transition

Life-expectancy (years)Type of trend (no. of countries)

Decline � Decline – No change1990 1995 2001 Rising surpass part recovery Decline (1)

CE 67.4 68.2 70.0 6 – – – –Baltics 66.0 63.8 66.2 – 3 – – –SEE 68.5 68.5 69.1 3 – – – –CISM 66.2 64.1 65.1 1 2 4 1 1CISL 65.3 63.3 64.7 1 1 1 – –

Notes: 1: The trend of life-expectancy rates is characterized as follows: rising � continualincrease from 1990 value; decline � surpass � decline after 1990, but recovery to reach value in2001 higher than 1990; decline – part recovery � decline with recovery to 2001 still below 1990;decline � continual decrease 1990 to 2001; no change � stable level in all 3 years.

Source: UNDP, Human Development Report, 2003.


Report (1997), UNDP Poverty in Transition (1998) and Micklewright (1999) areexcellent sources pointing to the following combination of key factors: thelarge drop in income, reduced budgetary resources for healthcare, poor pol-icy coordination of the change in healthcare ‘from excessive security to theuncertainties of the market’ (UNDP, 1998, p. 39, the stresses associated withjob-losses) or non-payment of wages even for the employed, high levels ofalcohol and drug abuse especially for males, and increasing incidence ofHIV/AIDS.

To summarize in brief, life-expectancy trends again demonstrate thatCentral Europe, and to a lesser extent South-East Europe and the Baltics,either avoided altogether any deterioration of health standards as experi-enced a relatively minor decline, fully recovering or even surpassing initialstandards. It is the CIS countries that experienced the greatest deterioration,again confirming the hypothesis that partial reforms may cause greateradjustment costs than more rapid and more complete reforms. Unlike eco-nomic indicators, however, life-expectancy trends in those countries whichhave undertaken least reform, the CISL, deteriorated about as much as in themore advanced CISM reformers.

One interpretation of the latter is that the official output was upward-biased in the CISL, either through data manipulation, or the simpler possibil-ity that valuation with unreformed Soviet prices and accounting overstatedthe value of production. In any event, the evidence on life-expectancy addsto that from Gini and poverty-ratio data, to suggest the CISL group’s model ofvery gradual market reforms either shows much overstated output perfor-mance, or is unable to translate this better output performance into signifi-cantly better results for the well-being of the population compared to theCISM group.

Education indicators: enrolment ratios

A well-known criticism of Russia’s approach to economic reform byReddaway and Glinski (2001) describes the result of applying the ‘shocktherapy’ supported by many Western institutions and advisors as follows:‘Russia from at least 1990 has been sinking into turmoil or decay’. In otherparts of this book, I take issue with their assessment of the extent of decayand in particular the reasons for it. But on one dimension, the deteriorationof the existing human capital base and the educational infrastructure thatprovided it, their concern is entirely valid: Russia which is their focus, as wellas the other non-Baltic countries successor to the USSR, have been ‘precipi-tously losing … status as intellectual powers … and losing huge numbers ofscientists’. In Russia they note the number has fallen from 3.4 million in thelate 1980s to 1.3 million at the start of the current decade. Indeed, educa-tional standards appears to be the one area, even more than with the caseof life-expectancy, where data is not easily exaggerated in favour of one’sargument, and hence critics and proponents of rapid reform are largely in

agreement.31 But yet again deterioration in education is barely visible, if atall, in the countries that have moved fastest on economic reform, CentralEurope and the Baltics. The CIS countries are the ones where the problem ismost acute, as seen in Table 3.11.

There are some problems with the CIS data in particular, as the UNDPsource gives values in 1995 and 2001 of 73 per cent and 72 per cent in theCISM, and 78 per cent and 81 per cent in CISL. Averaging the two givesabout 76 per cent and 78 per cent respectively. These are not only implausi-bly higher than in the Central European and Baltic countries, but are incon-sistent with the EBRD (2002) values for the CIS as a whole of 60 per cent and62 per cent (whose source is UNICEF). There is no inconsistency between thetwo sources for the other country groups. Given this anomaly, I consideronly the EBRD data for all CIS, which clearly show a sharp decline in grossenrolment rates for the CIS from a high of 80 per cent in 1990 to 73 per centin 1995. No recovery is seen using UNDP data, indeed a small drop occurs,while the EBRD data show a partial recovery. It is significant that the averagemasks some of the more dramatic declines in the war-torn and less-developed economies of the Caucasus, Central Asia as well as Moldova. Evenworse, the deterioration is greater at the basic education level as noted inEBRD (1999). In 1989, all the republics of the USSR enjoyed broadly similarbasic education enrolment percentages for the population between7–15 years, at about 94–95 per cent even before transition. These ratiosdeclined, in particular for the more agricultural economies such as Moldova,Caucasian countries and Central Asia, reaching levels of 84–85 per cent by1995. While they have begun to rise since, much ground has been lost.Enrolment at the secondary and tertiary level has not fallen as much, henceif the numbers used to calculate the gross ratios shown in Table 3.10 are

114 Background

Table 3.11 Gross educational enrolment ratios (%)

1990 1995 2001

CE 71 72 78Baltics 70 71 86SEE 66 64 67CISM 80 73 72CISL [84] [78] [81]CIS (EBRD) 66 60 62

Source: 1995 and 2000 figures are from UNDP, HDR, 2003; the value forTurkmenistan is about 90%, far higher than any other: it was adjusted to 80%,similar to data in other sources. The values for 1990 were not given in eitherUNDP reports on a comparable gross basis, and hence these were inferred from theEBRD Transition Report 2002. For CISL this implied 86–89% for this group, farhigher than values shown for other USSR republics before dissolution. The 1995and 2001 values also seem higher, hence all are square bracketed for caution.


correct, enrolment overall appears to have recovered to pre-transition levels.For some countries, the downward trend continues: Armenia, Georgia,Moldova and Azerbaijan. But the recent numbers are cast in doubt by theanomalies and inconsistencies among sources noted above.

What about quality of education, and its reorientation to the differentskills a market economy might demand? This is of course much more diffi-cult to assess and indeed it may be too early to do so. There are three impor-tant dimensions to the quality of education: the quality of basic education,the role of private education opportunities and the reorientation of skills(see Spagat, 2002, for a good conceptual analysis of human-capital issues intransition, as well as the comprehensive review of skills and labour-marketchanges in EBRD TR, 2000). Several recent studies are beginning to throwsome light on this, with some positive conclusions and some less rosy ones.On the positive side, Munich, Svejnar and Terrell (2002) show for the CzechRepublic that reorientation to newly needed types of skills has increased thereturn to human-capital investment. UNICEF (1999) and Micklewright(1999) indicate there is a recovery of enrolment at least in tertiary education.But on the gloomier side, EBRD (1999) worries about the continued slump inenrolments at basic education levels, especially in the less-industrializedcountries, and Micklewright (1999) notes the inequality of opportunitycreated by the development of private education.

In as much as a broad and high-quality education was, by wide consensus,a positive legacy of the communist period – as was universal healthcare – itis surely a welfare-reducing effect that the process of transition to open-market economies, which was intended to bring improvements, should resultin a deterioration of this legacy. Reversing this decline must be very high onthe priority list of remaining challenges for transition country governments.

Summing up the evidence on social costs

No-one writing about or involved in the transition issue at the beginningexpected immediate benefits; everyone agreed there would be a period ofdisturbance, decline in output, unemployment, and costs imposed uponsome segment of the population. After about five years, many analysts (andpoliticians) began to make early assessments, which generally tended toemphasize the much higher than expected output and job losses, the sharpdeterioration in various social indicators of health, the growing inequalityand poverty. There was somewhat of a rhetorical tilt to these conclusions,and they seemed apparently to confirm the worst fears of the gradualismproponents that too-rapid reforms would generate huge cost which a moregradual pace was intended to minimize. The evidence in this chapter suggesttwo difficulties with such a conclusion. First, a closer look at what happenedin the first half of the decade and how this varied across country groupsstrongly suggests that the costs were far less in the early and rapid reformers

than they were in the delayed and partial reformers. Second, given the con-sensus that the process was going to be a U-curve, any assessments done aftersuch a short period would necessarily emphasize the cost phase, and beunable to capture much of the benefit phase. This is not to say it was wrongto do interim assessments – it was absolutely legitimate and indeed useful.But it should have been clear that there would be a bias towards the cost sideof the balance. Some remarks on each of these points.

There is a large amount of evidence now available for social, political andeconomic results to about the mid-point of transition, mid-nineties whenmany countries had recovered and the laggards were bottoming out. Themajority of performance measures is much more consistent with the viewthat more rapid reforms give earlier results and minimize costs, than it iswith the opposing view that gradual reforms can minimize the costs. Theadvanced reformers of Central Europe and the Baltics had by 1995 donemuch better than the slower reforming CIS – with South-East Europe some-where in between – on such indicators as output recovery, inflation, FDI percapita, democracy indicators, the broad measure of well-being captured bythe UNDP’s HDI, as well as the more specific social indicators such asinequality measured by the Gini coefficient, increases in poverty ratios, life-expectancy and educational enrolment.

For some indicators, the CISL group does better than the CISM, but in nocase do any of the indicators show CISL countries doing better than theadvanced reformers. Furthermore, not all the indicators are consistent onthe CISL’s ‘superior’ performance: on output recovery, the HDI index, theycome out slightly ahead, but on all the other indicators covered they do not,or indeed on some like inflation and FDI per capita are behind the CISM. Evenapart from the possibility that the official GDP data for them is particularlyuncertain, there is the deeper question of whether their lesser costs are becauseof successful implementation of a gradualist adjustment strategy or simplyreflect a postponement of any advance towards a market economy and neces-sary restructuring. Certainly, their scores on the TPI measure show minimalprogress in that direction, and indeed some reversal since the mid–latenineties. Even more indicative is the measure of TPI values for institutionalelements, or as the EBRD labels them, second-phase reforms. A true gradualistpolicy would imply delaying liberalization and privatization while putting inplace good institutions; the evidence in Chapter 2 is clear: they are fartherbehind on institutional development than they are on liberalization reforms.

The quantitative evidence leaves no doubt that considerable costs wereincurred even in the most advanced reformers – the high unemploymentstill persisting in most of Central Europe and the Baltics is undeniable. At thesame time, critics of the transition who attribute such social costs to rapideconomic reforms must think again: it is too obviously the case that deterio-ration in education, health and other social measures has been largely

116 Background


avoided or quickly reversed precisely in the countries of Central Europe andthe Baltics where reforms were the most rapid and consistent. To say, as forexample Reddaway and Glinski do, that even proposing rapid reforms inRussia was a mistake as they would inevitably degenerate into abuse byelites, and neglect of the mass of the population, is not enough: the caseneeds to be made how gradual reforms would have avoided such abuse.There are possible answers, and I will return to this debate in the central partof this volume which argues that it was not too-rapid reforms but preciselythe opposite, too-gradual, which allowed new capitalist elites – oligarchs inpopular parlance – to capture the process, abuse it for their own interests andneglect the negative impacts on the larger populace. It is no coincidence thatthe countries which were the slowest in achieving economic recovery, andexperienced the largest non-economic costs, were the countries with slowestreforms and greatest degree of non-democratic capture by a relatively smallnewly-minted or regenerated economic elite.

On the timing issue, that the interim assessments were necessarily incom-plete, it is now much easier to update most of the indicators to about 2000or later, and to show that the worst of the deterioration is over and for allindicators the bottom has been reached with many but not all of them turn-ing up. For the advanced reformers this means that the initial levels have notonly been recovered, but have been surpassed. For the SEE and CIS countries,this is not yet the case, but the gains are now clearly visible. This conclusionis not as positive as it sounds at first sight: where the recovery of losses is notyet complete, there is clearly a situation one can compare to Latin America’s‘lost decade of the eighties’ reflecting their debt-crisis.

Beyond this is the result that losers are in such cases probably not yetbeing compensated even if there are many gainers and sufficient gains tohave a compensated Pareto optimum. As a last word on costs, consider thePareto-optimum criteria set out at the beginning, and how different coun-tries have fared. Central European countries and perhaps the Baltics seem tofall on the margin between a compensated and an uncompensated Paretooptimum, while the SEE are surely still uncompensated for the most part,and the CIS even more so. The difficulties of interpreting the minimalistreforms of CISL countries preclude even a reasonable guess on this issue. Forcomparison, how about China? Its rapid growth, high growth of incomeacross all income categories and declines in poverty ratios, suggest thatdespite a widening of income distribution it is surely in the zone of a Paretooptimum, perhaps even a more felicitous than a compensated one since, inabsolute terms, there is no evidence that anyone was made literally worse-off, in an economic sense, though the loss of democratization and the con-tinued rule of the communist party may be the trade-off. Does this meanEuropean transition countries should have followed the Chinese model?I argue later that it does not.

Appendix: estimates of output loss under different assumptions

To measure the overall output decline from the start of transition is difficult enoughgiven the data comparability problem noted in the main text. To measure the cumu-lative loss over the transition period is even more difficult because there are two addi-tional problems: GDP values given in the standard sources are inconsistent over time;and one must assume a counterfactual of what GDP would have been under contin-ued socialism after 1989 or 1991 respectively.

The data problem becomes evident if one attempts a comprehensive but simpleapproach: measure the loss as an output index in year ‘t ’ minus the initial year index(say 1989). EBRD Annual Transition Reports provide the best comprehensive time seriesof GDP indices and growth rates from 1989 to 2003. Taking the index value shown foreach year from individual Annual Reports, unfortunately often gives rather differentvalues than using the latest Report index value. This (not unusual) discrepancy reflectsthe substantial and frequent revisions of these estimates by country authorities andoutside observers as statistical agencies struggled with the conversion to UN account-ing standards, informal economy incorporation, and other problems. Arguably thelatest revised data are generally considered the best; but given that the objective wasto give only an approximate range of these losses, the decision I made was that usinga year-by-year estimation with such uncertain data had no advantage over a cruder‘triangle area’ approach described below.

I will exclude the third counterfactual that is least favourable to socialism, but willinclude the one most favourable, SCFI, even though it flies in the face of the histori-cal reality that brought these regimes to an end. Assuming the area Z1 and Z2 inFig. A3.1. are roughly equal, one can approximate SCF2 using a simplification: outputis flat from 1989 and then in the year that actual bottom was reached (low-point), itcrashes to the level of actual output. This gives the SCF2 path shown in Fig. A3.2.Analogously, if in Fig. A3.1 the two areas between actual and SCF1 are about equal,one can simplify SCF1 as a straight line at the initial level of GDP. Note that this isexactly what is implicitly assumed in most discussions of GDP loss due to transition,which typically take the difference between actual GDP and the initial level. Thus forexample, the EBRD Transition Report updates GDP annually as an index value, with1989 � 100. The EBRD may not wish to suggest that the actual loss is the simple dif-ference, but it leaves the door very wide open to precisely such an interpretation.

First, however consider the range of possible counterfactuals for the path of GDPhad the socialist regime continued. Given the nearly universal consensus that social-ist efficiency was deteriorating well before 1990, the most favourable assumption(SCF1 in Fig. A3.1) might allow for some modest GDP increase under stimulus of sub-sidies for a few years followed by a gradual decline over several years, perhaps fallingto the actual levels observed. The least favourable assumption (SCF3 in Fig. A3.1)starts from the reality that GDP in most countries began to fall before the regimechange. Arguably, continuation of socialism would not have been able to halt thistrend, hence GDP would have suffered about the same decline actually observed in thetransition period, hence in effect no GDP loss attributable to transition. I also proposean intermediate assumption of delayed deterioration in which socialist output wouldhave continued to grow very modestly with artificial support for at most 1 to 2 years,but would eventually crash dramatically as it in fact did in the early 1990s withnegative growth rates in the double digits (SCF2 in Figure A3.1).

118 Background

119

GDPindex

Z1

Z2SCF2

SCF1

Actual

100

Years oftransition

Full recoveryLower point

SCF3

100

Initial level of GDPActual path of GDPSocialist counterfactual 1: Sustained but declining performanceSocialist counterfactual 2: Briefly sustained growth, eventual crashSocialist counterfactual 3: Decline approximately as in ‘actual’

Figure A3.2 Schema for calculating GDP loss under various counterfactuals

Figure A3.1 Actual GDP and socialist counterfactuals

GDPindex

SCF2

SCF1Actual

100

Full recoveryLower point

A B

100

SCF1 as in Fig A3.1SCF2 as in Fig A3.1

Here, I propose to calculate a range of loss values based on the triangles A and B inFig. A3.2 as follows:

A � Loss in decline period � [1/2 loss in through x years of decline];B � Loss in recovery period � [1/2 loss in through x years of recover].

The value of Minimum Loss shown in Table A3.1 simply represents the intermediatecounterfactual, SCF2, and is equal to the triangle A; under this counterfactual periodof recovery does not imply any further transition losses, since by then the socialistregime would have crashed to about the same lows actually observed. I calculate aMaximum Loss based on the assumption most favourable to socialism; SCF1 in FigureA3.1, approximated by the flat line all the way across in Figure A3.2. Hence anadditional output loss occurs in the recovery period equal to the area of the triangle B.This, added to the minimum, gives the maximum loss of output. Since the period ofrecovery appears to be longer than that of decline (see Table A3.1), the maximum ismore than twice the minimum.

Under the assumptions made here, even the resulting minimum is arguably on thehigh side because it ignores the decline actually experienced in the last years of thesocialist period, and ignores the SCF3 counterfactual least favourable to socialism.Certainly, the maximum gives an extremely favourable tilt towards the ability ofsocialism to continue a decent ecnomic perfomance beyond 1989–91. To balance thistilt, Table A3.1 shows the maximum and minimum under different actual values, thatis official data, one-half of the Aslund adjustment, and the full Aslund adjustment. Itdoes not show values for SCF3, but the reader can easily imagine such a column addedto Table A3.1, which would in fact show values of about zero as explained earlier.

120 Background

Table A3.1 Estimates of cumulative loss of output

Per cent loss FullPer cent loss using using 1/2

Aslundofficial data Aslund adjustmentadjustment

Years Peak Min. Max. Peak Min. Max.decline loss % loss % loss % loss loss loss Min.

Central 3.7 18 33 100 13 24 72 11Europe

Baltics 2.7 48 65 194 34 46 138 27SEE 6.2 39 121 363 29 90 270 62CISM 7.2 60 216 648 45 162 486 108CISL 6.7 40 134 402 35 117 351 101

Note: For Central � SE Europe this is the number of years 1989, though the cases of countries inthe former Yugoslav Federation might, arguably, take 1991 as a start. For the Baltics and all CIS, thestart is taken as 1991.

Source: Author’s calculations as explained in the text using official GDP data from EBRDTransition Report 2003, and adjusted index values from Aslund (2001).

Part II

An Ex Post Transition Paradigm: Uncharted Waters, Pirate Raids and Safe Havens

4The SS Transition NavigationModel

123

‘The fox knows many things, but the hedgehog knows one big one.’Archilocus 7th c. BC1

Introduction

Preceding chapters have described progress in transition towards a marketeconomy for 27 countries, concluding that the degree of progress has varieda great deal. Most Central European countries – Croatia, the Czech Republic,Estonia, Hungary, Poland, Latvia, Lithuania, Slovakia and Slovenia – haveessentially become fully operational market economies with little vestige ofthe communist period to distinguish them from other middle-incomemarket economies. Though all of them retain some degree of governmentinvolvement in prices of critical consumer goods and varying degrees ofstate-ownership especially in utilities, a few may in fact have surpassedtypical market economies in the extent of liberalization of the market andprivatization of utilities. Estonia, and to a lesser degree Latvia and Lithuaniahave had to reduce liberalism recently to accord with the norms of the EU asthey entered this club, especially on external trade.

But most others are still far short of the typical market economy withconsiderable state-ownership even outside of utilities, and substantial stateintervention in the economy both formal and informal. This is more true forCIS countries than for those of South-East Europe, and in the former groupthere is furthermore a significant gap between a small group who havemoved only marginally from the regime of socialist central planning –Belarus, Uzbekistan, Turkmenistan – and the other nine where economicreforms, while far short of the transformation achieved in Central Europe,have gone far enough to dramatically alter the nature of these economies.Their transition is far from complete, the role of the market is not alwaysdominant, free-entry and competition are often trumped by oligopolisicand lobbying powers, but they are nonetheless economies with reasonablyfunctioning markets.

Having explained what happened in Part I, we now turn, in Part II, to thesecond objective, to give an explanation of why it happened as it did, inparticular why the considerable differences in progress described in Part I.Our preceding discussion already contains hints, suggesting broadly thatcountries which undertook relatively earlier and more rapid policy changesnot only achieved a higher degree of market functionality, but generally didbetter on performance indicators in the areas of economic, social and demo-cratic transformation. Part II puts this into a more complete logical frame-work, providing a parsimonious synthesis of the most important determinantsof a country’s progress in transition. The present chapter begins by layingout a theoretical model or paradigm explaining the path of progress towardsa market economy in a post-communist society.2 The model comprises threeinteracting elements or variables that determine progress in transition: thedelay in starting reforms; the risk of rent-seeking vested interests capturingstate policy; and the potential of EU membership acting as a guide to trans-formation. Chapters 5 through 7 provide an empirical application of themodel, discussing how each of these elements played a role in the transitioncountries and how it leads to a better understanding of the actual outcome.

Let me briefly motivate the model. The vast literature on transition ofpost-communist economies summarized in Chapter 1 contains a wide vari-ety of explanations for the differences in progress among countries. Some ofthese differences arise between economists and political scientists or histori-ans (many of the latter often arguing that a template approach such as theWashington Consensus ignores the political, historical initial conditions),but the spectrum of views is very large even within each discipline, and aparsimonious model to explain differences in progress does not readily offeritself from this literature.3 I will argue in the next section that the literatureprovides, if anything, too many explanations.

One might try to reduce this vast diversity of views by focusing on the mostimportant debated issues noted in Chapter 1: gradual v. big-bang reforms; thesequences among privatization, market liberalization and macro-stabilization;the relevance or irrelevance of how privatization is done; the timing of poli-cies to promote competitive market institutions relative to other economicreforms; the potential conflict between market liberalization and democratiza-tion; the role of initial historical conditions v. new policy initiatives;the importance of EU membership prospects; the pros and cons of externalinvolvement by international organizations and individual country assistance;and what may be the current most important question, whether the newoligarchic capitalists are a voice for rule of law and democracy, or the opposite.But even this approach risks losing perspective, because a comprehensiveanalysis for each of the outstanding issues leads to many explanatory factorsand results in losing sight of the forest for the trees. The alternative followedhere is to identify a parsimonious paradigm with sufficient explanatory poweracross countries to identify the most important common elements.

124 An Ex Post Transition Paradigm

The SS Transition Navigation Model 125

The rest of the chapter is structured as follows. The next section discussesthe pros and cons of what one may call an ‘encyclopedic approach’ toexplaining progress, concluding that the approach is much too broad andindeed either confusing or inconsistent. I then set forth the parsimoniousparadigm of the transition process (the navigation model), to provide aframework for the large number of factors delineated in the previous section.The model is built around three elements: the nature of the early debate onhow to sail out into the ‘uncharted waters’ of market transformation; theextent of capture of the 27 ‘transition ships of state’ by rent-seeking ‘pirateraids’; and the impact on the process of having or not having a potential‘safe haven’ such as EU membership or other external commitments like theWTO, the IMF, the World Bank, and so on (the sea-voyage allusions providea useful mnemonic, and lead to the shorthand reference of a NavigationModel). I then go on to derive a number of testable empirical hypothesesfrom this model, and indicate the possible ways of measuring the relevantvariables. In subsequent chapters existing quantitative studies are brought tobear on these hypotheses, and some new statistical analysis is presented.This provides results consistent with the Navigation Model. The final sectionconcludes by noting how the model relates to the existing literature on tran-sition, and what it offers that is new and unique; and an Appendix illustratesin broad terms how the paradigm might be formalized in a mathematicalmodel.

The value of comprehensive explanations

The process of transformation involves so many dimensions of theeconomy, polity and societal relations that one is easily tempted to concludeno simple systematic explanation of the process is possible because eachcountry has a set of unique features which influenced the decisions andchoices made. One way to deal with this is the case-study approach; that is,a detailed analysis of a single country or perhaps a small number of countriesin a comparative analysis.4 Another is to include in cross-country quantita-tive studies a large number of characteristics of each country and allow thestatistical analysis to tell the story of which ones matter most. In literatureexplaining growth in transition using econometric analysis one frequentlyfinds initial-condition variables alongside policy-choice variables (Chapters 1and 2). Were one to simply make an inventory of factors adduced in thetransition literature, the list would be very long: in Figure 4.1 I illustrate thispoint with a listing of only some of the factors that such an ‘EncyclopedicApproach’ would produce.

This already long list is easily extended in two ways. Additional factorshave been considered, such as prior national status in history (Romania yes,Moldova no), intensity of national sentiment (Baltics high, Belarus modest),quality of transport infrastructure (Central Europe good, Western FSU

moderate, Central Asia poor). The reader will easily be able to add stillothers. For each of the factors listed, it is possible to elaborate or measurethem in different ways: natural resource base can distinguish energy, mineraland agricultural wealth; overindustrialization can be measured as share ofGDP, or the excess of this share relative to some benchmark; rule-of-law canbe measured using one of the broad institutional indices mentioned earlier,or proxies, such as the length of time for commercial cases to be settled, thecontinuity of judges from communist period, and so on. The possibilities areendless and mind-boggling, pointing simultaneously to the major pro andcon of this approach. On the one hand, a detailed recounting of how thesemyriad factors play out in a given country can give a rich and full under-standing of a specific country, and in particular how its initial conditionshave influenced policy choices. On the other hand, there may be so muchcountry specificity that one loses all comparative perspective and fails toobserve important common elements across countries.

The Reddaway and Glinski (2001) study on Russia is best viewed as asuperb historical account of why the initial conditions made it so difficult tosuccessfully implement a rapid reform and stabilization policy agenda, butits limited comparative analysis precludes an understanding of why rapidreforms worked poorly there but far better elsewhere. Similarly, Van Zon(1998) attempts to explain the slow reforms in Ukraine by the need to builda nation state and the difficult inheritance of the Soviet period, but fails toaddress the question why other countries with the same national need andburdens – Moldova, Kyrgyz Republic, Georgia, not to speak of the Baltics –moved much faster.5 Kolodko (2000a) is right to note that the diverse


Initial conditions● Natural resources*● Landlocked status*● Distance from Brussels*● Years under communism● Level of per cap. income● Share of industry● Concentration of heavy/military ind.● Type of first gov. (communist/other)● Ethnic homogeneity● Major religion● Educational attainment● War/civil conflict● Int’l organization membership● Continuity of old elites

Transition policies● Gradual or big-bang● Year stabilization starts● TPI value reached by year t● Share of private sector by year t● Effectiveness of rule-of-law● Type of gov.: right, left● Frequency of gov. change● Freedom of press● Degree of democracy● Amount of foreign aid● Involvement of IMF/WB● Performance on IMF/Bank

programmes

Figure 4.1 An encyclopedic approach to explaining diverse transition results (someillustrative factors)

* Zinnes et al. (2001) usefully label these as immutable conditions, that is they cannot be changedby policy or time.


circumstances of countries makes the analysis of transition difficult, and isparticularly prescient in saying ‘it becomes even more difficult if one looksnot only at common denominators, but also searches for more detaileddescriptions’.

The advantages of the encyclopedic or comprehensive approach are clearand must be recognized: to understand well an individual country, the morecomprehensive the story the better. A further advantage is that countryspecifics can be very helpful in understanding outliers in systematic cross-country models. Thus, very early economic recovery in Poland compared toall others was not only due to its rapid stabilization and structural reforms,but was helped by the easing of its external debt burden and the fact of apopulace that had been allowed to travel in the Socialist period much morethan most others.6 The landlocked status of Kyrgyzstan may be the reason itsgrowth has not been as strong as expected despite an early advance on reformsand it being the first in the CIS to attain WTO membership. Analogously,Moldova with a strong comparative advantage in food products was heldback by the lack of access to the EU’s protected markets.

But when it comes to comparative analysis across countries, the disadvan-tages of the encyclopedic approach outweigh its advantages. First, there maybe too many factors which allow an enormous number of permutations, andin the end it becomes difficult to follow a line of logic applicable acrosscountries, and difficult if not impossible to discern common factors. Second,it often happens that the same factor has opposite effects in different countriesor over time,7 or an advantage works well for some and not at all for others.Third, there is a common tendency to fall into the ‘special case’ or ‘unique-ness’ trap for each country. The first point was illustrated above in referenceto some well-known studies on Russia and Ukraine. The other two points areexpanded upon below.

There are numerous instances of inconsistency of explanatory effect fora certain factor. Overindustrialization and a strong emphasis on militaryproduction is most commonly considered as a negative initial condition, forexample in the pioneering study on growth by De Melo et al. (1997). Butwhy should this be so? The need to reduce overemployment in inefficientindustries is indeed a burden, but their technical and human-capital capaci-ties are a positive inheritance. Estonian industry very quickly developed asub-contracting relation for electronic goods with Scandinavia, while Ukrainianfirms in Kharkiv specializing in space electronics did not; Hungarian pro-ducers of Soviet military optics now produce Zeiss-Ikon binoculars for theworld market, while Russian firms still struggle to convert. Geographic loca-tion was for centuries a bane for Poland, lying between Russia and Germany,but as Kolodko (2000b) points out, suddenly became an economic asset.Slovenia has benefited greatly from historical and geographic proximity toAustria and Italy, but Albania has not exactly thrived on its connections withItaly and Greece. The dominance of protestant religion as a factor explaining

prosperity was argued by Weber more than a century ago, and has foundsome echo in the transition literature: for example Wolf (1997) uses thisvariable in equations explaining the degree of reform progress (TPI ) chosen,and finds some significance. But the explanations are not easy to interpretconsistently; while most protestant countries did move quickly (Hungary,Czech Republic, Baltics), so too did many Catholic countries (Poland, Croatia,Slovakia). While a lot of the slower reformers are Orthodox or Moslem, theexceptions are too many to buy into the religion explanation: Russia, KyrgyzRepublic, Moldova, Armenia and Albania at least undertook early efforts atrapid reform even if these were not fully successful or sustained.

On the political-institutional side, consider three further examples:the preexistence of a strong nationalist commitment; the ideological orien-tation of the first government, that is, was it communist or not (a relatedhypothesis applies to subsequent governments as well); and the effectivenessof rule-of-law.

A strong national commitment in the Baltics is often adduced as perhapsthe key factor behind rapid and sustained economic liberalization; in thepresent volume, this factor and its relation to EU membership are given duerecognition when discussing the Baltics, but as a systematic explanation inother transition countries it works poorly because it does not have the samedirection of causation. Thus,8 the nationalist commitment in Ukraine is morean explanation for slow reforms, based on the argument that one neededfirst to build the institutions of a nation-state.9 In Belarus and Central Asia aweak sense of nation is part of the problem of slow liberalization, while inthe Caucasian states the conflicts and turmoil, coming from very strongnationalism, also shows reforms – Serbia is doubtless another such example,although in Croatia the effect was the opposite.

Continuation of the previous communist regime after transition begins isthought to be a barrier to market reforms; but why then did Slovenia steadilymove towards a market economy and adoption of the EU’s Acquis commu-nautaire conditions despite the fact that the government in power saw verylittle change in personalities after independence in 1991? The answer lies inthe fact that its communist (or better socialist) government in the Yugoslavperiod was already highly committed to a market economy, had indeedalready promoted a much more market-oriented regime well before 1991,and simply continued on the same path.10 Somewhat analogously, in Polandand Hungary the communists who had lost power in 1989–90 returned in afew years as a social democrat party of the Western European variety butmade no significant directional change in economic reforms.

In Poland, the Solidarity government which began the transition lost theOctober 1993 election to the Democratic Left Alliance, and Leszek Balcerowiczwho led the big-bang programme of January 1990 was replaced by a sharpcritic of his, Grzegorz Kolodko. But Kolodko’s (2000a) criticism of the big-bangnevertheless recognized that ‘a radical approach is practicable in the case of



liberalization and macroeconomic stabilization’, and was concerned mostabout what he and others considered an ‘overkill’ of monetary tightness andconsequent social costs.11 Thus, the fundamental transformations, liberaliza-tion and stabilization were not reversed and indeed continued, with the TPIvalue rising steadily from 3.0 in 1993 to 3.4 in 1997, at about the same paceas in the other transition leaders (Czech Republic and Hungary). Of coursethis was less of a jump than the first four years, but it is a mathematicalnecessity for an index that is capped at some maximum value that its paceeventually slows after an initial sharp rise.12

In Hungary, the Communist Party which had reconstituted itself as theHungarian Socialist Party (HSP) already in October 1989, lost the first freeelections in April 1990, to the new Hungarian Democratic Forum (HDF) andsome allies. The latter’s platform of ‘anti-Communist rhetoric and a gradualist[my emphasis] program of economic reform’ (Pittaway, Chapter 2 in Heenanand Lamontagne, 1999a), nevertheless led to a substantial increase of the TPIfrom 1.5 in 1989 to 3.1 in 1994, hardly gradual. In June 1994, electionsbrought what appeared to be a sharp political reversal with the HSP winninga majority of seats. And what did they do about economic policy? Theyappointed as Finance Minister Lajos Bokros, an outspoken reformer, andpursued a vigorous continuation of market reforms even sharpening thefiscal stabilization with ‘a dramatic package of tax rises combined with sav-age cuts in welfare spending. Although Bokros resigned in February 1996his policies were continued and augmented with … widespread and rapidprivatization’ (Pittaway, ibid.).

This story of former-communist parties continuing market reforms oreven speeding them up, is found repeatedly in Central Europe and South-East Europe, making it difficult to correlate market reform with the label orcolour of the political party in power.13 One could of course still argue thatthe labels and former associations of the new parties do not reflect thechanges in the nature of the party and the real platform is what matters. Butthis is circular logic, as the measure of a party’s platform has to be what itactually does in government, rendering useless the argument about continu-ity. The inability to correlate economic policies with the nature of the firstgovernment, or the type of party (or for that matter the stability v. turnoverof governments as witness the frequent changes in the Baltics), has stymiednot only economic analysts but also political scientists who might beexpected to better understand how polities function historically and glob-ally. The difficulties for that discipline are encapsulated by the followingcitation from a scholar of transitology, Valerie Bunce (1999, p. 779) trying toexplain the diverse patterns of development of democracy in the region:

There are a variety of factors that while logical and suggestive, do notseem to provide a robust explanation. For example, while all the stableand fully democratic cases are by regional comparative standards, both

rich and homogeneous [au; ethnically] some relatively rich states scorelow on both democratization and political stability (Slovakia, Croatia,Ukraine) [au: paradoxically these last serve to emphasize Bunce’snon-robustness point, as all have since seen changes in government andsubstantial improvements in their democracy ratings] and some homoge-neous states do the same (Albania, Armenia) Moreover, state age is not allthat helpful a factor, given, for instance, the inclusion of Slovenia and theCzech Republic in the group of stable and fully democratic orders. Finally,such factors as religion, imperial lineage (or inclusion in the Habsburgversus the Russian or Ottoman empires), and institutional designs (parlia-mentary government versus forms of presidentialism) do not seem toaccount all that well for these differences.

Let me turn to the example of effective rule of law which is now universallyagreed to be an important factor in transition. Broad measures based onsubjective perception do correlate well with the TPI as shown in Chapter 2,but most analysts consider these measures are too broad and not helpful forpolicy. Consider a more specific measure sometimes discussed in the litera-ture on law and the transition: how many judges are carried over fromthe communist period? Figure 4.2 shows a three-way grouping for progressin transition based on Chapter 2, and a three-way grouping of continuity ofjudges based on the Freedom House Nations in Transit publication,Karatnycky et al. (1997). Clearly no good correlation exists, perhaps becausethis is simply not the right way to measure this phenomenon. But that only


CONTINUITY

High Medium Low

Low Uzbekistan Belarus Turkmenistan

Medium Georgia Azerbaijan AlbaniaKazakhstan Russia ArmeniaMacedoniaRomaniaUkraine

High Latvia Estonia Czech RepublicSlovakia Hungary LithuaniaPolandSlovenia

Figure 4.2 Transition progress and continuity of communist-era judges

Source: A transition progress based on Chapter 2; continuity of judges based onKaratnycki et al., Nations in Transit (1997), response to query: ‘How many judgesremain from communist era?’ The responses did not use the same wording and thethree categories are the author’s interpretation.

PR

OG

RE

SS


leads to the conclusion that for many of these factors it is very difficult tofind a good quantitative measure, and hence it is risky to do any analysisbased on such factors as there are simply too many measurement andinterpretation problems.

Finally, the worst outcome of the encyclopedic approach is to fall into thetrap of seeing each country so narrowly that it appears to be unique in beingwell-positioned or poorly positioned for this difficult historical transforma-tion, and therefore unable to follow the standardized path of transforma-tion. This may be the logical outcome of observing so many different factorsand the conclusion that they do not necessarily have a systematic effect indifferent countries, but it is a mistaken point of view. Kolodko (2000b) isclear and firm that the constraint of initial conditions ‘should not be seen asan excuse for a lack of comprehensive reform and institutional change …[but in fact ] has to be offset by still more determined attempts’.

It is not surprising to find the ‘uniqueness’ story most commonly toldin countries which have lagged in reforms or performance, because it is ofcourse a position likely to be promoted by the leadership in less successfulcases. This is not to deny that each country does have some special attributesthat differentiate it from others and that these must be taken into account indeciding how quick is ‘as quick as possible’, as Klaus, 1995, defined rapid; butan overemphasis on uniqueness tends to hide the important commonalitieswhich this work tries to argue were in the end more important than thedifferences. The fact that uniqueness is abused comes out very clearly whenone observes that it is not two or three cases of uniqueness that are claimedin the literature, but almost all countries, especially in the less successfulgroups, are thought by many analysts to be unique. It is possible logicallyfor all countries to be unique, but a closer look at the literature reveals thatin most cases these countries are said to be unique in the same way! Onewonders, were Tolstoy writing about this today, might he perhaps say that allunfortunate transition countries are unlucky in the same way, while all thefortunate ones are lucky in different ways?

This sameness is exemplified by the following citation from the CISHandbook (1999), chapter 9, for country ‘XXX’ left unnamed:

XXX was a state that had never existed in history, with borders not of itschoosing and one of these (an important piece of territory) in dispute,and with a population and a political elite that had no experience offunctioning in a democratic fashion and very little experience of per-forming the functions of modern government. There was a pool of talentand scholarship on which to draw for advice, but the administrative elitehad no training for running an independent state, managing a currencyor dealing with the rest of the world – or for doing all of this in conditionsof severe economic dislocation, exacerbated by internal conflict. In addi-tion given the complex ethnic composition of the population and the

patently confused question of relations with [neighbour states ] A,B,C,the task of creating a distinct XXX identity and allegiance naturallyfeatured near the top of the political agenda … Further, the political cul-ture of the communist system had not trained politicians to seek com-promise which may be regarded as the essence of democracy. If theperformance of the political and administrative elite has been less thanwas hoped for or expected, that is hardly to be wondered at.

Any member of the unfortunate elite of this country facing so many ‘unique’hurdles would be proud of the above apologia. But what is truly interestingabout the citation is that the initial conditions described fit in large part somany transition countries that even an expert on the region would be hardput to guess what country name to replace for XXX. The reference to a non-existent historical state narrows it, though the remaining list of problems iscommonly cited even for countries with pre-existing states. The presentauthor, whose early work on transition focused a lot on Ukraine, found thiscitation extremely familiar. Those working on Caucasian, Central Asian orany Balkan states would probably recognize most of the hurdles cited asarguments of the special or unique problems faced by these other countriesas well. In fact the country analysed in this citation is Moldova; but theliterature on individual countries is replete with very similar lists of the spe-cial difficulties the country faces, particularly for countries that have sofar not been as successful in the transformation as the Baltics and CentralEurope.14

A truly unique twist on the ‘unique’ label comes in a description ofBulgaria (The CEE Handbook Chapter 5 (Heenan and Lamontagne (1999a)):‘Bulgaria faced no problems that were not experienced elsewhere in centraland eastern Europe: it was simply that the country had all of them at once,and to a degree not shared in other countries.’ The present author disagreescategorically that Bulgaria was somehow more unlucky than all the others,and the evidence of the literature also suggests so too would many analystsof other countries. But it is surely not a useful exercise to claim one’s coun-try was the most ill-favoured by history, or buffeted by historical storms fromall sides,15 because in many cases most of the same ills are shared by manyother countries, or as I illustrated above, the same condition has oppositeeffects in different countries. The unconvincing nature of the ‘unique andunfavourable conditions’ approach gives greater credence to the approachchosen here of a parsimonious paradigm applicable to all, complemented bya subsequent use of country-specific factors where needed.

To show the difficulties of the encyclopedic approach is an easy game toplay, and one could go on with examples of inconsistencies in an explanatorydirection for the factors in Figure 4.1. But the main lesson to draw from thediscussion here is that the explanations of the inconsistencies and excep-tions comes from some other not always easily identifiable combination of



factors, which of course means that one quickly becomes lost in the thicketof explanations as easily as a city-slicker in the boreal forests of NorthernOntario.16 I have gone on in great detail here to demonstrate that thisapproach provides rich and very useful historical detail on transitionprogress but, in the end, becomes so complicated, or so country-specific,that it fails to satisfy. That is why I propose the simplified framework of thenext section.

The upshot of our discussion will be that for present purposes, explainingthe difference in progress on the TPI measure across 27 countries using theencyclopedic or comprehensive approach will create more difficulties forunderstanding what happened than will reliance on a simpler paradigmconsisting of a small number of factors. But this parsimonious approach stillallows one to adduce the more detailed list of factors as underlying explana-tions. For example, the lengthy period before undertaking reforms inUkraine may be explained by the fact that the independence movementRukh agreed to have the communists conduct economic policy in exchangefor a commitment to independence. Similarly, one can turn to the list inFigure 4.1 for explanations of outlier cases like Slovakia’s temporary slow-down on institutional and especially democratic reforms in the mid-1990s:the greater concentration of heavy-military industries contributed to theelection results bringing in temporarily the slow-reforming vested-interestparty of Meciar.

A parsimonious approach: the navigation model

The reality of the historical process of transition in each country is far toointeresting to be adequately captured by a simplified theory, and hence therecannot be perfect applicability of the model to each of the 27 countries. Butthe aim is to capture the most important common elements, and to providea model based on a consistent theory of economic behaviour which allowsprediction of how transition may develop further. The core theoreticalidea is found by investigating the fact that countries with least progresstowards the market, are generally those where the power of economicvested-interests or oligarchs is greatest and the reason for this is that profit-maximizing behaviour leads capitalists to oppose open competition if theycan. This is most clear in the transition literature’s recent debate between the‘transition inevitable’ (TI) and the ‘transition frozen’ (TF) views described inChapter 1.

The TI writings reach the optimistic conclusion that a large enough privatesector will ensure demand for institutions of liberal competitive markets anddemocracy. In contrast, those in the TF school argue this will not happen ifownership is too greatly concentrated. But different authors give differentreasons for this cause and effect relation, and may differ in their explanationsof how this concentration came about. Buiter (2000) notes that popular

opinion considers the privatization to have been illegitimate or at a mini-mum very unfair, hence even if oligarchs were ready for legitimization andexpressed a demand for secure, transparent property rights in keeping withthe Coasean axiom that underlies the TI arguments, political sentiment mayprefer renationalizing the assets. But renationalization would only createmore uncertainty, thus a dilemma faces governments. Barnes (2003b) arguesthat the oligarchs are not yet ready for secure property rights because therecontinues to be a struggle for redistribution for privatized assets and opportu-nities for still unprivatized ones; for example in agriculture and energy dis-tribution. The well-publicized case of Yukos in Russia is a focus for many ofthese debates, and I will return to it later in discussing prospects in capturedstates (Chapter 8) and policy implications of the model proposed here(Chapter 9).

While both of the above arguments in favour of the TF view are valid forthe present historical moment, they only give reasons why the transition istemporarily delayed. A more fundamental theoretical reason for permanentfreezing of economic liberalization (that is a sustainable equilibrium) lies inthe possibility that oligarchs find the non-transparent, non-competitiveenvironment to be optimal because of the large rents to be had from retain-ing and even enhancing monopoly power. Such a view has deep roots in thehistory of economic thought, going back at least to Adam Smith in his manywarnings that capitalists may need markets to maximize their profits, butthey seek at the same time to limit competition in those markets since thesize of profits is of course larger the smaller the extent of competition. Thisis the principle that underlies the rent-seeking theories of Krueger (1974) andothers described earlier, wherein existing vested interests oppose trade liber-alization to limit competition and maximize profits. The same principleunderlies the work of Rajan and Zingales (2003a), which describes theoreti-cally and historically the efforts of existing vested interests in the financialsector to oppose development of transparent and competitive financial mar-ket institutions. Their simple and succinct explanation is worth citing: ‘Theeconomically powerful are concerned about the institutions underpinningfree markets because they treat people equally and make power redundant,[and] the markets are a source of competition forcing the powerful to provetheir competence again and again.’

Morck, Wolfenzon and Yeung (2005) have, in a recent survey, elaboratedthe concept of ‘economic entrenchment’ by powerful vested interests, argu-ing that while opposition of the powerful to free-market institutions maynot always be the equilibrium result, there are many reasons to believe thatin the majority of historical circumstances, the gains to be had from rent-seeking in an environment of weak rule of law favour the powerful incum-bents and outweigh the risks to them of insecure property rights as these canbe purchased non-transparently from governments. Both of the above worksare oriented as much to developed as developing countries, and the reality of



lobbying power in the most highly developed and democratic marketeconomies is illustrated in the work of Faccio (2003). The applicability of thisidea in countries with varying degrees of market development makes it evenmore appealing as an explanatory factor.

In the transition literature, a theory of optimal behaviour of ‘oligarchs’ ismost fully developed and explicit in Polischuk and Savvateev (2004) who posita formal model in which oligarchs seek to maximize their profits by trading-off the benefits of fully secure property rights against those of rent-seekingprofits. It is not difficult to imagine real circumstances in which theywill favour what the authors call a ‘hybrid equilibrium’ – that is, the non-competitive, non-transparent environment currently prevailing. When poten-tial rents are still very large because some assets remain unprivatized, orgovernment policy is inclined to continue distorting prices, and in additioninstitutions are so weak that the wealthy can buy security of property fromgovernment agents, it is highly likely the optimal behaviour of the oligarch isto support the status quo. Hellman (1998) had discussed this possibility earliernoting that the new capitalists soon turned to lobbying efforts protecting theiracquired assets against the threats of free, competitive markets, and hypothe-sized they would oppose further liberalization. Hellman and Schankermann(2000) provided empirical evidence for this hypothesis, using an index of‘state capture’ developed at the World Bank by Kaufmann and others.

The role of economically powerful vested interests in opposing furthercompetition once they have attained their position is an important part ofthe explanation for transition progress in the second half of the 1990s, butcan it also help one understand the first half? Certainly the preexisting vestedinterests, the political and technocratic ‘nomenklatura’ of the communistperiod would surely attempt to behave in the way described above and sim-ply transfer their power to a new capitalist regime.17 Some of this did happen,but if it were as simple as old communists repainting themselves capitalistblue, we would not see the huge differences in the extent of power they wieldin Russia and Ukraine at one extreme, and much of Central Europe and theBaltics at the other extreme. Indeed, even within the CIS the vested-interestprinciple may help explain why Russia’s 1992 radical reforms failed, but notwhy they were attempted in the first place, or why Armenia and the KyrgyzRepublic undertook early big-bang measures and Ukraine did not.

The framework I propose here incorporates the vested-interest principleand adds another link to the logic: historical or external factors that facilitatethe rise of the oligarch economy, in particular explaining why it occurred insome transition countries but not in others. I do not develop a formal modelhere, though formalization is not difficult as an Appendix to this chapterwill illustrate. Even without formalization, it is possible to derive severaltestable hypotheses later in the chapter.

The ‘Navigation Model’ takes as the independent variable progress intransition (TPI), and not the resulting success in terms of economic growth,

democratization or other performance measures. However, as Chapters 2and 3 have amply demonstrated, there are very strong statistical correlationsbetween the TPI and these various measures of success, so in that sense thenavigation model can be a building block for any explanation of perfor-mance differences as well. The explanatory factors are perhaps best thoughtof as three proximate causes of progress in transition:

● The length and intensity of debate on the ‘navigation charts’ that is thestrategy of reform programmes.

● The vulnerability of a country to ‘pirate raids’ on state assets by rent-seekinginterests.

● The availability of ‘a safe haven’ such as EU membership prospectsdisciplining the process to achieve a non-oligarchic transition.

These three factors are not independent of each other, and indeed anyformal specification would comprise a set of simultaneous equations (seeAppendix) including some exogenous variables taken from the list of initialconditions in Figure 4.1. Some have argued that the prospect of EU member-ship is an exogenous variable,18 but I argue below why it was not so simple;the exogenous offer of membership (independently of the model’s othervariables) was limited to at most two or three countries, far fewer than the eightthat have already become members and the additional three or more withnow-high prospects of membership before 2010.

The model’s line of reasoning can be stated briefly as follows. The longerand more intense the debates on how far and how to transform the economyinto a market-oriented one, the greater the opportunities for old and newvested interests to obtain large rents via continued distortions and also toconcentrate the transfer of state assets in a few hands; the more concentratedthe new ownership the slower will be subsequent stages of liberalizationand the less likely the development of new entrepreneurial activity. This isbecause of the vested-interest principle described above, that capitalists pre-fer less competition, not more. They also prefer less transparency, as this bestensures continued high profits from rent-seeking and state capture. Similarlywith property rights, new oligarchs may be willing to trade-off rule-of-law(ROL) security for the benefits of rent-seeking, using their financial influenceto ensure property rights informally. In a word oligarch capitalists behaveoptimally when they oppose full liberalization, democratization and rule-of-law. The TPI rises, but stops short of its full value.

The degree to which rent-seeking interests can be successful in their effortsto influence government policies varies, however, hence not all the post-communist states experience the extreme degrees of lobbying that resultin dominant state capture including the ability to determine results ofnominally-free elections. An important factor working against vested interests



is the strength of desire and prospects for EU membership, since this requiresautomatic implementation of many liberalization measures that limit theopportunities for rent-seeking. But other historical factors may also play arole here; a radical change in government towards one that has a strong mar-ket commitment, or a powerful and charismatic leader who leads such avisionary change in the society.19

A commitment to some outside organization or institution such as the EUor others can discipline the process of economic as well as political liberal-ization, helping to curtail debate and get started quickly, and helping toreduce the power of vested interests. I argue below why EU membership wasthe only such commitment to have any significant impact, though in prin-ciple membership in NATO, the WTO, the IMF, the World Bank or the EBRD,all provide vehicles for disciplining reform paths and all were influential inpractice. An important finding of this book is that the force of this disciplinedepended not only on the warmth of the EU’s ‘invitation’ to join, but on thecountry’s own intensity of desire to do so. A strong ‘return to Europe’ com-mitment may have been the way this effect worked in the Baltics, leading toa rapid rise in the TPI which helped overcome an initial reluctance by the EUitself to extend an offer.

Chapters 5–7 elaborate on the dynamics of each of these three factors inthe actual transformation and reform process observed in different coun-tries. Here, it will be useful to give a brief description of four categories ofcountries according to the outcome, measured by the degree of state captureand oligarchic development.

● Captured states. Countries with a high degree of concentrated ownershipin a few hands, that is a very powerful ‘economic oligarchy’ able to capturestate policy and virtually determine election outcomes: Russia, Ukraine,Kazakhstan, Azerbaijan, Moldova and Armenia, and probably Georgiaand the Kyrgyz Republic are in this category, though the last two may beon the margin of the second category. Serbia was very likely in this categoryduring the Milosevic years (but not Montenegro as it was able to conductan independent economic policy even within the ‘union’) but appears tohave moved into the second group.20

● Partial oligarchies. Countries with an intermediate situation, a consider-able degree of oligarchic development, but declining in recent years oralways lower than the first group hence state capture less complete:Albania, Croatia, Bulgaria and Romania are or were in this category untilrecently, as was Slovakia for an interim period in the mid-1990s.

● Competitive market economies. Countries that have largely avoidedthe development of an oligarchic concentration of power strong enoughto virtually determine state policy and election outcomes: all threeBaltics, the Czech Republic and Hungary are in this category, withBulgaria, Croatia and Romania moving into it recently.21

● Lagging reformers. Countries where market transformation has not gonefar enough to result in any significant change in elite-power structure:Belarus, Turkmenistan and Uzbekistan are in this group, with the originalnomenklatura and its role little changed, except for taking advantage ofalso becoming a capital-owning class in the margin where privatizationhas occurred. But that is limited and in any event the role of state guid-ance in the economy remains dominant. Tajikistan, despite greateradvances in economic reform recently, has not yet undergone enoughdevelopment of elites to be easily classified and may still be in this groupor the second, intermediate group.

The navigation model may have very little relevance to this last groupbecause it has seen too little change in economic and political structures.The second, intermediate group comprises countries which are not yetclearly in the state-capture or competitive-market category, though muchcan and will be said later about the tendencies over time in these cases. It willbe most useful to illustrate the interactions of the variables in the model bydescribing the first, state-capture group in a vicious circle of oligarchic devel-opment (Figure 4.3), and the third competitive market group in a virtuouscircle that avoids the extreme effects of oligarchic development (Figure 4.4).


Creates rent-seekingopportunities / old elite

revived

Delayed reform

EU membershipoffer (weak) EU membership

desire (weak)

START

Captures state policyfor self-interest

Oligarchy develops

New entrants’ SMEsface difficulties

Weak rule of law

Weak support for EUmembership

Against competition,prefer status quo,

prefer non-transparentprocedures

Fear EU membershipdiscipline

Figure 4.3 Vicious circle of delayed reform and oligarchic development


It is useful to begin the story of Figure 4.3 by assuming that reforms aredelayed or extremely slow, in any event very partial; the reasons may behistorical continuity of anti-reform governments, the lack of any externalpressure such as a mutual understanding with the EU on a high probabilityof EU membership, a consensus in new nations that state-building takespriority over economic transformation, or even a sincere intellectual convic-tion that market reforms are not the optimal strategy in the long term. Aslong as some new private ownership is made possible legally, this situationcreates capitalism with distorted markets and government regulation, cre-ates an ideal breeding ground for rent-seeking lobbies to operate, influencinggovernments to obtain privileges which generate huge profits. Buying stillvery cheap energy and reselling it at the world price either in a domesticliberalized market or abroad was one of the most common and dramaticvehicles, but as Chapter 6 describes, there were many other ways to accu-mulate the first large fortunes. It is important to note this was very differentfrom the Soviet period when underground ‘mafia’ operations as described inHandelman (1994) were widespread, but these were illegal. The new oligarchsbecame capitalists legally, and indeed most of the wealth accumulation that

Limits rent-seeking

Early & steady reforms

EU membershipoffer (strong) EU membership

desire (strong)

START

Limits development ofpowerful vested

interests

Facilitates SMEdevelopment & new

entry

Small entrepreneurssupport liberal

democracy, openmarkets

Support for globalintegration, EU

Transparency, even-handed rule of law

High level of marketcompetition

Figure 4.4 Virtuous circle of timely reform and development of competitive marketand rule of law

was so morally abhorrent to the populace – and with good reason – wereformally legal. This is what makes partial reforms so risky: it effectivelylegalizes what should be theft, but of course limits access to this process to aprivileged few.

If the reforms do not advance and cut off the distorted price and regula-tions vehicle for rent-seekers, a few of them may become wealthy enough tomerit the title of oligarch, and at that point become powerful enough tocapture or at least hugely influence state policy, certainly on economicaspects that affect them, but perhaps also on broader political direction aswell. If there is a quick-enough advance even after an early delay, the processmay limit the ability of powerful economic interests to capture the state andone moves into the virtuous circle of Figure 4.4. Where state capture prevails,the interests of the oligarch are clearly to limit competition, and even rule-of-law transparency. This last goal of powerful vested interests can in the extrememean they are more likely to support an autocratic or quasi-democraticregime than a truly democratic one. In a word, they support the status quowhich gives them maximum benefit, and the transition to market (anddemocracy) is frozen part-way to its end goal. It goes without saying theyalso oppose a return to a communist regime and are more likely to favourwhat text-books on comparative economic systems used to label a capitalist-command economy.22 Is there a link from this to the EU prospects? On theone hand, any capitalist wants to have access to a wider market, and hencebecoming part of the EU may seem like a good direction for them. On theother hand, membership only marginally improves access which is alreadythere in a globally open economy (especially for any product other than agri-cultural goods), while the exigencies of membership include substantiallyincreased competition which sharply reduces the margin of profits. Oligarchopposition to EU membership leads them to support of politicians who mayexpress pro-EU rhetoric but will not work hard in that direction; additionallythe limited economic and democratic reforms cools the desire of the EU sideto offer membership.

All this leads back in a circular causation to limited economic reforms andan enhancement of the power of oligarchs. The views held by oligarchs onthe policy orientation of a country will not of course be openly expressed.23

But this is not news in the history of the interactions between powerful cap-italists and the society nor is it by any means a monopoly of Soviet-formedpersonalities, rather it is the inherent nature of capitalists. It has been arguedthat even in the country where the degree of openness and competition infinancial markets was highest, the United States, the 1934 Glass–Steagall Actwhich did so much to cut back competition in this sector, was played uppublicly as having precisely the opposite purpose, to curb ‘abuses in thefinancial system’ (Rajan and Zingales 2003b, p. 222).

On the virtuous circle side, Figure 4.4, the storyline follows an analogouspath with the major difference being that sufficiently early and sufficiently



advanced liberalization of the market closes off the many opportunities forrent-seeking that exist under a mixed economy of partial reforms. This limitsbut does not preclude concentration of ownership, because the latter underany privatization process may be a natural tendency driven by efficiencyconsiderations. But the limited influence of such a group in the middle yearsmeans that a committed government can establish better rules for trans-parency of privatization, for institutional rule-of-law development that iseven-handed, for easier entry of de novo enterprises, external competitionand so on. Indeed, allowing for small-scale private ownership to flourishearly creates a political middle class that is likely to support more liberalpolicies and act as a counterweight to large vested interests.

It is not, in retrospect, entirely a coincidence that large-scale privatizationin the more successful transitions of Central Europe did not come that muchearlier – indeed sometimes even later – than in the CISM group of interme-diate reformers where the problem of state capture was so much moreprevalent.24 In this scenario, EU membership prospects could again havebeen both cause and effect in the model. Certainly, for three countries,Czechoslovakia, Poland and Hungary, the offer of membership while notexplicit or guaranteed at the beginning, was signalled in much warmer tonesthan for anyone else. But as Chapter 7 will show, that was not the wholestory by any means and there is good reason to believe all three countries(later four with Slovakia’s emergence) would have done much the same inthe first two to three years without such a signal. I will for the sake of argu-ment provocatively suggest that they might have moved even faster withoutthe offer, because like the Baltics their inherent demand for membershipwould lead to a more vigorous effort to show Brussels they deserved the invi-tation. In the event, the fact that they did move forward on economic andpolitical fronts towards a liberal and transparent polity, greatly enhancedtheir EU prospects and evoked a much stronger ‘offer’ from the EU side bythe mid-1990s, which in turn fed into the virtuous circle by discipliningfurther advances on reforms under the AC guidelines.

It should be clear from the above analysis using the navigation model thatthe three proximate factors are partly underlying causal variables (and can inprinciple be given quantitative value as shown later), but partly they arethemselves dependent on some other initial conditions. In addition they areinterdependent, as for example a strong prospect of EU membership resultsin a curtailing of debate and an early start on transition. Such a formulationof the explanatory model allows the incorporation of the most relevant ini-tial conditions discussed in the previous section, but only where these workclearly to affect the intensity of debate, vulnerability to rent-seeking lobbies,and prospects of EU membership. In that sense, the three factors provide aframework, or hat-rack on which to hang other determinants when relevant.Subsequent chapters discuss in detail each of these three factors, and illustratehow this simplification can help to incorporate a large number of relevant

initial conditions without the degeneration into inconsistent explanationsnoted earlier, without, that is, losing sight of the forest for the trees.

Finally, a few words on the methodological nature of the NavigationModel. The terms ‘framework’, ‘model’, ‘theory’ and ‘paradigm’ are powerfulanalytical concepts because they imply something much more than a meredescription of what happens. They imply a cause–effect relation of asystematic nature, reliance on an underlying theory or paradigm of social/societal behaviour, and the capacity for predicting future developments.While there is no rigorous definition in social sciences differentiating thefour terms, there exists a generally understood rank-ordering. In economics,in particular, a good rule of thumb is that the more formal the mathematicalspecification the safer the use of the term ‘model’ or ‘theory’, because theinternal consistency of the mathematical system of equations allows peers tosee immediately if there is an overall logic. In social sciences generally thisis more relaxed, but nevertheless there is some consensus that the higherthe degree of analytical sophistication and comprehensiveness of the exp-lanatory story, the higher up the rank order of the above four terms one cancomfortably go.

So what then is the proper label for the parable of the ‘Ship of StateTransition’ that I attempt to tell here? It is not a fully developed mathemat-ical model, nor does it capture broadly enough the evolution of differenteconomic systems in history to be a ‘systems’ paradigm. It is at a minimuma framework for organizing a large number of disparate causal relations inthe process of transition. It also provides a good mnemonic device to keepsome order in the storytelling and avoid the confusing, contradictory assess-ments illustrated in the first section of this chapter. But it is much more thana framework, because it has at its core two extremely important theoreticalpoints – I will venture to label them as axioms – from the received wisdomof the economics discipline: only competitive markets result in the social optimum;and incumbent capitalists may oppose competition because their private optimumcomes from having an established position in the market. It is also much morethan a framework because the parable explains how these theoretical axiomsinteract with different initial conditions to give different results for variouscountries. Finally, it is more than a framework since it provides a mechanismfor projecting history forward and indicating under what conditions theincomplete transition may move forward or be arrested.

The central role of rent-seekers, now known as oligarchs, and the strongcoincidence that many from the old nomenklatura (or their children fromthe Komsomol) comprise the oligarchy, leads many analysts, in academe andthe all-wise street, to see all this as a simple conspiracy theory. It would benaïve to exclude the thesis of the red elite in the late 1980s confabulating toseek ways of remaining powerful – though evidence for this awaits stillunwritten memoirs from many of these individuals – but this does not tell thefull story. In another work which also bases itself on the above two economic



axioms, Rajan and Zingales (2003b) note that ‘for this to be more than aconspiracy theory, it has to be useful in predicting when and where marketswill develop’. Thus, if the conspiracy of the economically powerful at thestart were the whole story, one would have seen all of the former communiststates develop the oligarch class. Indeed many predicted this,25 but it clearlydid not happen to anything like the same degree in all countries. Thenavigation model shows why; in a word, at this historic moment it was pos-sible to make a break from the past and avoid the path-dependency result ofthe conspiracy, and this was possible – but not automatic – where other fac-tors such as the attraction of the EU, a national vision of independence, orthe happenstance of a committed leadership outweighed the efforts of theconspirators.

In the end then I choose to use the term ‘model’ even if a mathematicalformalization is not given, because the three-factor explanation I posit doesmeet the three criteria noted: it contains an accepted theoretical axiom, iscomprised of behavioural cause–effect relations, and has predictive capacity.For the less-academic reader, it may be easier and less esoteric to think of thisapproach as a parable: the SS Transition undertaking its voyage on unchartedwaters and seeking a safe-haven from pirate raids and other dangers of thehigh seas.

Testable hypotheses

One of the important values of a behavioural model is its predictions orhypothesized relations to empirical facts. This section first discusses theproblems and possibilities for measuring quantitatively each of the variablesin the model, and then derives a number of testable hypotheses. The remainingchapters in Part II bring to bear empirical evidence from the vast transitionliterature to test these hypotheses. While no direct statistical estimate isundertaken for the parameters of the model itself, it will be shown that theempirical evidence on the transformation process is consistent with its logic.

The dependent variable, Transition Progress is measured by the TPI valuescompiled by the EBRD. The end goal of transition is improved performanceof the economy and polity, but Chapters 2 and 3 have demonstrated amplythat these final goals, which can be measured by several different variables,are closely correlated to the degree of market transformation captured by theTPI, which can be considered as the necessary means towards the ultimateaims. Furthermore, the point of the model is not to explain performance assuch, but to answer the simpler question: what explains the fact that after15 years there is a substantial variation among countries in the movement towardsa complete transformation to a liberal market economy with a high degree of opencompetition? TPI is not a perfect statistic and can be criticized on manycounts including its subjectivity in general, and under or overestimates inparticular countries. However, it is the only comprehensive measure of this

sort covering the full period of the transition, and despite its shortcomingsit appears generally consistent with all other, partial indicators of markettransformation or subsequent performance.

The three independent variables are more difficult to quantify, partlybecause some of them are not directly measurable, thus must be proxied,and while there may be different ways of doing so it is certainly possible tofind values. Thus, the concept of delay in beginning reforms could bemeasured literally as the time between the change in political regime whichis easily identified, and the date of a stabilization programme, which wasfirst specified in the work of Fischer, Sahay and Vegh (1996) (shown in thirdcolumn of Table 5.1). Broadly this does separate countries as between rapidor big-bang and lagging reformers, but there are several cases where the firstattempt failed to achieve stabilization and consequently led to a halt or evenreversal in liberalization as well. This is true not only of the renownedinstance of Russia’s January 1992 programme, but also of Croatia, Bulgaria,Albania, Armenia and others. In the logic of the navigation model, a failedearly attempt at reforms might result in the same opportunities for rent-seeking as a delayed reform. Alternative measures are: the amount of timefrom regime change to the sustained stabilization: (first column in Table 5.2)attainment of some threshold level of TPI (columns 3 to 5 in Table 5.2); andthe level of TPI reached after four years (TPI � 4), a measure of reformspeed.26 The threshold might be taken as TPI � 2.7, which on the EBRD’sscale is half-way to a market economy, or a tougher standard of ‘near-market’operations of TPI � 3.3 reflecting an economy where the role of the marketis dominant and likely entrenched.

There are two measures relating to rent-seeking effects that might beconsidered: the fact of eventual state capture, and the prior vulnerabilityto rent-seeking pressures. The first of these has been estimated by the exten-sive World Bank study cited in Chapter 2 summarized by Hellman andSchankermann (2000), and is measured as a continuum. In later chapters Iwill use their measure and label it SC99, the first year for which the surveydata was compiled.27 This also permits testing the link between state-captureby the late 1990s and continued progress of the TPI values, in effect the ‘frozentransition’ hypothesis. To test the link between reform delay and the degree ofstate-capture, it is important to condition this as well on the vulnerabilityof the polity to pressures from preexisting lobby groups.

Initial vulnerability to rent-seeking is increased by reform delays, but theremay also be exogenous, underlying characteristics of a society that makes itparticularly vulnerable to rent-seeking efforts. All these countries had apreexisting rent-seeking history in the Soviet period, with a powerful nomen-klatura, sometimes working with an illegal ‘mafia’ of economic operatives,skimming-off substantial personal benefits. This is well-documented inHandelmann (1994) or Anderson and Boettke (1997), and it is often said inthe literature that this was more deeply ingrained in the USSR than the



others, hence years under communism might be a good proxy for this.Indeed it has been used in the political science literature in particular. Butthe alleged difference between the USSR and others, while plausible, isunfortunately never empirically demonstrated in the literature, nor indeedthoroughly argued from first principles. Time under communism (seeTable 5.1) is a particularly easy trap in this regard. It was no less than 40�

years or two full generations in Central and South-East Europe, 50 years forthe Baltics, and 70 for the now-CIS countries or three generations (withexception of Moldova, 50 years). That one additional generation may addsomewhat to entrenchment of communist rent-seeking is plausible, but thatit would make a dramatic difference is harder to accept.

A variant of this is the notion that people have forgotten how to operate inthe market. This too is not compelling. Even 40 years was long enough forthose who remembered markets in the 1930s to be over 60 in 1990, and byall evidence in Central Europe they were not the generation that leaped tosupport reforms and participate in new entrepreneurial activities such as ahot-dog stand outside Warsaw airport. Another difficulty is that the degree ofideological commitment varied by country regardless of the tenure of theregime. It is well-known that Czechoslovakia, Bulgaria and, arguably, Serbiaexperienced more rigorous communist regimes than did others in CentralEurope with the same tenure; indeed, under Gorbachev the USSR was arguablyfar less devoted to communist ideals than the aforementioned countries.

Alternatively, the continuation of a communist government at the start oftransition (changed perhaps in name), could better capture the notion ofstronger vested interests from the very beginning. Broadly, virtually all theCIS countries saw a continuation of the political nomenklatura; minorexceptions such as the choice of Akayev, an academic, as president in theKyrgyz Republic do not comprise a significant deviation since the bulk ofgovernment figures and parliamentary majorities still came from the list offormer communists. Also, broadly, the degree of state capture is higher in theCIS than in other countries where a change in government was most com-mon. But in Central Europe this was not universal, with some cases of theformer political elite (‘communists’) continuing in power (Slovenia, Croatia,Romania), and many cases of transformed communists returning in thesecond elections (Bulgaria, Hungary, Poland, Slovakia). In general theydid not pursue a transformation strategy markedly different from the non-communist first governments, which suggest that their conversion was sin-cere, or indeed that their commitments to communist ideals was low alreadyin the 1980s and Gorbacchev’s liberal stance permitted taking a new positionfavourable to markets. The ideal measure of proclivity to rent-seeking is notdirectly available: the elite’s commitment to market reforms. If one is carefulin understanding the above nuances, the ideological nature of the first gov-ernment, and not simply its label, can indeed be a reasonable proxy for theproclivity of the political establishment to be pressured by existing and new

vested interests. Measures for this variable can be found in two publications:Heenan and Lamontagne (1999), Central and East Europe Handbook, and CISHandbook (CEEH and CISH henceforth).

Thus ‘period under communism’ seems a poor measure of proclivity torent-seeking, nor is ‘first-government communist’ any better. However,Chapter 5 will show that a deeper analysis of the underlying causal factor,ideological commitment for or against liberalism, can yield meaningfulquantitative tests of the relevant hypothesis.

The prospects of EU membership can be thought of as a probability on acontinuum of zero–one, but this is not subject to quantitative measurement.At best, it is possible to infer at different times (early, mid, late-1990s)whether the prospect was high, medium or low. Chapter 7 will argue thatthis probability results from the interaction of a country’s own desire or‘demand’ for membership and the EU’s inclination to invite the country or‘offer’ membership. One point merits comment here: this interaction typi-cally implies some circular causality: the more intensely a country wantsmembership the faster its transformation, and the more likely the EU willextend an offer; the stronger the signal of an offer by the EU, the stronger themotivation to proceed with transformation There is no ready-made quanti-tative measure, and I will infer this qualitatively from various analyses in theliterature.

There are four sets of hypotheses derived from the navigation model thatare testable using at least bivariate correlations of the above variables, withsome being testable in more than one way; this is summarized in Figure 4.5.

The Appendix lists several possible ways of quantifying the variables of themodel. But for many concepts (strength of EU offer, or intensity of a coun-try’s demand for membership), quantitative measures can be at best onlyrank-ordered (High, Medium, Low or 1, 2, 3) based on interpretations ofexperts in the field. However, all the hypothesis tests can be supplementedby a more qualitative, less-rigorous assessment based on country case studies –which of course reemphasizes the value of the encyclopedic approach as acomplement to the systemic explanatory approach of the Navigation Model.

What is new in the navigation model

So what is the one big thing the hedgehog knows? It is that Adam Smith’sfelicitous result of the Invisible Hand providing the Benthamite ‘greatestgood for the greatest number’ comes not simply from having private owners(capitalists) seeking to maximize profit, but from the discipline that an openand competitive market imposes upon them, preventing excess profits andaccumulation of monopoly power.28 A different way of putting it is thatprivate ownership is a necessary means to the end of optimal social welfare,but it needs two additional (related) elements: a competitive open market;and transparent rule of law to ensure the security of property rights on a



‘level-playing field’ for all capitalists, big and small, established andinchoate. The ‘level-playing field’ term is new, but Smith might well haveused it too if football were an eighteenth-century game, as it captures per-fectly his oft-expressed concern that the established and powerful capitalistsuse their influence to prevent competition from new and unestablishedmerchants.

This simple way of putting the matter highlights the main distinctionbetween the ‘transition frozen’ school of thought and the ‘transitioninevitable’ view. TI concentrates as it were upon one of the hands inthe Invisible Hand paradigm, the fact of private ownership, and errs by notpaying due attention to the other equally necessary hand of competitive dis-cipline. Smith’s emphasis on the interplay of motivated private owners withthe discipline of competition is an important part of neo-classical dogma. Ifthe market is not competitive but monopolized, profit maximization doesdo good things as it results in technically most efficient production processesyielding the greatest profit for the owner-capitalist, but not necessarily forsociety at large.29

H1: Reform delay increases probability of state capture*● Degree of state capture (SC99) is negatively correlated with some measure

of delay, such as time from regime change to start of reforms, or to achieve-ment of a certain threshold.

H2:The greater state capture the more likely reforms are frozen● TPI late-1990s is negatively correlated with degree of state capture (SC99).● ‘Oligarchs’ less supportive of reform parties than small entrepreneurs.● ‘Oligarchs’ prefer easy rents to efficiency-based profits.

H3: Higher EU prospects reduce delay in reforms● Favourable ‘offer’ signals in early-1990s encourage early reforms.● Strong ‘own demand’ for membership drives early reforms regardless of sig-

nal from EU.● Strong progress in reforms induces more favourable stance of EU.● Reluctance of EU in face of strong own ‘demand’ induces speed-up of

reforms.

H4: Other miscellaneous hypotheses● Stronger political commitment to liberal vision, delay is shorter.● Stronger independence/nationalism sentiment, delay is shorter unless

violence results in which case the liberalization is even slower.

Figure 4.5 Testable hypotheses of the navigation model

* Possible ways of quantifying the variables of this list are shown in the Appendix to the chapter.

In this connection it is useful to comment on the frequent comparisons ofpost-communist oligarchs with robber barons of the nineteenth century,which often draw the mistaken lesson that with time oligarchs too will haveamassed enough and become good capitalists scattering cornucopias ofphilanthropy and accepting to live in a state with equal rule of law for all.There is one big similarity and one important difference between the twogroups. They both amassed a lot of wealth and hence had similar oligopolis-tic or even monopoly powers. But the robber barons did not get their wealthprimarily by underpriced transfer of preexisting state assets, rather theycreated value-added. It is also important to remember that they did notbenevolently and willingly push for Anti-Trust legislation which limitedtheir profit-making. (Acemoglu, Johnson and Robinson (2004) give thoroughaccounts of this.) The real lesson from the robber baron period is that drawnby Rajan and Zingales (2003b): powerful capitalists do not willingly giveup their positions of wealth, influence and future monopolistic profits orrents, and not only do they fight against any legislative and implementationinitiatives for a level playing field coming from the body politic broadly writ,but in fact often initiate government regulations masked as benefiting soci-ety but in fact resulting in limited competition. This is not a matter of an evilclass conspiracy, but a simple matter of economic behaviour: capitalists (bigand small) worship maximum profits and abhor any competition whichreduces them.30

The above theoretical basis for the navigation model is therefore not new,but it is often necessary in economic writing to keep reviving forgottenprinciples. That capitalists seek to limit and not expand competition makesa large difference in the eventual outcome, and such behaviour during thetransformation has to be part of the analysis. Furthermore, the three explana-tory factors in the model are also not novel ideas and have all been discussedin the literature in many different contexts. But there are three importantnew points that the model does put on the table.

First, these three causal factors suffice to explain a substantial part of thevariations in results, hence they provide a useful framework for understandingcross-country differences, allowing relevant reference to country specifics,but avoiding the impenetrability of comprehensive explanations for eachcountry. Second, this parsimonious model provides a simple but powerfulresult: an explanation of an important new divide that has been noted bymany observers: an incomplete and halted transition in many CIS countriesand perhaps some Balkan countries compared to the advanced and continu-ing progress in Central Europe. The latter ended up with a partially reformedeconomy subject to ‘capture’ by powerful vested interests. This state ofaffairs has generated a hotly debated question. Is ‘the increasing strengthof Russian capitalists essentially a liberalizing force’ as Aslund (1997), Aron(2003), Arvedlund (2004 and most forcefully Shleifer (1997) argue, presumablynot just for Russia. Or alternatively could ‘the power of concentrated vested



interests could seriously compromise institutional and regulatory develop-ments that underpin good governance’ (Hellman and Schankermann, 2000),and freeze transition as the current author argued in the book.

This leads to the third new element: the navigation model provides a basisfor discussing the TI v. TF debate. Though it is conceptually consistent witheither result, its logic strongly suggests that the TF conclusion is both theo-retically and empirically more compelling. Once concentrated ownershipand oligarchy comes about, these new vested interests have good reason tooppose further economic liberalization and democratization, which resultsin an equilibrium half-way between plan and market, half-way betweenautocracy and democracy, or a ‘frozen transition’ (Havrylyshyn, 1994).

Appendix: an illustrative formalization of the navigation model

This Appendix illustrates how the navigation model might be formalized into asystem of four equations for the variables discussed: transition progress, delays inreform, proclivity to rent-seeking, and EU membership prospects.

Definitions and possible measures of variables Endogenous

TPI04 � EBRD Transition Index in 2004average of six indicators, excluding infrastructure.

DELR � Delay in starting reforms(i) months from regime change to start of stabilization,

(ii) or months from regime change to 5% monthly inflation,(iii) or years from regime change to reach TPI � 2.7; half-way to market economy,(iv) or value of TPI four years after start.

TPI�4 � Value of TI four years after regime change (alternative to DELR)benchmark year see Table 6.1 ‘Transition Star’, 90 for POL, 91 for other CB and SEE,92 for war cases of YUG, 92 for 15 successors of USSR. Coefficient signs opposite tothose for DELR.

SC 99 � Index of state capture 1999World Bank and BEEPS survey (as in Hellman, Jones and Kaufmann 2003)Belarus � Uzbekistan are anomalies, BEEPS shows very low value of state capture,but this is misleading because it is the state that holds captive enterprises, not viceversa. Therefore, our rationale that high capture by the state is equivalent to highcapture of the state, I would set their index at an average of the highest five.

EUM � Prospects of eventual EU membership about 1990–94(i) strength of ‘offer’ by Brussels assigned values of 3, 2, 1, 0 (high, medium, low,

zero) based on formal and informal signals from EU,(ii) strength of desire by society.

H: Visegrad four � three Baltics, and SloveniaM: Bulgaria, Romania, Croatia (?)L: SEE, Ukraine, Moldova, Armenia, Georgia0: Russia, Belarus, all Central Asia

Exogenous variables

� Society commitment to dual liberal vision at start(H, M, L) a composite index of strength of nationalist sentiment, First Govt.Non-Com, Committed Reformers before transition.Possible values:H: Pol, H, Czech, Slovenia, BalticsM: Slovakia, CRO, MDA, Russia, KYR, ARM, GEOL: AZ, KAZ, TAJ, UKRO: BEL, UZB, TRK

� Happenstance of strong leader with liberal visionMeasure as dummy 1, 0(1 � Poland, CZR, SVK (89), EST, LVA, Russia, KYR0 � all others)

� Strong vested interests inherited from Soviet periodComposite variable:First Gov. not communist: VESTED � yrs of com. � 1.0First Gov. coalition: VESTED � yrs of com. � 1.5First Gov. communist: VESTED � yrs of com. � 2.0

� Dummy for Slovakia� 1, others � 0; to reflect short period of democratic reversal in Méciar years.

� Dummy for Croatia� 1, others � 0; to reflect quasi-liberal democracy in war period of Tudjmangovernment despite liberal economy.

� Geographic, contact or historical-cultural distance from EuropePossible measures:GEO. DIST � kilometers from Brussels as in Fischer, Sahay and Vegh (1997)CONTACT � distance for each country to closest European country as in Kopsteinand Reilly (2000)CULT. DIST � years under communism or average of yrs � CONTACT index

The navigation model (in all equations, DELR could be substituted by TPI � 4)

(�) (�) (�)

(�) (�) (�)

(�) (�) (�) (�)

(�) (�) (�)(4) EUM � a4 � b4 DELR � c4 DIST � d4 LIBVIS

(3) SC99 � a3 � b3 DELR � c3 VESTED � d3 SVKD � d4 CROD

(2) DELR � a2 � b2 EUM � c2 LIBVIS � d2 LEADER

(1) TPI 04 � a1 � b1 DELR � c1 SC99 � d1 EUM

DIST

CROD

SVKD

VESTED

LEADER

LIBVIS


5The Search for a Navigation Chart: Legitimate Debates, Vested Interests and ReformistCommitments

151

The release from communist captivity: carpe diem or follow the money?

When the Berlin Wall fell on 9 November 1989, the euphoria of the satellitecountries was palpable, and the rejoicing at the opening of prison gates formillions of repressed people was shared by the population of the non-communist world, particularly in Europe. In the USA, to sympathy wasadded a strong dose of victorious pride, since Americans saw this as an act ofcapitulation to President Reagan’s challenge, ‘Mr. Gorbachev, tear down thiswall!’ The citizens of the Soviet Union remained uncertain of their fate foranother two years until the convulsive events of mid-August 1991 firstbrought the putsch of anti-Gorbachevites, then within a week of almostvaudevillian dynamics, its termination, ‘not in a bang but in a whimper’.Declarations of independence followed in a flood from all republics includ-ing de facto Russia, most of them preceding the formal dissolution of theUSSR in mid-December 1991. The euphoria had now spread to the whole ofthe Soviet bloc, and was virtually universal save perhaps for some high-rankingmembers of the nomenklatura who faced an uncertain future.1

While the level of expectations may have been unrealistic as is character-istic of the state of euphoria, it was not unreasonable to expect that quickand dramatic steps would now be taken to implement democratic and free-market mechanisms in these societies. The euphoria certainly served to gen-erate a strong consensus in the populace for such changes, an enablingenvironment for policy-makers that Balcerowicz (1993) has labelled a periodof ‘special politics’. But, the opportunity to put in place radical and oftenpainful changes during the honeymoon – for example closing inefficient

factories which put people out of work – was not seized equally rapidly bythe leaders of all countries. The main objective of this chapter is to under-stand these differences, why some countries started the march towards a freemarket economy within months after the beginning of a new regime, andothers delayed this for as much as three–four years. The central thesis is thatthe extent of delays and the subsequent pace and trajectory of market imple-mentation were related to the strength of commitment to the dual vision ofa liberal economy and a liberal polity. In one sense this is almost a truism,but it will become clear why it is so important to understand the intensityand nature of the commitment. The statement ceases to be a truism whenthe forces behind different degrees of commitment are understood. Foursuch forces are explored: the desire to break with the communist past;nationalist sentiment; the hidden agenda of those previously in power andable to retain at least some influence in the new regime; and the vision ofintegration with major international institutions, in particular the goal ofEU membership – this last is postponed to Chapter 7, while the first three arethe subject of this chapter.

This chapter is structured as follows. The next section first addresses thequestion: was there a real problem of ‘uncharted waters’ or a lack ofroadmaps forcing policy-makers to take more time to debate an optimal planof action for setting out into uncharted territory? The answer given here isthat not only were there several clear charts, but if anything there were toomany options to choose from. That leads to the second question: was it notsensible to delay taking action and to analyse the options closely to decidewhat was the optimal path? To answer this question, we move on to attemptto lay the empirical groundwork for each country by tracing the timeline ofevents: the appropriate dating of the regime change in each country,the dates of the first steps in economic stabilization and transformation, andthe dates of milestones reached in progress towards a market economy. Thisclearly points to the historical fact that many countries started almost imme-diately and proceeded at a steady pace, while many others delayed the begin-ning and/or proceeded at a very slow pace. I then turn to the core of thehistorical analysis, assessing the role of the above-noted forces in explainingdelays in starting and the differences in strategies chosen. A final sectionsummarizes how the historical facts of the search for a ‘navigation chart’accord with the navigation model.

Availability of navigation charts

At independence, facing the dual task of building a new state and amarket economy … Ukraine was entering uncharted waters … The slowprogress of economic reforms was understandable given the policy-maker’slack of knowledge of a market economy. (Shen, 1996, p. 187)


The Search for a Navigation Chart 153

Although the sea was essentially uncharted, the prior experience ofEastern Europe had produced a veritable Babel of voices proffering strate-gies on how to get from here to there. (Schroeder, 1996, p. 12)

Although parts of the transformation represent uncharted territory (massprivatization), many other aspects … are quite familiar. Many othercountries once cut off from the rest of the world by immobilizing author-itarian regimes have successfully integrated in to the rest of the world …the proto-typical one in Europe is Spain. (Laar, 2002, p. 13)

Experience shows clearly that the major difficulties were not because ofa lack of the theoretical knowledge … but because of the inability ofgovernments to carry out social policy based on this knowledge.(Kolodko, 2000a, p. 5)

The above citations are respectively from two outside academics, and twokey policy-makers in Estonia and Poland. Where does the truth lie? Itdepends on the meaning of the word ‘uncharted’, to paraphrase a former USPresident. If it means narrowly that no country had ever shown the wayfrom plan to market, indeed this was virgin territory and even Spain’s exampledid not help much, at least on the economic dimension.2 If it means thecountry’s leaders had absolutely no idea how to proceed, it is simply untrueand ‘uncharted’ takes on the obfuscatory valence of President Clinton’s nowfamous syllogism. Schroeder is certainly right that the popular notion of‘uncharted waters’ resulted in an unfortunate surfeit of ideas, strategies andprogrammes – debate on which seemed justified by the laudable goal ofseeking ‘the optimal route which minimizes social costs’.3 Unfortunately,the Babel of voices included not only those sincerely seeking the social sci-entist’s Holy Grail of optimality, but also most of the nomenklatura whowere opposed to reforms altogether, fearing for their privileged positions,and thus delighted to find a ‘scientific’ reason for delay.4

Some of the many alternatives available and well-known at the time arelisted in Figure 5.1; its contents clearly contradict the plaint that there wasno chart or roadmap. Any leaders and governments committed to begin rea-sonably quickly had plenty of choices with clear and concrete details. Thelist can easily be expanded – for example it has sometimes been argued thatthe substantial liberalization of distorted markets in Asia, Africa and LatinAmerica in the 1970s and 1980s provided considerable lessons if not aprecise precedent for the transition.5 But the five broad approaches shownsuffice to demonstrate that plenty of choice and flexibility existed wherepolicy-makers were committed to move forward. It is useful to discuss brieflythe nature of the five main options.

Consider the first option, ‘Goodbye Lenin and just do it’, which in effectmeans reversing one-by-one the main elements of communism. A popularaphorism at the beginning was that you can make fish-soup from an aquarium

but you cannot make an aquarium from fish-soup. Such glib commentsmay have contained a smidgen of truth but were at best a diversion and atworst nonsense. The process of turning a capitalist economy into a socialistone from 1917 involved very concrete steps, many of which could withoutmuch trouble be reversed: central plan and supply-price bureaucracies couldbe abolished rather than just renamed ‘Ministry of Economy’ and‘Committee for Price Regulation’, the infamous Collectivization Brigadesthat helpfully came to the peasants to show them how to collectivize couldbe mimicked in the form of ‘Decollectivization Brigades’ to help land-ownersbegin serious private farming. The list is endless, and the feasibility of suchan approach is seen in Poland’s rapid abolishing of many economic laws


Rapid reform charts

1 Goodbye Lenin, and just do it

● Reverse the actions taken in 1917● Denationalize (privatize, restitute)● Decollectivization Brigades on land

2 Mimic market economies

● Washington Consensus (some flexibility of timing and sequence)● EU’s Acquis Communautaire (limited flexibility by negotiation)● Best-practice standards (banking from Basel rules; corporate governance from

OECD; budget operation IMF standards)● Borrow legislation/institutions (Baltics from Scandinavia)

Gradual reform charts

3 Mimic China

● Liberalize agriculture first● Maintain state firms but ease new entry● Liberalize sector-by-sector, but do so radically● Was Perestroika � China model?

4 Import-substitution strategy

● Private ownership with infant-industry support● Gradual opening-up of foreign competition

5 Institutions precede market liberalization

● Stabilization rapid, liberalization gradual● Market and legal institutions of minimal adequacy before liberalization and

privatization● Synchronized liberalization and institutional progress

Figure 5.1 Uncharted waters or too many charts?


replaced by reinstatement of the prewar commercial code; and Estonia’snearly Xerox-like adoption of the German commercial code, in contrast tothe bit-by-bit efforts in the Balkan countries, Ukraine and even Russiadespite its mini big-bang in 1992.

Mimicking market economies was a second possibility, not inconsistentwith the first but more forward-looking and with more specific and detailedcontent. There were a number of specific charts for this approach. TheWashington Consensus (WC)6 remains the best-known and was based onthe similar but more limited liberalization experience of developing coun-tries adapted for the comprehensiveness and size of the transition task. Themany elements of such a reform path could be flexibly and experimentallyapplied, adapting the pace and sequence of implementation, with the excep-tion of an urgent priority for financial stabilization. Despite some flexibility,in spirit the WC emphasizes the need for a comprehensive package and earlyachievement of some critical mass to ensure it is irreversible and effective;hence it is not wrong to allege it promotes a rapid and big-bang approach.

There were other mechanisms for copying market economies, thoughthey coincide or overlap with the WC in many dimensions. The vast body oflegislation required of all potential members of the EU, the AcquisCommunautaire, was freely available to all countries as a maquette to put inplace the institutions of a market economy, though it did not contain allinstructions, for example how to build or rebuild the private sector. Thedominant role of this mechanism for many Central European and Balticcountries is best understood in the context of their quest for EU membershipexplored in Chapter 7. Still other mechanisms for emulation of marketsexisted, such as adoption of best-practice standards in finance, budgetaryoperations and the like. This could be done from the blueprints (charts)established by various international bodies including the IMF, the OECD, orthe International Accounting Standards Board. Borrowing with minimaladaptation from an existing market economy was also an option followed bymost countries in Central Europe. In the newly-independent countries ofCIS, in contrast, the most common approach was first to re-adopt the Sovietconstitution and laws, then begin to modify it.

The third option was to mimic China, whose economic reforms hadstarted more than a decade earlier and were certainly already known for thesuccess in stimulating rapid growth, considerable inroads of private-sectoractivity and attractiveness to foreign investors. I argue elsewhere that almostnone of the Eurasian countries could have followed the same route success-fully, but here the point is merely that the example did exist and China’sexperience was available as a chart. Many participants in the early debatespoke of China, including outside academics who argued for a gradualapproach, but in practice little serious attention was paid to this possibleroute by the leadership. Central Europeans who chose a more rapid route

ignored the China example or even spoke of it dismissively, as Klaus (1995),lumping it in with all other ‘unique’ or third-way approaches as ‘the way tothe third-world’. For them the lack of any political changes in the Chinesemodel was enough to cast doubt on its applicability.

Others in the Balkans and the CIS who did not jump into early reforms didnot seriously follow the examples of China in practice, but tended to use itspossibilities rhetorically as a reason for continued delay in reforms. Inparticular note that CIS countries, with the exception of Armenia, Georgiaand Moldova, did not emulate China by undertaking land privatization as afirst measure. Another reason for not following the China model was that ineffect Gorbachev’s perestroika tried to do so and failed miserably. Most of theleaders of the CIS countries may have wanted to see the partial and slow paceof perestroika continued, but seeing the writing on the wall, feigned supportfor an eventual transformation to a full-fledged market; perhaps onlyNazarbayev was courageous enough to argue openly for a continuation ofthe perestroika strategy.

The import-substitution (IS) model – sometimes called the industrial-policy model – was the fourth possible alternative. Its philosophy isantithetical to the WC whose genesis was the search for a strategy to over-come the import-substitution regimes of many developing countries. It isnevertheless a market-oriented model, for in theory at least the markets areliberalized, private-ownership prevails, and the government only selectivelyand temporarily assists infant industries to get on their feet. Regardless ofone’s view of its efficacy, it has a long history and was a ‘chart’ with manyprecedents for implementation, going back to Alexander Hamilton’s strategyof protecting Yankee industry at the start of America’s national experience,and goes forward through the cases of Meiji Japan, Witte’s industrial promo-tion policy for Tsarist Russia, Ataturk’s industrialization strategy for post-Ottoman Turkey, even France’s Indicative Planning. There is no shortage ofhistorical precedent, yet it was not until several years into the transition thatthese ideas were trotted out as the ‘unique’ path for Belarus, Uzbekistan,Ukraine, Kazakhstan and, most recently, Russia. Why not earlier? Could it bebecause at the beginning in these countries, most of the governments werelargely a continuation of the previous communist nomenklatura who foundit advantageous to say they were in favour of a market economy but that thebest way forward needed a lot of thought – and time. In effect they wereimplementing a hidden agenda as the resulting delay in reforms would leadto better opportunities to convert themselves into a capitalist elite. The oneexception with some claim to sincerity, or at least consistency, may beKarimov in Uzbekistan, whose strategy of promoting domestic industrywas explicitly formulated early on and has been more or less consistentlymaintained.7

The last group of available charts are those that emphasize the primacy ofmarket and legal institutions. It is not entirely distinct from either the WC



or gradualist prescription. It agrees with the WC about the need for immediatefinancial stabilization and eventual market liberalization and private sectorexpansion, but distinct from the WC proposals (which, as Chapter 2 noted,also list this element) is its insistence that some minimal amount of institu-tional change is needed before the other liberalizing reforms are undertaken.The roots of this view are in the work of Douglass North (1995), analysingthe several-centuries history of the development of effective market institu-tions one sees in advanced economies. Its adaptation to transition countriesis best exemplified in the work of Murrell (for example in Clague and Reuser1992). Though he did not of course propose waiting centuries or evendecades for all institutions to be fully developed, but warned that one shouldnot leap forward with a big-bang until some minimally adequate institu-tional framework exists to ensure competitive markets, free entry, and fairtreatment of small enterprises.

On the face of it this is in principle not only eminently sensible but seemsclearly superior to taking the risk of a big-bang leap towards the market witha copy of the WC in hand. Indeed no-one today argues that institutions donot matter. This debate has seen enough ink in preceding chapters, but hereit is worth adding one point. The institutional strategy was never applied,but was more honoured in the breach and much abused for political advan-tage. Many countries did delay their reforms giving precisely such reason-ing, exemplified by popular phrases such as ‘we must first build theconditions for a market’, ‘our goal is a social-market economy’, ‘we must pro-tect the poor’, or ‘the people are not ready for the market’. Had these beenmore than slogans, then countries that took a gradual approach, as did mostof the CIS and some in the Balkans, should have been implementing insti-tutional reforms while delaying the liberalization of markets and privatiza-tion. The stark truth is that none of them did, as already hinted earlier inChapters 2 and 3 and explicitly demonstrated below. One is hard put toresist the conclusion that there were no sincere gradualist policy-makerscommitted to properly applying this particular theory of the academicscribblers.

In practice, the debate on alternatives quickly ‘coalesced into two polarizedstrategies, big-bang v. gradualism’,8 probably for worse not better. Broadly,the first two strategy types – reverse communism, mimic the market – can beconsidered as big-bang approaches while the other three – mimic China,import-substitution, institutions first – are typically thought of as implyinggradualism. The match is not perfect, but close.

The claim that there was no theory or historical experience to guide theprocess of transition from central plan to market appears in retrospect tohave been a red-herring, perhaps understandable and harmless were it onlyan academic debating point. But unfortunately it became an extremely use-ful tool for opponents of reform whose real reason was the fear that anychange would be harmful to their privileged pre-transition positions. In the

end, two things mattered most about these debates: how long they went onand delayed the start of the transition; and whether they resulted in aradically different ‘third-way’ strategy.

In retrospect, it is impossible to argue that delays in reforms were due tothe lack of precedents, the non-availability of charts or roadmaps, for ifanything the problem was not too few charts to follow but too many. Thatcould still mean that delay was justified to allow an adequate debate anddiscussion of what was the best path, what was most suitable to the socialpreferences of each nation. It should be unnecessary to demonstrate indetail that in societies truly open to democratic debate, the choice at leastinitially was to move relatively fast. In countries where reforms weredelayed, open democratic debate was not really possible. Indeed in mostcases this ‘debate’ was abused by the hidden agenda of potential losers fromtransition, the economic and political elites of the nomenklatura. To makethis case it is useful to quantitavely measure the delays in getting started,and identify the strategies actually chosen – which is the task of the nextsection.

A timeline of economic transformation events

Logically, categorizing countries by strategy must be based on the historyof the main events of economic policy and the subsequent results. For pre-sentation purposes, it will be easier to follow the events chronology byturning the logic around and showing first the outcome, grouping coun-tries into the five strategy types of Figure 5.2.9 The 10 countries of CentralEurope and the Baltics enjoyed a wide consensus and hence experiencedvery limited debate, and proceeded to early reforms. But among them somewere already more advanced and did not move as quickly (Type I:Advanced Start, Steady Progress), while others starting farther back under-took a more rapid pace, closest to the big-bang concept that is so often butso loosely used in the literature (Type II: Sustained Big-Bang). A third, smallgroup of countries, some from South-East Europe, some from the CIS, alsostarted early and moved fast but did not sustain the pace and should beconsidered as aborted big-bang cases (Type III: Unsustained Big-Bang).They differed significantly from the clear big-bang cases, both in the extentof change in the first years and, most importantly, in not sustaining thisstrategy but experiencing policy reversals and resurgence of inflation. In afourth group of gradual reformers the start was delayed for a long timeand/or put on a very slow-track as a result of the extended debate (Type IV:Gradual, Delayed Reforms). Some countries of Type III suffered renewal ofthe debate and later fell into Type IV: Russia, Moldova, perhaps KyrgyzRepublic. Finally a fifth group is identical to the limited reforms group inChapter 2 though two of these countries, Belarus and Uzbekistan, did show



some modest early progress similar to many in the gradual reformers, butlater this was halted and indeed reversed (Type V: Limited or ReversedReforms).

In popular memory and as a historic symbol, 9 November 1989, the daythat the Berlin Wall fell under the euphoria of German youth with hammersof many sizes and shapes, will doubtless mark the beginning of transition.For a more analytical comparison among the 27 countries it has to be recog-nized that this event did not immediately mark an opening of new opportu-nities, as these openings came in steps over the next year for most of Centraland South Europe and only two or more years later for the 15 countries aris-ing from the ashes of the USSR. Table 5.1 lists the most important suchevents reflecting the beginning points of transformation.

Consider first as a measure of reform delay the time between the regimechange (defined as independence or change from communist party control)and the start of financial stabilization efforts, which virtually all observersagreed should be undertaken quickly. For the majority of countries of thefirst three types in Figure 5.2, this was almost immediate, and certainlywithin a half year. The exceptions are Croatia and Macedonia where war orembargo kept the situation unsettled for at least one or two years, andRomania where political struggles and debates went on even longer. For theCIS this delay was far longer, from 20 to 40 months except in Russia with itsalmost immediate big-bang effort to stabilize.

I Advanced Start, Steady III Big-Bang IV Gradual, DelayedProgress Unsustained reforms

Croatia Albania AzerbaijanHungary Bulgaria Armenia(Poland?) Kyrgyz R. Georgia(Macedonia?) Macedonia KazakhstanSlovenia (Moldova?) Moldova

Russia UkraineTajikistanRomania

II Sustained V Limited or ReversedBig-Bang Reforms

Estonia BelarusCzech R. UzbekistanLatvia TurkmenistanLithuaniaPolandSlovakia

Figure 5.2 Transition countries and type of reform strategy

Table 5.1 also shows the number of years under communism, often notedas a determinant of the transition path proxying for ‘market memory’. It isa matter of interpretation whether 70 years v. 40 years is a large difference,and Chapter 4 has already argued that perhaps three generations is not


Table 5.1 Chronology of events: post-communist countries

Years under Regime Transition Stabilization Europecomm. change start start (months) Agreement

Central EuropeHungary 42 Oct 1989 1990 March 1990 (5) March 1992Poland 41 June 1989 1990 Jan 1990 (7) March 1992Czech Rep. 42 Dec 1989 1991 Jan 1991 (1) March 1992Slovakia 42 Dec 1989 1991 Jan 1991 (1) March 1992Slovenia 46 June 1991 1992 Feb 1992 (8) June 1996Croatia 46 June 1991 1992 Dec 1992* (18) June 2004

BalticsEstonia 51 Aug 1991 1992 June 1992 (10) June 1995Latvia 51 Sept 1991 1992 June 1992 (9) June 1995Lithuania 51 Aug 1991 1992 June 1992 (9) June 1995

SEEBulgaria 43 June 1990 1991 Feb 1991 (7) Jan 95Romania 42 May 1990 1991 Jan 1993 (32) Feb 95Macedonia 47 Dec 1991 1992 Jan 1994 (24) –Albania 47 March 1992 1991 Aug 1992 (5) –

CISMArmenia 71 Sept 1991 1992 Dec 1994 (38) –Georgia 70 April 1991 1992 Sept 1994 (38) –Kazakhstan 71 Dec 1991 1992 Jan 1994 (38) –Russia 74 Dec 1991 1992 Jan 1992* (2) –Kyrgyz R. 71 Aug 1991 1992 April 1993 (20) –Moldova 51 Aug 1991 1992 Sept 1993 (24) –Ukraine 72* Aug 1991 1992 Nov 1994 (38) –Azerbaijan 70 Oct 1991 1992 Jan 1995 (40) –Tajikistan 71 Sept 1991 1992 Feb 1995 (40) –

CISLUzbekistan 71 Sept 1991 1992 Nov 1994 (38) –Belarus 72 July 1991 1992 Nov 1994 (40) –Turkmenistan 71 Oct 1991 1992 1999?? (90�) –

Note:* Three dates have been arbitrarily changed by author. For Croatia, the October 1993 date inIMF source is the finally successful effort; the first in August 1991 was stillborn due to war financ-ing, while a second shown here in December 1992 was temporarily successful though reversed.Russia’s first stabilization undoubtedly started with Gaidar’s reforms in January 1992 even if theyfailed. April 95 dates the first large IMF standby which was more successful. Ukraine had a social-ist government as early as 1917, but Republican governments contended for power until 1919.

Sources: Years of communism, transition start, stabilization start from IMF (2000), Tables 3.1 and3.6; regime change, Europe Agreement from Heenan and Lamontagne (1999a).


really very different from two generations. Since the Second World Warperiod did not exactly provide a normal market experience for the popula-tions of Central Europe, only those in early adulthood or older couldremember the market from the 1930s. They would have been well over 60years of age at the start of transition. On the other side of the argument,there is no denying the broad correlation: countries with 40–50 years undercommunism generally moved more rapidly on economic transformation,those with 70-plus years moved more slowly. The exceptions suggest otherfactors at work. Russia with the longest communist history at least attempteda stabilization and big-bang within a few months of its regime change, whileMoldova with a communist period about the same as in the Baltics delayedthis for two years. Kyrgyz Republic was the third Former Soviet Union (FSU)case with a relatively early stabilization effort within 20 months, though itdid begin other reforms even earlier. For Russia and Kyrgyz this was mostlyattributable to ‘accident’ of personalities,10 while Moldova was likeRomania where political instability put off serious efforts for nearly threeyears. War conditions in Croatia and Macedonia caused some delays albeitmuch shorter ones than seen in the war-torn Caucasus and Tajikistan.Bulgaria had a shorter history of communism than USSR successors, but itwas well-known as the most fervent adherent to the communist ideologyamong the satellite countries, demonstrated symbolically by its formalrequest to become a republic of the USSR.11 Furthermore, at the start of tran-sition its first government was not radically changed, being comprised ofcommunists albeit younger and mildly reformist. Nevertheless it undertookan early first effort at stabilization. The number of exceptions to the com-munist continuity thesis is large enough to conclude at a minimum that itis only a partial explanation. The next section will attempt to reconcilethese inconsistencies.

An equally important measure of delay concerns the deeper reform ofthe economy, liberalizing prices, markets and private activity. The move-ments and milestones of the EBRD’s TPI which capture these changes areshown in Table 5.2. While Table 5.1 focused on starting points and an indi-cation of how soon first efforts at reform were undertaken, Table 5.2 givessome measure of how quickly results were achieved: how many monthsfrom regime change to inflation stability defined as 5 per cent month infla-tion (while this seems still high recall annual inflation in many countrieswas in the thousands of per cent); the year that the private-sector share ofGDP reaches respectively about half and two-thirds; three milestones ofprogress to a fully-functioning market economy as measured by the TPI;and the points change of the TPI index in the first three years after regimechange (del TPI).

The three milestones have the following rationale. A TPI value of 2.5represents the average reached by Central European and Baltic countriesthe year before their GDP started to recover, reflecting the notion that at

least this much progress was needed to turn around the transitional reces-sion. The second I define as ‘half-way to market’ with a TPI value of 2.7,halfway on the EBRD scale from central plan (1.0) to full-market (4.3). Thethird milestone with a TPI value of 3.3 I define as ‘near-market’, reflectinga situation where market forces start to become more important than gov-ernment plans or directives. This is not simply a mechanical and arbitrary


Table 5.2 Milestones of transition progress

Months to Year priv. Year Year TPI Year TPI del TPI5%/months sec. share TPI half-market near-market first

inflation 50%/65% �2.5 �2.7 �3.3 3 yrs

Central EuropeHungary 0 1993/96 1992 1993 1995 1.0Poland 12 1993/97 1992 1993 1995 1.2Czech Rep. 0 1993/94 1992 1993 1994 1.6Slovakia 0 1994/96 1992 1993 1995 1.6Slovenia 10 1995/00 1993 1993 1998 0.9Croatia 25 1996/03 1994 1995 2000 0.6

BalticsEstonia 14 1994/95 1993 1994 1996 1.7Latvia 14 1995/98 1994 1995 2000 1.4Lithuania 10 1994/95 1993 1994 2000 1.6

SEEBulgaria 45 1995/98 1995 1997 2002 1.0Romania 48 1996/01 1995 1997 [2003 � 3.0] 0.9Macedonia 28 1996/03? 1995 1996 [2003 � 3.0] 0Albania 15 1994/96 1995 2000 [2003 � 2.8] 0.9

CISMArmenia 45 1996/02 1996 1998 [2003 � 3.0] 0.5Georgia 45 1996/02 1996 1997 [2003 � 3.0] 0.4Kazakhstan 39 1997/02 1996 1996 [2003 � 3.0] 0.8Russia 44 1994/97 1995 1995 [2003 � 2.9]* 1.3Kyrgyz R. 29 1996/02 1994 1995 [2003 � 3.0]* 1.5Moldova 32 1998/03 � 50 1994 1995 [2003 � 2.8]* 1.1Ukraine 51 1996/02 1996 2002 [2003 � 2.8] 0.4Azerbaijan 40 2001/03? 1998 2000 [2003 � 2.7] 0.3Tajikistan 60 2002/03 � 55 1998 2003 [2003 � 2.7] 0.3

CISLUzbekistan 43 [2002 � 45] [1995 � 2.4] [2003 � 2.2]* 1.0Belarus 42 [2002 � 25] [1995 � 2.1] [2003 � 1.9]* 0.6Turkmenistan 70 [2002 � 25] [1997 � 1.6] [2003 � 1.3]* 0.1

Notes: * denotes a TPI score constant or reduced in recent years, as shown in CISL cases with maxi-mum under col. 3–4. For Russia the maximum was 3.0 in 1997; Kyrgyz has remained at 3.0 since1997, and Moldova at 2.8 since 1998. ? indicates data not available after 2002, but recent trend if pro-jected would give 65% by 2003. [x] denotes latest or closest relevant value of TPI achieved.

Sources: Months to reach 5% monthly inflation taken from Kraft (2000) for 10 cases, IMF reports forthe rest. Other data author’s computations based on EBRD Transition Report 2003 and PI values as inChpater 3 Appendix.


choice of a value much closer to 4.3 than 1.0, but reflects the qualitativedescriptions the EBRD gives for the value 3.0 in different dimensions:12

For example, under enterprise governance ‘significant and sustainedactions to harden budget constraints’; under prices ‘significant progress inprice liberalization’; under trade ‘removal of almost all import and exportrestrictions’; under competition ‘substantial reduction of entry barriers’.One might have chosen an even higher value for ‘near-market’ as theEBRD also notes that at 3.0 ‘substantial state procurement continues’ andthat small-scale privatization is complete only at 4.0. But there is nointent here to be precise or categorical in any of these definitions. Besides,as seen in Table 5.1, only 10 of the 27 countries had reached the 3.3 levelby 2003, thus setting a higher value would not be very illuminating or dis-criminating.

The delay in achieving stabilization is measured here as the sum of thedelay in undertaking an effort and the time to reach 5 per cent inflation permonth on a sustained basis for at least a year;13 later reversals are discussedseparately. All but one CEB country achieved this within about a year;Croatia had two unsuccessful attempts in 1991 and 1992 as war activitiescontinued, but the October 1993 programme achieved an almost immediatedrop of inflation well below 5 per cent per month. The SEE and CIS groupsgenerally suffered through extremely high inflation levels for three to fouryears with the exception of Albania (15 months), Macedonia (28) and theKyrgyz Republic (29). Several countries saw reversals. Bulgaria and Romaniawho only reached this level about 1994 continued deficit financing due todirect and indirect subsidies of state firms (most importantly farms inRomania) so even that achievement was reversed with a resurgence of infla-tion in 1996–97. Albania’s reversal was due to a crisis of confidence wheninadequate control on new private investment funds allowed many shadypyramid initiatives which exploded in 1997 with spillover to legitimatefunds. In Russia after the 1998 default and sharp devaluation there was a rel-atively mild and short-lived resurgence of inflation which spilled over tosome of the neighbouring countries, but since then inflation has fallen tosingle digits annually in most CISM countries, and not above 15 per cent. Inthe CISL group, Belarus and Uzbekistan continue to have annual inflationwell above 20 per cent. Their initial stabilization efforts were about as effec-tive as of the CISM countries, but have not continued towards the ultimategoal of single-digit annual inflation seen in virtually all other transitioncountries.

The difference between the advanced and lagging countries are equallystark on the milestones for private-sector share. For Central Europe and theBaltics, 50 per cent was achieved within two–four years of the start of transi-tion, and the two-thirds mark in another two–three years, that is in wellunder a decade for most of them. Even Poland, where large-scale privatizationwas intentionally delayed, reached the two-thirds point by 1997, much of

this attributable to the boom in de novo enterprises. In the SEE region thiswas only marginally slower, whereas in the CISM (the CISL needs littlecomment, the figures fit well the label ‘limited’) some interesting exceptionsexist. Both Croatia and Slovenia were much slower privatizing but for differ-ent reasons explored later in the book. Albania and Bulgaria, despite theirproblems of achieving stabilization, were not much behind the leaders inCentral Europe. Nor indeed was Russia, but its rapid ‘insider’ privatization strat-egy may have had severe negative effects leading to oligarch development; thisis elaborated in Chapter 6. Indeed, it is reasonable to ask if rapid privatiza-tion is a positive sign of progress; the answer is not necessarily so, but in thischapter the aim is merely to describe the different pace and trajectory ofreforms pursued by countries. The judgments on the consequences comeelsewhere in the volume.

The most dramatic difference between leading and lagging reformers isseen in the trajectory over time towards the goal of a functioning marketeconomy, in the TPI. Whether it is the growth-recovery threshold, the‘halfway to market’ point, or the ‘near-market’ one, the CEB group is clearlyfar ahead of all the others. While they long ago passed the ‘near-market’point, none of the others come close except Bulgaria which changed courseafter the United Democratic Front victory in 1997 and the adoption of asuccessful currency-board regime. Three countries in the CISM group wererelatively quick starters: Russia, the Kyrgyz Republic and, to a more limitedextent, Moldova. They can be considered to have taken a big-bang approachinitially, and indeed the last column of Table 5.2 shows that their TPI indexincreased in the first three years from the start of transition by 1.1 to1.5 points, little different from the values for many of the early rapid reform-ers in the CEB groups. However, the process was not sustained; in all thesecases stabilization took much longer, and structural reform progress stalled oreven reversed somewhat so that none of them had reached the ‘near-market’TPI even as late as 2003. Russia reached its maximum (3.0) in 1997 and mayhave slipped marginally since; Moldova has not moved forward from a2.8 score in 1998, and Kyrgyz has similarly stalled at 3.0 since 1997.

The choice of strategy for economic transformation

This brief review of the events and milestones provide the rationale for thecategorization of countries by type of transformation strategy as shown inFigure 5.2. The five groups are based on specific values in Tables 5.1 and 5.2as follows:14

I Advanced Start, Steady Progress: those which may not have needed a bigleap or big-bang given an advanced start but pursued steady progress tostay in the forefront [TPI(0) � 1.5� and TPI(�4) � 3.0�].

II Sustained Big-Bang: those which undertook and sustained an initial big-bang [TPI�3 � 3.0� and no inflation resurgence].



III Unsustained Big-Bang: those which undertook an initial big-bang butaborted it and/or were unable to sustain it [TPI(�4) � 2.5; but stalledand/or followed by resurgence of inflation].

IV Gradual, Delayed Reforms: those which delayed the start of reformsand/or pursued a gradual and not always steady pace [TPI(�4) � 2.5;stalling, reversals, restarts].

V Limited or Reversed Reforms: those which undertook very limited reformsor reversed early progress to remain at low levels [TPI(�4) � 2.0; at notime reach 2.5].

The clear cases of countries with an advance start are Hungary and the for-mer Yugoslav republics with estimated TPI values in 1989 of 1.5 or more.Macedonia may not be appropriately classified here despite its initial TPIvalue because it did not – or was unable given the conflicts and embargo ofthe Yugoslav region – pursue steady reforms for the first few years. It may bemore correct to put it in the intermediate category with Bulgaria, whose tra-jectory it follows closely (see Appendix, Chapter 2). On the other side,Poland which by wide consensus was a sustained big-bang case, might alsobe classified as an advanced-start case, with TPI value of 1.3. Many analystshave in fact argued that Poland, while perhaps not as advanced as Hungary,owed its rapid progress to considerable partial reforms in the late 1980s.15

While noting this, I have chosen to keep Poland in the sustained big-banggroup with the three Baltics, the Czech Republic and Slovakia. Note inTable 5.2 that the change in the TPI value over three years was well above 1.0points for the big-bang cases, and even above 1.5 for the Baltics and formerCzechoslovakia; in contrast it was 1.0 or less for the advanced starters.

In Type III, about five–six countries must be considered as attempting abig-bang but unable to sustain it – though the start was early in only thecases of Russia and Albania (and Bulgaria?), and much delayed in Romania,Moldova and Kyrgyz Republic. It is a toss-up whether Moldova falls in thisgroup or the delayed-reform one; while it started stabilization and reformssomewhat earlier than the other CIS countries and achieved a lot in the firstthree years, by 1997 its reforms stalled and moved haltingly so that after2000 many of the late starters had caught up. Some major reversals in thelast few years – partial renationalization of airline and power companies,reversal of agricultural marketing liberalization, would argue for groupingMoldova with the gradual, unsteady reformers.

Most important here is that no country of Type III was able to sustainreforms, as inadequate hard budget constraints prevented full attainmentof stabilization objectives, and continued political struggles over the pace ofreforms and the redistribution of wealth led to reversals of the process. Ofthese, eventually only Bulgaria was able to get back on track as of about1997–98, to a considerable extent in reaction to the external commitment ofEU membership desires.

The gradual reformers comprise most of the CISM group, countries thatgenerally delayed a serious stabilization effort for more than two years, thateven then pursued a pace that was much less rapid than the preceding cases.Considering only the recent years, Russia and Kyrgyz Republic, whilemoving forward much earlier, appeared to have exhausted themselves by1997, from which time as noted the first actually undertook many reversals,especially in the area of external trade restrictions, while the second appearsto have stalled since 1997. Nevertheless, both remain at the top of the CISMgroup as seen in Figure 2.1.

Finally and unsurprisingly, Belarus, Uzbekistan and Turkmenistan haveclearly chosen a strategy of not simply slow, but minimalist market reforms.In fact early progress in Belarus and Uzbekistan has been gradually erodedand reversed.

I end this discussion with the question: have any countries pursued grad-ual reforms according to the institutionalist model? Figure 5.2 does not showthis as a type for the simple reason that no country appears to have pursuedthat sort of strategy, notwithstanding much rhetoric to the effect that thesociety needs first to put in place market institutions. The hard evidence forthis conclusion is simple; under a pure institutionalist rationale, marketliberalizing reforms are delayed while institutional reforms move ahead toreach some level considered minimally necessary, or at least the two moveforward simultaneously.

The EBRD Transition Report 2002 investigated the different paths for thesetwo categories of reforms which they define as first and second-phasereforms,16 and reach two conclusions. First, in all transition countries the lib-eralization reforms came first, with a considerable lag for the institutionalones. Second, the lag of institutional reforms behind liberalizing ones ismuch greater in CIS countries where the gradual reformers are – it is particu-larly large in the CISL group – than in Central Europe and the Baltics wherethe rapid reformers are found. This was already shown in Figure 2.4, and inthe Appendix data of Chapter 2, which confirm there was no single casewhere institutions came first, or even in parallel with liberalization reforms.This would seem to put the lie to assertions by many policy-makers andadvisers in the early 1990s that they were delaying market reforms for thesake of putting in place the proper legal and institutional framework neededin a market economy. If, for the sake of argument, the three CISL countriesare taken to best represent gradualism at work, it is striking that not onlydid the TPI average for institutional reforms lag behind the liberalizing ones,but that the group average has remained unchanged at 1.5 since 1994(Figure 2.4). In contrast, the rapid Central European reformers had by 1994already a much higher value (2.7) and continued to improve with a 2004 TPIvalue of 3.3. Even the CISM, which in 1994 was at the same level as the CISL,progressed slowly but steadily on institutional reforms. This claim to begoing slowly while preparing institutions was surely one of the biggest and



most scandalous canards that were trotted out in the early debates, and theevidence 15 years later strongly confirms there was a different hiddenagenda of policy-makers.

Commitment to a liberal market economy

To say as did Bunce (1999) that countries with a commitment to a ‘visionthat is fully anti-communist and hence fully liberal’ will move earlier andmore forcefully on economic transformation is fundamentally correct but,without some understanding of the forces that lie behind the commitment,it is an incomplete explanation. Here, two such forces are discussed: a renun-ciation of communism including a concomitant breach with its political rolein society; and pursuit of a national vision integrating or reintegrating thestate into the global economy and the liberal, democratic assembly ofnations.17

Renouncing communism

At one extreme, Baltic countries exemplify societies with a very strongcommitment to ‘renounce communism’ (Laar, 2002). There are no clearcases at the other extreme of countries fully retaining the system of commu-nist central-planning, but several are not far from this. Belarus, sinceLukashenka took over in 1994, and Turkmenistan, from the beginning underNiyazov (Turkmenbashi), exemplify autocracies in which the communistleadership has continued in power and practised a version of the old systemonly marginally modified. Uzbekistan began a bit differently. Despite thecontinuing rule of former communists under Karimov, at first it moved atleast as fast on the economy as several CISM gradualists, but this began to besteadily reversed in the second half of the 1990s. Comparing only these twopolar groupings, the much greater success of the Baltics compared to the CISis unquestioned and finds few if any counter-arguments, and hence itappears to confirm the popular hypothesis that ‘the best predictor of eco-nomic reform [progress] … is the outcome of the first election’.18 But oncethe countries between the Baltic–CISL polarities are considered, the use offirst-election results as a proxy for the commitment to a vision of a liberalsociety in fact does not work all that well. Nevertheless, I will show belowthat the fundamental idea of Bunce about anti-communist hence liberalvision is correct. The problem is measuring this concept.

Consider the three groups that Bunce identifies. First, with a clear non-communist victory in first elections are the Czech Republic, Estonia, Latvia,Poland and Lithuania; added to this are Hungary and Slovenia because‘reform communists’ had been in power for some prior time. This groupis almost identical to the two first groups in Figure 5.2 with a rapid andsteady path of reforms. At the other extreme she lists countries with a clearfirst win for former communists: Serbia, Uzbekistan, Kazakhstan, Belarus and

Azerbaijan, again as predicted the slow-movers. However, in between Buncelists cases where neither communists nor opposition won a clear victory andhence the path of reforms was indecisive: Russia, Ukraine, Romania, Slovakiaand Croatia.

There are several problems with the Bunce test. The list includes only 17 of27 countries and misses out on the actual beginning at the time of regimechanges where no elections took place; she does not consider the veryimportant consequence of early second elections; and miscategorizes severalcases where a renamed communist party wins the first election (for exampleUkraine); and groups several countries incorrectly as having ‘indecisivereforms’ (Russia, Ukraine, Romania, Slovakia and Croatia surely do notbelong together). All of this is easily corrected on the basis of informationavailable in Table 5.1, and compendia on the transformation process in theregion (Heenan and Lamontagne (1999a) and (1999b)). Doing so results inthe groupings of Tables 5.3 and 5.4. Consider some of the changes this bringsto bear on her categories.

Belarus can be considered to have had at least a coalition government atfirst, where the new Chairman of the Supreme Council in September 1991,Stanislau Shushkevich, was a centrist politician with a reform agenda whowas often supportive of the unelected popular front, and able to begin aprocess of gradual economic reform at least equal to the delayed slow-moverslike Ukraine.19 It was only the ‘election’ of 1994 that brought in a clearlycommunist-oriented regime led by Lukashenka after which (as the Appendixfigure in Chapter 2 make clear) there began a reversal on economic reforms.

Bunce’s intermediate list also contains three countries whose experience isnot properly characterized. Characterizing Ukraine as a ‘mixed result’ isquestionable; the Presidential elections were easily won by Kravchuk, theformer First Secretary, and the renamed but non-ideological communists.20

The Rukh opposition and its allies formed a sizeable opposition and agreedto a co-habitation, but were not greatly influential in the government’sagenda, particularly on economic matters.21 Making this correction on Ukrainebrings the results more in line with the original hypothesis: Ukraine’s veryslow progress on the economic front is related to the continuation of the for-mer communists in power. Croatia might be characterized as a nationalistindependence-oriented government, but there was considerable continuityfrom the pre-transition socialist party governments22 who must be consideredas ‘committed reformers’ no less than those in Slovenia.

These adjustments and additions to Bunce’s grouping result in Table 5.3,which also cross-classifies by the type of reform strategy of Figure 5.2. The‘first-government is non-communist’ explanation would imply a diagonalgoing from top-left to bottom-right in the table; but this is only weaklyevident with many exceptions in the opposing corners. The communistor coalition cases which followed rapid reforms are the three countrieswith a clear history of pre-transition commitment to a market orientation(Hungary, Croatia, Slovenia). The three cases of non-communist or coalition



governments which did not pursue rapid reforms were not clear-cut cases ofan anti-communist and liberal vision, but also in each case there is a simplespecific explanation. Armenia and Georgia had strongly nationalist govern-ments that pursued the earliest effort at land privatization, but they weresoon embroiled in civil wars and cross-border conflicts which drained ener-gies and slowed reforms. The case of Belarus was explained above: an earlysecond election solidified the dominance of the former communists andquickly reversed the mild early reform.

Bunce’s attempt to use first-government as a proxy for liberal vision doesnot work well on a strict definition of communist or non-communist, asTable 5.3 makes clear. Thus, she tries to rescue the test by two stratagems:allowing Hungary and Slovenia to be ‘non-communist’ because their com-munists (or socialists in the latter case) were reform-oriented; and she lumpstogether in a too cavalier fashion a lot of intermediate cases. These errors ofcategorization make the test appear less convincing and certainly very soft.But in fact, once these categorization problems are dealt with carefully andone recognizes it is not the party name but the commitment to liberalism that

Table 5.3 Type of first government and reform strategy

Non-communist Coalition* Renamed-communist

Advanced Start, Hungary SloveniaSteady Progress CroatiaSustained Big-Bang Poland

CzechoslovakiaEstoniaLatviaLithuania

Unsustained Big-Bang (Russia?) Albania(Bulgaria 1992) Bulgaria 1991

MacedoniaRussiaKyrgyz R.

Gradual, Delayed Armenia Georgia RomaniaReforms (Georgia?) Kazakhstan

MoldovaUkraineAzerbaijanTajikistan

Limited or Reversed Belarus UzbekistanReforms Turkmenistan

* Coalition here means with a communist or renamed communist party. Coalitions that wereentirely non-communist fall under ‘non-communist’.

Source: Type of government as described in Heenan and Lamontagne (1999 a and b), Handbooksfor Central Europe and CIS. Some cases were marginal or unclear or very short-lived; these arerecorded twice, without brackets in most likely interpretation, in brackets for alternative. Strategyas in Figure 5.2.

is the real driving force, the evidence is found to be very consistent with thethesis. Table 5.4 adjusts Table 5.3 categorizing by degree of commitment:strong (countries with clearly non-communist first-government plus thosewith a strong reformist orientation pre-transition); moderate (coalition gov-ernments plus ‘continuity’ governments with a visionary leader, that is,Russia, Kyrgyzstan23); weak (clear-cut ‘continuity’ governments of formercommunists regardless of party name).

Not all the categorization of commitment is clear, but the few cases whereit is on the margin do little to change the strong implication of Table 5.4.When liberal commitment is defined combining as best as possible actualelection results, pre-transition reformist tendencies and the happenstance ofa visionary leader in a weakly-committed polity, the expected diagonal is farmore clear; that is, the stronger the commitment the more likely reforms willstart early and proceed steadily. Where commitment is weak or wavering,reforms will tend to move slowly, or at best initial early efforts will not besustained, will stall and in some cases reverse. The inadequacy of the party-label as a proxy for commitment is even more clear when one goes beyond


Table 5.4 Liberal commitment at start and reform strategy

Commitment

Strong Moderate Weak

Advanced, Croatia*Start Steady Hungary*Progress Slovenia*

Sustained Poland*Big-Bang Czechoslovakia


Unsustained Bulgaria? MacedoniaBig-Bang Albania

Kyrgyz�

Russia�

Gradual, Delayed Armenia Georgia RomaniaReforms Kazakhstan

MoldovaUkraineAzerbaijanTajikistan

Limited or Reversed BelarusReforms Uzbekistan Turkmenistan

* Reform oriented leadership even before transition starts.� Visionary leader in a weak commitment polity.


the first government and observes the effect of government changes onreform strategy. This is best discussed in the context of the Przeworski thesis.

The prediction of Przeworski (1991) that the pain caused by too-rapid eco-nomic transformation would quickly lead to a replacement of democraticfirst governments is realized in several countries as Table 5.5 shows and forthe reasons he raised. Kim and Pirtilla (2004) provides a nice analysis of vot-ing patterns in second elections confirming that people were most con-cerned about the loss of employment and reduced social benefits caused bytightening budgets. But Przeworski’s associated prediction that these gov-ernments would reverse economic and democratic transformation did noton the whole come true. The clear exception is Belarus where both of thesevisions were indeed sharply reversed, and to a lesser extent Azerbaijan wherethe short-lived hope for democratic development was nipped in the bud andeconomic liberalization was slowed but not reversed.24

In Hungary (1994), Poland (1993), Slovakia (1994) and arguably Albania(1997) one sees the clearest cases of former communists returning to somedegree of power in government – sometimes in a coalition as in Poland – butnot changing the course of policy very much, except perhaps to ease a littlethe pace of economic transformation and its first negative impacts on thepopulace. The trajectory of the EBRD Transition Index is not perceptiblydifferent after these changes than it had been before (Appendix, Chapter 2),except slightly in Albania where it is slowed but certainly not halted orreversed. For the first two the democratic liberalization also continuedunabated. In Slovakia, this dimension is the one that suffers, but not for longas the same voters that may have reacted to early pains and voted in theMeciar government of reconstituted communists in 1994, within four yearsreturned a more democratic government again. In Albania the return offormer communists is attributed by Clunies-Ross and Sudar (1998) to very

Table 5.5 Countries with reversals in government

Non-communist to Communist tocomm. or coalition non-communist

Advanced, Start, Hungary (1994)Steady ProgressSustained Big-Bang Poland (1993)

Slovakia (1993) Slovakia (1998)Croatia (2000)

Unsustained Big-Bang Albania (1997) Bulgaria (1997)

Gradual Reforms Georgia (1992)Azerbaijan (1993)

Limited Reforms Belarus (1994)

Source: As for Table 5.3.

different concerns; the populace was not at all unhappy with the privatiza-tion of land and new-found freedoms after the dungeon-like years underHoxha, but was enraged at their financial losses in the crash of pyramidschemes of 1997.

The degree of conversion of former communists to a liberal vision wasperhaps strongest in some Yugoslav republics and in Hungary and Poland.I discuss the last two. The Hungarian Socialists (renamed-communists) tookpower in coalition with the Alliance of Free Democrats in June 1994, and,despite the platform of easing the transition pains caused by the previousgovernment, appointed as Minister of Finance a well-known reformer andbanker, Lajos Bokros, who proceeded to implement an even harsher budgetarypolicy (see Pittaway, Chapter 2 in Heenan and Lamontagne, 1999). Thisechoed the Polish result in 1993 where the Democratic Left Alliance incoalition with the Peasant Party replaced Solidarity, but named as its DeputyPrime Minister for the economy Grzegorz Kolodko, who though a vocalcritic of the too-harsh measures of Balcerowicz, was an unquestionedproponent of market transformation.25

Slovakia’s first steps on the economy took place before the Velvet Divorcein 1993, and cannot be ignored since they were not reversed by the new1994 Slovak government of a more mixed nature. Furthermore, the govern-ment changed once again to a non-communist one in 1998, with expectedeffects on the pace of economic (and political) liberalization. Croatia, too, issurely in the category of committed reformers at least on the economic side.Its first government was indeed composed of many former members of theYugoslav (and Croatian) Socialist Party, but so too was that of Slovenia andHungary, both of which Bunce correctly designates as having been committedreformers in the late socialist period.

All of this leads to the following conclusion: the fact of a non-communistfirst-government is not a good measure of the degree of renunciation ofcommunism. There is no objective metric available to measure the strengthof such renunciation and one must inevitably seek to understand the deeperbut unfortunately vaguer notion of commitment to liberalism. In doing so, itis useful to consider party identification, but only in combination with theother forces mentioned at the outset: nationalist sentiment; external visionof a European anchor; the negative force of the hidden agenda behind for-mer nomenklatura expressions of support for the market and democracy;and the happenstance of a visionary leader.

Nevertheless, there is something to the simpler view that replacing thecommunist governments with new ones is a recipe for success; indeed Laar(2002) insists it is necessary for success because ‘the state apparatus from theold regime is unsuitable to implement the appropriate path. It can onlytransplant corruption and “telephone rights” [Laar means insider privileges:Au.] from the old system to the new.’ In the context of Estonia and the otherBaltics it may indeed have been impossible to proceed with any significant



involvement of the nomenklatura because it shared neither the importantnational vision nor the liberal market one. In other Central European coun-tries much of the old nomenklatura was much less conservative and readyto ally with others, or even on their own (Slovenia) move with determinationon the vision which ‘bundled two projects: democratization and economicreform’ (Bunce, 1999). This is seen not only in the cases at the upper-right-hand corner of Table 5.3, but even more so in the Central European caseswhere second elections replaced non-communist governments withrenamed-communists, sometimes in coalition.

National vision and economic liberalism

I have often criticized Baltic nationalism, but in the post-communistworld it is of irreplaceable importance to providing some sort of hedgeagainst blatant corruption and in mobilizing a sense of service andsacrifice. (Lieven, 1993)

Few observers would disagree with the above prescient remark on the posi-tive role played by nationalism in the Baltic transformation. But manywould and indeed do emphasize the negative consequences that national-ism may have, ranging from the vicious fighting and resulting bloodshed ofthe post-Yugoslav space to the same in the Caucasus, and including manycases of mistreated minorities throughout the region. The negative effectsare those more often evoked when the word ‘nationalism’ is used, despitethe historical evidence that it can also be a force for good. Just as the ‘nation-alism’ of Allied nations in the Second World War provided the strength tocombat the negatives of Hitler’s nationalism, Lieven argues nationalismcontributed to the bloodless dissolution of the Soviet empire in the late1980s. As Tsygankov (2002) shows, it was not only the Baltics who were kepton a clear path to democracy and a free market by this force, but to somedegree, this also played a role in the CIS countries, with those having agreater ‘degree of national identity strength’ following a stronger ‘foreigneconomic policy sharply away from … the Soviet empire and toward newpartners’.26

That the negatives of nationalism are more often studied than the positiveones is an issue well-beyond the present study, but two points merit atten-tion as they help understand the link between nationalism and commitmentto liberalism. First is the historical reality that ‘old’ nationalisms which havebeen solidly entrenched for centuries as in Western Europe or NorthAmerica, have, at least since 1950, generated much fewer sharp disputes andviolence,27 while unfulfilled nationalisms often do. The differences betweenentrenched and embryonic nation states is often reflected in the terminol-ogy used. Nationalists of long-ago revolutionary wars were called by theimperialist force rebels, traitors and the like; today they are patriots.Nationalists taking up armed struggle for independence or autonomy are

today called patriots by their supporters, rebels and terrorists by the rulingpowers. ‘La plus ça change!’ That ruling powers easily abuse this value-ladenterm is too well-known to need examples.28

Second, it is important to distinguish the difference between cases whereviolence or negative effects come because nationalism is resisted by imperial-ism, and cases where nationalism leads to mistreatment of minorities. If alegitimate nationalist desire for independence is stymied by an imperialistforce which resists autonomy or independence, the ensuing violencecannot be primarily attributed to the evils of nationalism, but should equallybe explained by the evils of imperialism. The refusal of Milosevic’s rumpYugoslavia to accept the independence of Slovenia and Croatia was surelypartly responsible for the initial fighting. At the same time, a comparison ofthese two illustrates well that a strongly nationalist environment can simul-taneously contain the positive and negative forces. Since, as Croatia’s strugglewith Yugoslavia (Serbia) went on much longer than that of Slovenia, it led tomutual atrocities and spillover of harm to minority Bosnians in the contestedterritories. This simultaneity in much less violent form was observed in theBaltics and their treatment of Slavic minorities, who had come there duringSoviet times. No matter the correctness of the historical argument about this‘colonization’, the Baltic countries eventually had to yield to pressures of theEU to soften their citizenship laws for the minorities. Herein lies a nice para-dox of how positive nationalism drove the Baltics towards European values,and how the demands of EU membership led them to attenuate the negativeelement of minority treatment. Mature nation states or national groups thathave attained secure status are not immune from this simultaneity. Accordingto The Economist Catalan nationalists in Spain, who have for years had theright to the use of their language, are said to oppose recent proposals of thecentre to accord the same privilege to other languages, in particular toValencian which some argue is simply a dialect of Catalan.29

The possibility of negative forces in nationalism should not obscure the(possibly simultaneous) positive content. The point here is that in the post-communist transformation, these positive nationalist elements were formany countries an integral part of the strong commitment to the dual visionof democratic and economic liberalism. The logic is simple. A national desirefor self-expression could only be accepted by the global community if itemulated the established nation states with a liberal democracy and openfree-market economy; hence this drove much of Central Europe and theBaltics to emulate such modern nation-states. In South-East Europe and theCIS the force was less pronounced, but not totally absent. Bulgaria andRomania had a strong desire to be taken in by the global community and,despite a much slower pace, did move eventually in that direction. In theCIS, Tsygankov notes that the very weak national basis in Belarus comparedto Ukraine explains why the latter did move forward, albeit slowly, and the



former has remained virtually unchanged. Even in Central Asia one sees avariant of outward looking vision for the nation. Gleason (1997) describeshow the more ‘socialist cosmopolitan’ leaders of Kazakhstan and KyrgyzRepublic (Nazarbayev, Akayev) have over a decade led their country to a farmore advanced state of market operations than was the case in Uzbekistan,or in Turkmenistan.30

An interesting example of how strength of nationalism led to greater com-mitment on economic policy is provided by Kraft (2000), who shows thatthe stronger the nationalism movement, the sooner it led to achievingfinancial stabilization. In his sample of 14 countries, six countries withstrong nationalism all had very early and successful stabilization (Slovenia,Croatia, the Baltics and Moldova), while of five countries with weak nation-alism only one (Kyrgyz Republic) was moderately successful in stabilization,while the others (Kazakhstan, Uzbekistan, Turkmenistan, Belarus) tookmuch longer to stabilize. His measure of success, months from independence(or regime change) to reach 5%/month inflation was used in Table 5.2.

Clearly, strength of nationalism alone does not discriminate adequatelybetween slower and faster reformers, but its positive incentives add anotherbuilding block to the understanding of the underlying force: commitment toa liberal polity and economy.

It is also an incomplete explanation because there are instances where,arguably, nationalism led to less liberal commitment and not more, in par-ticular the diversion of energies from economic reforms to armed conflictseen in extreme form in Serbia until Milosevic was replaced, and in lesser butalso damaging form in Armenia and Azerbaijan fighting over Nagorno-Karabagh. Finally, to show as is done here that a positive drive may havecome from nationalism, is not to diminish the potential and actual instan-ces of its becoming a negative force, if it turns inward from a strugglewith an imperialist resistance to the mistreatment of minorities. There areplenty of examples of this, but the simultaneity of opposing forces fromnationalism should not be considered as too complex to deal with, for it issurely not rocket science to judge when a legitimate national desire forindependence is resisted by an imperialist force or when that desire isunjustifiably aimed at internal minorities and in effect becomes itself a formof imperialism.

Historical evidence and hypotheses of the navigation model

There was no lack of possible navigation charts to follow, so the delay instarting reforms cannot be attributed to this. Figure 4.5 included a hypothe-sis (H4) that delays were more due to a lack of national commitment to a lib-eral vision, and another that a visionary leader could add to a weak national

commitment and push forward early reforms. This chapter has demon-strated three things. First, that a non-communist first-government is not byitself a good measure for such a commitment. But, second, if judiciouslyqualified by an understanding of prior reformist inclinations and supple-mented by evidence of positive nationalism forces, the nature of first-government can be redefined as to the degree of its commitment. Such ameasure, albeit soft in its quantitative rigour, does allow one to confirm thepositive correlation between commitment and short delay. Third, a review ofstated commitments to market reforms contrasted with actual underlyingcommitment and actual steps taken, leads to the conclusion that many policy-makers who argued for the theoretical advantages of gradualism did so notfrom a sincere desire to find the optimal strategy of transformation, but torationalize a long delay creating opportunities for personal enhancement ofthe previous nomenklatura and its transformation into the new capitalistelite. How this was achieved is the subject of the next chapter.


6From Rent-Seeking to Oligarchy to State Capture

177

Overview

All but a handful of the post-communist countries have experienced asubstantial increase in the role of private-sector activity, which means a lotof new capitalists are now active in these economies, many with small enter-prises, but a few owning and controlling vast amounts of industrial, finan-cial and even agricultural assets. The term oligarch is now almost universallyused to describe these few whose holdings are estimated to have values inthe billions of dollars, and unsurprisingly they attract the greatest attentionof popular and academic commentary. Popular interest in the notoriety sur-rounding oligarchs is not simply following the lives of ‘the rich and famous’.Taras in Sumy, Ukraine, reading complex accounts in Post-Postup of how30-plus Rinat Akhmetov became a multi billionaire, is not to be equatedwith Arthur of Eau Claire, USA, absorbing all the juicy details in People mag-azine about Brad Pitt’s latest off-screen romance. Akhmetov’s past and futuredealings doubtless have far greater impact on the well-being of Taras and hisfamily than do Pitt’s twists and turns on Arthur’s life, and they both knowthis. Serious analysis of such oligarchs in the post-communist region has, infact, great importance not only in understanding this episode in history, butalso in countries where they have become politically powerful it is essentialfor a proper perspective into the future. The goal of this chapter is to eluci-date the evolution of this new capitalist class, to understand what conditionsduring the transition nurtured the evolution of such great concentrations ofownership, to investigate who they are and who they were in the Sovietperiod, and finally to make clear why the term is more typically applied tocountries in the CIS group and less so, if at all, in Central Europe.1

These issues are analysed in the rest of this chapter under three headings.The next section describes the process of new capitalists evolving, andargues that the most powerful paradigm for explaining this evolution wasthe extent to which delayed and partial economic liberalization created rent-seeking opportunities. But the rise of the oligarchy is more interesting than

a one-note tune. In concert with the central theme of partial reforms therewere a number of other forces which made a particular country more or lessvulnerable to rent-seeking and eventual state capture by oligarchs: priorhistorical conditions such as a stronger and more united nomenklatura; hap-penstance of a bolder or more cautious leader; the effect of internationalcommitments such as EU association or membership. These complementaryforces are considered in the second section, before moving to the most criti-cal piece of the story: how did rent-seekers become powerful enough to meritthe title ‘oligarch’ and how did they then go on to ‘capture the state’?

The fact of a handful of powerful rent-seekers or lobbyists established in aneconomy does not by itself make Kazakhstan, Ukraine or Russia uniquelydifferent from other developing or even developed economies. As Shleiferand Treisman (2004) correctly observe, in Brazil, Mexico, Italy, Sweden andothers, ‘a few family-run firms often control a large share of national pro-duction’. Indeed, rent-seeking activities were, and are, common to all thetransition countries to some extent, even the more advanced ones. ButShleifer and Treisman’s argument that all of this makes Russia a normalcountry misses an important distinction. A large-enough difference indegree of such power and influence may result in a difference in kind: theoligarchs can not only influence specific policies to favour themselves (thisis how best to define rent-seeking), they may in fact be powerful enough incollusion with each other to ‘capture the state’ in the sense of virtually con-trolling the direction of general policy or even election outcomes. Finally, tobetter understand what is unique about vested interests in these capturedstates, an Appendix to this chapter summarizes the literature on lobbying,entrenched interests and buying political favours in non-transition societies.

Partial reforms and the evolution of rent-seeking

Principal methods of capital accumulation

In almost all of the transition countries, laws prohibiting private enterprisewere changed relatively early, indeed in a literal sense even before the tran-sition truly began. In much of Central Europe it was possible to set up a smallbusiness well before 1989, and in the USSR the key laws on co-operativescame in 1997 and 1998. The combination of legalizing private business withcontinued government regulation of prices, export, import and internaltrade licensing, immediately opened up possibilities for arbitraging or rent-seeking. A new private firm that could obtain a licence to export Ukrainianbeer at still-low Soviet prices in 1992 to Poland where prices had long agomoved towards European levels, stood to make a lot of money very quickly.Seeking out differences in purchase and sale price of beer in different marketsis what the term ‘arbitraging’ means in economics, and is a perfectly validand justifiable way of making profits, as long as the opportunity to buy low


From Rent-Seeking to Oligarchy to State Capture 179

and sell high is available to all. When the low price is set by government, andthe transaction is regulated, not all can participate. Obtaining the govern-ment privilege (not granted to everyone) of buying the beer and exporting itto Poland or anywhere that it fetched a higher price yields a special kindof profit economists call ‘rent’; while influencing the licence-giving agencyis the activity of ‘rent-seeking’. I will also refer to this dual situation of gov-ernment regulated prices and government-assigned transition licences as‘insider arbitrage’. The beer to Poland example is simple enough to illustrateclearly the process of rent-seeking, but in reality it was penny-ante stuffcompared to the major ways in which the first post-communist capitalistsaccumulated wealth.

The incentive for a new capitalist to make maximum profits was universalin all countries because, no matter how rapid the reforms, some time elapsedbetween legalization of private activity and the elimination of governmentregulations which created a gap between market prices and official prices.But the extent of such opportunities and the magnitude of potential rentswas greater the longer the delay and the more partial the reforms. Thus,countries in the Figure 5.2 categories of gradual reformers and unsustainedbig-bang cases (IV and III) became most vulnerable to rent-seeking. Thesedelays allowed a continuation for many years of fiscal and monetary insta-bility, consequent inflation and exchange rate crisis, a plunge in credibilityand a much deeper fall in output; that is, conditions ideal for arbitrage,abuse of insider contacts, and favouritism in granting government privi-leges. This was typical of all CIS countries in varying degrees, but also manyin SEE. The period of instability and extended debate on how best to reformin fact provided a perfect opportunity for development of a rapacious newcapitalist class based on insider-privileges, regardless of the formal mode ofprivatization. The legal opportunity of private enterprise provided byGorbachev’s 1988 law on co-operatives and reinforced in most of the newstates by legalization of ownership in the early 1990s, formed the initial basisof capital accumulation.2 But the truly big opportunities came after 1991, inparticular using four broad vehicles. Three of them were included in the firstphase: direct budget subsidies for state firms and the skimming-off of assetsto associated private firms; borrowing from the government at low interestin times of hyperinflation; and buying energy and other natural resources atstill-controlled low prices and reselling them at far higher market prices.A second phase was particularly focused on the fourth: ‘insider’3 privatiza-tion of large-state firms. Consider each of these vehicles in turn.

Direct subsidies to state firms continued for several years in most of theCIS countries and several large ones elsewhere like Bulgaria, Romania and fora short time in Slovakia. This was not simply the inertia of Soviet-periodoperations, but also a reaction to the heated debates on how to reformwithout generating large social costs like unemployment. Since in all thesecountries private firms could be established, many of the members of the old

nomenklatura did so, including managers of firms or Red Directors, whilethe debate and subsidization went on. The Director and other membersof upper management of a Russian machinery firm like UralMash or aBulgarian industrial conglomerate like Metalchim could and did establishprivate firms that did business with the still state-owned firm: buying andselling products, providing services such as accounting, marketing, organizingthe sports-teams sponsored by the firm, running the subsidized cafeteria –the list is simply endless. Unsurprisingly, the deals were very favourableto the private firm and caused a financial loss to the state firm, often makingit effectively bankrupt. This in turn, at the time of economic crisis andincreasing unemployment, provided the Director with the argument that‘for the good of his workers and local community’ there should be help fromthe state budget. The venality of this position did not go unnoticed to thegeneral public which, incorrectly but understandably, blamed capitalism forthe whole packet of problems.4

High inflation and low interest rates combined with privileged relationsbetween government officials and owners of the co-operatives (often called‘Konserny’) in the late Perestroika period, allowed them to borrow at hugenegative rates – in fact often not even repay. Sometimes this was indirectborrowing, that is by the state firm whose management also happened tohave some new private ventures, and used these funds to ‘buy’ goods orservices from the private firms at highly inflated prices, or to sell at belowmarket prices, in effect transferring resources to that firm. When the timecame to repay after two or three years of hyperinflation, the amount was solow in real terms that it was often forgiven by the government. In Ukrainewith 10,000 per cent inflation in 1992–93, such credits were supplied toindustry and agriculture at interest levels of 20 per cent in a magnitudeestimated at 50–60 per cent of GDP.5

Controlled prices of energy and raw materials at well below world levels,again with privileged access to trade licenses, gave the first large dividends torent-seeking activities. The narrative of Freeland (2000b) on this ‘Sale of theCentury’ is particularly good reading with great personal detail in suchstories as the formation of Lukoil in Russia, United Energy Systems ofUkraine, the analogue in Moldova, and others. There was a special twist inUkraine and Moldova, both energy importers. Unlike Russia, rents could notbe made from exploitation of oil resources, but instead elaborate schemeswere established starting with privileged licensing for imports of Russian gasand oil at ‘government’ prices (low of course) with a government guarantee,and rights for resale at higher prices either on the open market includingexports, or to budget agencies (city heating plants, hospitals, schools) againat higher prices but with government energy subsidies going to these agencies.The government guarantee was often invoked, as some of these licencees,having siphoned off considerable financial capital, were, surprise, surprise,unable to pay the Russian supplier: the government picked up the debt.



Balmaceda (1998) and Puglisi (2003) describe how these operations evolvedand changed hands over time, starting with several firms sharing this rentand moving to greater concentration. The individuals are well-known toobservers of Ukraine including Scherban of Industrial Union of Donbass(later apparently murdered) and Ischenko of Ol-House, followed by virtualmonopolization of the sector by United Energy Systems under Lazarenko, aPrime Minister who soon fell out of favour and lost the privilege to NeftehazUkrainy within the circle close to the President, including Surkis andPinchuk,6 and in the South-East regions to a new rising star Rinat Akhmetov,then only 30 years old.

Once financial stabilization and inflation control began to be achieved bytighter fiscal and monetary policies, and many of the price distortions werereduced, the process of accumulating wealth entered a second phase: obtain-ing favourable conditions for the privatization of the largest state firms.In the CIS countries in particular this was very much a ‘political-insider’process in the broad sense of favouring those who were close to the politicalpowers, even if they were not literally ‘firm-insiders’, that is Red Directors.This non-transparent process was not significantly affected by the choice offormal modality – public vouchers, auctions, direct sales, privileged sales tofirm-insiders to workers or managers – as every one of these modalities couldand was easily abused.

As many have shown (Reddaway and Glinski, 2001, for Russia; Barnes,2003a, for several countries), such unfair privatization favouring the oldnomenklatura and new young entrepreneurs who developed connectionswith the nomenklatura began with the ‘spontaneous’ transfer of assets fromstate entities to associated ‘Konserny’ which accumulated sufficient assets toplay the rent-seeking games above, and by the mid-1990s had amassedenough to be first in line for the privatization of the huge firms of 10,000-plusemployees and potential value in the billions. However, the 1990s privatiza-tion was different from that of the 1980s in two respects. It was much biggerin scale than the early process, and it involved not only the old nomen-klatura but also new young capitalists – though often not ‘as new’ as theymight seem, since many of them were high-level Komsomol members whowere closely associated with the former elite.7 Who the new capitalists wereand who became part of the very small coterie of oligarchs is addressed in thenext section. Even if there were some new ones not from the old nomen-klatura, the delays in stabilization and liberalization in most countries outsideCentral Europe and the Baltics allowed the disoriented top nomenklaturaechelon to regroup politically, relabel themselves using the word ‘democra-tic’ somewhere in the party name, recolour themselves from red to green(not as in environment but as in ‘buksy’), and by quickly accumulating largesums of capital regain influence and indeed control of the political appara-tus of the state regardless of open democratic elections, that is to ‘capture thestate’ as described later in the chapter.

Reform dynamics and potential rents

In Chapter 5 the transition countries were categorized in five groups accord-ing to the type of economic reform strategy (Figure 5.2). For countries ingroup V – Belarus, Uzbekistan and Turkmenistan – with very limited liberal-ization progress or significant reversals, ‘the private sector is very small [and]important elements of the command system [remain], state capture can onlybe minimal’.8 These countries are therefore excluded from the analysis inthis chapter. To understand the link between the strategy of reforms (timing,speed, sequencing) and state capture it is most useful to consider only twotypes:

● Type A: high vulnerability to rent-seeking: countries pursuing gradualreforms, and countries experiencing an unsustained big-bang (approxi-mately groups IV and II in Figure 5.2).

● Type B: low vulnerability to rent-seeking: countries with advanced startand subsequent steady progress, and countries undertaking a sustainedbig-bang approach (approximately groups I and II in Figure 5.2).

Many analysts of transition recognized very early the risks of rent-seekingopportunities arising with too-slow reforms, though the precise balancewas of course not easy to define. The broad relationship was noted inHavrylyshyn (1994 and 1995) and Hellman (1998), the slower the elimina-tion of distortions in the economy, the larger the potential rents. In retro-spect, one can paint a much clearer picture of this causal link.

In type A, high vulnerability countries (Figure 6.1), the formal right ofprivate activity comes early, but other reform elements are delayed for a yearor more. They come in the following approximate sequence: partial price lib-eralization excluding key raw materials, energy and some consumer items;partial liberalization of external trade; small-scale privatization; financialstabilization based on gradual tightening of fiscal, monetary and firm creditpolicies; limited institutional changes such as formal anti-monopoly laws;large-scale privatization.

There is a considerable variation across countries, as for example muchearlier liberalization and stabilization efforts in Russia and Kyrgyz Republic,though these were not sustained, and the much later large-scale privatizationin Ukraine compared to Russia. For Ukraine, in the end this delay made nodifference as the process was no less ‘insider’-oriented than Russia’s loans forshares.9 Also common to this group of countries is the very long delay oninstitutional measures which stimulate de novo small enterprise: simplifica-tion of licensing and tax regulations, and equal access to transparent com-mercial adjudication process. Such a combination resulted in a rapid increaseof rent opportunities and an extended period for accumulating considerablewealth from these rents prior to the sales of the very large state firms.



In the first few years (1989–92 in the USSR successor states) rents were easyto generate, but limited because the economy was still too much state-owned, and the lack of early stabilization sharply reduced the economic basefor these rents. But once small-scale privatization added to the real assetsavailable for purchase and stabilization began to take hold, the rent poten-tial took off (at the inflection point about year 4–5, Figure 6.1) with wide-spread use of the mechanisms described above, and continued to rise aslarge-scale privatization provided the new capitalist class an even greaterbase for profit-making. At some point (notionally years 12–14 in Figure 6.1),a maximum level of rents is achieved, by which time the massive privatiza-tions led to a highly concentrated ownership structure because the numberof individuals allowed insider privileges was necessarily small. If liberalizingreforms were to continue, with elimination of all price and regulatory dis-tortions, removal of barriers to entry and establishment of a level rule-of-lawplaying field, the potential rents would begin to shrink and eventuallyapproach very low levels again. It is precisely to prevent this decline in rentsand maintain a status quo near the maximum that the largest rent-seekers,with enough financial power to wield great influence, put on the mantle of‘oligarch’ and go beyond lobbying governments to controlling them and‘freezing’ the transition at a partially-liberalized state.

Type B countries – Central Europe and the Baltics and some in SEE – areshown as low-rent cases in Figure 6.2. The strategy of reforms also begins

0

2

4

6

8

10

12

14

16

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Years of transition

Ren

t lev

el

Figure 6.1 Rent-seeking type A: gradual or aborted big-bang reformers, highvulnerability

with an early legalization of private activity, indeed in several (Hungary,Poland, the successor states of Yugoslavia) this preceded the beginning oftransition by many years. But in these countries some marginal liberaliza-tion of prices and external trade came earlier, already reducing distortionsand therefore potential rents. The main difference vis-à-vis type A is thatwidespread liberalization followed very soon after and over a short period:quick stabilization, extensive price liberalization including for key rawmaterials, small-scale privatization, and institutional changes encoded incommercial codes. The result was that while the rent curve rose, at first per-haps even more sharply than in the type A, it very soon reached an inflec-tion point in year 3–4 after which the opportunities were rapidly curtailed byremoving the distortions underlying insider arbitrage and closing off thevarious state privileges of explicit and implicit subsidies, monopoly licensingand so on. Rent opportunities reached an early and low-level peak and con-tinued to decline as the increasingly open and competitive environmentlimited any excess profits or rents.

Finally, the much more open, transparent and disciplined privatization oflarge-scale enterprises virtually shut the door to the huge concentrations ofwealth, and reinforced the establishment of a small group of oligarchs intype A countries. Just as later large-scale privatization in Ukraine than Russia


0

2

4

6

8

10

12

14

16

Ren

t lev

el

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Years of transition

Rent HIRent-LO

Figure 6.2 Rent-seeking, types A and B


had little effect on the final outcome, so too earlier rather than later privati-zation in the Czech Republic had some but limited effect: a somewhatfuzzier governance situation ensuing in a hiccup of economic stagnation inthe second half of the 1990s. In the grand scheme of things this was notenough to put the country into the same state capture league as the CISones. As noted several times, the anchor of EU membership also played a roleto limit the damage.

There were cases in type B of falling behind and only catching up much later(Croatia and Slovakia), as well as cases of gradual reformers with a more suc-cessful second start – Bulgaria and Romania, who may have shifted from thehigh-rent to the low-rent curve. For the most part, however, Central Europeminimized rent-seeking activities early by quickly reducing the major distor-tions of government policy which provide the opportunities for privilege-based profit-making. It is not surprising that while the key phrase ‘oligarchs’yields numerous internet entries for Russia, Ukraine and other CIS countries,for Central Europe the first reference I found with Google is to the eleventh-century magnates in Hungary and Poland! This is not to say there was no rent-seeking and corruption, or that it is eliminated fully. All of these countriessuffered from these ills early, and all of them continue to experience somedegree of corruption, because the roots of bribery and lobbying efforts remainin all societies: there is some minimum amount of government regulation ineven the most marketized economies (see the Appendix). But the differencein degree is so large that it is in fact a difference in kind. In Central Europethere are new millionaries and enough insider privatization scandals. Butthese involve magnitudes in the tens and rarely the hundreds of millions invalue. As an illustration, the Forbes (2005) list of world billionaires show onlyPoland with two. In contrast, Ukraine, with a GDP about half Poland’s (due tomuch lower per capita income level) appears with three billionaires, andRussia, with a GDP 2–3 times that of Poland, has twenty-six billionaires.Nissinen (1999) for Latvia, one of the advanced reformers with the highestcorruption indicators, explains succinctly that difference: ‘Pressure groupactivity is focused rather than universal. Private companies have neither theincentive [my italics] nor the means to interfere with the formulation of gen-eral direction of the [policy] … They become active on a specific questionwhere their business interests are directly involved.’ This activity may belabelled as lobbying, which was, is, and will continue to be common in allmarket economies, as distinct from state capture, which is limited to a fewpost-communist states and perhaps post-colonial African ones as well.

Other forces affecting vulnerability to rent-seeking

The evolution of oligarchs

While the pace and consistency of reform strategy followed by each countrygoes a long way to explain why some fell prey to considerable rent-seeking

and oligarchic developments while others did not, this was not the onlyforce at play. The effects of external commitments, such as EU membership,are left to Chapter 7. Here, a few observations are warranted on specificcountry cases where two other influences affected the relationship betweenreform delay and state capture: prior historical conditions, and the happen-stance of a visionary leader.

It is often said that in the Soviet period the nomenklatura was morepowerful in the USSR than the satellites, and this was reflected in a higherlevel of informal networks, including the Soviet-era mafia.10 If true, thiscertainly implies a greater historical vulnerability in the CIS countries – withthe Baltics eliminating this effect immediately by removing virtually all for-mer communists from important positions. But it is not clear why this wouldbe so, or if indeed it was. A study of informal or underground activities in theSoviet period shows that it was at about the same level in Central Europe, theBaltics and western USSR republics, 20–25 per cent of GDP, and much higheronly in the Caucasian and Central Asian republics.11 This does not seem tobe the difference which can explain why the nomenklatura in Czechoslovakiayielded power to the opposition so much more easily than, say, the one inUkraine; one must rely on saying the opposition itself was not as committedto economic reform as was the one in Czechoslovakia. Nor does it tell theright story about why Russia with perhaps the strongest most centralizedand disciplined nomenklatura, and Kyrgyz Republic, allowed a radicalinsider to take over. Happenstance and personalities are another part of thestory here: Yeltsin was at least passionate about, if not in the end intellectu-ally committed to, the market (Shevtsova, 1999); Havel and Klaus usuallyagreed on little, but instinctively and firmly agreed on the need to turn west-ward; Chornovil and his intellectual dissident colleagues were adamantabout independence for Ukraine, but fundamentally indifferent to economicsystem issues; Akayev, a renowned Academy of Science physicist, initiallyruled from an intellectual attitude and not as an apparatchik or Asian auto-crat. In a word, these inertial facts of history are often the most importantfactor explaining a slight deviation from the central tendency of the naviga-tion model, but do not themselves provide a consistent enough result acrosscountries to support path-dependence interpretation.

The special case of Yeltsin deserves some elaboration because he was notafter all an outsider even to the extent that Akayev was, but very much pastmember of the top echelon of the nomenklatura, a member of the Politburo.His early ambitions for taking over from Gorbachev led to loose allianceswith emerging democratic forces and young technical reformers, all culmi-nating in the famous leap upon the tank of August 1991 which was symbol-ically at least a revolutionary action. Biographers do suggest Yeltsin was amuch more emotionally driven personality than Gorbachev, and it is notunreasonable to believe that the way in which he came to power, not as inSoviet history through Politburo intrigue, but in fact by seizing the opportunity



of disarray in the party and the country, instilled in him a vision of a saviourwho would lead society towards the radical changes it needed. His choice ofGaidar and other young Turks for economic portfolios and his blessing forthe initial big-bang efforts of January 1992, and indeed his later decisions tobring back Gaidar, Chubais, Fyodorov and others, are all actions consistentwith this vision.

But as a child of the nomenklatura, he may not have been completelyconvinced intellectually, and certainly was not cured of his politburo habitsof intrigue and compromise by the tank-epiphany. When the resistance toGaidar’s programme became too strong, Yeltsin retreated at least temporarilybut kept trying to push the reform agenda in fits and starts. In the Loans-for-shares episode, he compromised with the Red Directors lobby by givingthem privileged access to the privatization process. This might be seen as afull retreat to the ‘nomenklatura-Yeltsin’; in fact, it fit perfectly with his splitpersonality of nomenklatura–revolutionary. Loans-for-shares was touted byits proponents and outside advisors as the perfect compromise, because theopponents of economic transformation would be bought-off, while privatiz-ing the bulk of state assets would move the economy far forward on itspath to the market, creating a new capitalist class that in the long run wouldbe the staunchest defender of market institutions. How could Yeltsin notgo for this perfectly balanced tactic? It worked in the short run, but I willargue in Chapter 8 that these new capitalists have not become proponentsof completing the creation of market institutions, but instead powerfulopponents of further liberalization.

Who are the oligarchs?

In the countries where rent-seeking was substantial and an oligarchy hasdeveloped there is one important and fascinating continuity from history,which is the lineage of the new capitalists, and eventual oligarchs. Thereappear to have been four main sources, with some overlap: the upper-echelon nomenklatura including Red Directors of the Soviet period; youngermembers of the nomenklatura dynasties still in the Komsomol in the late1980s; newcomer entrepreneurs seemingly unaffiliated with the nomen-klatura, but known to have made connections to someone in the nomen-klatura early on; and members of the Soviet period underground economy,the ‘mafia’ network of traders. Many books and articles have been writtenon the new capitalists and how they became so wealthy so quickly and thereis no need to repeat such details, fascinating as they are, or even to system-atize precisely the personal history of these individuals. Here I wish only toillustrate with a few examples from these writings.12 Naturally, the informa-tion is often incomplete and unverifiable, especially for the last group, andespecially on the possibilities of criminal wrongdoing being involved atsome stage of an individual’s evolution as a major capitalist.

The upper-echelon nomenklatura provenance was most significant forsome of the Central Asian states, with the formal ownership of large compa-nies in the hands of relatives and close associates of the respective presi-dents, be it Nazarbayev in Kazakhstan or Aliyev in Azerbaijan. The regionalclan structure of the new oligarchs is not really a new fact: the communistnomenklatura there was also reflecting clear leadership powers. What isprobably new is the much smaller role of ethnic Slavs in the upper ranks.The nomenklatura provenance is also seen in Ukraine with family membersbut also close associates of President Kuchma, like Pinchuk, Surkis andMedvedchuk now controlling large ‘holdings’ in many sectors. It includesregional politicians from Dnipropetrovsk, like the erstwhile Prime Ministerand oligarch Lazarenko in the energy business, now exiled and doing timefor money-laundering in the USA. In Russia this provenance of big capitalistsis not as important as the next one, the Komsomoltsy, but does include somevery important players, like Chernomyrdin and Viakerov of Gazprom.

Some middle-level apparatchiks did become important, like Potanin, aDepartment Director in the Ministry of Trade (Souyzpromexport to be pre-cise) who set up the trading company Interros later to become a bank which,‘in a lucky break’ (Brady, 1999, p. 136), took over management of the COME-CON bank’s assets and clients when it was disbanded. How ‘lucky’ or‘insider’ this break was may never be known, but Interros is ranked today bythe World Bank (2004) as the fifth largest of two dozen oligarch holdings. Atabout the same time, Vagit Alikperov, then Deputy Minister of Finance ofthe Oil Ministry also got a ‘lucky break’ according to the interview he gaveto Rose Brady as written up in her book: the government rejected his sug-gestion to set up an independent oil company so he did it on his own, andthus was born Lukoil, today number six in the World Bank ranking. As withPotanin, we will never know how much was luck and how much was theinsider connection.

Arguably, the Komsomol may have been the largest source of this newentrepreneurial talent in Russia, though the definitive PhD thesis quantify-ing the provenance of oligarchs is yet to be written. Certainly a very largenumber of the new capitalists were much below the average age of thenomenklatura members, though by the late 1980s dynastic tendencies wereevident as one of the privileges of the nomenklatura had been to place theirkin in the top schools, in top Komsomol positions, and so on up the ladder.When the definitive thesis is written, it will be very important to identifyany family links of seemingly non-nomenklatura young capitalists to theupper echelon of the party. From the existing accounts, it is clear that severalof today’s major oligarchs had precisely this provenance.

Khodorkovsky of Yukos, now jailed for tax evasion, began business asDeputy Secretary of the Komsomol at Moscow’s prestigious MendelevChemical Institute, using 7,000 roubles from the Komsomol to organizegraduate students who prepared special studies for enterprises ‘at cost’.



This yielded a hefty 169,000 roubles (evidently not a not-for-profit exercise)allowing the network to expand hugely, and eventually generate enough toset up a new bank, MENATEP, in partnership with a small state bank;MENATEP still exists and the fate of Yukos is well-known. Very entrepre-neurial for a 26-year-old indeed. Gusinsky had organized music festivals forKomsomol, and went into a construction co-operative instead of completinghis theatrical studies. This co-operative became another of Russia’s majorbanks, MOST. The Komsomol source includes many others, like Aven ofAlpha, Deripaska the aluminum and nickel king, Bendukidze, Dubinin.In Ukraine this was also evident at least among the ‘mini-garchs’ likePoroshenko in food products, or the deposed banker Zherdytski, now in jailin Germany for fraudulent use of German reparations funds for Ukrainianvictims of war crimes.

The third category, new unaffiliated entrepreneurs, is on the face of it theonly one suggesting pure competitive entrepreneurship. It is not as importantas the first two, and may in fact be less pure than appears at first sight. Thebiographies are not always clear, and often, in the otherwise excellent bookssuch as Brady, are based only on interviews with the individuals.13 Perhaps,most important, in many cases the history of these capitalists shows that veryearly on they turn to the established institutions and hence to partnerswithin the nomenklatura with offers of setting up big enterprises or banks.Nevertheless there are many capitalists who at least began outside the estab-lishment. In Russia there were scientists whose future turned very bleak after1991, and therefore decided to go into business. Aven and Berezovsky are twosuch examples who still figure among the big players a decade later. Butagain, very early on, they became partners with establishment enterprises ornomenklatura individuals to leverage their first big earnings into truly bigopportunities when privatization started. Many others who became wealthyin early years are noted by Brady, but not all were able to make the big leapinto oligarchy, failing to find an insider connection for big privatizations.

Soviet-period price and supply distortions generated many undergroundcapitalists, who in a market economy would probably have been legaltraders. Under Soviet laws prohibiting ‘speculation’ defined as any privateeconomic activity, they were not legal and indeed often engaged in othercrimes besides ‘economic speculation’. They generally had much morewealth than most top nomenklatura people, and therefore should have beenin an ideal position to jump into capitalism. Indeed Handelman (1994)describes in detail how they did try to do this. But it did not work out wellfor most of them because the insiders, who had tolerated the mafiosi in theSoviet period for the financial benefits these could provide ‘informally’when being a capitalist was illegal, wanted to become capitalists themselvesonce it became legal.

It is striking that in the index of Handelman’s book, not one of the bigmafia players of the Soviet period is recognizable today as an important new

oligarch. This could of course reflect anonymity of ownership, but a morelikely explanation is that most of them were unable to go clean quicklyenough to avoid the fate of one such important mafiosnik, Smolinsky, whowas arrested on criminal charges in 1991. An alleged exception noted inPuglisi (2003) may have been, indirectly, Rinat Akhmetov, now regarded asthe top oligarch in Ukraine’s Donbass region. Though the early biography isunclear, many references are made (including in Forbes, 2005) to his inher-iting leadership of the economic fortune built up by Akhat Bragin, allegedlya ‘Mafia’ boss of Donetsk, who was killed in 1995. The 29-year-old Akhmetovtook over the major companies of Lyuks and the local football club, Shakhtar(Miner), but like many Russian ‘new comers’ he decided early on to link upwith powerful insiders, including then governor of Donetsk ViktorYanukovich, to found a bank whence business took off to reach an estimatednet worth over $2 billion.14

In conclusion, it is clear that being part of the upper-echelon nomen-klatura did not automatically lead to oligarchy today. Goldman (2003)emphasizes that in Russia only two came from this source and the rest heclaims, somewhat unconvincingly, ‘were ne’er do wells, from the fringes ofsociety, engaged in black-market speculation’, while Brady (1999) concludes‘some did come from the black-market, but the majority surprisingly wereprofessors, scientists, managers – in other words from the intellectual elite’.Neither of these characterizations is altogether correct, for they seem to sug-gest most new capitalists were new faces unaffiliated with the party. In fact,this was only so on the surface. A large number of these newcomers were inthe Komsomol, the party’s training school,15 many were in middle manage-ment positions; these positions were traditionally the middle rungs on thenomenklatura career ladder. Thus, the second conclusion is that many ofthose who may have been outsiders were only eventually successful in build-ing very large business empires after establishing some insider contact withan existing state bank as firm, or with the patronage of someone in theupper-echelon nomenklatura.

It seems clear that in the CIS countries at least there was a considerablecontinuity from the political power-holders of the Soviet period to the newcapitalists, rent-seekers and eventual oligarchs. Does this mean the new cap-italists were not entrepreneurs in the Schumpeterian sense, but simply inher-ited their power? They were most certainly very entrepreneurial, as some ofthe short stories above illustrate and, more important, far from everyonewith inside connections exhibited such energy, ingenuity and industriousness:that is, entrepreneurship. But equally important, insiders had a great advan-tage over those on the outside, as was best demonstrated by the cases of thelatter seeking early on to bring on board an insider. In a word, in countrieswith partial market reforms to become a big capitalist it was not sufficientto be on the inside, one had to be entrepreneurial enough to seize newopportunities. But neither was it sufficient to be entrepreneurial, because the



continued regulations by the state meant one had to have an insider con-nection as well. The new capitalists, and certainly the powerful oligarchs, arebest characterized as the most entrepreneurial insiders. An unfortunateimplication of this was that entrepreneurial outsiders with no connectionscould not go very far in business. This is elaborated in Chapter 8 under therubric of small–medium enterprise prospects.

From rent-seeking to state capture

Rent-seeking opportunities were not only greater in the gradual reformersoutside Central Europe, but lasted much longer allowing new capitalists toaccumulate considerable wealth. But, until the large-scale privatizationbegan in the mid-1990s, this did not differ greatly from the situation inmany market economies where powerful enterprises lobby for favours (seeAppendix to this chapter). The transfer of these vast state assets to privatefirms not only hugely increased the holdings of the most successful rent-seekers, but led to even greater concentration of ownership, since the largelynon-transparent process favoured the largest of the early rent-seeking capi-talists. The World Bank’s April (2004) study of ownership concentration inRussia finds that about two dozen oligarchic groups account for at least35 per cent of sales, and in key sectors such as energy, chemicals, metallurgyand banking this share is much higher. The high degree of concentration ofownership in Russia at the top is also suggested, approximately, by a simplecount of billionaires in Forbes (2005). The 26 individuals in Russia (it may betwo or three more if some former emigrants and recent exiles domiciledelsewhere are included) out of a 700 total in Forbes’ count is an extraordi-nary number for an economy which accounts for 1–2 per cent of worldGDP.16 Russia, in fact, ranks third in the billionaire count after the USA with340 and Germany 41, but it ranks first if measured proportionately, using theratio: share of Forbes billionaires to share of world GDP. For Russia this isabout 2 to 4, for the USA about 1.5, and Germany about 1.0. In other large,high-income economies in Western Europe (the UK, France, Italy) withworld GDP shares around 5 per cent, the number of billionaires is 10 or lessand the ratio noted above is far lower, about 0.25 to 0.30. The extraordinaryconcentration in Russia is even more clear if compared to some economies ofroughly equivalent GDP size (1–2 per cent of world GDP): Australia, Brazil,Canada, Mexico, Netherlands, Spain, Sweden. For these, the numbers ofbillionaires range from 16 in Canada to four in the Netherlands, and,with the exception of Canada and Sweden, the ratio defined above is wellbelow one.17

Only three other transition countries appear in the Forbes list: Ukraine withthree billionaires, Poland with two, and Kazakhstan with two. In all cases, theindividuals are at the very bottom of the list with at most $1–2 billion; whilein Russia, 15 of the 26 have fortunes ranging from $2.0 to $13 billion.

The much smaller fortunes in other transition countries are to some extent areflection of much smaller economies and do not reflect the state-capturephenomenon. In Moldova, fortunes of less $100 million or even less areproportionately already very large and suffice to provide influence of anoligarchic type. It is nevertheless striking how large are the disproportionsin Russia and the much more limited evidence of such wealth concentrationin non-CIS transition countries. That Poland with a GDP about one-halfof Russia’s counts only two billionaires with assets of about $1.0 billion isillustrative of the much more restricted opportunities for quickly amassinglarge wealth based on rent-seeking and political influence.

The related index of ‘state capture’, devised by a World Bank team led byHellman and Kaufmann, attempts to directly measure the extent to which asmall number of enterprises is powerful enough to not only lobby and occa-sionally influence government measures, but in fact to capture and controlgovernment policy and its general direction. A summary of these results for20 transition countries in 1999 and average by country group is given inTable 6.1.18 Belarus and Uzbekistan can be ignored for the reason cited earlier –they are still closer to central-plan than market economies. For the rest, it iseasily demonstrated that Hypothesis 1 of the navigation model in Chapter 4,that the degree of state capture is strongly correlated with delay in under-taking stabilization and reforms, is consistent with this data. Figure 6.3


Table 6.1 State-capture index, 1999

Hungary 0.10 Armenia 0.11Poland 0.17 Georgia 0.34Czech Republic 0.16 Kazakhstan 0.18Slovakia 0.34 Russia 0.45Croatia 0.43 Kyrgyz Republic 0.41

Central Europe 0.20 Moldova 0.52Ukraine 0.45

Estonia 0.14 Azerbaijan 0.58Latvia (0.22) Tajikistan —Lithuania 0.17 CISM 0.38Baltics 0.18

Uzbekistan 0.08Bulgaria 0.40 Beralus 0.12Romania 0.30 Turkmenistan —Macedonia (0.35)

CISL 0.10Albania (0.35)Bosnia-Herzegovina (0.40)Serbia-Montenegro (0.50)

South-East Europe 0.35 (0.38)

Source: Author’s calculations averaging two concepts of state capture, pervasiveness andconcentration, as reported in Figure 6.3 of Hellman and Schankermann (2000). For bracketedcases, no values are reported in the 1999 study; 1999 values are pro-pated using a 2003 World Bankstudy of ‘crony bias’ (a related but different concept; Sce Hellman and Kaufmann (2003)), Chart 2.


groups the countries into high (Kyrgystan and up), medium (Kazakhstan toBulgaria) and low state capture and correlates this with two measures ofdelay: months from regime change to start of stabilization, and years toreach a TPI value of 2.7, designated in Chapter 5 as the mid-point from acentral plan to a market economy.

There is a clear progression: countries of low state capture began stabiliza-tion efforts no later than 10 months, those with medium capture rangedfrom seven months to 32, and those with high state capture were in therange 20–40 months. There are two outliers, Armenia and Russia. Armenia isan outlier where stabilization came very late (38 months) but the state cap-ture index is suspiciously low. There may be a measurement problem. It is onthe face of it implausible that its state capture is so much lower than eventhe medium level of Georgia, and indeed lower than all but Hungary andSlovenia in Central Europe. Russia with very high state capture was very fastoff the mark on stabilization, four months, but this was not a successfulstabilization and the big-bang effort of January 1992 was aborted in less thanone year as noted earlier.

A more precise correlation between delayed reforms and degree of statecapture is seen in Figures 6.4 and 6.5. The first is a scatter diagram showingthe number of months from the start of transition to inflation stabilization(5% per month) with a correlation of �0.4882. The implausibly low valuesfor Armenia (0.11) and Kazakhstan (0.18) are set at a level 0.25 (similar toBulgaria, Slovakia and Croatia but still lower than other CIS countries) thecorrelation is �0.719. Slovakia and Croatia are also outliers, with an imme-diate stabilization but a quite high state-capture index of 0.34 and 0.43. InSlovakia this reflects a short period of the Meciar government and a tendency

Degree of state capture

Low Medium HighMonths tostart ofstabilization 1–10 7–32 20–40

(Arm.: 38) (Russia: 4)Years to reachhalf-point tomarket 2–3 5 3–11

Figure 6.3 Delay facilitates state capture

Notes: Months to start of a stabilization programme from IMF (2000); half-pointto market economy as measured by European Bank for Reconstruction andDevelopment, Index of transition. A full-market economy � 4.3, a central-planeconomy is 1.0, hence I take the index value of 2.7 as the half-way point.As examples, all of Central Europe reached this level by 1993, the Baltics by1994–95, Russia and Kyrgyz Republic by 1995, Ukraine in 2002, Tajikistan in2003, while Belarus, Turkmenistan and Uzbekistan are still not there.

194

MDA

KGZ

MKD

AZE

RUS UKR

GEO

BGR

ROM

ARMKAZ

HRV

ALB

POL

ESTLTU

SVK

CZE

HUNSVN

0.7

0.6

0.5

0.4

0.3

Sta

te c

aptu

re in

dex,

199

9

0.2

0.1

00 5 10 15 20

Months to 5% / month inflation

25 30 35 40 45 50 55 60

y = 0.0057x + 0.143R2 = 0.4882

Figure 6.4 State capture and delay in stabilization

0.7

0.6

0.5

0.4

0.3

Sta

te c

aptu

re in

dex,

199

9

0.2

0.1

01.5 2 2.5 3 3.5 4

Transition progress indicator after 4 years

y = –0.2306x + 0.9239R2 = 0.4545

MDA

KGZ

MKD

AZE

RUSUKR

GEO

BGR

ROM

ARM KAZ

HRV

ALB

POL, LTU EST

SVK

CZE

HUN

SVN

Figure 6.5 State capture and delay in reform progress


to rent-seeking and vested interest favouritism, which were quickly reversedafter 1998. The values for them are set at the top of the CIS range, 0.20.

A broader measure of early reform achievement is the actual level of TPIreached four years after the start of transition (1996 in CIS and Baltics, 1994elsewhere), which is to say it is not how fast a country moved initially buthow far it had got in closing the rent-seeking opportunities within fouryears. The correlation with state capture is shown in Figure 6.6, with a nega-tive coefficient of �0.593. �0.4545 again adjusting for Armenia, Kazakhstan,Slovakia and Croatia. Other cases also appear to be mismeasured, and in gen-eral the direction of bias understates the correlation between slower reformsand greater state capture. Armenia and Kazakhstan have been mentioned,and also Slovakia under the Meciar government 1994–98 which did indeedexperience a lot of insider privatization resulting in the temporarily highstate-capture index. But the latter’s degree of market liberalization as mea-sured by TPI did not reverse in this period, indeed it kept increasing, albeitslightly slower than in other rapid reformers. When the Dzurinda govern-ment brought back a more democratic and market-oriented policy the back-sliding under Meciar was easily overcome, as there had not been enoughtime to enshrine the new policy.19

Croatia like Slovakia had a high level of market liberalization progress, butthe insider dealings for privatization created some bias towards state capture.This was curtailed by the new government in 2000, determined to changeCroatia’s image in Europe and jump-start the process of EU membership.Something analogous occurred in Bulgaria in 1997 with the replacement ofthe renamed communists (Socialists) by the democratic and market-orientedUnion of Democratic Forces. All of these changes since the 1999 index wascompiled by the World Bank go in the direction of reinforcing the hypothesisthat state capture is least where market reforms move fastest to close theopportunities for rent-seeking.

Kyrgyz Republic and especially Russia both pursued an early start and didreach the ‘halfway’ point within three–four years; however, both as discussedearlier then stalled and even experienced some reversals in their liberaliza-tion. More important, Russia’s large-scale privatization was explicitly insider-oriented, which perhaps unintentionally but inevitably led to developmentof oligarchs and state capture. Privatization in Kyrgyz Republic did not in theend differ much from the clan-based distribution of state assets in CentralAsia, despite the early liberalism of Akayev. It is noteworthy that the masspopular demonstrations in March 2005 were first of all directed towards thelarge retail establishments owned by the President’s family.

The bivariate correlations shown above illustrate the hypothesis that thegreater the delay in stabilization and reform, the higher is the degree of even-tual state failure. In Appendix 4.A, equation (3) suggests a formalizationwhere delay is measured both for the stabilization elements (M 5% INF) andthe broader measure of marker reform (TPI � 4). A regression analysis using

both of these variables, and adjusting for the four countries with question-able SC99 values gives even stronger statistical results than the separatebivariate correlations, and the expected positive coefficient for M 5% INF,negative for TPI � 4.

SC99 � 0.55 � 0.004 M 5% IMF � 0.129 TPI � 4(0.23) (0.002)* (0.072)**

R2 � 0.569 F � 11.24n � 20 significance � 0.0008***

Values in brackets are standard errors.Significance: * � 10%; ** � 5%; *** � 1%.

A regression without adjustments gives lower R2 (� 0.324) and less statisticalsignificance, but with such a small sample the most important test is theconsistency of the signs of coefficients; these remain for all variants of theregression as above. Interestingly, if the questionable cases are simply excluded,despite the resulting very small sample of 16, the statistical fit is stronger(R2 � 0.676), and significance (F, � statistics) remains much the same. Forsome observers, the statistical link between delayed reforms and state cap-ture will remain unconvincing and a simpler and historically deeply-rootedthesis is preferred: Central Europe and the Baltics had a European history towhich they always wanted to return, and this meant that even in the Sovietperiod they were already more liberally inclined. This set the stage for a rad-ical change in governments from communist to democratic, and enabledthem to apply rapid reform measures that the countries farther East could notafford politically. Such an interpretation implies the following hypothesis:the process of rent-seeking, insider privatization and state capture wasinevitable in the CIS and perhaps in some SEE countries, but unlikely tooccur in Central Europe.

This is one of the most interesting hypotheses for historians to debate, buta difficult counterfactual to resolve. This book argues that in some of thelarge post-USSR countries it was not inevitable, even if the forces of nomen-klatura inertia were more powerful than in Central Europe. For Ukraine,Havrylyshyn (1997) argued it may have been avoided if the independencemovement (Rukh) had not explicitly and willingly made what Kuzio (1997)called a ‘Faustian bargain’ with the former communists led by Kravchuk.The latter guaranteed to support independence and the former agreed to letthe old communist hierarchy form a government. In addition, Rukh was amovement dominated by writers and dissidents with few technocrateconomists, and perhaps because of this it did not press for rapid economicreform but was satisfied in the bargain with the nomenklatura’s guaranteedsupport of independence – which has so far been upheld both by Kravchuk



and Kuchma, but at the cost of a strong tendency to oligarch development,state capture, and even a deterioration of democratic standards. Did Rukhhave a choice? This continues to be debated by scholars of Ukraine, but thereis at a minimum agreement that the communists, even while remaining for-mally in power, were at that time uncertain of what the future held, andfeared the opposition movement’s ‘ability to unleash social unrest on a mas-sive scale’.20 This strongly suggests they could have been pressed for muchmore economic reform in the first years.

In Russia, the bargain was between reformers around Yeltsin and the RedDirectors, part of the nomenklatura. Compared to Ukraine, it was moreimplicit and focused on the loans-for-shares process in 1994–95. Both theearly privatization and certainly those in the mid-1990s were intentionallyinsider-oriented to coopt the opposition of the Red Directors. Boycko, Shleiferand Vishny (1995), as well as many others, consider this to have been a wisetactic to achieve a rapid privatization and reach the critical-mass of privateownership needed to avoid reversal. Could this insider relationship have beenavoided? Reddaway and Glinski (2001) argue yes, if Yeltsin had incorporatedmore of the early democratic movement into his government and not reliedso much on young technicians and former nomenklatura. Other critics focuson the fact that before the big-bang was aborted, a lot of progress hadbeen made. TPI had reached the half-way mark by 1995, almost as quickly asin the Baltics, and hence no special coopting was needed. Odling-Smee andThomsen (2003), while urging continued reforms, recognize performancesince 1998 has been very good. There remains much for historians to sort out.

In Armenia, ‘insider-arbitrage’ was avoided for a while with an early liber-alization including the first land privatization in the CIS, but it soon fell intoline with others as a result of the need for support of the war with Azerbaijan.The very early years in Belarus and especially Kyrgyz Republic with relativeoutsiders as new leaders, Shushkevich and Akayev, also demonstrate economicliberalization could be done early in an attempt to cut off the rent-seekers.Thus, there is plenty of evidence for the argument that choices could bemade and path-dependence broken, as this was after all a historical momentof opportunity to greater or lesser degree in all transition countries.

Freezing of transition

To conclude, it will be useful to test the hypothesis in Figure 4.5 that once anoligarch class captures the state, the transition process is frozen (H2). One ofthe tests suggested is that the higher the state-capture index in 1999, the lessthe advance in further transition. It’s not so straightforward to measure theextent of continued progress in transition, since the TPI index has an upperlimit of 4.3 and the uncaptured states had already reached very high levelsby 1999. Mathematically, this means they could not by definition experi-ence such rapid advances as in early years.21 Thus, the actual level of TPIreached in the latest year is in fact a better measure than the change in TPI.

Figure 6.6 shows the clear negative correlation; even without any adjust-ments for the questionable SC99 values noted earlier, there is statisticallysignificant regression coefficient of 0.4878.

Chapter 8 will explore this correlation further and address the question ofreversibility of state capture and a frozen transition; here it is useful todescribe a handful of cases where it was in fact reversed, but at a relativelyearly stage of intermediate state capture. Bulgaria is an excellent example ofoligarchization being reversible: the capture of the state was not unlike thatin the CIS and some other countries in SEE (Hellman and Schankermann,2000, put it in that category in 1997–98). But it was quickly reversed after the1997 elections which brought in a highly reformist government. In the back-ground of this was the desire to be European and the clear signals fromBrussels of what it desired to see. But the most intriguing question is why theSocialist Party consented to a sufficiently open election to risk losing – whichit did. It is also interesting to note the unique process of negotiations withthe IMF on a badly needed stand-by loan: both the Socialist and the UDFpublicly agreed to the conditions of a programme to be implemented afterthe elections.22 In Romania, no sharp return to a liberalization policyoccurred, though the desire for eventual EU membership probably mitigatedthe excesses and kept the liberalization process moving ineluctably forwardalbeit with fits and starts. Slovakia may be the best example of the latterforce. Recall that all three countries were included in the early groups ofpotential candidates under Europe Agreements – Slovakia in 1992, the othertwo in 1995. Most recently, Croatia has seen a similar development motivatedalso by EU membership desires, and in 2004–05 one sees this force pushingrecent development in Ukraine and Georgia and providing a potential escape


3.5

4

4.5

3

2.5

2

1.5

Tra

nsiti

on p

rogr

ess

indi

cato

r, 2

004

1

0.5

00 0.1 0.2 0.3 0.4 0.5 0.6

State capture index, 1999

y = −1.9478x + 3.8244R2 = 0.4878

HUN

SVNARM

ESTCZEPOLLTU

KAZROM

SVK

MK

GEO ALBBIH UKR

BGRHRV

KGZ RUS MDA AZEYUG

Figure 6.6 State capture freezes transition


route from state capture. The EU membership effort is explored in Chapter 7,while Chapter 8 considers more explicitly the prospects for states whichremain under state capture.

Appendix: rent-seeking, vested interests and entrenchment in non-transition economies

Efforts to lobby and influence the government for economic favours by the vestedinterests of the nomenklatura or any new post-communist capitalists are by no meansunique to the period of transition from socialism to capitalism. Economic theorygoing back to Adam Smith has always been clear that profits are greater in a non-competitive environment, and that those with greater economic power prefer to keepthat power than to face open competition. Smith’s famous warning to governmentsthat they should never promote a gathering of merchants of the same trade was meantto emphasize that such a gathering would inevitably result in efforts among them tocollude against new competition and self-serving proposals that government, in theinterests of the common good, promote that trade by one or another mechanism, sayprotection against damaging foreign interests. Subsequent development of imperfectcompetition theory by numerous economic thinkers of the early twentieth century –Marshall and Robinson are examples – made clear that monopolists and oligopolistshave an interest in collusion and lobbying of governments to prevent competitionwhich may erode their excess profits. Smith’s marvel of the Invisible Hand of themarket giving the greatest welfare benefit to all society requires that there be perfectcompetition, and not monopoly or oligopoly, and not only private ownership. Smith’sfelicitous result comes in fact from the interplay of two hands: the incentive to great-est efficiency from private ownership (capitalism), and the discipline of competition(an open market economy) to prevent monopoly profits. The role of competition hasoften been overlooked or at least underemphasized in the transition, with the conse-quence of too much effort devoted to the process of disposing state assets into privatehands – that is, creating capitalism – and too little effort to the development ofcompetition – that is, creating a market. The evidence presented so far in this bookstrongly suggests that most post-communist societies have gone a long way to creat-ing capitalism, but unfortunately many of them have not gone very far in creating anopen market economy. The history of capitalism without competitive markets isreplete with the kind of lobbying, rent-seeking and oligarch phenomena that havegripped the slow reformers of the post-communist region.

Post-communist oligarchs are often compared to American robber barons of thenineteenth century. Are Russia’s Abramovich and Potanin, and Ukraine’s Pinchuk andAkhmetov the same as Rockefeller, Vanderbilt and J.P. Morgan in the USA? Throughthe narrow lens of economic theory of profit maximization, they all look likeextremely wealthy capitalists, they act like powerful capitalists and they try to influ-ence governments like powerful capitalists, so they are the same. But looked at in awider and historical perspective, there is one, very big, difference. The initial wealthaccumulation of the robber barons came from entrepreneurial value-added activity,not from bargain prices for preexisting state assets in privatization, or privileged pur-chase of energy at regulated prices and resale at world prices five–ten times higher.Goldman (2003) makes the same point and shows that most other powerful capitalistsin history – European industrialists like Krupp or Agnelli, chaebols in Korea, even LatinAmerican padrones – differed greatly from the oligarchs in this way.23 The similarity only

begins once both have become wealthy and powerful: robber barons are then bestseen through the eyes of Mark Twain’s The Gilded Age, or Thorsten Veblen’s VestedInterests, as financial giants with the resources and connections to influence govern-ment decision in their favour, or in Twain’s phrase, ‘get to know a Senator, sell aproject’. Another similarity of great importance to the debate on frozen v. inevitabletransition: the robber barons did not easily or willingly go along with the variousefforts to restrict their monopolistic power, such as anti-trust legislation, or the estab-lishment of regulatory agencies (such as the Securities and Exchange Commission) toprevent non-transparent insider-trading – and neither have the oligarchs shown theyare willing to give up power. In a recent book of great relevance to this matter, Rajanand Zingales’ (2003b) Saving Capitalism from the Capitalists, the point is made suc-cinctly: ‘those in power, the incumbents, prefer to stay in power’ (p. 10). In this samecontext, one should note an important difference: before the robber barons, theAmerican economy had developed a very large middle class of entrepreneurs and, at aminimum, some rudimentary contract-enforcement institutions, that is rule-of-law.This is far from the case in the CIS today.24

America’s powerful vested interests – or for that matter those in other parts of theworld – did not ever give up their efforts to lobby for protection against competition,even after the high concentration of wealth were reduced by anti-trust and other insti-tutional measures. Rajan and Zingales (2003b) provide many examples for the finan-cial sector in several countries, essentially arguing that these markets were more openand competitive in 1913 than they became over the next decades thanks to such lob-bying. One example for the USA will suffice. In 1932–34 the US Senate BankingCommittee (the Pecora hearings after the chief counsel) investigated abuses in thefinancial system by commercial banks in the issuance of securities for firms that werealso their banking clients: clearly a conflict-of-interest risk. The outcome was twopieces of legislation, the Securities Act and the Glass–Steagall Act which, as the authorsexplain in detail, prohibited commercial banks from underwriting securities and frompaying interest on demand deposits. The winners? The traditional investment banks –like J.P. Morgan and Kuhn Loeb – which, it turns out, instigated the hearings as theywere feeling the competitive pressure of commercial banks in the issuance of new cor-porate securities. For the sake of avoiding conflict-of-interest (which the authors notewas not necessarily there) the legislation favoured the incumbents and restricted com-petition in the financial sectors. Some part of the profits of the traditional investmenthouses was now gratis the legislation, hence, rents.

A more recent example of successful rent-seeking is described, wittingly or unwit-tingly, in the autobiography of Lee Iacocca, president of Chrysler Corporation in the1970s at the time of the first big threat from Japanese imports. Iacocca – who made hisreputation as a dynamic manager and eventually CEO at Ford, including introductionof the legendary and profitable Ford Mustang model – was also entrepreneurialenough to sense that considerable ‘profits’ could be made by obtaining certain gov-ernment privilege. His book describes how, as new CEO at Chrysler in the early 1970s,he travelled out of Detroit to make many strategic speeches around the country on theimminent demise of the American auto industry, and meet with politicians inWashington to push for legislation that would give low-cost loans to Chrysler thatwould restrict ‘temporarily’ Japanese auto imports to give American producersbreathing-room for restructuring. One can literally count in his autobiography thenumber of days spent on this versus those in Detroit fixing the internal problems ofChrysler. The former dominated by far and to good effect: in early 1972 the US Congressimposed a temporary quota on Japanese imports, and the following day the side-window



price stickers in automobile showrooms of America (and Canada for that matter) hadan extra line added: producer surcharge �$500 to $1,000. The following year’s finan-cial statement showed Chrysler moving from a loss of about $1 billion to a profit of $1 bil-lion. The reader now familiar with the difference between profits made by bettermanagement v. ‘rents’ should have no difficulty calculating that Iacocca made forChrysler a one-year rent of $2 billion, and well-deserved the hefty (for those days)bonus of a million dollars – indeed, I would have thought it right for the shareholdersof GM and Ford to award him a bonus also, because of course these rents of up to$1,000 per automobile were reaped by them as well.

The power of wealthy capitalists in advanced market economies has been reviewedmost thoroughly by Morck, Wolfenzon and Yeung (2005), which reviews the literatureon corporate governance and what they define as ‘entrenchment’, in effect the abilityof the incumbent big-players to remain in power. They note the common phenome-non in many countries of a small number of individuals or families controlling largeparts of an economy, often with relatively small holdings in pyramidal structures,cross-holdings and super-voting privileges. They also observe that the entrenchedelites ‘to preserve the status quo … are able to influence public policies so as to curtailprivate property rights development, capital market development, and economicopenness’. This, too, bears upon the frozen-transition inevitable debate.

In developing countries the rent-seeking phenomenon was closely studied byeconomists in the late 1960s trying to demonstrate that the prevailing policies ofimport-substitution were damaging the prospects of growth. The key point is that theprotectionism of these regimes generates rent opportunities for those who are able toobtain the licence to import, export or produce regulated goods. Krueger (1974) laysout the theoretical basis for rent-seeking activities using the example of importlicences, though it applies to any situation where governments regulate economicactivity creating a distortion that can yield excess profits for those permitted to engagein that activity. The model describes why and how economic agents compete for theselicences devoting resources up to the value of the rent itself, hence resulting in a wasteof resources. This can, however, be avoided if the licences are granted by governmentsto whom they choose, a situation closest to what one sees in many of the big vested-interest stories described above. As Krueger notes, doing so redistributes incometowards the beneficiaries of this favouritism, which in turn affects societal perceptions;this is surely what one sees happening in post-Soviet transition. The magnitude ofthese potential rents was shown to be very large – for India and Turkey 7 per cent and15 per cent of GDP respectively – and numerous detailed studies of the costs of tradeand other restrictions at that time confirmed such high values. Equally important, theliterature of the time generally concluded that liberalizing trade regimes was best donerapidly to reduce the time available for the rent-seeking vested interests to mount aneffective opposition to the policy.25

Clientelism is the term applied to the analogous phenomenon in post-colonialAfrica. That some of this favouritism and pork-barrel politics was necessary and served‘as the glue that binds together weakly integrated societies’26 does not alter the factthat regardless of the geographical, social and historical environment, where govern-ments regulate the economy, rent-seeking opportunities arise and the rights to theserents are distributed somehow. Africa differs from the transition countries in not hav-ing a strongly entrenched pre-colonial power elite that stood ready to benefit from theopportunities, hence in many cases the clan or tribal affiliations and happenstance ofwhich group took power served as the analogue for the communist nomenklatura.In Central Asian countries the two were already combined in the Soviet period, hence

it is little surprise that their new big capitalists are the leaders of regional clans. TheAfrican post-colonial experience which now covers several decades bears a lesson forthe future of captured transition states. Sandbrook (2000) refers to the theory of the‘new broom’ that many writing on Africa adduce as the way in which the clientelismcan be broken up and a more liberal, competitive and fair environment can be created.He notes that so far the evidence is that new brooms may begin well – the early liber-alizing efforts of Lt Rawlings in Ghana – but that in most cases with the possibleexception of Mauritius, ‘the new leaders soon fell into old neo-patrimonial patterns’as in Zambia, Niger, Madagascar and Ghana. This serves to highlight the nearlyaxiomatic incentive for anyone with a chance to obtain economic privileges, to go forit, and to predict that any new leaders in the captured states may simply sweep awaythe existing oligarchs and sweep in their own favourites.27

This brief review of how the rich and powerful in non-transition economies becameso, and how they are all driven by the same principle of using that power to makemonopoly-like profits, might be taken as confirmation of the Shleifer and Treisman(2004) argument that Russia is just a normal country. This is not quite justified, for itonly shows that Russian (and Ukrainian or Kazakh) new big-capitalists behave exactlythe same way as big-capitalists have always behaved historically and around the globe:in their own interests. The difference in degree of such power and influence, as arguedin the main text, is big enough to become a difference in kind: the oligarchs are ableto influence not only economic measures affecting them directly, but are powerfulenough to influence the general direction of economic policy and even the outcomeof elections. That is what one means by state capture. Historically, in the Middle Agesin Europe or post-colonial Latin America, or in East Asia, state capture may haveexisted. Today, and certainly in the advanced countries, the power of the entrenchedis less extensive, as an analysis of lobbying for 42 countries by Faccio (2003) shows.She demonstrates that for low corruption countries the benefits of a political connec-tion are often small and statistically insignificant, with the exception of making for ahigher market share, where the effect is high and significant. But the more corrupt acountry, a proxy for good rule-of-law, the greater the benefits for firms of makingpolitical connections. It has been demonstrated in Chapter 6 that the captured statesof the CIS rank near the bottom of the Transparency International corruption index.The implication is that to keep benefits high, oligarchs will want to maintain the non-transparent high-corruption environment. It also follows, as Morck, Wolfenzon andYeung (2005) suggest, that in low-corruption countries entrenched elites may wish tosee less openness, but it may for these cases be too late to reverse the institutionalachievements of even-handed rule-of-law, as political influence that is purchased isoften found out and punished, and the role of middle-class civil society is asentrenched as the big-players. While the highly corrupt societies of the world, includ-ing some post-communist ones, may be in a low-level equilibrium of illiberal politiesand economies, the advanced countries may be in a high-level equilibrium of liberalpolities and largely but not fully liberal, competitive market economies.


7Safe Havens for Market Reforms:Membership in the EU and OtherInternational Organizations

203

Courtship games of the EU and post-communist countries

The Community has taken the necessary decisions to strengthenits co-operation … with States which intend their founding princi-ples to be democracy pluralism and the rule-of-law. It … willcontinue its examination of the appropriate forms of Associationwith the countries which are pursuing the path of economicand political reform. (Statement of the EU Strasbourg Summit,12 December 1989)1

Was the above an invitation to EU membership for the countries to the eastof the Berlin Wall? Of course it was, albeit its terms were already a bit moreguarded and constrained than earlier political statements welcoming thesecountries and calling for new initiatives to respond to their needs. Thus, forexample, Chancellor Kohl insisted at the G-7 Summit in June 1990 onincluding a reference to East European integration and more concretely:‘to ask the European Commission to take the necessary initiatives … and toassociate … all interested countries’. The Commission’s first response was toestablish the PHARE programme of economic assistance to Central Europe,but it then went further, preparing the Strasbourg Statement cited above. Asa good bureaucracy with the duty of reining in the flourishes of politicians,the Commission enshrined in the Strasbourg Statement an initial cautionaryprinciple: ‘the legal basis of forthcoming negotiations was to be Article 238(Association); not Article 237 (Accesssion) as many East European govern-ments had hoped’.2 The intention was to dampen expectations of a rapidprocess leading to eventual membership, and to establish three importantprinciples of the Association process: integration into Europe was to be step-wise and flexible; it was conditional on a country’s progress along democratic

and economic dimensions; and Accession did not automatically follow. Thusbegan the courtship games between the EU and the aspirants.

The very fact that such carefully crafted statements were necessary todampen membership expectations is virtually prima facie evidence that forthe post-communist societies themselves membership was in fact the finalgoal.3 The lack of explicit formal statements of this desire was not because ofambivalence about the final goal, but because the leaders understood thenature of the game: do not expect explicit public reference to membershipfrom the EU side too early. The actions of governments on legal and otherchanges attest even more strongly to the Accession intention. Thus, Czaba(2004) points to Hungary’s 1990 Law on Legislation requiring any proposedlegislation be pre-screened to ensure it did not contradict European legisla-tion. Similarly, Laar (2002) emphasizes for Estonia that borrowing Germanlegislation was done to avoid future changes required by harmonizationwith EU requirements (the early approach of all the Baltic countries is elabo-rated further in the Appendix to this chapter). Kolodko (2000b) also makessimilar points: there was no question about the unilateral wish of all thesecountries to join the community, but neither was there doubt about theinherent value of the reforms that followed. If the reforms were also consis-tent with EU accession, so much the better. Myant in Heenan andLamontaigne (1999a) notes that as early as the 1992–93 Slovak debates con-cerning independence, one important argument was how this would affectEU membership, with some Slovaks proposing a postponement of indepen-dence until after accession.

To the candidate countries, the hurdles established by the three principles ofthe Strasbourg Statement, which did not explicitly mention Membership butonly Association, were not an overwhelming deterrent as they well under-stood the game. After all, the Statement did not explicitly exclude Membershipas had been done in the case of some earlier Association Agreements for theLomé countries or the Barcelona Agreements for the Mediterranean countries.While ceding this first round of the game, the aspirants gained some purchasein a statement of the External Relations Commissioner, Frans Andriessen, whoas early as January 1990 elaborated: ‘As democracy and economic liberaliza-tion take root the agreements can be applied flexibly, and further developed toprovide for a form of Association corresponding with aspirations [my italics] inEast Europe, and the Community’s own interest.’ The word ‘aspirations’ wasunderstood to be code for ‘accession to membership’. This was made official atthe April 1990 Dublin Summit of the EC Council which incidentally also reaf-firmed the wide scope of the potential offer ‘to all the countries of Central andEastern Europe’.4

Thus long before the first formal Association Agreements (known asEurope Agreements) were signed with Czechoslovakia, Hungary and Polandin December 1991, the aspirants had won an important victory: recognitionby both sides that while the attainment of membership was conditional on


Safe Havens for Market Reforms 205

progress, and this was to be assessed by the European Commission (EC)experts, it was indeed membership that was the end-goal and not anythingshort of that. True, the EU won the right to make the judgments virtuallyunilaterally, but that is hardly surprising since the incumbents would surelyhave to be the final arbiters on admitting new members. This meant that thenegotiations were hardly that, but rather as a Bulgarian negotiator put it ‘aset of instructions for us to follow’. For countries which had achieved aninternal consensus and strong commitment to the dual liberal vision any-way, this was not giving up much and progress towards accession was thecarrot of reward for good behaviour they broadly intended to pursue any-way. For countries with a weaker commitment this was much more of a bur-den, and the EC’s power was in several instances applied as a stick to punishthem with postponement.

The critical references to eventual accession were made even more explicitin the Europe Agreement preamble which recognized accession as the candi-dates’ ‘ultimate objective’, though at the same time it reaffirmed that the EUmade no commitment to ‘automatic’ passage from Association to Accession.The Agreements themselves did not fully spell out the precise conditions forallowing progress towards membership; this was done in the June 1993 deci-sions of the Copenhagen Council, which in the first instance removed theveil from the meaning of ‘aspirations’, but also set the very tough agenda tofollow including the vehicle for the Commission’s power of judgment,annual Progress Reports for each country. Czaba (2004) describes how thesereports led to ‘a beauty contest’ among aspirants as to which country wouldbe put on the fast-track for accession, and emphasizes how this provided astrong incentive for reformist momentum. The destiny of different countriesas dictated by these reports is explored later in this chapter.

The 1991 Europe Agreements were followed by a series of others. After thevelvet divorce of the Czech Republic and Slovakia they had to be redone astwo separate Europe Agreements in late 1994, sometimes leading to the mis-perception that these two countries came later than Hungary and Poland. Inearly 1995 Agreements were finally signed with Bulgaria and Romania, andin mid-1995 with the three Baltic states and Slovenia. It is notable that bythis time the Copenhagen criteria were in place, hence the ultimate goalof accession was no longer implicit. The last Europe Agreement came forCroatia in June 2004. For each of the countries, a formal application tomembership was filed at various times, usually sometime after the EuropeAgreement rather than before, reflecting the reality noted by Schimmelfennig(2003) that: ‘states may want to become members long before they formallyapply but they wait for favorable circumstances’.5 Only after the approvalby the EU was a schedule set for beginning negotiations on the 34 chaptersof the Acquis. The first group of countries completed their negotiations by2002–03, and in the Athens Summit the EU officially declared Accession forthem in May 2004.

In the interim, a lot of negotiation, postponement and restarting tookplace, and game playing continued until the very end. Indeed, in this gameBulgaria and Romania were held back for a very long time, despite a start in1995. At the 2002 Athens Summit they were excluded from the May 2004accession, but were still rewarded for recent improvements with a newnegotiating schedule and indications of accession about 2007–08. The game-playing for a new set of countries in South-East Europe was started in theAthens Summit with varying ‘offers’. Croatia, was given the warmest new‘offer’ with a very fast-track – an early signing of a Europe Agreement inJune 2004, application immediately after. But Acquis negotiations, to startin May 2005, were delayed as a political condition concerning war-crimeswas not met. Nevertheless, Croatia may still catch up with the slow-movingBulgarian and Romanian cases. Albania, Bosnia-Hercegovina, Macedoniaand Serbia-Montenegro (with Kosovo unmentioned) were more vaguelypromised good ‘prospects’ and ongoing discussion, but no indications ofa timetable. Nevertheless, their formal inclusion in the South-East EuropeCo-ordination Commission vehicle is implicitly understood as a pre-Europeagreement process.

Reviewing this entire process, it is evident that the candidates had a greatdeal to gain from this game, economically, politically and for future security.But many analysts have argued that the gains for the EU were marginalat best, except possibly on the security element. Security gains are to beinterpreted very broadly and mean far more than the uncertainty aboutthe future direction of Russia and its possible renewed dominance over thecountries of the region. The dissolution of the Soviet empire provided someassurance of a much less confrontational relationship for some time tocome. Overtime, global integration would lead to economic prosperity andpolitical stability in the periphery countries, and for some EU membershipwould contribute considerably to stability. Papadimitriou (2002) andSchimmelfennig (2003) both give considerable weight to the security argu-ments, but insist that was not the whole story behind the EU’s eventual will-ingness to accept so many new members. Papadimitriou emphasizes a sort ofGrand Game involving a large number of elements with attendant costs andbenefits – economic, political, moral obligations, security – in which anyeconomic costs for EU (subsidies, low-cost labour threats) were too marginalto matter in the end. Schimmelfennig gives especial emphasis to the moralobligations of Western Europe to offer a democratic home for the releasedcaptives of communism. In a game-theory negotiation, he notes how thisgave the candidates a great bargaining tool of ‘rhetorical actions’, whichthey used to great advantage. After the early political calls for ‘maximumhelp’ (Papadimitriou, 2002, p. 73), ‘the CEE governments and their [EU]supporters based their claims for enlargement on the collective identityand … the liberal values of the community … The opponents of eastern



enlargement found themselves rhetorically entrapped’ (Schimmelfennig,(2003), p. 5).

How is this ‘courtship game’ to be interpreted in the navigation modelof Chapter 4? In the jargon of economists, the ‘variable’ EU membershipprospect or probability might be considered exogenous – that is not deter-mined by the other variables of the model but simply given from the outside –or it might be considered endogenous – that is caused by one of the othervariables, as for example progress on the TPI measure. In fact it is some ofboth. The very early rhetorical signals from European leaders can certainlybe considered as exogenous ‘invitations’ or offers to join the club, albeit withconditions. These conditions motivated countries that desired or demandedmembership to move more expeditiously on political and economic liberal-ization. The membership probability variable becomes endogenous at thispoint, as the continued willingness of the EU to proceed towards accession isincreased with greater reform progress, or reduced with lagging reforms, inthe circular causation shown in Figures 4.2 and 4.3. But endogeneity mayhave come even earlier if the desires of a country to turn away from com-munism and the dominance of Moscow had led to a vision of EU member-ship (also NATO and other groupings in parallel) which in turn had led to aconsensus on needed reforms well before early signals from the EU or thestatement of Copenhagen criteria. One way to think about this is that can-didates demand membership and the EU ‘supplies’ it simultaneously: theinteraction determines the equilibrium result, which here would be not somuch a financial price, but the waiting time before accession.6

The remainder of this chapter shows that the degree of exogeneity v. endo-geneity of EU prospects varied for different countries and over time. Thus,the next section will discuss EU membership being used sometimes exoge-nously as a carrot providing ex ante enticement towards liberal reforms andex post reward for progress on reforms, and sometimes as an endogenousstick to punish poor performance with the hope of herding-in these straysand laggards. We then turn to a handful of countries where an EU offer waswithheld or denied despite desires of the country, and analyse the counter-factual of what might have happened to reform progress in some CIS coun-tries like Ukraine (or even Moldova and Belarus) if the EU had made a moreexplicit early offer of the membership carrot. Moving then beyond the EU toother international organizations we ask the question: did they have similareffects to that of EU membership, did any of them provide the same ‘safehaven’ for liberalization? The final section returns to the explanatory modelof Chapter 4 to test the hypothesis that the stronger the prospects for EUmembership in the early years, the less the delays and the quicker the paceof transformation policies. An Appendix provides a more detailed illustra-tion of the dynamics between EU membership desires and offers in the Balticcountries.

EU membership incentives: sometimes a carrot, sometimes a stick

Ex ante enticement: Visegrad countries

It will be useful to start with a brief discussion of the ways in which EUprospects influenced reform strategy. The effect of potential EU membershipcan be separated into two parts: the beacon effect attracting the country – ormore precisely its citizens – to a safe haven of calmer, richer and more demo-cratic waters; and the navigation chart effect provided in the details of theAcquis Communautaire (Acquis) the concrete instructions needed to transformthe economic and political institutions.

The beacon effect has both a demand and offer side as defined above, butthe influence on internal policy is in principle similar whether it is thedemand or the offer that is stronger. The beacon of a liberal democracy andmarket that the EU’s early warm signals ‘offered’ to the Visegrad countrieswas the same beacon that the Baltic countries envisioned for themselvesdespite the lack of any such early warmth from Europe. The Baltics in factprovide the best example of countries that chose to follow the beacon ontheir own with very little being offered by the EU initially. This is not to sug-gest that the Visegrad countries only moved quickly on the liberal visionbecause the EU enticed them with an early membership offer. The evidenceof Chapter 5 is clear that their commitment and consensus on the directionof change was second to none and doubtless strong enough to keep up themomentum of liberalization. Even if there had not been an EU offer theywould surely have proceeded, as did the Baltics, trying hard to show the EUthey were worthy candidates.7

But the EU offer did no harm in this respect, and in the words of one of thenegotiators for Hungary, Bela Kadar, ‘it certainly added to the resolve andarguments of the rapid-reform proponents within society, especially to keepup the momentum’.8 In all cases, these countries organized their reformefforts with the goal of EU membership in mind, with Hungary doing so inthe most rigorous way with its aforementioned Law on Legislation requiringconsistency with EU legislation. The early candidates with an ‘offer’, albeit anon-automatic one, were enticed ex ante to move forward by the formal andinformal nature of such an offer, and were rewarded along the way withmany carrots: the political and symbolic value of an early signing of theEurope Agreements; its immediate economic benefits; an early start on acces-sion negotiations; generally positive annual reports by the EC;9 and anassured accession in the first wave.

The value of the interim rewards before accession is not to be underesti-mated. The free-trade part of the agreements helped spark an export boom tothe EU and a shift in trade patterns much earlier than for FSU countries,which in turn contributed to an earlier economic recovery. The strong signalof the agreements and generally positive messages of steady progress in the



annual EC reports greatly ‘improved the region’s reputation both in interna-tional forums and among foreign capital investors’.10 The high level of FDIper capita in the early candidates is evident in Table 2.1. This was also relatedto the more advanced status of economic liberalization, but Havrylyshyn(2004) shows in a regression analysis that the prospect of EU membershipwas as important to foreign investors, and in particular it is this factor ratherthan TPI differences that explains the wide gulf of FDI per capita betweenprospective members and those with no such prospects in the foreseeablefuture. It is notable that these large FDI flows were anticipational and camewell before the actual accession in 2004.

Ex post reward: the Baltics

The Baltics were not given the same ex ante incentive signals as the Visegradcountries (or for that matter Bulgaria and Romania discussed later), otherthan the faint hope signalled by including the Baltics under the PHARE pro-gramme of assistance for Central Europe and not the TACIS programmedevised for the successor states of the USSR. They only benefited from an expost reward for doing what the EU was asking of the earlier candidates. Whydid they do it without the EU membership enticement? Because they had astrong desire for membership which served as well as a strong offer to moti-vate policy in the liberalizing direction. Like the old marketing story of Aviswho as No. 2 in the car-rental business behind Hertz made a lot of its slogan‘we try harder’, so too the Baltics in the period 1992–95 tried harder and, asseen in Chapter 2, essentially caught up to the leaders by 1995 on the TPImeasure of economic transformation. They were indeed rewarded with aEurope Agreement in 1995, subsequent negotiations and, with some ups anddowns, accession in the first wave.

The Appendix to this chapter elaborates on the hypothesis of how theearly reform policies in the Baltic states were influenced or at the leastinformed by the desire for eventual membership. The lack of an early offer,equivalent to that received by Visegrad countries, meant this linkage was notas precise; in the beginning, it was explicit only in Estonia but became muchmore explicit over time in the other two as well, even before the signing ofthe 1995 Europe Agreements. As early as 1991–92, there is evidence of thesecountries orienting the process of legislation and policy formulation towardsthe EU membership goal, again especially in Estonia. Prime Minister Laar’smonograph cited earlier avers to this, and it is more fully analysed by Pisuke(2001) who concludes that, as preparation for potential membership, thenew legislation was modelled on EU member states. Levits (2001) suggeststhat in Latvia the orientation of conforming to EU standards came slightlylater, but was in evidence in 1993–94 well before the signing of EuropeAgreements. At the beginning Lithuania may have been somewhat lessintensely focused on EU membership than the others, perhaps because, asLane (2002) argues, it promoted and had the best relations with Russia.

Nevertheless, this did not preclude following the same model of borrowingfrom European law as in the 1992 Law on Principles of Accounting, atleast suggesting the same motivation: ensuring future concordance with EUaccession requirements (O’Rourke, 1995).

The differences began to be minimized and the common objectives high-lighted in the June 1993 joint declaration of the three Baltic presidents callingon ‘EU countries to initiate negotiations on the granting of associatemember status to the three Baltic states’.11 In fact this was not the first stepin the courtship game; the broad invitation of the April 1990 Dublin Summitto ‘all countries of Central and Eastern Europe’ may or may not have beenintended by the EU to include the Baltics who were still de facto part of theUSSR then, but for the Baltic states who insisted successfully that they werenever de jure part of the USSR, Eastern Europe by definition included them.And so they acted, so they spoke publicly during 1990–91, including theirdeclarations of sovereignty and independence of mid-1990. Their first suc-cess was the signing of Trade and Cooperation Agreements with the EU inMay 1992, very soon after the dissolution of the USSR, and well before anal-ogous agreements for the CIS states (1994–98). This was followed by a FreeTrade Agreement in July 1994, and finally the Europe Agreements in June1995 which put them on the explicit track of potential membership, albeitagain non-automatic as per the principles established in 1990 by theStrasbourg Statement. The arbitrary power of the EC to decide on timetablesled to the inclusion of Estonia in the first wave, but a postponement for theother two. In the event they too caught up and acceded in May 2004.

The Acquis as a navigation chart

The ‘navigational effect’ provided by the Acquis is extremely detailed andcomprehensive, comprising 34 chapters and over 80,000 pages of require-ments a country needs to fulfil in legislation, institutions and procedures toaspire to accession. The overwhelming negotiating position of the EC waspartly due to the volume and detail of the Acquis – any country truly seriousabout membership had little room for manoeuvre here apart from somemarginal ability to negotiate interpretation and timing of implementation.Three points are worthy of attention. First, there is an important similarityor at least consistency between the Acquis as a chart towards the market, andthe elements of the Washington Consensus (WC), though the Acquis har-monization process has virtually no prioritization of elements, no specificsequencing or recommended timetable. But the annual reports from the ECabout how countries were progressing to meet the Acquis and the require-ments listed under the Copenhagen criteria, implicitly supposed the WCapproach, with frequent references to having and implementing a programmewith the IMF and World Bank – or failure to do so. Second, the process of nego-tiating the Acquis chapter by chapter permitted the EC to differentiate amongcountries and in effect rank their progress, with the possible consequence of



delaying the accession timetable or even suspending temporarily negotiations;this ‘stick’ was applied in several cases.

Third, it should not be forgotten that while the Acquis is a formal require-ment only for those on the path to EU membership, it is a public documentavailable to any government that in the early 1990s was debating alternativepaths to a democratic market economy. For those desiring EU membershipbut not immediately receiving welcoming signals, there was nothing to pre-clude demonstrating their seriousness of purpose by proceeding on theirown to implement Acquis components, at least in spirit if not precisely tothe letter. Some like the Baltics did, even to the point of setting up govern-ment units for eventual implementation of the Acquis before the EU clarifiedits ‘offer’; others – like Ukraine before Yuschenko – professed a desire formembership but did not try at all hard to follow such a strategy.

EU membership as a stick to herd-in strays

There were several cases of an offer being given to countries that had a weakinternal commitment to the dual liberal vision and therefore did not respondto the enticement. The earliest examples may be Bulgaria and Romania. Theywere already in the list of countries mentioned in various documents as earlyas 1990, for example Commissioner Andriessen’s clarification of 4 January1990, President Delors 17 January speech to the European Parliament, andothers. By August of 1990, however, concerned about the continuation ofgovernments with a heavy communist presence, the Commission proposedputting Bulgaria and Romania in a second wave behind the Visegrad coun-tries.12 This was the first application of the stick as a punishment for poorresults on the democratic dimension. Interestingly, this was soon reversed byan EU Council decision in September 1991 to reopen negotiations for anAssociation Agreement, that is the offer of a carrot.

But the game of carrot–stick–carrot has continued for these two countriesto the present day. The slow progress on economic reforms delayed the sign-ing of the Europe Agreement until 1995, four years later than the Visegradcountries. But even then delay continued and negotiations also started onlyin 2000, much later than for the first wave applicants. This second post-ponement ‘stick’ came as a result of the resurgence of inflationary pressuresand financial instability in 1996. This was followed by a new and more suc-cessful stabilization – particularly in Bulgaria with the election of a democ-ratic government in February 1997 and the successful establishment of acurrency-board regime three months later. Nevertheless, both countries hadfallen so far behind that they could not catch up to the first wave. In con-trast, several others that were in a second wave – Latvia, Lithuania, Slovakiaand Malta – did catch up and were included in the May 2004 accession.

Another episode of the stick being applied came later and was muchbriefer: Slovakia. During the Méciar period (1994–98), considerable doubtswere raised about regression in democratic practices and to a lesser extent on

economic reforms, especially favouritism in privatization. The EC droppedSlovakia from the first wave and suspended negotiations, with the intent notonly of punishing but hoping to bring policies back on track. Slovak analystsgive a lot of weight to voters’ concerns of not joining the EU with theirneighbours as a factor leading to Méciar’s defeat in 1998, and the restartof reforms under Dzurinda.13 Less dramatic were the cases of Latvia andLithuania. The 1997 Reports of the EC warned both of lagging behind. UnlikeSlovakia, neither of these had been on the list of first-wave countries, withEstonia always ahead of them. But the impact of this warning was consider-able, and sufficient to spark an acceleration of efforts which in the end suc-ceeded putting these countries in the May 2004 accession group. Sloveniaalso started much later on the EU path, but that was perhaps less a case of theEU wielding a stick and more of a case of the Slovenians being confident thatthey were doing the right things all along, even before 1990,14 and didnot need to bend over backwards to satisfy the EC. In the event they accededto the EU with the first group and the highest per capita income of all thetransition countries – about $9,000–10,000 at the time.

The stick effort in Croatia differed in many ways from all the others. Itwas excluded from the beginning – despite an advanced level of economicreforms, market operations (like Slovenia) and steady progress – because ofEuropean concerns about war practices and democratic violations. Until atleast 1995, the nation was reasonably united as the independence war wasbeing conducted. Economic liberalization was proceeding perhaps moreslowly than in the rest of Central Europe, but was proceeding from a higherbase and without reversals; on this there was an internal consensus andstrong commitment, as indeed there was on the desire for EU membership.But as long as the war President, Tudjman, was alive, any efforts by opposi-tion forces to win elections failed, and the stick of withholding the EUmembership beacon had little effect. With his passing and the 2000 electionwin of Mesic, discussions began immediately with the EU, resulting in aStabilization and Association Agreement (a sort of precursor to a full EuropeAgreement) in May 2001, and a Europe Agreement in June 2004 withimmediate application for membership. Echoes of the past sensitivity to war-crime issues continued, however, and have slowed the pace of negotiationsdespite the sense on all sides of the very high prospects for successfulprogress towards Accession.

Unrequited desires for membership

A number of countries on what has become the new eastern periphery of theEU, on the other side of the new ‘Schengen Curtain’, have at different timesand with varying intensity stated a desire for EU membership: Ukraine,Moldova, Belarus before Lukashenka, even some Caucasian countries. Theresponse of the EU has been lukewarm at best and categorically negative at



worst; economic and democratic transformation of these countries have ingeneral lagged considerably behind the new members, which on the face ofit vindicates the lack of reciprocation by the EU. However, an important his-torical counterfactual question may be raised: could these countries havebeen motivated to greater advances if the EU had responded more favourably,that is used more boldly the carrot of enticement and as needed the stick ofpunishment which we saw had such a powerful effect in the central swatheof Europe? As a counterfactual this can only be analysed in a speculativefashion, but it is possible to derive examples from the successful cases whichlend support to such a hypothesis. I will not consider the Caucasian caseshere, and concentrate on Ukraine, partly because of its much larger role inthe region, and partly because it suffices to make points applicable to theother two periphery countries: Belarus and Moldova. Most important, ofcourse, the issue is once again on the table for Ukraine: can a warmer EUoffer help buttress internal commitment to reforms?

There has also been something of a membership dance between the EUand Ukraine over the past decade, but it has been less like courtship andmore like shadow-boxing with no real blows struck and no real achieve-ments to note. Kuzio (2003) characterizes the result as a stalemate attribut-able to ‘virtual policies adopted by the EU and Ukraine towards each other.The EU has never adopted a clear strategy towards Ukraine [and] Ukraine hasespoused “European choice” rhetoric [while] adopting domestic policies thatundermine these goals.’ He goes on to argue that had the EU been bolder andmade a warmer offer with some hope of eventual membership despite theunquestionably inadequate reforms in Ukraine, this might have tipped thescales in the direction of greater progress on the dual liberal vision. The EUposition is not without weighty counterarguments, two of which are explicit.Countries need to demonstrate greater action as had been done in much of theregion to the west of Ukraine, and not just express good intentions. Ukrainehas been engaged in the Partnership and Cooperation Agreement since 1994,and can participate in the new initiative of the European NeighbourhoodPolicy (ENP). Two others are implicit such as enlargement fatigue and the viewthat Ukraine and the other eastern aspirants are really outside of ‘Europe’ insome geographic and/or societal sense. I elaborate on these four below.

The EU rationale that Ukraine has not done enough to merit more of anoffer is very powerful, but at the same time there are clear precedents amongthe successful aspirants that the EU can, by extending the hope of member-ship, have an extremely powerful influence reorienting a country’s policiesin a more liberal direction. The recognized democrat in the candidacy for theUkrainian presidential election of November 2004, Viktor Yuschenko, madea passionate appeal (Herald Tribune, 9 September 2004) for the EU to take aclear position for integrating Ukraine, including eventual membership. Hereferred to the 1998 precedent of the ‘clear European position that helpedthe democratic forces in Slovakia defeat an authoritarian regime’. Slovakia

under Méciar was indeed straying from the path, and the stick of postponingthe membership process did have the desired impact. Perhaps an even betterexample – which would have been absolutely impolitic for Yuschenko to usebut an academic tome can – is Romania. It was, as noted, on the very earlymembership lists despite the political instability and slow economic reforms,and while punished by postponements it remained on a tentative member-ship track. In the second half of the 1990s, as the Economist 6 November2004 notes,

The EU has treated Romania quite generously … Setting out to reunitepost-communist Europe it tried hard not to omit Romania, even when itseemed to be going backwards. [Now] democracy seems to be working andthe direction of the country in the next election is not fundamentally in question.[My italics]

The italicized part contains the key lesson: EU membership has made a bigdifference in at least three critical cases already – Slovakia, Bulgaria andRomania – and Croatia looks set to become a fourth example, which suggestsit could also have done so in some of the cases of unrequited desires. Theseprecedents give some credence to the counterfactual: had the EU in its owninterest for stability on the periphery offered a warmer welcome to Ukrainedespite its reform shortcomings, the direction Ukraine took might have beenmore felicitous. This might have been done at two moments of importantadvances in reforms: in 1994–95 when President Kuchma first came in andbegan to demonstrate reform zeal, and more recently when Mr Yuschenkobecame Prime Minister in 2000 and reinvigorated the reform process thathad stalled after 1997, moving reforms forward smartly in a period whendemocratic standards though beginning to falter were far from reaching theugly depths of the 2004 campaign.15 The second opportunity with a reformistdemocratic PM, oriented to and popular in the West, could easily have beentaken by the EU without fear of losing credibility on the principles of EuropeAgreements.

Of course these are counterfactuals, the EU judgment that a membershipoffer in 1994–95 was too early, and one in 2000 while Yuschenko was PM toorisky because the real power had by now been consolidated in the new oli-garch clique with Kuchma as the political if not financial leader. But even athird opportunity was not taken: a response to Yuschenko’s plea for somesort of signal, a little flash of the EU beacon somewhat analogous to whathad been done prior to the 1998 Slovak elections. The EU could not and wasnot being asked to suggest a Yuschenko victory would reopen the issue ofmembership, just as it could not in 1998 openly point to Dzurinda as apreference. But just as then it indicated a return to a more democratic and lib-eral path after the elections might allow a retracking of Slovakia, it could havesignalled that pursuit of an open democratic process and continued economic



reform progress might lift Ukraine out of the ENP group with no expectationof future membership. In the event, democracy prevailed without such acarrot from the EU, on the basis of the home-grown urges of the peoplemanifested in the well-known events of the Orange Revolution.

A fourth opportunity, after the 26 December 2004 presidential electionof Yuschenko, offers the EU an even more auspicious moment. The oligarchpower has been defeated, a democratic and civil society has demonstrated itscapacity (even if it is not yet fully entrenched), and a new government withan evidently much stronger commitment to the dual liberal vision is in place.

What about the arguments that Ukraine had nevertheless been givensomething in the Partnership Agreement and the new ENP arrangements?The evidence of this chapter is clear that nothing short of a possible futureaccession, no matter how unclear the timing, has much effect. Some of thefinancial and trade-investment facilitation assistance can generate positivebenefits for the partners, but as Wolczuk (2004) notes, these sort of non-membership agreements ‘are too weak to make any difference on the generaldirection of policy’. The EBRD Transition Report 2000 is somewhat moreguarded in the use of words but the meaning is clear in the following: ‘In theabsence of prospects for closer integration, EU association agreements havenot exercised significant influence on the reform process in the CIS.’ Indeedthe ENP goal of ‘creating a ring of friends’ may be counterproductive sincethe message it sends is not a faint hope of future membership but itsopposite, an implicit but clear ‘NO’. The choice of the word ‘neighbourhood’was effective in making the message clear, but may have been unfortunateif it launches a self-fulfilling prophecy that ‘bad fences make bad neigh-bours’. The European Neighbourhood Policy risks reinforcing the feeling ofthese countries being spurned, already fed by the existence of a new curtainreplacing the old Iron one, the Schengen Curtain; it is a paper curtain, butsurprisingly strong paper.

What about the implicit geographic argument that Europe cannot beextended indefinitely eastward. For the western CIS countries, the relevanceof this argument was completely undermined by the correct, laudable andbold decision of the EU to finally clarify the possibility of membership forTurkey. It extends further east than several aspirants in the CIS, it has a less-European historical, cultural and religious nature. The implication that allthese differences did not in the end matter surely strengthen the case forpotential inclusion of Ukraine, Moldova, Belarus and even the Caucasus.

The only argument that has some objective weight is expansion fatigue,but it is unimaginable that any opponents to further enlargement woulduse this argument explicitly. It is so contrary to the high-minded principlesof common democratic, liberal values that opposition is forced to use false orweak arguments, which of course strengthens the hand of the aspirants.

The counterfactual question of whether things may have gone better inUkraine, Belarus or Moldova (and possibly farther east as well) had the EU

been bold enough to offer the enticement of membership will remain spec-ulation, but the evidence of the carrot–stick approach for many other coun-tries clearly shows the line of causation: EU membership strongly contributesto more liberal policy directions. At a minimum, such evidence should lead toEurope keeping an open mind about new, additional suitors. In Ukraine, theremay be a new opportunity since Ukraine’s large size and strategic role areboth advantages and disadvantages for its EU desires: it is too big to ignore, butso big it is not easy to absorb. Similarly in Belarus, the only past opportunityin 1992–93 with a reformist Prime Minister, Shushkevich, is long past and itis inconceivable that the Lukashenka administration would respond properlyto an EU offer; a future change may or may not generate a new opportunity.Moldova may be more susceptible to such an influence even under the latestcommunist-dominated government, because its small size means that itsbenefits from membership would be huge, but the costs to Europe tiny. Butgiven the very political nature of the courtship game within Europe, it isunlikely that either of these would be brought on board before Ukraine; andfor all three enlargement fatigue may result in continuing the status quoeven if the aspirants radically change their policy. It is not inconceivable thatcountries last in the SEE queue – Albania, Bosnia-Herzegovina, Serbia, asJudah (2005) pointedly describes the driving force pushing both a strongdemand and an eventual reluctant offer is the ‘need for security of EU mem-bership in the EU to keep them from slipping back into violence and hatred’(p. 76), and a new Kosovo – would become members before Ukraine, Belarusand Moldova.

Other international beacons

In principle there are other safe havens for a transition country apart from theEU: NATO; the WTO; international financial institutions (IFIs), in particularthe IMF; the World Bank; the EBRD; as well as regional trade or common-market arrangements, such as the Single Economic Space (SES) under discus-sion for some CIS countries. The primary concern in this discussion is thebeacon and navigation chart effects discussed above – not the economic ben-efits of free trade or free economic zones, but some indication of how largethese last might be will be useful.

NATO needs little comment. First, it is unlikely to be relevant for countriesoutside the CEB and SEE with two possible exceptions: Ukraine, where itis still conceivable to see NATO membership without EU membership (likeTurkey had been for decades), and Georgia. And, second, while NATO mem-bership has effects beyond security issues by drawing countries closer to theclub of liberal market economy and functioning democracy, it has virtuallyno requirements or levers to motivate market reforms, or even ensure demo-cratic ones. Arguably, it can help inculcate such a spirit by osmotic exampleand informal peer pressure, but there is nothing in NATO even approximating



the Acquis requirements for membership that would make it as powerful.The long-standing fact of military governments, de jure or de facto in Turkeyuntil the early 1980s is indicative of the lack of a democratizing influence.This holds even more true for liberal economic institutions.

WTO membership16 almost certainly will have more influence on economicpolicy than NATO, although it falls far short of the impact EU membershiphas had on liberalizing market institutions, and has virtually zero effect ondemocratic standards. Its attraction to potential members has to do with thetrade-creating opportunities it provides by lowering barriers to its exports inall the incumbent members; along with this, the new member receives yetanother ‘certification’ of approval by an international organization whichraises somewhat its status in the eyes of international investors, financiersand traders.

But some doubt has been raised in the literature whether the WTO hasmuch of a trade-creating effect. Rose (2002a) shows econometric results thatit does not, but Subramanian and Wei (2003) counter this with more positiveresults for advanced countries, and accept trade creation is limited for devel-oping countries. The unhappiness in Kyrgyz Republic and Moldova that WTOmembership has not generated a trade boom would find a resonance withthese results. Nevertheless, it is clear that the WTO could potentially pro-vide the magnitude of economic benefits that EU’s very large economy can,17

and that the continued efforts of countries not yet members are justified. Onemay disregard the sceptical view of Rose (2002a) about the trade-creatingeffects of the WTO, but it is far more difficult to disregard the arguments byRose (2002b) and Irwin (1995) that for developing countries with high-tradebarriers WTO membership is not that powerful an influence on liberalizingtrade comprehensively. Partly this is due to the historical inertia of the GATTthen WTO of giving developing countries more leeway to protect domesticindustries, and partly it is due to the bilateral process of negotiating proto-cols between the candidate and interested incumbents on just how muchliberalization and on which goods it applies. There is no standard require-ment of all new members, hence, the result is likely to be uneven and cer-tainly incomplete liberalization. In addition, there is very limited possibilityin the WTO either before membership or after to exert liberalizing influenceon domestic markets, unless a member puts forth a dispute claiming thatcertain internal regulations – say energy subsidies or provincial governmentsupport – indirectly create an unfair trading advantage. The dispute thengoes through a lengthy process of adjudication with no certainty about thedirection of the results. On balance, therefore, the WTO impact on liberal-ization is necessarily far more limited than that of EU membership. Otherregional trade arrangements are even less likely than the WTO alternative toprovide an EU-like effect for two reasons. First, any trade-creation benefitswill be much smaller and may indeed be outweighed by trade-diversioneffects in a very small regional grouping where each country does a lot of

trading outside the group. Second, the motivation and discipline for thebroader political and economic transformation under the EU is virtuallynon-existent.

As to the IFIs, the major shortfall compared to the EU is that mere mem-bership provides neither the beacon nor the navigation-chart effect. Theminimum requirements for continued membership are a minor financialcontribution, and agreement to undergo regular reviews of policy – such asthe Article IV Consultation at the IMF. It is only if the country wishes to bor-row that any serious policy conditionality is applied, with a firmness that isbroadly proportional to the size of the borrowing but which may also besoftened by the relative political importance of countries.18 Since Sloveniahas never wished to have an IMF programme, its slow pace of privatizationwas not subject to any formal conditionality. Turkmenistan’s total disinter-est in IMF financing left it free to undertake whatever level of reforms itwished, with the only consequence being a critical Annual Report during theArticle IV consultation. As these reports cannot carry significant sanctionsby the institution itself, even if they have some impact on perceptionsof outside business interests, their recommendations are applied entirely vol-untarily. In this case the authorities maintained the formal dialogue andconsultation, but with significant delays on data provision and timing ofmissions. In principle, the IMF and the World Bank could in extreme casesimpose a sanction for non-cooperation on data provision, for example pre-cluding the country from borrowing. The irrelevance of this for Turkmenistanis obvious. The story can go in the opposite direction – the unwillingness ofBelarus to undertake significant liberalization and monetary stabilizationmeant it could not receive the large financing it did desire and continuallyrequested. But wanting financing while not wanting to do what is necessaryto obtain it ends up at the same point as not wanting financing at all; that is,no conditions are imposed, hence no leverage can be applied by the IMF.Much the same logic applies to the operations of the World bank and theEBRD, and, not surprisingly, the same cases of ineffectual leverage on slowreformers are echoed in those institutions.

The only relationship powerful enough to approximate the effect of theEU is the programme conditionality of the IMF, the World Bank and regionalbanks, the EBRD in this case. But conditionality only applies if a countryalready wishes to carry out such reforms (most CEB countries did follow verysuccessfully such programmes with and without lending), or has a deepfinancial crisis and must accept the conditions to receive the financing.Many analyses of IFI programmes over decades, including econometric esti-mates, have shown that IMF programmes generally do have a positive impacton both inflation and growth, but the effect is not overwhelming and notalways statistically significant.19 The problem for statistical analysis is two-fold: failure of a programme is sometimes due to poor implementation of theconditions; and it often takes a long time for the effect to work through, by



which time many other influences are at work and the programme’s contri-bution is not easily separated out. Many reluctant borrowers negotiate hardfor the least difficult conditions, do not implement them well, and negotiateafter the fact for waivers of the initial conditionality; thus many programmesgive poor results because they are not well-implemented. When this can becontrolled in the analysis the programme’s effect is seen to be stronger. Inany country trying to overcome a crisis, IFI programmes and their financingare only a part, and sometimes a small part, of the environment determiningsuccess or failure; econometric analysis cannot always separate out all theseinfluences.

It is notable that the leverage of IMF/World Bank conditionality was infact enhanced by EU membership dynamics; in most cases, the EC’s annualprogress reports encouraged candidate countries to do better on IMF standbyprogrammes, or praised them for good performance.

Outside programme conditionality, all the IFIs have probably made avery large contribution that is virtually unmeasurable through the technicalassistance and general policy advice provided to government officials. Odling-Smee (2004) describes the nature of this contribution for Russia by the IMF.The initially limited knowledge of market-economy institutions and conceptsin the post-communist countries was fertile ground for such knowledge-building, exemplified by the establishment of the Joint Vienna Institutesponsored by the Austrian government, the three main IFIs and other bilat-eral donors. It has in 10 years provided practical economic training forthousands of officials from the region. It has of course been said numeroustimes that the advice of the IFIs was incorrect and misguided, but such criti-cisms are only worth debating in reference to actual programme conditionsas in Chapter 1; for technical and policy advice the contribution is morelikely to be positive as it is but one of many channels through which varyingadvice was provided to authorities, who then chose to follow it or ignore it.The knowledge transfer came from many other sources as well, includingindependent academic or business community advisors financed by foreignassistance programmes of major public and private donors, short-term train-ing programmes in donor countries, a large inflow of students from theregion into western universities for undergraduate and graduate programmesand so on. With such a wide variety of vehicles for knowledge transfer, it isunlikely that any bias in the IFI contribution could have overwhelmed theprocess.

It is perhaps paradoxical that the benefits of IFI general policy advice andtechnical assistance, outside programme conditionality, was generally great-est in countries with a strong inherent commitment. This was so not only forthe obvious reason that committed reformers were far more receptive tomarket-oriented advice, but also because in the early years of transition therewas limited local knowledge of market operations. Over time, the broad con-cepts became well-known, but – even to the present day – detailed technical

assistance on day-to-day operations of governments’ rules, regulations andprocedures has in general been welcomed. Even in countries where the com-mitment of the leadership was low or limited, operational assistance wasoften very well met by the middle-level bureaucracy.20

In summarizing the evidence of beacon and navigation effects provided bydifferent international organizations, one must conclude that no alterna-tives come close to having the effect of EU membership. NATO simply hasno mechanisms for leveraging policy on economic or democratic dimen-sions. The IFIs when applying programme conditionality may approximatefor a time the effects of the EU, but their influence is far more limited if amember does not need or want financing and therefore does not enter intoa programme-conditionality relation. The WTO has a very focused influenceon trade liberalization, that is often limited, but not on any other dimensionof reform. Regional trade arrangements are even less important. It bearsrepeating that various EU cooperation arrangements which explicitly excludemembership as a possible end-point are also of limited impact.

EU membership prospects and reform delays

While academic analysts have engaged themselves in endlessdebates about the proper choice of ‘East European capitalism’, andits alternative models, for both Left and Right policy-makers mostdecisions have been either determined by or suggested by the exi-gency of adjusting to the formal and informal requirements of theEuropean Union.

(Czaba, 2004, p. 338)

In Chapter 5 it was argued that internal commitments of a country, suchas a stronger renunciation of communism and more intense nationalistsentiment, generally led to an earlier start on economic stabilization and lib-eralization. These forces can be summarized as the degree of internal com-mitment to the dual vision of a liberal market and democracy. We can nowblend into this the influence of the external commitment to eventual EUmembership. In the illustrative formalized version of the navigation model(Chapter 4, Appendix) each of these, and possibly other historical conditionssuch as the number of years under communism, would be the independentvariables of the equation determining the delay between regime change andthe start of an economic transformation programme. While it is particularlydifficult to find measurable proxies for the initial prospects or probability ofEU membership, a number of testable hypotheses relating EU membershipto delays in reform start – as set out in Figure 4.5 – can nevertheless be tested.The analysis in Chapter 5 and Chapter 7 provides qualitative evidence usingcategory groupings or so-called Latin-square tests.

Consider first the hypothesis that a more favourable ‘offer’ of membership inthe early 1990s encouraged earlier reforms, as illustrated in the relationship of



Table 7.1. On the basis of the preceding discussion, countries are put intothree broad categories according to the strength of the initial ‘offer’ of mem-bership from the EU; that is about 1990–91 for non-USSR cases, about 1991–92for the others:

● Strong: Czech Republic and Slovakia, Hungary, Poland.● Moderate: Bulgaria, Croatia,21 Romania and Slovenia; the three Baltic states.● Weak or zero: all others.

Over time the strength of the initial offer changed, and the impact of suchchanges has already been reviewed. But the early decision to begin reforms –typically not more than two–three years even in most of the laggards – wasaffected only by the strength of the initial offer. The delay measure used isthe number of months from regime change to achievement of the 5% permonth inflation, as defined in Chapter 5. If it is the start of stabilization thatis thought to matter, one could use that as an alternative measure of delay;

Table 7.1 Strength of EU ‘offer’ and reform delay

Scale of theno. of months Strength of EU offerto 5%/monthinflation Strong Moderate Weak

60 Turkm(70)Taj

50 UkrRomBulg Arm/Geo

Rus/UzbBel

40 AzeKaz

Mol30 Kyr

MacCro

20[Alb]

Est/LvaPol Slv/Lith

10HunCzech/Svk

Source: Based on the no. of months to 5%/month inflation,Tables 6.1 and 6.2. For details of the EU offer see text.

in fact, this does not materially change the correlation that emerges. As athird alternative, delay might also be measured as the time to attain a certainTPI level of liberalizing reforms, but again this does not affect the correlationmaterially. That the time to achieve stabilization proxies well for other reformmeasures is not surprising. In virtually all cases, programmes of reform oncestarted were quite comprehensive and included stabilization, liberalization,privatization and some institutional change; where stabilizations failed itwas also generally the case that other reforms stalled or even reversed.

The pattern of the Latin square in Table 7.1 conforms well to the hypoth-esis, with three exceptions which are easy to explain: Bulgaria, Romania andAlbania. The countries with a strong prospect, Poland, Hungary and then-Czechoslovakia, had a delay of no more than 10 months.22 Slovenia, Croatiaand the Baltic states in the moderate group had only slightly longer delays of10–15 months. All the countries with weak or effectively non-existent offershad delays typically in the range 30–50 months and even higher, with theexception of Albania at 25 months. This last outlier is easiest to deal with: itis square bracketed to reflect the eventual breakdown of the stabilization by1997 due to the unregulated financial-pyramid schemes, which suggests itsplacement ought to be higher up on the vertical axis. As to Bulgaria andRomania, their weak commitment to a liberal agenda overwhelmed their desirefor membership, and the failure to sustain stabilization stemming from inad-equate progress on structural reforms (especially a continued soft-budgetattitude towards state enterprises) resulted in the EC wielding the stick ofpostponement of negotiations well beyond the mid-1990s. Thus even a verysimple picture relating only two variables, reform delay and EU membershipoffer, needs only minor adjustments for three outliers to yield a strongcorrelation consistent with the first hypothesis.

But this chapter has emphasized that the membership influence was notsimply as an exogenous offer from the EU side, but had also a demand side,that is the desire of the aspirant countries themselves. The relevant hypothe-sis was stated in Figure 4.5 as follows: strong demand for membership drives earlyreforms regardless of the signal from the EU. In Table 7.1, nine countries judgedto have strong internal commitment and related desire for EU membership(as described in Chapter 5), are noted in bold. Only four of them had a strongoffer early on. Thus, statistically, it would be very difficult, if not impossible,to separate out the respective influences of commitment and EU offer andthus assumes the counterfactual that they would have done exactly the sameeven without a strong offer.

However, this hypotheses appears to be confirmed by the experience ofthe group with strong commitment and strong EU desire, but at best a mod-est initial offer: this certainly comprises the three Baltic states, Slovenia, andarguably Croatia. For the first four the preceding discussion concluded thatstrong commitment and desire for membership drove their early and steadyreform progress. The end-goal was to demonstrate to the EU that they merited



inclusion in the first wave of accession – and they succeeded. There is also nodoubt about Croatia’s strong desire for membership or its demonstratedcommitment to economic liberalization. While throughout the decade itwas last on the TPI indicator among the Central Europe group and was sur-passed by the Baltics in 1995–96, it continued to keep pace with progress inCentral Europe and remained clearly above all the rest of the post-communistcountries, including in the effectiveness of stabilization and economic reforms.The issue in Croatia has been almost entirely one of democratic standardsand judging them in the sensitive context of a war with mutual abuses ofvarying degree.

The evolution in the two other countries with a modest offer, Bulgaria andRomania, is also supportive of this hypothesis. Their weak commitmentmeant in effect that the demand for membership was weaker than in theothers, hence, they did not proceed to early and steady reforms in the earlyyears despite having a modestly positive signal even earlier and stronger atthe outset than the other four modest cases.

The relatively early upgrade of the EU offer for the Baltics and Slovenia, aswell as the much delayed upgrade for Croatia, supports the third hypothesison membership: strong progress in reforms induces a more favourable stance bythe EU. The first four were able to sign a Europe Agreement by mid-1995 withEstonia and Slovenia, the stronger reformers, leaping ahead of Bulgaria andRomania to join the first wave. The 2003 decision to designate Bulgaria andRomania as the next wave of accession with a 2006–07 time horizon isanother example of the hypothesis, though I have noted above that the evi-dence of actual progress is far stronger for Bulgaria than Romania, suggestingthat for the latter it is perhaps more enticement than reward. The EU has notbeen as generous with Macedonia and Albania despite the fact their progresson the economic dimension is not much below that of Romania; the con-tinued instability in the region and the shadow it casts over democratic prac-tices is no doubt the explanation – in all of these hypotheses reform progressmust not be interpreted as economic liberalization alone, but must accountfor democracy status as well, because the EU places such great weight on this.

Consider finally the fourth hypothesis: negative stance of the EU (use of a‘stick’) in a country with strong demand for membership, induces acceleration ofprogress. Several cases of this sort have been described earlier, and need onlybe recapped here. Latvia and Lithuania were not put into the first wave withEstonia, and this ‘stick’ clearly was taken seriously by the leadership andled to accelerated efforts to comply with EC concerns expressed in annualProgress Reports in the years 1996–99. It worked, as the EC began to soften itsstance in 2000 and eventually consented to include them in the first wave.Analogously, Slovakia was dropped from the first wave after the Méciargovernment undertook some regressive steps especially on democratic rights.In the 1998 election, the opposition stressed EU membership as an impor-tant campaign issue, and interpreted the veiled signals in EU Reports as

saying a democratic win could bring Slovakia back on track. This greatly con-tributed to the opposition’s eventual victory and consequent policy retracking.Analogously, the long period of withholding a membership offer from Croatiaalso contributed to the election victory of the opposition forces in 2000 and thealmost immediate warming of relations under Prime Minister Mesic.

To some analysts all of this is too sophisticated by half, and many suggestone can cut through the dynamics of this courtship game and the argumentsabout greater or lesser commitment to a liberal vision with one word: geography.Though this has been addressed earlier, it again merits attention to under-stand what geography can and cannot tell us. The attractiveness of geographyas a simple and powerful explanation is understandable. All of the countriesthat have been or will be included in the EU, and have been pulled to it bysome interplay of their own desires, philosophical commitments and Europeanon–off offers, are undeniably geographically close to Western Europe. All ofthe countries given the cold shoulder and/or showing little credible com-mitment are much farther away. Though this last point is now made moredebatable after the decision on Turkey, one cannot deny the broad validityof proximity as an explanation.

The simple distance measure also turns out to work very nicely in econo-metric analysis of post-communist evolution, as for example in an early workof Fischer, Sahay and Vegh (1996) explaining growth performance, where thedistance from Brussels was a very powerful and statistically significant vari-able. But this does not satisfy analytical curiosity for two reasons. First itimplies a very rigid historical-geographic determinism with little room forpolicy choice; why bother with all of the analysis, discussions and resourcesdevoted to designing policies in the countries close to Europe? No policy rec-ommendation automatically follows from proximity, and it is easy to pointout exceptions where equal proximity leads in very different directions. IsAlbania and Moldova that much farther away from its European neighboursthan the Baltic states? No, but they went very different ways, including thefirst two not getting the warm European welcome that the Baltics did. It isessential to understand the much stronger Baltic commitments and imple-mentation of policies in order to know why they went different ways despiteequidistance. The second reason distance does not satisfy as an explanationis that kilometres tell one nothing about the behavioural mechanisms throughwhich proximity influences attitudes, commitment and policy choice.

The geography approach only begins to take on weight, help understandevents, and – not least important in social science – result in sensible recom-mendations for policy, when such mechanisms are explored in detail.Fortunately, this has been done in the study of transformation by manyanalysts – typically political scientists rather than economists. One recentstudy provides a succinct example of the way in which the analysis can bemade satisfying and useful. Kopstein and Reilly (2000) go well-beyond show-ing that distance from the nearest large European city (Berlin and Vienna are



the two chosen) has strong and significant explanatory power in regressionfor economic and democratic freedom. They peel back the first layers of theonion to investigate how proximity enhanced diffusion of knowledge andattitudes, creating a composite variable of ‘openness’ comprised of measur-able elements including television-ownership, newspaper-circulation, inter-national telecommunications, inbound tourism, FDI, and trade. Greaterproximity gives greater trade, FDI and so on, but its diffusion is also multipliedby greater availability of communication modes.

This already allows one to understand that the much earlier opening-up ofPoland and Hungary mattered a lot, as did the Estonian’s ability to watchFinnish television well before 1989. Further layers are peeled to reveal thatnot only does proximity to Europe (or contiguity more usefully) give positiveliberalizing impetus, but, symmetrically, the contiguity of Kyrgyz Republicto others in Central Asia may give the opposite impetus – which is a possibleway of understanding the eventual deterioration of the liberal start under-taken there by a visionary leader, President Akayev. Even deeper layers areexplored, for example with the explanation that in Hungary EU proximityand strong membership prospects did not automatically determine all policy.Internal politics provided a strong consensus against foreign ownership ofland, but the inevitability of this under EU requirements was understood byall sides and was in the end reluctantly accepted.

In summary, EU membership clearly played a very large role in determin-ing the rapidity of initial reform actions, in the pace and steadfastness ofpolitical and economic liberalization, and on the final outcome after about15 years. But it is imperative to emphasize that the dynamics of this were nota straightforward relation of the EU inviting a country to membership andthus motivating the progress. The EU invitation came in varying strengths,it came sometimes earlier, sometimes later, it came as reward for self-generatedliberalization in some countries, and it was very effectively withdrawn andused as a stick to herd-in strays. But, also, it was not a force unto itself, butwas intertwined with the internal push to liberalization which was higherthe greater the degree of a non-aggressive national vision, and the greaterthe commitment of society and its leaders to the dual liberal vision.

Appendix: EU membership as a factor in political and economic reform in the Baltic republics

The prospect of EU membership has provided the Baltic republics with a strong incen-tive to economic, political and administrative reform in the post-communist period.However, this incentive has not been equally strong throughout, nor has it beenequally strong in all three countries. Table A7.1 shows the timing of formal events inthe process. Prospective membership became much more important in the reformsof Estonia, Latvia and Lithuania only after the signing of the Europe Agreements(Association and Accession agreements) in mid-1995. Prior to that, the republics hadmade only preliminary attempts to approximate their socio-economic structures to

that of the EU member states. After the signing of Europe Agreements, a more system-atic approach was undertaken in each instance. Furthermore, even prior to 1995 Estoniadedicated more resources and was more advanced in its efforts to obtain membershipin the EU than were the other two republics. Thus, while all three have declared theirdesire to work for EU membership even prior to gaining firm signals from the Unionon the viability of such expectations, real internal efforts commenced only in thesecond half of the 1990s, intensifying with each new achievement. This Appendixdocuments these efforts and illustrates the differences between countries and betweentime periods. It draws heavily on Basta (2004b).

EstoniaOn the ideological level, Estonia was as dedicated to integration with the Westerneconomic, political and security organizations even before gaining independence in1991 as were its Baltic counterparts. Membership in the EU was part of this strategic(re)orientation. Presidents of Estonia, Latvia and Lithuania met in June of 1993 to signa joint declaration calling on ‘EU countries to initiate negotiations concerning thegranting of associate member status to the three Baltic states’ (Bleiere (1999), p. 50). Inthe case of Estonia, this initiative was accompanied by the public statements reiterat-ing the importance of integration with the EU as part of a guarantee of the country’ssecurity23 (Park (1995), p. 33). Although the request was not granted, the EU indicatedthe conditions for the signing of Association agreements, including the harmonizationof legislation in line with that of the EU (Bleiere, op. cit).

On a number of accounts, Estonia took EU countries as role models for reform. Oneneeds to be careful, however, not to overinterpret the Estonian government’s actionsas being motivated exclusively by the prospect of membership. Rather, many reformpackages may have been informed by EU legislation because Estonia had little in termsof its own legislative heritage and experience to draw upon. For instance, Estonia’slegal reform starting in 1991 drew on the laws of Germany, Austria, the Netherlandsand Denmark (Pisuke (2001), p. 192). Pisuke concludes that new Estonian legislationwas modelled on national laws ‘of the EU Member States or the states (Austria, Finland,and Sweden) who were in the stage of approximating their laws with EC law at thetime that Estonian laws were being drafted’ (ibid., p. 194). The latter might be taken tomean that Estonia wanted to prepare for membership even without any assurancesthat membership was likely and that this is why it took laws of current and prospec-tive EU members as a template for reform. On the other hand, these same countrieswould seem to be logical models for emulation as it is, since Estonia wished to attaintheir level of prosperity and efficiency. Furthermore, it is possible that both reasonsplayed a part simultaneously. However, it is unclear which consideration or combina-tion of considerations had primacy in decision-making.

What is clear is that Estonia implemented institutional provisions aimed at prepa-ration for EU membership, and approximation of legislative and administrative frame-works of the republic to those of the Union, as early as 1993. In December of that year,the government established a working group ‘to review issues concerning Estonia’spossible accession to the European Union and to prepare the corresponding docu-ments’ (ibid., p. 196). The working group included government bureaucrats, membersof parliament and academics. It outlined a theoretical foundation for Estonia’s reformsin line with EU legislation (ibid.). The establishment of such a forum by the govern-ment with EU membership in mind is something that was, as far as this author hasbeen able to ascertain, present only in Estonia, not in Latvia and Lithuania. Beyondthis, the government adopted a resolution for the establishment of a system aimed at



coordinating the inter-ministerial activities related to potential accession to the EU.This took place in December of 1994 (ibid., p. 197). Although the resolution failed tobe implemented, it signals the early dedication of Estonia to the adoption of EU lawseven prior to the signing of the Accession and Association Agreement.

Estonia signed the Europe Agreement with the EU on 12 June 1995, together withLatvia and Estonia.24 This Agreement, although not ratified until 1 February 1998,25

was a clear signal by the European Union of its intent to seriously consider Estonia asa potential member. The systematic efforts to approximate Estonian laws to the AcquisCommunautaire (AC) took place after this date. Beginning in January 1996, Estoniaestablished the institutional system devised to facilitate approximation of nationallaws to those of the Union (Maniokas (1999), p. 125). The following is a point-formrepresentation of that system:

● The Ministers’ Committee (or European Integration Committee) comprised of keyministers and in charge of setting the general policy principles (Pisuke, p. 198).

● The Council of Higher Civil Servants, comprised of members of most ministriesand the Bank of Estonia, coordinating inter-ministerial work related to issues ofEuropean integration (Pisuke, p. 198). It furthermore elaborates and implementspolicies related to the EU and it prepares meetings and drafts decisions for theMinister’s Committee (SIGMA 1, p. 22).

● Office of European Integration coordinates the activities of the above two organi-zations, arranges for information exchanges with the European Commission, coor-dinates technical assistance and trains public officials in EU matters, and monitorsprogress on approximation of national laws with the AC (Pisuke, p. 198).

However, individual ministries have the responsibility of complying with EU legisla-tion in their area of activity (ibid., p. 199). According to Maniokas (1999), this is thedistinctive feature of Estonian EU-related organizations (p. 125).

Having established the institutional infrastructure, the Estonian government adoptedthe first Activity Plan for Joining the European Union on 6 June 1996 (Pisuke, (2001),p. 201). The February 1997 Activity Plan was devised for the purpose of the imple-mentation of the European Commission White Paper which outlined the steps thatapplicant countries needed to implement in order to be considered for membership(ibid.). Like the other two Baltic republics, Estonia was subsequently subject to carefulscrutiny of the European Commission, which requested regular reports on, and thenproduced regular reactions to, the progress of each of the countries. As a result of thisprocess, the Commission decided in 1997 to open accession negotiations with Estonia.26

The EU’s influence from that point on was increasingly more direct and explicit. Likeother applicants, Estonia had to meet the demands of the Commission, or face thepostponement of the accession date (Beselaere (1999), p. 79).

Latvia and LithuaniaAs already mentioned, Latvia and Lithuania did not initiate the institutional reformsas early as Estonia. Indeed, Latvia devised a systematic foreign policy as late as early1995 (Pabriks and Purs (2002), p. 124). In his study of reform of Latvia’s legal system,Levits (2001) mentions vaguely that the reform ‘in the direction’ of conforming to EUquality standards of public administration and court system was begun in 1993 and1994, but that it was ‘flagging’ from 1995 to 1999 (p. 184). Again, as in the case ofEstonia, it is difficult to ascertain whether such reform was driven by the need forwhat were perceived as proper legislative arrangements solely on their own merits, orwhether the intent was to impress upon the EU the desire for membership. Certainly,

this author has found no evidence of a government-initiated institutional frameworkfor the approximation of Latvian laws with those of the European Union prior to the1995 signing of Europe Agreements between Latvia and the EU.

Latvia established the institutional framework for approximation of legislationto the Acquis Communautaire in 1995, and the following is a point-form schematicrepresentation of that framework:

● The European Integration Council is an inter-ministerial body comprised of therelevant ministers and headed by the Prime Minister (SIGMA 2, p. 23).

● Council of Senior Officials, acting as a preparatory body for the EuropeanIntegration Council, first met in February of 1998. It provides the ‘interministerialframework for dealing with EU co-ordination’ (ibid.).

● European Integration Bureau is an administrative unit providing policy and admin-istrative support to organizations involved in the integration process. Furthermore,it is in charge of the preparation of the National Programme for European Integration(SIGMA 2, p. 23). This body has been part of the organizational infrastructure since1995 (Maniokas, p. 124).

The Latvian Cabinet adopted the Latvian National Programme for Integration intothe European Union in December of 1996 (Levits, p. 185). However, despite the morefocused efforts on behalf of the Latvian government to satisfy the so-called CopenhagenCriteria of the EU,27 the Commission’s opinion on Latvia’s application for member-ship left it, together with Lithuania, out of the next round of accession negotiations.The stated reasons for the postponement of negotiations were the perceived incapac-ity of both countries’ economic sectors to cope with the competitive pressures of thesingle market (Bleiere, p. 51).

Lithuania differed somewhat from its Baltic neighbours in its approach to integra-tion into Western institutions. Namely, while it was in principle as dedicated to mem-bership in the European Union and integration into Western institutions as the otherBaltic republics, its relationship to Russia was initially more amicable than in thecase of Latvia and Estonia. In his treatment of the subject, Thomas Lane (2002) citesnumerous examples of this. Before the actual dissolution of the Soviet Union, Russiaformally recognized Lithuania’s independence on 29th July 1991, referring ‘uniquely’to the ‘annexation’ of 1940 as the legal basis of this recognition (Lane, 2002, p. 200).Furthermore, although politicians of all hues agreed on the necessity of extracting thecountry from Russia’s sphere of influence, ‘there was general agreement on the needto maintain good neighborly relations with Russia’ (ibid.). Efforts in this directionyielded the most favoured-nation status with Russia in November 1993 (ibid., p. 201).28

Lane furthermore emphasizes that although Lithuania continued reorienting its tradetowards the West, especially the countries of the European Union, ‘it was widely acceptedthat [economic relations with the West] would not completely displace Russian mar-kets and supplies’ (p. 202). Thus, early on, Lithuania’s foreign policy efforts weredivided between cooperation with the West and the East.29 This could explain in partthe reasons behind the apparent lack of urgency in conforming to the EU Copenhagencriteria in the early 1990s.

Like the other Baltic states, Lithuania, too, modelled its legislative reform on thelegislative systems of the EU member states. For instance, Lithuania’s law on Principlesof Accounting, passed in 1992, was in line with EU company law (O’Rourke (1995),p. 121). However, as in the above cases, it is difficult to establish whether this was merelya case of legislative and policy learning or a matter of trying to increase the chances of EUmembership in the future. The institutional framework for the approximation of



national legislative and policy structures to those of the Union did not exist longbefore the signing of the Europe Agreement. Lithuania’s system of EU-related institu-tions was established a month before the signing in 1995 (Maniokas, p. 125), and thefollowing is a point-form outline of the system:

● The Governmental Commission of European Integration, established in 1995, isthe institution in charge of the strategic direction of Lithuania’s EU integration. Itdevelops the ‘basic principles of the integration policy pursued by the government,and … [coordinates] and [supervises] the activities of the ministries and govern-ment agencies in this field’ (SIGMA 3, p. 25). Chaired by the Prime Minister, it iscomprised of key ministers in charge of the integration process (ibid.).

● The Delegation for the European Union Pre-accession Negotiations is an interiminterdepartmental entity in charge of coordinating the work of the departments inpreparing Lithuania for membership. It is comprised of senior public sector offi-cials and related persons. It was established in 1997 (ibid.).

● The Ministry of Foreign Affairs establishes policy guidelines in the integrationprocess, coordinates economic policy development in line with EU criteria and is theliaison between Lithuania and the EU (ibid.).

● The European Committee under the Government, established in 1998, ensuresthe implementation of the National Programme of the Adoption of the AcquisCommunautaire, and the Institution-Building Programme (ibid.).

Just like with Latvia, Lithuania’s negotiations for accession were postponed by theCommission in its 1997 opinion. In addition to having been considered poorly pre-pared for the competition of the single market, both Latvia and Lithuania were foundto have had an inadequate administrative capacity to implement the Acquis.30 However,the postponement provided a strong impetus for further reform to both countries. AsBungs (1998) states, ‘within weeks, however, the Latvians and Lithuanians were backat work, striving with even greater intensity than before to raise their countries’ stand-ing vis-à-vis the membership criteria’31 (Bungs, (1998) p. 13). The eventual full com-pliance with the EU criteria has led to the membership for all three Baltic republics inMay of 2004.

ConclusionThe prospect of EU membership has had significant impact on the course of politicaland economic reform in the three Baltic republics. However, this influence was muchmore explicit and easily discerned in the aftermath of the 1995 signing of EuropeAgreements, and especially in the aftermath of the 1997 Commission opinions on themembership applications, and the decision to open negotiations with Estonia. Inthe pre-1995 period, the influence is less clear. Although there is evidence that allthree states used EU member states’ legislation as a template for reform in various sectors,it is not clear whether this was because they wished to put pressure on the EU formembership, or whether it was due to the lack of indigenous knowledge on how toorganize democratic polities and market economies. Certainly, it is possible that bothconsiderations played a part. However, no definitive evidence is available in the exist-ing literature to make a proper assessment in this matter. What is clear is that Estoniawas the only state systematically preparing for EU membership negotiations evenbefore having signed the Europe Agreement, having initiated the development ofinstitutions to coordinate the approximation of domestic legislation to that of the EU.That all three countries have eventually completely reformed their administrative,

legislative and economic systems in as little as a decade is a testament to the power ofthe EU membership prospects in the post-communist era.


Table A7.1 Contractual relations between the EU and the Baltic republics up to 1995

Estonia Latvia Lithuania

Establishment of 27 August 1991diplomatic relationswith the EuropeanCommunities

Signing of Trade 11 May 1992and CooperationAgreement with the EC

Free Trade Agreement 18 July 1994 (entered into force 1 January 1995)signed with theEuropean Union

Signed Europe 12 June 1995Agreements

Applied for EU 24 November 27 October 1995 8 December 1995membership 1995

Source: Commission Opinion on Estonia/Latvia/Lithuania’s Applications for Membership of theEuropean Union. Accessed at europa.eu.int/comm./enlargement/candidate.htm

Part III

A Summing Up

8Future Prospects for Captured States

233

A fork in the transition road

‘Forecasting is very difficult – especially for the future.’Mark Twain

The 27 post-communist countries have diverged considerably in their progresstowards open markets and democracy, but the two dimensions which haveperhaps the greatest implication for future prospects are the degree of oli-garchic development and the potential for EU membership. These dimen-sions in fact overlap closely. One group in Central Europe, the Baltics andsome of South-East Europe have either already achieved membership as ofMay 2004, or are in the process of negotiating accession in the near future.Most of them overcame strong rent-seeking and state-capture risks early on,while others like Bulgaria, Croatia and Romania are well on the way toreversing such effects. The transformation in these countries is for all practi-cal purposes complete, any final refinements of market and democratic insti-tutions being more or less assured as EU standards kick in. Their futureeconomic problems are those of middle-income emerging-market economies,plus all that is associated with working in a new enlarged EU. The remainingcountries in South-East Europe have been provided a conditional offer fromthe EU without a clear time-horizon, and while they face the challenge ofcontinuing to move forward in their transformation reducing any remainingstate-capture elements, they do this with a strong beacon to motivate thepolitical consensus and the clear guidelines of the Acquis. For them the trans-formation is not over by any means, but the light at the end of the tunnel isstrong and almost certain to grow stronger year by year. The future for thesetwo groups is fairly clear even if not assured, as it rests within the sphere ofEU integration and accession. It will not be explored in any greater detail here.

It is in the third group one finds the most important and interestingissues, and of course uncertainties. With the possible exception of Ukraine,the 12 countries of the CIS face a zero or low prospect of EU accession in the

foreseeable future, and have become burdened with an oligarch class thatdominate the economy and to varying degrees have captured the polity. Forthem, the central issue is whether the transformation is for now sloweddown or temporarily stalled, or whether it had been frozen in a political-economy equilibrium somewhere between plan and market, between social-ism and capitalism, no longer a communist dictatorship but far from aliberal democracy. Three of these countries, Belarus, Turkmenistan andUzbekistan, are clearly much closer to the original starting point of a com-munist regime, and their future is much too uncertain to permit usefuldiscussion of prospects until a radical change in leadership occurs. It istherefore the nine CIS countries with moderate but significant progress intransformation that form the focus of this chapter.

For them I propose the most useful and critical aspects of discussions ordebates (as they have already started) can be grouped under two schools ofthought. The first argues that once a minimum of stabilization, market lib-eralization and privatization is achieved, further progress in transition isinevitable (TI); indeed, it will also help eventually in furthering democraticprocesses. The second takes a conditional view of this inevitability, arguingthat where the reform process allows vested interests to build up quicklyand benefit from rent-seeking opportunities of partial liberalization, theyacquire a concentration of state assets in an opaque privatization andbecome what is popularly known as the ‘oligarchy’ which captures gover-nance of the state. Their interests are not to liberalize or democratize further,rather exactly the opposite; the transition is frozen (TF) into a capitalist butnot competitive economy and an autocratic polity perhaps in superficial dis-guise of a voting democracy. The arguments pro and con for these two viewsare explored in the next section. This is followed by a discussion of the con-sequences for a country where transition is in fact frozen by vested interests,and a final section will consider what policies might be undertaken to over-come the equilibrium of frozen transition and to free the captured state. Oneof the theoretical possibilities, a new revolution to overthrow the oligarchy,has become quite realistic, as already three instances of Gandhian non-violentpeople-power have materialized: Georgia, Ukraine, Kyrgyz Republic. Whilein all three cases it is much too early to understand well these episodes – notto speak of declaring success – the ‘Revolutions of Colour’ can be usefullyanalysed in the context of the Navigation Model.

Is further transition inevitable or is it frozen?

The theoretical arguments

The TI view is more than a simple optimistic assessment on the future behav-iour of oligarchs, or a mechanical ‘critical-mass’ argument that private own-ership and market rules have passed a threshold and are irreversible. It has amuch deeper economic logic. First, it argues that the critical mass of private

234 A Summing Up

Future Prospects for Captured States 235

interests and market mechanisms has reduced the power of the bureaucracyso much it no longer threatens continued liberalization. Second, the newprivate interests, in particular the big capitalists or oligarchs, now want tosecure the rights to their property and will begin to demand good rule-of-law, transparency and so on. Thus, for example, Shleifer (1997) argues‘Russia’s experience shows how privatization, combined with equity incen-tives for insiders, transfers control rights from the bureaucrats and stimulatespolitical and economic pressures to protect property rights.’ Aslund makesthe same case that ‘the strength of capitalists is essentially a liberalizingforce’ (1995) because ‘capitalists want to be independent of bureaucrats andsafeguarded by a system of law’ (1997). Many have extended this logic to ademand for democracy; thus, Aron (2003) sharply criticizes the jailing ofthe head of Yukos, Mikhail Khodorkovsky, stating that ‘oligarchs can helpadvance the cause of democracy’.1

These views often lay behind the arguments for privatizing rapidly andearly, even at the cost of using ‘insider co-opting’ approaches, as for examplein Boycko, Shleifer and Vishny (1995): ‘the main contribution of large scaleprivatization is to jump-start the demand for institutional development’.Gaidar was a very forceful proponent (and of course implementer) of theneed to break the overweening power of the state and bureaucracy by rapidand substantial privatization, and to overcome what he called the deep-rooted historical ‘eastern’ vision of private property (always) being subordi-nate to the state and its bureaucracy. However, Gaidar was not necessarily inagreement with the prediction that new capitalists could soon demandbetter institutions. Indeed, he worried as early as 1994 whether Russia was‘moving into a free and open Western-type market, or into nomenklaturacapitalism’ (Gaidar, 2003, p. 82), and worse that a renewed desire for a‘firm hand’ was leading in the direction of the traditional strong state inthe ‘eastern’ mode (p. 109). In that caution he seems to lean towards the TFarguments.

A similar ambivalence is reflected in Buiter’s (2000) article. On the onehand it provides one of the clearest expositions of the argument based onthe Coase proposition that efficiency only requires unambiguous assign-ment of property rights, and the derived implication that all capitalists wantsuch transparent property rights law. On the other hand it expresses greatconcern about the ‘predation’ seen so far and is sceptical that there is a good-will by oligarchs now that privatization is done to press for a future of rule-of-law in Russia and other CIS states.2

The efforts for oligarchs to become legitimate and gentrified are clearlyincreasing, as exemplified by the recent comment of Vladimir Potanin ofRussia: ‘We want to be normal and socially acceptable. Let us pay our per-sonal taxes and show how much money we spend on sponsoring culture.’3

This is consistent with the TI prediction that once they have ‘stolen’ enough(or all there is), oligarchs wish to normalize their status enshrining property

rights. But such gentrification is also consistent with maintaining enoughpolitical influence to prevent full liberalization with open entry for new,small capitalists who receive equal treatment from governments and thelegal system.

Most of the TI arguments are made in reference to Russia, but they clearlyapply to all countries where a concentrated ownership and formation of afew powerful capitalist groups has occurred. The World Bank (1996), in itsfirst review of transformation progress, on the one hand expresses the samecaution as did Gaidar about the process of such accumulation, but on theother hand elaborates in a more positive tone the ideas of further develop-ment of good institutions in response to the demands for secure propertyrights. The hopeful tone of the 1996 Report is not repeated in the 2004 studyof Russia’s privatization and the concentrated ownership structure that hasresulted.4

The counter-argument of the TF view can be put most simply as follows:capitalist-oligarchs will demand general improvements in rule-of-law notout of benevolence but only if that is what gives the optimum outcome fortheir profits. In the current circumstances of oligarchs in captured CIS states,there are many reasons to believe that the optimum for them means acontinuation of the status quo. Some, such as Barnes (2003b), argue thatprivatization opportunities still exist, hence oligarchs still prefer a non-transparent environment for the near future. But the substance of the TFargument extends beyond this, for even when all is privatized there are sub-stantial rents to be obtained by the large and influential oligarchs; for exam-ple, Puglisi (2003) in an interview with the head of Neftehaz Ukraine,learned that the additional profits to be made from their unique privilege ofbuying gas below world prices were as much as a third higher.

The historical evidence of the Chapter 6 Appendix makes clear thataround the world, over time, there is considerable value to the rent-seekingefforts of ‘entrenched vested-interests’ in economies at different stages ofdevelopment. As long as there are rents or excess profits to be had, it makesmost sense to analyse the behaviour of the entrenched post-communist oli-garchs as a choice to be made trading-off the benefits of secure propertyrights in a transparent rule-of-law environment, against the benefits of con-tinued rents from lobbying and influence in a non-transparent environ-ment. In that context, the TF views agree with those of the TI school thatthere are benefits for oligarchs from improved rule-of-law, and if these bene-fits are large enough they will indeed demand such improvements. The TFviews, however, deny that this is automatic.

A number of recent theoretical analyses of this trade-off have been under-taken and, though the mathematical models developed show that the choicecan go in either direction, their authors first point out that the automaticityof the TI school’s predictions is not theoretically justified, and second tendto lean in the direction of the TF school that for the near-future it is more

236 A Summing Up


likely that oligarchs would choose to continue the status quo. Thus,Polischuk and Savvateer (2004) show that the more unequal ownership, themore likely that the choice goes in the direction of the status quo; the evi-dence suggests ownership is in fact highly unequal. They conclude that‘some wealthier agents would prefer a hybrid equilibrium with informal pro-cedures and rent-seeking opportunities [my italics] to the market one and thuswould resist secure property rights’. Their model also shows a very importantresult that the status quo – where oligarchs retain their privileges, competi-tion is restricted and property-rights security does not prevail – can be anequilibrium state and can continue for some time contrary to the predictionsof the TI school.

Sonin (2003) goes even further by including in his model the possibilitythat security of property rights can be bought individually rather than pro-vided as a public good. The very rich can buy a lot of such security, hiringguards for their physical property (which is no small deterrent in such anenvironment even against unfriendly government inspectors), hiring thebest lawyers, buying out high-level bureaucrats and politicians. Doing this,rather than pressing for rule-of-law as a public good, means not having togive up the privileged position of protection from competition. Sonin thuscomes to a result that, for a long time to come, the trade-off will not lead tooligarchs demanding rule-of-law, but rather to support for a non-competitiveand non-transparent status quo. The strength of Sonin’s variable – purchaseof property rights – is surely even greater when one thinks of the combinedeffect of oligarchs in collusion going beyond specific narrow acts of lobbyingfor individual privileges, to the broader notion of state capture and controlof electoral outcomes and powerful influence on general policy direction.

An important and pioneering contribution to the TF school is Hellman(1998), who confirms the fears that privatization which coopts, politicallyand financially, powerful individuals by giving them a quick and low-costinsider access to privatizing assets, did indeed result in a concentration ofassets that would be inimical to further liberalization. He underlined a para-dox in the matter: from the mid-1990s, the strongest and most effectiveopposition to further reform in many CIS countries came not from thelosers, the population which felt the pain of unemployment and lower liv-ing standards, but from the ‘winners’, the new capitalists benefiting from anon-transparent and generally inequitable transfer of state assets.5 A keyconclusion of Hellman’s article is that these oligarchs have the ability to cap-ture the state and ensure its policies are favourable to them and not to open,competitive markets.

A small qualification is in order concerning support for reforms. The newcapitalists, unlike the soviet bureaucracy, were far from categoricallyopposed to all economic reforms. Havrylyshyn (1995a and b) recounts thehistory of what they favoured and what they opposed. They enthusiasticallysupported Gorbachev’s perestroika efforts which legalized private activities,

such as the 1988 Law on Co-operatives, and continued to favour freeingmarkets and privatization of state assets, even the general liberalization ofmost goods other than energy and raw materials. They were happy to seeonly a quasi-liberalization for the latter, as the wedge between still-controlledand world prices on such key goods became a major source of the first largeaccumulation of wealth. They were also not opposed to the continued fiscalsupport for state firms as most of the new private entities were implicitlyconnected to them and could easily practice transfer pricing which siphonedoff funds to the private firm. Nor did they oppose the monetary looseness ofcheap credits to help enterprises keep paying their workers but often joinedin the chorus of demands for such ‘socially-oriented’ policies. To them, thehigh inflation this generated provided yet another form of capital accumu-lation allowing borrowing at negative rates of interest.6 By the time they hadaccumulated very large amounts of wealth however – about the mid-1990s –they reversed position on inflation which would in the long-term under-mine growth prospects and kill the goose that lays the golden egg. Butthey continued, perhaps even more forcefully, to oppose further economicliberalization which could lead to growing competition from newcomers,domestically and externally.

The historical evidence of non-transition economies

Powerful economic interests influencing governments in their own favour isnot a ‘monopoly’ of the post-communist states, but is historically wide-spread, as summarized in the Appendix of Chapter 6. This historical experi-ence is, unsurprisingly, adduced by proponents of both the TI and TFschools. The former, for example, will sometimes describe the Russian oli-garchy as a passing phase analogous to that of the Robber Barons in the USA.In Chapter 6, I have already noted how the oligarchs differ from these andfrom East Asian crony capitalists, who accumulated their first wealth byvalue-added activities, as Krueger (1999) notes, rather than privileged accessto privatization of state assets. Where there is a similarity, the historicallesson favours the TF conclusions: powerful entrenched interests resistliberalization rather than support it.

Both the theoretical rationale for this last proposition and considerablehistorical evidence is to be found in overview studies of entrenched interestsor powerful incumbents such as in Rajan and Zingales (2003b) and Morck,Wolfenzon and Yeung (2005) summarized earlier.

An application of the historical lessons for transition economies byDurnev, Li, Morck and Yeung (2004) is worth citing at some length:

There is thus a potential danger that some transition economies are intransit from communist dictatorship to economic dictatorship by a smallclique of politically well-connected, entrenched insiders. Once installedthe elite undertake further rent-seeking to lock in the status-quo. Thisrent-seeking aims to limit outsider’s property rights, public investor’s

238 A Summing Up


rights transparency, and openness to the global economy. It induces cap-ital misallocation and therefore slows growth. It seems likely that a lack ofupward economic mobility for latecomers should also result though thereis no evidence for or against this at present.

The present author finds arguments such as the last much more realistic andmore compelling than those of the TI school. Certainly, the historical evi-dence of how powerful capitalists have behaved in market economies doesnot lend support to the proposition that their post-Soviet reincarnations willwant to gentrify and therefore begin to demand a generalized improvementof property-rights institutions. The Robber Barons, or Latin American tycoons,did not yield up their power willingly, but had it reduced by externalpressures of populist government policy, or financial crises catalysing deepstructural reforms.

Today, the difference between the captured post-communist states andothers is not of course that the latter are free of corruption, powerfulentrenched capitalists and government lobbying; the phenomenon has andwill continue to exist everywhere. The real difference is that the degree of theoligarch power is so much more than elsewhere that it becomes a differencein kind. More precisely, restating the points made in Chapter 6, governmentlobbying by large firms typically involves selected actions when an issueis on the government agenda affecting that particular firm. However, asNissinen (1999) notes for Latvia, ‘they have neither the interest, the time northe resources to try to influence the general directions of policy’. Capturedstates are different in that the oligarchs there do have the interest, the timeand the resources to influence the general directions of policy up to andincluding colluding on efforts to determine election results. The election ofPutin in 2004 is clouded by the fact of his disfavouring some of the oligarchswhile having the support of others, and of the populace seeking the stabilityof a strong hand. The election of Ukraine, with its egregious and in the eventmuch more transparent manipulation and fraud, provides a clear examplewhere the candidate of the incumbent party-in-power was supportive of andsupported by virtually all the major oligarchs.7

Empirical evidence of frozen transition

Both the TI and TF schools make predictions of what will happen next inthe captured states, and only time will tell which is more correct. However,there are already indications of the choices being made by the oligarchs atleast to the present TF position. In Chapter 4 three testable hypothesesrelated to the debate were posited:

● A high level of state capture freezes or slows transition.● Large firms tend to be much less supportive of liberalization rule-of-law

than are small and new firms.● Large firms generate less efficiency improvements and innovation than

do small and new firms.

The first two are explored below, while the third is discussed in the nextsection under the rubric of growth prospects in captured states. The firsthypothesis can be found very early in the literature. Both Havrylyshyn(1995a) and Hellman (1998) argue that new powerful capitalists at firstsupport reform progress, but at some point further reforms will, by reducingdistortions (recall Figure 6.1), begin to reduce their rents. At this point theywill attempt to ‘freeze’ the reforms as this maximizes their profits. The state-capture index (SC) compiled by the World Bank for 1999 (SC 99) and theannual TPI measures of reform progress of the EBRD permit a simple test ofthe hypothesis broadly illustrated already in Fig. 6.7. Some further statisticalrefinements are given here.

As noted in Ch. 6, some of the SC99 values might be questioned. The laterestimates of ‘corny bias’ (Kaufman, Kraay, and Mastruzzi, 2003) do in factput Armenia much higher up the scale and Croatia much lower (thoughKazakhstan remains low and Slovakia high). As before, one can test for theconsistency of statistical results by making adjustments for outliers orexcluding them. In Fig. 6.7 all available data is used; Fig. 8.1 sets the valuesfor Armenia and Kazakhstan at 0.25, and those for Croatia and Slovakia at0.20 for reasons given earlier. The statistical fit is subtantially stronger withR2 rising to 0.720; if instead the four countries are simply excluded, the R2

rises even higher to 0.793 and despite the smaller sample the statistical signif-icance remains high.

Indirect supportive evidence for this hypothesis is also to be found innumerous studies describing the process of privatization in the region,8

the ways in which insider-favouritism was likely to lead to much greater

240 A Summing Up

Figure 8.1 State capture leads to frozen transition

3.5

4

4.5

3

2.5

2

1.5

Tra

nsiti

on p

rogr

ess

indi

cato

r, 2

004

1

0.5

00 0.1 0.2 0.3 0.4 0.5 0.6

State capture index, 1999

y = 2.4736x + 3.9725R2 = 0.7201

MDAMKD

AZEYUG

KGZ RUS

BIHUKRGEO

BGRROM

ARMHRV

SVK

KAZ ALB

POL

LTU

ESTCZEHUN

SVN


concentration of ownership in a few hands,9 and the subsequent ways inwhich these powerful owners could use their influence to reinforce theirposition. The most thorough such analysis is the World Bank 2004 study ofownership accumulation, concentration and consequences in Russia. It iden-tifies the 23 largest Financial–Industrial groups (FIGs) and lists the names of36 individuals as the major shareholders, a list quite similar to that found inBarnes (2003b). The study estimates that at a minimum they control 35 percent of sales in the entire economy, and perhaps as much as 60–80 per centin the key sectors of autos, metallurgy and energy. The degree of such con-trol is unparalleled elsewhere, except perhaps in the other captured states ofthe region where such a thorough estimation remains to be undertaken.

The second hypothesis posits that small, medium and new entrepreneursare more likely to support liberalization and even-handed rule-of-law thanlarge oligarchs. This is not new to transition, and as Aslund and Johnson(2004) outline is based on Olson’s 1971 theory of Collective Action (repre-sented with application to post-communist countries in Olson (1993)). Ifthere are very few small firms, there is no incentive for anyone to lobby thegovernment for a public good shared by all, and each is too small to get thesort of insider privileges an oligarch can. Once the group becomes largeenough it can form coalitions to lobby governments for their interest. Foreach small entrepreneur the goal of profit maximization is the same as foroligarchs, as is the wish to have less competition. But while the oligarch canreasonably expect success in efforts to restrict competition, a large number ofsmall entrepreneurs cannot and can only work to establish the public goodof fair and transparent laws to ensure no-one is given more privilege.

As of now, the extent and political power of small and medium-sizedenterprises (SMEs) in captured states remains very small; Aslund andJohnson (2004) estimate that while in Central Europe this has now reachedlevels where 50–60 per cent of GDP is produced by SMEs as in the industri-alized countries, it is only about 20 per cent in CIS countries. Thus, it is fartoo small to have much impact even if their efforts are devoted to supportimproved institutions. Are they? A number of quantitative studies of votingbehaviour in transition economies provide the closest approximate test forthe hypothesis. While to my knowledge no evidence exists correlating sup-port for better rule of law by small enterprises or efforts against competitionby oligarchs, the voting pattern analyses do show that the strongest sup-porters of reform-oriented parties and liberalizing policies are indeed SMEs.10

The third hypothesis is discussed below under the rubric ‘Growth Prospects’.On balance, the theoretical, historical and direct empirical evidence in

transition countries tends to be much more consistent with the TF view.Where does the TI proposition err? Certainly not in the theoretical conceptof Coase-theorem market efficiency and the values of secure property rightsfor all capitalists small and large. As described in Buiter (2000), if there arisesa demand for good institutions there will indeed be a supply response from

the government and the society. But the reality on the ground in capturedstates is that so far, and for the near future, there is no evidence of such ademand coming from the oligarchs, and there is evidence that small entre-preneurs, who may demand this, are too powerless as a group, for now, toachieve results.

Consequences of a frozen transition

If transition is indeed frozen for this group of about 10 states, what are theconsequences for their further evolution? I consider here four dimensions:enlargement of the SME sector; institutional development; growth prospectsboth in the short and long term; and the prospects of future EU membership.

Enlargement of the SME sector

In a captured-state equilibrium where powerful oligarchs can influence eventhe general direction of policy, their interests would likely result in a muchless favourable environment for small and new enterprises. Competitionrestrictions would continue, the legal environment would not be very trans-parent, consistent or even-handed, and SMEs would face greater burdens oftaxation, bureaucratic harassment and the need to engage in bribery. In factsuch effects prevail already in the relevant countries.

As noted, the sector remains very small in the CIS countries, accountingfor only about 20 per cent of GDP compared to the 50–60 per cent norm ofadvanced market economies; the latter norms have incidentally beenreached by now for many Central European and Baltic countries (see Aslundand Johnson, 2004; World Bank, 2002). How far behind the CIS group ofcountries finds itself is dramatically pointed out in Frye (2003) and Frye andShleifer (1997): Poland, with a much smaller population and economy,already by the mid-1990s had six times the number of registered SMEs thandid Russia. The importance of this for development of civil society and apolitical force for improved rule-of-law cannot be overestimated; the WorldBank (2002) not only discusses this linkage, but suggests that historically acritical mass of about 40 per cent of GDP is needed before the politicalweight of this sector begins to be felt. If the transition is frozen, the contin-ued growth of SMEs may not be halted, but it will certainly be slowed, andreaching the critical mass may require not years but decades.

Actual tax payments by SMEs appear to be higher than for large enterprisesas indicated in many survey results for these countries (EBRD TransitionReports 2000, 2003; the 2002 World Bank study; an IFC 1999 study forUkraine). In Ukraine, for example, the average tax load was 30 per cent ofgross sales. In addition, the ‘bribe-tax’ businesses are forced to pay in orderto be in business is also much higher. The rate for small firms in all transitioncountries is on average 5.4 per cent of sales, twice the 2.8 per cent rate forlarge firms. In the CIS states, subject to greater state capture, figures are even

242 A Summing Up


higher, as much as 8 per cent of revenues and in effect ‘a third or more oftheir profits’.11

Parallel to the higher tax and bribe payments, SMEs face much greaterbureaucratic harassment, time-costs of dealing with inspections, amorphousand changing regulations. It is reported in Kaufmann and Kaliberda (1996) thatfor small enterprises senior management spent an extraordinary 18 per cent oftheir time dealing with government officials. On lack of consistency, Puglisi(2003) cites a business consultant in Ukraine: ‘nobody plans long into thefuture … you may start by playing chess, then in the middle of the game youfind you are playing basketball or football’. The bureaucratic burdens, bribecosts and changeable rules, rather than the formal tax levels, are widely agreedto be the main reason behind the large increase of underground economicactivity in many CIS states, even beyond the high levels in the Soviet period.Kaufmann and Kaliberda (1996) and Feige and Urban (2004), as discussed inChapter 2 Appendix, amply demonstrated the very high levels (50–70%) andthe predominance of harassment as the reason for going underground.

Most analysts attribute the poor environment facing SMEs to the fact ofoligarch interests and influence on policy. Hellman and Schankermann(2000), in the key article reporting on state capture, make this link explicitand explain the higher burdens for SMEs noting ‘the interaction betweenfirms and the state is rooted in a bargain that is tailored to the … bargainingpower of the individual firms’. Ronnas (1996) describes how SMEs ‘are likelyto be most affected by poor law enforcement [while]; large enterprises areusually able to fend for themselves’. That this discrimination is not simplyan expression of the will of the government is argued forcefully in Puglisi(2003): ‘Political power allowed … big business to shape the rules of the mar-ket to fit their own preferences.’ But not all analysts agree. In a paper argu-ing compellingly that SME development is critical and recommending asimple lump-sum tax to stimulate it, Aslund and Johnson (2004) attributethe discrimination against SMEs to government policy and outside advisorslike the IMF, the World Bank and OECD, which insist the same rules shouldapply to enterprises of all size. They also note there is no direct evidence bigbusinesses who have captured the state want to block the development ofsmall enterprise, and make two contentions: small and large businesses oper-ate in different sectors, hence do not compete with each other; and Ukrainemanaged to liberalize small business operation at the height of oligarchicrule in 1998. On the first point, can one really expect to find direct evidenceof the oligarchs’ actions to reduce competition? That they currently operatein different sectors is not necessarily evidence of good intentions byoligarchs, it may in fact be evidence of the contrary, that oligarchs arerestricting competition.

The last point about Ukraine’s SME liberalization deserves more elabora-tion. It may be questioned just how great this liberalization was compared tothe desired end goal. The EBRD’s value for the index of competition policy,

which covers this area, shows for Ukraine a steady 2.0 from 1995 to 1999, amodest rise to 2.3 (� 2� in the EBRD Reports) and no further change sincethen. But, more important, some SME activity is not only unpreventable byoligarchs and their government captives, but it is desirable for them to havea small SME sector to demonstrate good will, and indeed to divert attentionfrom the oligarch dominance.12 Corruption has become a four-letter word inthe international community, and even the most corrupted and capturedgovernment has reacted by establishing anti-corruption laws, agencies andcampaigns. To make the practice of anti-corruption appear credible, theremust be enough small and medium-sized business to inspect, raid and pros-ecute on occasion, otherwise government would have to attack the largefirms.13 For a captured state this serves as an excellent diversion from thehigh level of corruption which is formally legal, but nevertheless results in ahuge legal diversion of state revenues by rent-seeking oligarchs.

Institutional development

There appears to be universal consensus in the literature on the propositionthat promoting SMEs is also good for institutional development. Aslund andJohnson (2004) may or may not agree that oligarchs try to block small busi-ness, but are unequivocal in the view that the most important effect of astronger SME sector ‘is to change the political equilibrium, creating a power-ful force for further institutional improvement’. The wider literature oninstitutions has for a long time expressed this consensus, but also frequentlypointed to the very different interests of large and small business. In the clas-sical work of North on institutions and economic development in a long his-torical perspective, he states clearly:

institutions are not usually created to be socially efficient, [but] are cre-ated to serve the interests of those with bargaining power to create newrules. (North, 1990, p. 16)

The role of institutions in the development of a liberal and market econ-omy, as well as democracy, has been much analysed by North and othersand, as Chapter 1 discussed, was a big part of the debate on how to under-take the transition. It has come back into fashion as the ‘missing link’ notonly in understanding the transition, but also the stubborn inability ofmany poor countries to take off on a sustainable growth path. In one suchrecent analysis, Acemoglu, Johnson and Robinson (2004) provided up todate evidence of how important institutions are, describe the history oftheir development in different parts of the world, but admit they are unableto provide an easy recommendation of how to put them in place where theyremain weak.

Laar (2002), the first Prime Minister of Estonia, in a book describing a suc-cessful transition including quick institutional improvements, comments on

244 A Summing Up


a failing case, Russia, and its strategy of quick privatization to insiders:

reformist politicians … were convinced that by giving the market econ-omy a chance, businessmen involved in making improper deals wouldsuddenly become proper businessmen involved only in honest business.It was also assumed these businessmen would soon support the govern-ment which had given them the chance to get richer. This was a gravemistake. The newly wealthy were the first to stand up against the policies[of the reformers] and opposition from industry to the further liberaliza-tion of the economy … was now blocking reforms. This was actually stilla minor problem. The attitude had even more serious consequencesbecause it slowed the construction of civil society.14

A more powerful statement of the frozen transition argument cannot beimagined, and more authoritative practitioners than Laar are few in number.The critical role of small business elites in ‘forming the backbone of civilsociety’15 goes back to Weber if not earlier, and so too does the counterposingof interests of the large and small entrepreneur in this regard. That oligarchdevelopment is inimical to good institutional development has becomenearly an axiom in the historical literature. Nothing appears to suggest thatthe cases of post-communist captured states will become an exception.

Growth prospects

Growth in the short and medium term may not be much impeded by theilliberal environment of a captured state, (the same point is made in Johnsonand Subramanian (2005)) but the analytics of such a situation as well asthe considerable historical evidence of analogous situations of dominanceby entrenched big business generally point to unfavourable prospects foreconomic growth in the long term. Consider the long-term effects first.

Other things being equal, the oligarchs doubtless prefer higher growthwhich extends the base from which their rents are taken. This was alreadyseen in the mid-1990s when they switched from favouring high inflationpolicies to support stabilization, recognizing this would be better for growth.But things are not always equal, and as Engerman and Sokoloff (2003), twowell-known economic historians reviewing the role of institutions in thelong-run history of economic growth around the globe conclude: ‘elites mayprefer policies that raise their share of national income even if they reducelong-run rates of growth’. The circumstances under which such a trade-offmay arise for economic elites are precisely the ones elaborated in great detailin the preceding discussion: a more competitive, open, transparent environ-ment leads to more efficiency, innovativeness, productivity improvementsand hence higher long-run growth; but more distortions raises the share ofthe elite’s rents. The lack of open competition and the uncertainty of prop-erty rights, under the non-transparent institutions that prevail in a frozentransition, impede a healthy SME sector from emerging and operating. This

is already noted by many in the relevant transition economies: ‘Insecureproperty rights are the most important barrier to sustained growth, entre-preneurship, and private sector development’.16

That large entrenched business entities are less likely to be innovative andhighly productive was long ago recognized in the non-transition literature.This is summarized in the recent review work of Morck, Wolfenzon andYeung (2005): ‘growth is lower where inherited billionaire wealth is larger asa percent of GDP’. In transition countries there is already considerable evi-dence from econometric studies of firm behaviour after privatization point-ing to a strong consensus consistent with the above, and indirectly consistentwith the third hypothesis derived in Chapter 4, that is the link between statecapture and the freezing of transition: oligarchs, like all monopolists, prefereasy rent-seeking to the more difficult work of seeking efficiency improve-ments. The general conclusion of numerous firm-level studies is that all pri-vatized firms are better than state firms in productivity, higher growth andjob-creation, but that these effects are strongest in the following order: newfirms, small firms, large firms, still unprivatized firms. In parallel, cross-country studies show that the gains from privatization are greatest where thedegree of competitiveness and quality of institutions is highest.17

Buiter (2000) also expresses concern that the predation which led to statecapture inhibits secure property rights and thus depresses capital formationand growth in the long run. In the same vein, Yavlinsky (2003) makes apoignant plea for battling the oligarchy, a plea firmly planted in simple eco-nomic theory of competition: ‘the larger and more influential the group, thegreater are its opportunities to deviate from the universality principle of thebusiness climate and fair competition’.

In the short run, however, it is quite possible for oligarchic capturedstates to experience surges of growth. Since about 2000, the fastest growingeconomies in the post-communist world are in fact those with highest statecapture indices, providing the basis for an important counter-argument onthe impact of oligarchy on growth. Shleifer and Treisman (2004) correctlynote that the crony capitalism of East Asia coincided with ‘some of themost rapid growth ever seen’. For Russia, they point to the timing of thesharp output decline before the oligarchs emerged and the rapid growthafter. The timing of facts is correct in all cases, but the simple ‘post hoc’ cau-sation implied is arguable. For East Asia’s unquestioned miracle many empir-ical studies have found that the major driving force was not productivityimprovement, but very high investment ratios.18 This does not gainsay thereal growth and its real benefits, but it speaks to the long-term sustainabilityof high growth under crony or oligarch capitalism. In the case of the recentrecovery with East Asian-like growth rates, it is a bit of a stretch to attributethis only to the dynamism of the big new companies and their new invest-ments. Chapter 2 discussed the various factors that jointly explain this surgeof growth and the new confidence of domestic business of all sizes to invest,

246 A Summing Up


risk and expand is undoubtedly part of this. But so too are other elements ofluck and good policy: a natural bounce-back from the bottom, oil prices,devaluation, macro-stability, and minimal threshold attainment of structuralreforms.

The fact that oligarch companies have contributed to this growth does nottell us if this is proportionately more or less than SMEs, not until more stud-ies are done to determine whether the productivity improvement effectsvary by size of firm, and how. The recent performance does make clear thatin the short and medium term it is altogether possible for captured states, likeother economies which suffered from entrenched incumbents (for exampleLatin America), to have periods of high, even very high, growth. This maydepend more on the underutilized potential of the economy than on thedegree of capture or entrenchment. But the recent strong performance inno way assures a continued strong performance in the long run, indeed thehistorical evidence points in the opposite direction.

Future EU membership

For Russia the issue of future EU membership is arguably irrelevant; as thenominal inheritor of the USSR superpower status, its nuclear arsenal, and anuncertain but continuing regional power position,19 means it does not wishsuch membership and the EU will not initiate such an idea. The most likelyaspirants in the foreseeable future have been noted earlier and will not bediscussed specifically here. Rather, the issue is for those that may have a non-zero probability, what will the frozen transition do to that probability. Theanswer is surely to reduce the chances, because both the demand for mem-bership by the country and the offer by the EU will be lower. On the demandside, the reasoning has been outlined in preceding chapters (see the discus-sion of Figure 4.3) and is succinctly captured in Wolczuk (2004): ‘entrenchedbusiness interests would oppose the greater openness and transparency thatwould come with [EU-mandated] changes’.20 An indication of such attitudesis seen in Ukraine, under Kuchma, both in the vagueness and ambivalence itdisplayed on membership (recall the analysis of Kuzio, 2003) and the moredirect statement of disinterest and preference for the Single Economic Spaceconcept a few months before the 2004 presidential election.

The reaction of the EU itself is a mirror image: the less competition, themore oligarch dominance leads to halting or even reversing liberality in mar-kets and polity, the less welcoming will the EU be to any entreaties for mem-bership. This behaviour by the EU is already observable since the mid-1990s,when it used the stick of delaying membership for Slovakia, Bulgaria andRomania, then re-tracking them a few years later as these countries returnedto the path of liberal righteousness. The EU’s continued reluctance to givepositive signals to Ukraine as the political environment became less and lessliberal, and its parallel criticisms of these developments, are also evidence ofa negative response to a frozen transition.

Can anything be done to free the captured states?

If a country is ‘captured’ by powerful economic elites or oligarchs and afrozen transition prevails, what policies could one recommend to thegovernment? It may be an oxymoron to speak of recommendations to thegovernment of a captured state on how to unfreeze the transition; after all,the government that is captured by powerful economic interests may beunable and/or unwilling to change the status quo. But the grip of oligarchs –even in today’s post-communist captured states – is not that complete orirreversible, nor the institutions so solidly frozen as to render the situationhopeless. The Rose Revolution in Georgia in 2003 and the Orange Revolutionof Ukraine in 2004 clearly demonstrate the opportunities for quick change –and slow underlying change is surely even more possible. Whether thechanges made are quiet and slow or more rapid under a political ‘revolution’of some appropriate colour,21 it may be useful to group the recommendationsinto three:

● to reverse the most offending privatizations and take away the powerof the oligarchs;

● to restrict the power of the oligarchs by anti-monopoly regulation; and● to promote the SME sector by policies of open competition, easy entry,

fair and even-handed institutional development.

In addition, time may bring internally generated changes that could reducethe power of oligarchs, and these too need to be considered. Finally, a revo-lution of people power, as may have happened in a handful of cases, couldalso unfreeze the transition. These five points are considered in turn.

Reverse privatization

It is certainly possible to reverse the cases of privatization that were mostegregious and led to the overwhelming power of a few dozen individuals,but it is a risky approach despite the popular support it may evoke from themass of the population. First, it risks being not what it appears, but simplythe transfer of assets from disfavoured oligarchs to newly or more favouredones. The vastly popular Russian action to jail Khodorkovsky and punishYukos with ‘back-tax’ penalties large enough to threaten bankruptcy seemsso far to be an isolated action not affecting the operations of the remainingtwo dozen dominant Financial–Industrial Groups (FIGs) except to impartthe lesson that FIGs cannot ignore the authority of government. The non-transparency of a captured state hides well the possibility that the relationsbetween the government and the majority of the oligarchy remain as before,with the interests of the two being nearly indistinguishable or at least highlyoverlapping. In one interpretation the Yukos affair is a small correction of

248 A Summing Up


the general tendency, intended to reduce the excesses of consumerist andfinancial greed of the oligarchy, and rebalance towards ensuring the state hasenough resources to carry out its tasks of stability. Mid-2005 is too early totell either way, and the action may indeed be the start of deeper changeswhich truly will reduce the ability of oligarchs to virtually control govern-ment policy. In the more recent Ukrainian Post Orange Revolution case ofreviewing the privatization of Krivorizhstal, great care will need to be takento avoid the analogous appearance of punishing the oligarch, RinatAkhmetov, who supported President Yuschenko’s opponent. In particular,any action must avoid a result in which some of President Yuschenko’s sup-porters (‘lesser’ oligarchs like Mr Poroshenko, Ms Tymoshenko or others) inany way benefit from the resale.

The second risk is that any renationalization sends a confusing signal ofthe freedom of private activity and security of property rights: is this asincere attempt to counter the power of oligarchs, or is it a manifestationof what Gaidar feared in his 1994 book, another historical reversal to the‘eastern’ dominant state which is the final, and arbitrary, arbiter of propertyrights. The uncertainty this creates for producers and investors, includingforeign ones, is not good for the economy. A limited number of reversals, forthe most egregious insider transactions, can be done and give good results,as long as the process is very strictly legal, transparent and seen to be just, inthe sense of rendering justice to the nation and not just revenge by thosenewly in power.

Restrict power of oligarchs

It is certainly necessary to have in place and forcefully apply anti-monopoly,insider-trading and minority shareholder rights regulations to limit as muchas possible the excess profits and insider deals that reinforce the power of theoligarchs. But the problem is not one of legislation or formal institutions,but of enforcing these regulations without the political influence that verylarge interests can apply. The US anti-trust climate did not bring resultsovernight as has been discussed, and that was in an economy where an influ-ential small business class already existed; in the CIS captured states it doesnot, hence its additional political pressures to ensure regulators do their jobas per the regulations is missing. Quis custodiet custodies (who regulates theregulators) continues to be a problem in the most institutionally advancedpolities; in the captured states of the post-Soviet region the phrase has to becapitalized in bold. There, the risk obtains of continued arbitrary favouritismas with privatization reversals, enhancing the signal of confusion aboutproperty rights for all property owners, large and small. Small business may,as individuals, feel some satisfaction with actions to punish some oligarchs,but the general lesson they draw from this reinforces their mistrust of thestate apparatus: if they are willing to do this to some of the big guys, imaginewhat they can to do to me!

Promote the SME sector by easing its bureaucratic burden

The critical role of the SME sector, as a forceful part of civil society and apolitical counterweight tipping the balance towards liberalizing policies,is virtually universally cited in the literature on transition, the history ofinstitutional development, and growth prospects. I return below to the pos-sibility that it will increase on its own even under the difficult conditions itfaces in lagging reformers or captured states. Here consider the specific policiesthat might be implemented. The preceding analysis of what these impedi-ments are directly leads to the recommendations: lower tax-burden; simpli-fied regulations (the lump-sum tax proposal kills both these birds with onestone); ease of entry of new firms; reduced bribe-seeking harassment by tax,health and safety regulations; and, most important, fair and even-handed secu-rity of property rights under the law. The more of these impediments areremoved, the faster one might expect to see small enterprise evolve andincrease its share of GDP. At the same time, it may be the case that improve-ment on one of these dimensions is minimal and offset by worsening inanother – that is, less frequent inspections but higher bribes each time – andthe overall impact is negligible. If the lump-sum tax does not lead to worseningof other burdens, it may make a difference. Also, the phenomenon has beenstudied sufficiently to indicate what aspects have the most effect, with the con-sensus saying that complexity and unpredictability of government regulationsis more important than the actual financial burden. Thus, the simplicity of thelump-sum tax rather than a lower level may be what matters most.

The question remains whether a captured government can or is willing tofollow these recommendations. There is some hope that the oligarchs find ithard to publicly oppose measures of improved operations for private activ-ity, hence any remaining autonomy of politicians and policy-makers mightbe used to weasel in a few gradual changes to ease the operations of SMEs.Certainly, any leverage that foreign actors have to insist on continual improve-ments of this sort should be used to move as far as possible in that direction.But international organizations must be extremely careful here to avoidbeing captured as well and agreeing in a financial programme framework topartial changes that seem to be cumulative but are in fact devices containingthe sort of offsets noted above. Such episodes provide no benefit and onlycast doubt on the intentions of these organizations. Arguably, the strongestsuch case is the loans-for-shares episode in Russia, or the privatizations oflocal energy companies in Ukraine after 2000.

Time will heal some wounds

The contentions of the TI argument that time will correct the inequities andinstitutional shortcomings of insider-privatizations may not on balance beconvincing, but it is not entirely without merit. Some individual oligarchswill be the exception to the rule and seek to change their personal goals fromfurther accumulation of wealth to nation-building and favourable future

250 A Summing Up


recognition in history. It is of course far too early to see clear evidence of this.But could Russia’s Mikhail Khodorkovsky be sincere in his support for thevery liberal parties of Russia, and is it for this he is being punished? CouldYulia Tymoshenko, a key supporter of the liberal reformer Yuschenko inUkraine, be sincere in her claims that she is not interested in regaining herearlier economic power as the ‘Gas Princess’, but wants to work to democra-tize Ukraine and counter the dominant power of its oligarchs?22 Can the‘gentrification’ of Potanin and other Russian oligarchs lead to a gradualreduction of their enormous power? Time will tell, but if it is not these, therewill surely be some who enjoy a Pauline conversion on the road to Zurichand the Cayman Islands.

Other oligarchs, in fact probably many, will want to move from notorietyto perhaps a philanthropic reputation. But this is less likely to be a fully sin-cere conversion, as tax laws will probably provide incentive, as well as sim-ple personal ego requiring some good deeds that may return a notionalprofit in opening other opportunities. More effective will be the cases ofgoing global to increase the operation and/or to improve access to low-costinternational financing. Just one example is described by Gerth (2004): theefforts of Vagit Alekperov, CEO of Lukoil, to establish a beachhead in theUnited States. At a minimum, this will begin to force such companies toattain international standards of accounting, reporting, and minority share-holder treatment (I hope not Enron standards!). The reference to Enron ispartly serious – there is much such companies can do to create the appear-ance of new legitimacy without seriously undermining the power of theincumbent controlling interests. Furthermore, it is not clear that the attain-ing of such global standards will do much to reduce the ‘state capture’behaviour domestically in the near future.

The greatest effect of time passing is likely to come eventually with asteady and even, if slow, build-up of the SME sector. Despite the impedi-ments they face, small business operators continue to expand and grow innumbers, and only the near-Soviet conditions that prevail in Belarus,Turkmenistan and Uzbekistan might halt or freeze their development. Highgrowth rates in the CIS countries are the driving force, as is the hope thatconditions will change and their opportunities open up even more. Just asevery MBA graduate wants to be a CEO, so too every small entrepreneur inthe region wants to become at least a lesser oligarch with tens of millions ifnot hundreds of millions of worth. The more of this competition there is,the less likely there will simply be a replacement of present oligarchs byother individuals, and the more likely the power of the oligarchy breaksdown, for the various reasons adduced earlier. But such a natural evolutionof small business, in a captured state environment without any major policyefforts to help them, may be inevitable but very slow, taking not years butdecades to reach the critical mass when they become an effective politicalcounterweight to the influence of oligarchs.

Revolutions of colour

Until December 2004, the only reasonable possibilities for loosening statecapture by oligarchs comprised marginal and gradual effects. With thedramatic events of the Orange Revolution in Ukraine, the unthinkablebecame almost conventional as occasionally happens to turn the directionof history. While the colour revolution in Ukraine was not the first on thepalette, being preceded by that of Serbia in October 2000 (no colour wasaffixed) and Georgia’s Rose Revolution in November 2003, Ukraine’s size,strategic importance and massive popular participation fixed the world’sattention on the the power of the ‘demos’ as a systematic prospect. Thoughreference to Gandhi has not often been made, the historical similarity ofnon-violence and massive participation is striking. As this is being written inspring 2005, popular analogies are also being drawn with Lebanon, Palestineand elsewhere outside the post-communist region. These are well-beyondthe scope of this book, but the March 2005 events in Kyrgyz Republic (TulipRevolution) are surely within the same category. So, too, are the quieterrumblings in Moldova, Belarus and elsewhere in the group of captured states.

It is early days to discuss the success of these revolutions in breaking thestate capture by oligarchs, except possibly in Serbia where one could say‘so far so good’.23 It is even more premature to speculate on the prospects ofsuch revolutions spreading to the other states of the region. Here I proposeto focus briefly only on the connection between the navigation model andthese people-revolutions. Does the logic of the model lead naturally to suchpeople-power actions, or are the latter in fact contrary to the model’s predic-tions? Tentatively, I suggest a case can be made both ways: there is a system-atic causal relation from the model to these few cases of a people revolution;at the same time such revolutions seem contrary to the ‘freezing’ predictionsof the model, and one should seek other forces (or simply happenstance) asexplanations.

Consider the latter argument: non-violent revolt by the masses seemscontrary to the prediction of a frozen transition. Where state capture is mostsevere, the ruling coalition of oligarchs and their political satraps is far toostrong to be overthrown politically. The policies of frozen transition preventsignificant development of a middle class of entrepreneurs, of truly openand equal competition in the economic, or political market. The increasingvalue of oligarch assets and potential rents motivates them to increasinglystronger resistance to a change in regime. Given the inertial effect of Soviet-period complacency and fear in the populace, they are able to prevent anydemocratic-liberal opposition from developing sufficient power. The toolsare well-studied by political scientists,24 and include legitimate financial sup-port of favoured political parties, as well as various degrees of illegitimacythrough corruption of individuals, restricted access to media, voter testsand vote-rigging, purchase and falsification of votes – all the way to physicalviolence including assassinations.

252 A Summing Up


If the captured state logic in its most rigid interpretation precludes popu-lar revolt, what are the other factors early analysis lists as explanations? Twopreliminary assessments of Ukraine’s Orange Revolution in the Journal ofDemocracy (2005), Kuzio (2005) and Way (2005), list the following indicativefactors, also common to the other cases:

● A catalytic political event, typically a presidential or parliamentaryelection with a patently distorted result in favour of the ‘non-democratic’ruling clique.

● The presence of a charismatic opposition leader to rally anti-regimeforces.

● A high degree of resentment and frustration with the economic power ofoligarchs, sufficient to overcome popular complacency and lead to auniversal cry ‘ENOUGH’.

● A minimal level of civil-society institutions, sufficient to provide vehiclesfor mass demonstration.

● A sufficiently soft authoritarianism to permit some real political oppositionto exist.

The first of these, a catalytic election, has so far been a necessary conditionfor all such episodes and needs only one additional comment: it must beconsidered as a ‘necessary but not sufficient’ condition – many cases ofegregious elections, such as two recent ones in Zimbabwe, attest to its non-sufficiency. The second factor, presence of a charismatic leader, one sees inonly two cases – Saakashvili in Georgia and Yuschenko in Ukraine. Thus, itmay not be necessary as Serbia and Kyrgyz Republic suggest, though in theformer the attraction of EU membership were extremely powerful ( Judah,2005) and probably provided a good substitute.

A soft authoritarianism and some threshold level of civil society are essen-tially two sides of the same coin, and together with the resentment againstthe oligarch clique probably reflect major forces which led to the revolutionsof colour. But the resentment factor is not exogenous to the navigation model,indeed, it is very closely connected. Consider how these three factors arerelated to the logic of the navigation model and the creation of a capturedstate.

The first thing to note is that the egregious misconduct of the catalysingelections, that is the various distortions and manipulations, is as a manifes-tation of the profit-maximizing behaviour of oligarchs: they will go to greatlengths to ensure the status quo.25 The oft-cited financial support to the tuneof $600 million for Yanukovich by Ukraine’s number-one oligarch, RinatAkhmetov,26 exemplifies this. The yet unproven attempt at assassination bypoisoning of Yuschenko is certainly part of the public’s perception of theextreme behaviour possible to defend the power of the ruling clique.

Furthermore, the resentment of the oligarch’s ‘ill-gotten’ gains, and eventhe soft authoritarianism, are also consistent with the profit-maximizingaxiom of the model, when it is interpreted with some nuance. Levitsky andWay (2003) developed the concept of ‘competitive authoritarianism’ notonly for transition countries but also for other developing countries. In thetransition group, Way (2005) elaborates on this phenomena in Ukraine andnotes two pillars for it: the use of informal authoritarian levers in a contextof formal ‘competitive’ democracy; and a coalition of oligarchic forces in asymbiotic relationship with politicians. But Way may not go quite far enough,for he gives no answer to the question: why did Ukraine’s oligarchs concurwith a ‘soft’ authoritarianism and not press for a harsh and secure authori-tarian regime? One answer lies in the trade-off much discussed in Chapter 6.As long as their capture of the state is not threatened, a soft authoritarianism –including ‘pretend’ elections which can be manipulated – is more likely toensure a good image domestically and in the world, provide the political andeconomic stability that stimulates higher economic growth, and hencemaximize their long-term profits. Jumping forward to the time of actual massprotest, stability-seeking and profit-maximizing, oligarchs may have beenhesitant to favour the substantial force needed to overcome mass demon-strations, though they were surely altogether prepared for minor controlactions and violence in the shadows.

When the levels of extreme economic and political abuse reached the tip-ping point between political apathy and the passions of people empower-ment, the conditions for these revolutions were met and a change in regimeresulted, though a full reversal of state capture and unfreezing of transitionis far from being instantly achieved. Resentment was, unlike some of theother factors listed earlier, common to every recent episode of people’sdemocracy and was emphasized by many observers, for example by McFaulwhen he succinctly characterized the common element with one word‘enough!’27 That is, a critical mass of the populace was prepared to go on thestreets, risk the force of the authorities to express their view, to say enoughof this abuse of power, unfairness, corruption and so on.

In post-communist states that had attained a high degree of truly competi-tive democracy, this and other resentments were able to find vehicles of a lessradical, more conventional sort. Even where oligarch tendencies had gonesome way – Bulgaria, Croatia, Romania, Slovakia – sufficiently fair electionsprecluded the need for a colour revolution.

In these cases, the role of EU membership prospects, one of the pillars ofthe navigation model, must not be understated. Elsewhere, where this wasnot on the cards, and where state capture had reached extreme levels, theopposition and the populace had only two choices: continued political apa-thy (fed by fear of forceful suppression), or mass demonstrations. The weightof the ‘enough’ had in these cases become sufficiently large to overcome thefear-based apathy of the past.

254 A Summing Up

9Diverse Outcomes: Liberal Societies, Captured States andUndetermined Polities

255

Introduction

In assessing the outcomes of post-communist transformation to the presentdate, one should begin with the question: has the country completed thechange-over from communist central planning to a liberal market economy,in a word, is transition over? The first conclusion of this book points to thewide divergence of outcomes among the 27 countries, with one group havingessentially completed this journey, a middle group still short of the end-pointbut progressing steadily, and a third group not only farther behind but stalledat a stage of partial, frozen transformation. In the light of their new EU mem-bership, the transition in Central Europe and the Baltics is surely over for allintents and purposes; while in South-East Europe it is clearly not over yet,though steady progress is evident. For the rest, the CIS countries’ transforma-tion is definitely not complete in the conceptual sense of establishing a lib-eral economy and polity regime. But in a practical and unfortunate sense, itcould be said it is ‘over’ for now because they have become trapped in anoligarchic-autocratic regime of partial capitalism and a far from matureddemocracy. While the people-revolutions in a handful of these countries maybreak open this trap, it is far too early to judge their prospects, or the possi-bilities of a similar development in other countries. The rest of this chaptersummarizes this variation in outcome by recapping, in the next section, whathappened in the course of 15 years, followed by our explanations of why tran-sition evolved differently across the region. A final section summarizes the dif-fering implications for future prospects and policy.

Variation in transition progress

A first look at individual measures of progress such as indices of market oper-ation, democracy, economic recovery, poverty ratios and life-expectancy

trends shows a tremendous diversity of outcomes across the 27 countries.One might conclude from this that there are almost as many differentoutcome stories as there are countries and measures; however, it turns outthat a grouping of countries into four to five categories provides a simplifiedyet systematic picture of the dynamics of transformation in this period.Chapters 2 and 3 have shown that five internally homogeneous groupscan be identified in rank order of progress or performance in transition asfollows: Central Europe, the Baltics, South-East Europe, the Former SovietUnion (FSU) countries with moderate economic reform (CISM) and thosewith limited reforms (CISL). This rank ordering holds for nearly all reasonablemeasures of progress.

Using the synthetic transition progress indicator (TPI) of the EuropeanBank for Reconstruction and Development to rank order and group coun-tries, one finds the overlap among the groups is almost non-existent, exceptmarginally between South-East Europe and the CIS moderate group. Mostimportant and somewhat surprising, this measure, which was alwaysintended to be a summary or proxy only for the economic dimension of thetransformation, is highly correlated with a large number of other measuresthat transition scholars consider to indicate progress in transformation. Thedetailed analysis in preceding chapters is summarized in Table 9.1, whichshows selected proxy indicators for each of six dimensions. The first twoshould be considered as measures of the dual economic and politicaltransformation: progress towards a market economy as measured by the TPI,and the degree of democratization as proxied by the corruption index (CI) ofTransparency International, and the media freedom index prepared byReporters Without Borders. These three might be interpreted as policy inputsaimed at generating desirable outcomes or results in the society. The nexttwo indicators in fact do measure results: first, economic recovery from thetransitional recession using an index of 2003 output partially adjusted forchanges in GDP definition;1 and social well-being as measured by the overallhuman development index (HDI) of the UNDP. Finally, the chart shows twoimportant institutional outcomes: the extent to which the polity has beensubject to ‘state capture’ by powerful economic vested interests (oligarchs inthe vernacular); and the status of EU membership. The definitions and mea-surement problems for these and other data used have been discussed ingreat detail in the relevant parts of the book, and here it will suffice only toremind the reader of some of the key conclusions, and qualifications whererelevant.

The information in Table 9.1 reflects the earlier extensive analysis whichconcluded that the various measures of policy change (inputs) and of per-formance results were highly correlated, and that the rankings of the coun-try groups were essentially the same regardless of the measure used. The onlyexceptions to this were the ranking of the CISL group for the index of outputrecovery and the HDI. If one uses official GDP with a benchmark year of

256 A Summing Up

Diverse Outcomes 257

1989 for all countries, this index ranked the groups as follows: CentralEurope CISL, Baltics, South-East Europe and CISM. But the use of 1989 as abenchmark for countries other than the Visegrad four cannot possibly beright, as this attributes the output declines of the late years under commu-nism to transformation policies that did not begin until much later.Adjusting for this alone makes a big difference, bringing the ordering almostto that of other indicators, with CISL still appearing to do slightly better thanCISM. Some partial adjustment for differences in GDP measurement in thecommunist and market period brings the ranking even closer to that of othermeasures, with the values shown in Table 9.1. The CISL group still does bet-ter than CISM, although even that can be questioned: it may reflect overval-ued GDP due to barter trading at favourable terms; it may reflect a merepostponement of the necessary rationalization of inefficient producers; itmay be simple output overstatement in official data. Most important, allother economic performance indicators reviewed such as inflation and FDIper capita distinctly rank the CISL countries last, far behind the CISM group.

On the human development index (HDI), the CISL group also does some-what better than CISM, although again the detailed analysis of specific indi-cators in Chapter 3 suggests the deterioration may have been more thanofficial data show. In particular, more recent poverty ratio estimates andhence also Gini values for at least two of the CIS countries – Turkmenistanand Uzbekistan – suggest levels quite comparable to CISM countries. Buteven if the values of HDI in Table 9.1 are about right, the most reasonableinterpretation is not that CISL countries managed the transition successfullyby a gradual approach that minimized social costs. Rather, they have simplypostponed the transformation and at best postponed some of the inevitableadjustment pains that all others, including the CISM, have by now largelyabsorbed.

In the early debates, a number of important hypotheses correlating policyinputs with results were mooted. Two in particular merit attention here: toorapid market reforms threatened evolution of democracy, and rapid marketreforms would result in greater social costs. Consider the first. The well-known proposition of Przeworski, that rapid progress to market institutionsand democracy may be incompatible, is directly contradicted by the data asTable 9.1 recaps: countries that have moved fastest and advanced farthest oneconomic reforms have seen far greater progress in liberal democracy thanthose that have lagged behind. This is proxied in the media-freedom indica-tor of Table 9.1, and was shown in Chapter 2 by the strong correlationbetween the TPI measure and different measures of democracy, especiallyFigure 2.2 using an index of constitutional liberalism.

The prediction of Przeworski that rapid implementation of economicstabilization and liberalization policies would be unpopular, and lead toearly governments losing elections to more populist governments favouringgradual transformation, was only partly borne out. In many Central European

258

Table 9.1 Key indicators of transformation outcomes

EconomicDemocracy recovery HDI

TPI ● Corruption index ● 2002 GDP ● 2001 State2004 � Media-freedom rank � FDI per capita � Change capture index EU membership

CE 3.7 ● 5.9–3.6 ● 134 ● 845 15 5 acceded in May 2000;� 11–33 � $2,300 � �.030 Croatia high prospects

Baltics 3.7 ● 5.5–3.8 ● 105 ● 822 11 All 3 acceded in� 11–17 � $1,640 � �.010 May 2004

SEE 3.1 ● 3.9–2.3 ● 83 ● 772 24 Bulgaria and Romania� 34–59 � $470 � �.020 high prospects;

others distant butfeasible prospects

CISM 2.9 ● 3.0–1.8 ● 76 ● 737 30 Most virtually no� 73–148 � $340 � �.035 prospects;

Ukraine a possibilitysince OrangeRevolution

CISL 1.8 ● Uzbekistan 2.4; others n.a. ● 104 ● 760 n.a. Zero prospects� 151–158 � $170 � �.010

Notes: TPI � EBRD transition indicator 2004; the corruption index is taken from the latest values from the 2004 listing on the Transparency Internationalwebsite; the media-freedom rating is from Reporters Without Borders (2003); Economic Recovery lists the index of GDP where 1989 � 100; and Table 3.1; FDIis from Table 2.3, HDI � human development index, as described in Chapter 3, with the second value being the change in HDI from 1990 to 2001; the statecapture index is as in Chapter 6.


(but not Baltic) countries, former socialist-communist parties returned but,surprisingly, most of these second governments modified the policy onlymarginally if at all. The reasons could be many but include most importantlythe following: more populist parties were as convinced of the need forchange as the first governments and were only electioneering to get intopower; all parties had the same vision of the EU membership goal, whichmeant necessarily steady progress towards the market. These two motivationsare impossible to disentangle and both are part of a strong commitment tomarket reforms and liberal democracy, the ‘dual liberal vision’.

Another aspect of this is that the countries which delayed their economicreforms and moved slowly – countries south and east of those Przeworskianalysed – did not do so in concordance with his implied sequencing, that isfor the sake of first establishing democracies and then proceeding toeconomic liberalization, but to a large extent as a delaying tactic to allow theformer nomenklatura time to reorient itself to the new game of capitalism.The end-result was slow progress on both fronts or, worse, development ofoligarch regimes in which a short period of democratic progress was followedfrom the mid-1990s by reversal toward autocracy. This phenomenon isconsidered further below under the rubric of state capture.

Consider the second major hypothesis on the effect of rapid reforms uponsocial well-being. Recall the intense debate discussed in Chapter 2 on thespeed and intensity of reforms. Those proposing a gradual and one-step-at-a-time approach were primarily motivated by concern for the social costs tothe population that would result from a sharp adjustment in the highly inef-ficient socialist enterprises. The prediction was that greater suffering wouldcome from more rapid reforms. In fact, the evidence does not support theprediction and is strongly suggestive of the very opposite proposition:the earlier and faster the reforms were imposed, the smaller the costs andthe faster the recovery of well-being. It is certainly the case that CentralEuropean and Baltic countries which moved most rapidly not only experi-enced minimal social costs, but even these were more than recouped by theend of the decade.

There are two alternative interpretations of the data that may still supportthe original gradualist argument. First, some argue it is not the level of TPIachievement today that defines rapid v. gradual reformers, but the intensity ofthe initial ‘shock’ of stabilization. Thus, countries like Hungary and Sloveniamight be categorized by some as gradual reformers, while Russia and KyrgyzRepublic as big-bang cases. This is somewhat of a red-herring; true, as Chapter 5has demonstrated the first two did not leap forward in the first years, while thesecond two indeed did see an initial leap. But this leap was soon aborted andthey moved to a stalled and even reversed trajectory of the TPI index, whileHungary and Slovenia never wavered. Why did they not jump quite as fast asdid Poland, Czechoslovakia and the Baltics? Because they started so muchfarther ahead in 1989 there was less need for an immediate big-bang effort.

The second interpretation of data favourable to the gradualist position hasbeen that within the CIS, the three countries that moved much more slowly –Belarus, Uzbekistan and Turkmenistan – have maintained their social well-being better than the CISM with faster reforms. As already noted, updatedinformation of such social indicators casts doubt on such an interpretation;these countries may have fared a little better on such measures as life-expectancy and income distribution than did the extreme cases in the CISMgroup like Russia, but they did so at the expense of no progress on transfor-mation, simply a postponement. Furthermore, they did no better than theabove average CISM performers.

One last conclusion on outcomes merits emphasis in this summing-upconcerning EU membership status and state capture. EU status is shown hereon three levels: attained membership, on-track for membership, and verylow prospects for future membership. State capture is measured by the WorldBank survey described in earlier chapters. It is evident that the ranking ofprogress in transition is much higher the higher had been the early prospectsin the quest for membership. Similarly, the greater the progress on transitionthe lower the degree of state capture. What does this imply about cause andeffect? For the previous measures, the broad implication was that the fasterthe progress and the higher the achievement on transformation of theeconomy and the polity, the better the performance on economic and socialindicators. In the case of EU and state-capture status, the causation is not sosimple; indeed it is the central analytical proposition of this book that thepolicy choice on reform strategy, EU membership goals and prospects, andthe risk of state capture by rent-seekers, are interrelated in a circular causationwhich can lead either to a virtuous circle or a vicious one. This issue isaddressed in the next section.

But first consider again the question ‘is transition over?’ in somewhatgreater detail. For Central Europe and the Baltics transition is essentiallyover, save for a few minor refinements in legal and institutional regulationsthat were not fully completed at the time of EU accession, but were allowedshort delays. Of this group, only Croatia is not yet an EU member, but bymid-2004 a Europe Agreement was signed, and accession negotiations arelikely to start soon and proceed quickly. By all indications, it must now bethought of as a middle-income market economy rather than a transitionregime in most regards. A small exception is on privatization where it isslightly behind the others in this group: Croatia has about 65 per centprivate-sector share in GDP, the others are typically at 75–80 per cent or evenmore. A notable exception on the opposite end is Estonia, where eveninfrastructure like public transportation and utilities tends to be moreprivatized than in Western Europe.

South-East Europe has two sub-groups. Bulgaria and Romania are far aheadof the others, and the only ones in this region to have essentially completednegotiations of Acquis chapters. Of the two, Bulgaria has been distinctly

260 A Summing Up


more dynamic in its reforms since the financial crises of 1996–97. Allindicators of progress show that by 2003–04 it compared well with thecountries at the lower end of the Central Europe–Baltic group, whileRomania has been moving forward at a more deliberate pace. Nevertheless,the pull of EU-accession negotiations, which by early 2005 were virtuallycomplete for both, provides optimism for a rapid catching-up in both coun-tries. The others in SEE have fallen behind for various reasons, includingcontinued civil instability, and with less certain vaguer prospects for EUmembership their progress is less assured. Macedonia may be particularlyunfortunate in this regard, as the spillover effects of the Yugoslav Federationbreakup and the Kosovo conflict have greatly impeded economic reforms.On the EBRD index, it was well-ahead of both Bulgaria and Romania untilthe mid-1990s (reflecting its advanced start as part of the YugoslavFederation), but by 2003 despite some modest progress fell far behind bothof the leaders in this region. But even these countries have at least a positivealbeit vague indication that membership may be possible.

Nine of the CIS countries are well-advanced on market reforms, but never-theless they are far from the degree of market, and even more so democraticstatus, one might consider as a completed transition. While a form ofcapitalism thrives in these countries with private-sector GDP shares at least55 per cent and in many cases 65–70 per cent, the market economy is not anopen competitive one, but one dominated by very large often monopolisticentities and a very low share of small and medium-sized enterprise activity.Market and legal institutions are far from conducive to new small entrepre-neurs challenging the giants. Evenhandedness and transparency of regula-tions and legal actions are overshadowed by insider influence-peddling andrent-seeking behaviour of large politically connected entities. This state ofaffairs has come to be labelled as a regime of economic oligarchs who havecaptured the state, not only profiting from insider privileges, but largelycontrolling its economic policies and political nature. Paradoxically, one canat the same time say transition is far from complete in these countries, butalso that it is for all practical purposes ‘over’ because it has reached a political-economy equilibrium that is unlikely to allow further liberalization in thenear future.

Finally, there is the small group of three CISL countries – Belarus,Turkmenistan and Uzbekistan – where transition has barely started, andwhere the economic regime has not changed much from the Soviet period.In all cases the degree of privatization remains very low with a share of GDPat 25 per cent in the first two and 45 per cent in the third. As of 2003–04 allscored extremely low on the EBRD index of market development, thoughthere were differences in the time-path of market reforms. Turkmenistan hastaken very few steps towards a market economy since independence. Belarusmade a modest leap forward until Lukashenka became President in 1996,and reversed these achievements over the next four years. Some modest

liberalization, partial privatization and macroeconomic stabilization tookplace since 2000, but even that fell short of the progress marked in the pre-Lukashenka years. Uzbekistan’s initial reforms were even bolder and withinthe context of its ‘Industrial Policy’ strategy were until 1996 comparable tothose of some modest but gradual reformers like Ukraine and Kazakhstan.But while the latter subsequently accelerated their transformation efforts,Uzbekistan’s leadership began to worry about losing control in a more priva-tized and free market economy and slowly reversed many of the liberalizingmeasures. In these three countries transition is best described as barelybegun.

A useful way to sum up the outcome of transformation across the region isto describe four regime types now prevailing in the post-communistcountries. These four coincide closely but not precisely with the five groupsused throughout this book. While there is a clear geographic bunching, theunderlying criteria for these groupings are far more important than geogra-phy: the extent of liberality in both economic and political dimensions.

Liberal societies

At the more successful end of the transformation spectrum, one finds at leasteight or more countries that have steadily developed a functioning marketeconomy with open entry and considerable competition, and in parallel anincreasingly liberal and transparent democracy. They have largely avoidedthe development of an oligarchic concentration of power strong enough toinfluence state policy and election outcomes, although they were like allmarket economies subject to lobbying by economic interests seeking protec-tion or privilege. They comprise at a minimum all three Baltic states, theCzech Republic, Hungary, Poland and Slovenia, and by now Slovakia andCroatia. The last two for some period in the 1990s had been subject to somecaptured-state tendencies, but these are now being overcome. Bulgaria andRomania also experienced this intermediate phase, but have begun, albeitmore slowly, to move in the direction of liberal economies and polities. Forall of these countries, the degree of advancement to a liberal society is alsoreflected in better performance of most economic and social indicators. Alsosignificant is the parallel fact of eight countries having achieved EU mem-bership in the May 2004 first wave, with two to three others (Bulgaria,Romania and perhaps Croatia) on-track for the next wave.

Captured states

At the less successful end of the spectrum are several countries with a highdegree of concentrated ownership in a few hands; that is, a very powerful‘economic oligarchy’ which has captured state policy and has overwhelminginfluence on election campaigns and their outcome. Russia, Kazakhstan,Azerbaijan, Moldova and Armenia are in varying degrees in this category, as

262 A Summing Up


are for now Georgia, Ukraine and Kyrgyz Republic, though they may bemoving away from this status since their revolutions of colour. Serbia wascertainly in this category during the Milosevic years (but not Montenegro asit was able to conduct an independent and generally very liberal economicpolicy, even within the ‘union’), but now appears to have moved at least intoan intermediate group (described below). Tajikistan has not yet undergoneenough evolution to be easily classified and may be in this group or theintermediate group. Economic and social indicators in this group, while notnecessarily the worst among transition countries, have until recentlybeen on the lower end of the scale. The recent surge in economic growth hasbeen impressive, but given the low points reached during a decade longrecession, these countries are still far from catching up to the leaders in tran-sition. On democratic institutions, most of these countries reached a peak inthe mid-1990s and have seen some reversal to non-democratic norms.

Intermediate regimes

Countries with an intermediate situation still have a considerable degree ofrent-seeking and oligarchic tendencies, but generally lower than in the cap-tured states, and in several cases this has receded in recent years. Albania,Bosnia-Herzegovina, Macedonia and perhaps Serbia are in this category, anduntil recently so were marginally Croatia, Bulgaria, Romania and evenSlovakia for a short period of three–four years under the Meciar government.Their performance on economic and social indicators has also been in themiddle range of the scale. Also very important is the apparent growing seri-ousness of interest in eventual EU membership, which would appear to bethe main force pushing them away from the captured state type and towardsthe liberal society type.

Lagging reformers

A small group of three countries have barely moved forward from the pre-ceding communist regime; market transformation has not gone far enough toresult in any significant change in the role of open markets or even in elitepower structure. Belarus, Turkmenistan and Uzbekistan are in this group,with the original communist nomenklatura and its role little changed, exceptfor the name and its taking advantage of the spirit of the day by becoming acapital-owning class in the small margin where privatization has occurred.But all these changes are limited and the role of state guidance in the econ-omy remains dominant. Official data on economic and social indicators sug-gests somewhat better performance than that of the captured-states group,but there remains a lot of controversy on these data. Some suggest these areoverstated, others argue they confirm the gradualist hypothesis that one canavoid the worst of the social costs by going more slowly, still others say theysimply reflect the postponement of eventually inevitable transition costs.

Why such divergent outcomes?

The central analytical thesis of this book is that the diverse transformationoutcomes described above do not need elaborate explanations, but can belargely understood in the ‘navigation paradigm’ framework comprising threekey factors or determinants of transition progress. The three factors whichcan be considered as proximate causes of the transformation outcome are:

● The length and intensity of debate on the ‘navigation charts’; that is thereform programmes. The debate has most often been between proponentsof a rapid, comprehensive approach and those advocating a gradual andpiecemeal approach.

● The vulnerability of the government to ‘pirate raids’ on state assets andprivileges by rent-seeking interests, and the subsequent development ofoligarchs who capture the state.

● The availability and prospect of a ‘safe haven’ such as the EU, or otherorganizational membership which disciplines the process to ensure anon-oligarchic outcome.

The three factors are interdependent and a formal specification would com-prise a set of simultaneous equations, including some exogenous variablesreflecting different historical experiences and starting conditions. One obvi-ous candidate for an exogenous variable is the prospect of EU membership,but I have argued it is not so simple; the exogenous offer of membership waslimited to at most two or three countries, far fewer than the eight that havealready become members and the additional three or more with very highprospects of membership before 2010.

The navigation paradigm developed in this book aims to explain thedifferential progress in transition of different countries, and not necessarilythe resulting success in terms of economic growth, democratization or otherperformance measures. However, as Chapters 2 and 3 have amply demon-strated, there are in fact very strong statistical correlations between the TPIand various other measures of success, so in that sense the navigationparadigm can be a building block for an explanation of performance differ-ences as well.

Table 9.2 summarizes in qualitative terms the ways in which these threeexplanatory factors played out to give the outcome-types observed for dif-ferent countries: the ‘lagging reformers’ type is excluded since it reflectsessentially no significant change.

The model’s line of reasoning can be restated briefly as follows. The longerand more intense the debates on how to transform the economy into a market-oriented one, the greater the opportunities for old and new vested intereststo lobby the government for the rents available in a partially reformed

264 A Summing Up

265Table 9.2 Final transformation outcomes and principal determinants

Outcome Reform debates Proclivity to rent-seeking Eu beacontypes*

Liberal ● Almost all rapid reformers, ● Most had first government ● Visegrad four, Bulgaria and societies either advanced start and non-comm. or coalition, or Romania had early ‘invitations’

early progress or sustained reformed communist ● Baltics and Croatia had clear and big-bang ● Privatization generally open, strong vision, demand for

● Marginal: Bulgaria early transparent membership and eventually efforts but unsustained; ● Bulgaria and Romania convinced EU through reformRomania always gradual renamed-communist progress

● Slovakia short period of ‘crony’ ● In several cases, EU played ‘stick andprivatization carrot’ game effectively: Slovakia,

● Croatia somewhat greater Bulgaria, Romania best examples degree of insider-privatization

Captured ● All but 2–3 gradual ● Almost all renamed-communist gov. ● None enjoyed a warm signal from EU,states ● Russia, Kyrgyz, Moldova ● Privatization generally oriented and but own desire ranged from very low to

(perhaps Armenia) cases of non-transparent lukewarmaborted big-bang ● No willingness by EU to try ‘stick and

carrot’ in serious way

Intermediate ● Albania and Macedonia ● First governments generally renamed- ● All had long-term vision of EUregimes aborted big-bang communist. Albania exception, very membership, but no expectations of

● Others very late and mixed but lots of political instability early Accessiongradual ● Privatization very mixed, but with ● EU disinterest less categorical than for

exception of Albania late and not CIS countries, and by 2003 slightlyalways transparent positive

* The categorization of outcome types is as given in Chapter 9.

economy, and also to concentrate the transfer of state assets in a few hands;the more concentrated the new ownership the slower will be subsequentstages of liberalization and the less likely the development of new entrepre-neurial activity. This is because of the vested-interest principle that capital-ists prefer less competition, not more. They also prefer less transparency asthis best ensures continued high profits from rent-seeking and state capture.Similarly, oligarchs may be willing to trade-off the security of property rightsthat good rule-of-law (ROL) provides against the benefits of rent-seeking. Ina word, oligarch capitalists behave optimally when they oppose liberalizationand democratization.

The degree to which rent-seeking interests can be successful in their effortsto influence government policies varies, however, hence not all the post-communist states experience the extreme degrees of lobbying that result indominant state capture including the ability to determine results of nominally-free elections. One of the factors working against vested interests is thestrength of desire and prospects for EU membership, since this requiresautomatic implementation of many liberalization measures that limit theopportunities for rent-seeking. But other historical factors may play a rolehere, such as a radical change in government towards one that has a strongmarket-commitment, or a powerful and charismatic leader who leads such adirectional change in the society. This is the most plausible explanation forthe early big-bang efforts in Russia and Kyrgyzstan, but the reason for theirbeing aborted and unsustained are more difficult to pin down; in Russia itwas partly because Yeltsin’s commitment was a ‘heat of the moment’phenomenon, partly because the powers of non-reformers were overwhelm-ing. The latter factor may be the best simple way to explain the eventual stalland reversal in Kyrgyzstan, and its slippage into a captured-state mode.

A commitment to some outside organization or institution, such as the EUor others, can discipline the process of economic as well as political liberal-ization, helping to curtail debate and get started quickly, and helping toreduce the power of vested interests. In practice, EU membership was themajor such commitment to have had significant impact, though in principlemembership in NATO, the WTO, the IMF, the World Bank or the EBRD allprovide vehicles for disciplining reform paths and all had some influence. Itis important to note that the discipline came not only when there was an EU‘invitation’ to join, but equally importantly from the intense desire of acountry for membership, driving it to demonstrate an ability to do so evenwithout an invitation. A strong ‘return to Europe’ commitment may havebeen the way this effect worked in the Baltics, leading to a rapid rise in theTPI which helped overcome the initial reluctance by the EU to include themin early discussions.

The interplay of these three factors involves a circular causation startingwith the delay of reforms that may begin a vicious circle, or a rapid reformwhich leads to a virtuous circle. In the case of delays they may be partly due

266 A Summing Up


to low EU membership prospects or desire, but, regardless of the reason,delay creates greater opportunity for rent-seeking, allowing rapid accumula-tion of wealth especially by insiders, and a solidification of such vested inter-ests into an oligarch clique. Their desire to maintain a status quo in turnleads back to limited economic reforms and an enhancement of the power ofoligarchs. The views held by oligarchs on the policy orientation of a countrywill not of course be openly expressed, with one possible exception, reveal-ing their humanity if not humanness: they have often recognized publiclythe extent of their political power.2 But this is not news in the history of theinteractions between powerful capitalists and society, nor is it by any meansa monopoly of Soviet-formed personalities, rather it is the inherent nature ofcapitalists. As Rajan and Zingales (2003b) show, even in the country wherethe degree of openness and competition in financial markets was relativelyhighest, the United States, the 1934 Glass–Steagall Act which did so much tocut back competition in this sector was played up publicly as having preciselythe opposite purpose to curb ‘abuses in the financial system’ (p. 222).

How did these three proximate causes lead to the four regime types notedabove? It is easiest to deal with the last type separately and then focus on theother three. In the three lagging reformers, reform efforts from the begin-ning were at best very limited and even these were reversed later in thedecade. With the exception of a very short period in Belarus, there was vir-tually no change in the ruling power structure, other than the renaming ofthe communist party; the commitment to deep market and democraticreforms was extremely weak, instead their leaders were more inclined to verymodest reforms and some self-serving actions to preserve and enhance theirpositions. There was, furthermore, neither the prospect of EU membershipoffers nor a strong desire for such ties. All other international membershipswere either unavailable or had very limited influence to push in the direc-tion of more liberalization. None of these countries faced such a dramaticeconomic crisis that they were forced to seek external support from IFIs,hence even that form of leverage on policies was mostly absent.

For the other three types, the interplay of the three main factors – plusothers in certain cases – was not quite so simple, but nevertheless the forcesare evident; Table 9.2 summarizes the cause and effect dynamics until theonset of the colour revolutions in 2004, which are treated separately later.Consider first the Liberal Societies. All but two of the countries of this typeundertook early, rapid and comprehensive reforms, either with a big-bangeffort, like Czechoslovakia, Poland and the Baltics, or a more moderate pacebut from a higher initial level as in Hungary, Slovenia and Croatia (thoughin the last the progress on democratic institutions was much slower until2000). The quick start on liberalization precluded a lengthy period of dis-torted subsidies, pricing and administrative regulations and thus preventedany rent-seeking activities from attaining the magnitudes of the capturedstates. Privatization was not neat and transparent in all cases, with insider

preferences evident to some degree in Croatia, the Czech Republic and par-ticularly Slovakia for a few years in the middle of the decade. Nevertheless,in all cases the degree of liberality and competitiveness in the markets washigh enough to limit the damages of insider problems, and eventually cor-rections were made in the privatization process itself.

The other two factors played an important role in curtailing the period ofdebate on strategy. In most cases, the proclivity to insider privileges was lowas first governments tended to be very reform-minded non-communists orformer communists who had probably become converted to the dual liberalvision even before 1989. This reflected a strong commitment in society for aradical turn away from communism and to markets and democracy, and a‘return to Europe’. This alone was enough to ensure a rapid and transparentprocess of transformation. For several of these countries a clear signal fromthe EU was provided early on that they were welcome to apply for member-ship, and this helped cement the commitment. The Baltics did not enjoythis warm embrace immediately, but their own commitment and intensedesire for membership drove their transformation so effectively that by 1995they, too, were on the formal path to membership. These different pathsshow that the discipline of EU membership could come either from an offerby the EU, or a strong demand by the country even without an offer.

But sometimes even a clear offer of eventual membership was insufficientto ensure reform discipline, as exemplified by Bulgaria and Romania. Theycan only be included in the liberal society category in recent years and thisis still a marginal judgment. In the early years they were at best in theintermediate regime group, with tendencies to state capture. An abortedmini-bang for Bulgaria and a clearly gradualist reform strategy in Romaniathroughout kept them on a volatile path of transformation. Both were in for-mal discussions for the first steps towards membership, about as early as theVisegrad four, but the continuation of governments, which are best catego-rized as renamed communists with some democratic influences, meant theircommitment to the dual liberal vision was far weaker. Consequently the EUput them on hold for many years. Until a second financial crisis in 1996–97,they were moving more in the direction of developing rent-seeking, oligarchregimes. But, since then, a turn to more democratic governments meansboth have become much more strongly committed to the EU vision, andmore willing to accept the discipline of the Acquis. This has been especiallyso in Bulgaria with the victory of the Union of Democratic Forces in the 1997election.

None of the three factors by itself suffices to explain the path of thesecountries, and in all cases the prior history cannot be ignored as it is oftenpart of the underlying explanation for the initial choice of reform strategy.The variety of country experiences even within this group also demonstratesthat the pull of EU membership is not a simple deus ex machina force. Thelack of an early offer to the Baltic states did not reduce their zeal for a liberal

268 A Summing Up


strategy, indeed it may even have strengthened their resolve to show thatthey were as good as the Visegrad four. On the other extreme, the offer toBulgaria and Romania was not enough to overcome the lack of internal com-mitment and consensus until much later. Even for some of the others withmuch greater internal commitment, the EU had to play a game of carrot andstick to maintain the disciplining effect. This was the case for Slovakia in theperiod 1994–98, Latvia and Lithuania who had been left behind Estoniauntil the late 1990s, and later Croatia.

In the case of the Captured States, the storyline is close to a mirror image ofthe liberal societies. If one begins with the second factor, proclivity to rent-seeking, it is clear that even in CIS countries which have eventually achieveda moderate degree of reform progress, the leadership and perhaps evensociety did not have the same degree of commitment to the dual liberalvision. Almost all the new governments comprised in the majority renamedcommunist parties, whose members could not be compared to the reformedcommunists in Hungary and Slovenia, but instead sought to maintain theirprivileged position either by preventing radical reforms or by taking advantageof opportunities to become the new capitalist class. The latter force generallyprevailed and led to the choice of a strategy of long-delayed stabilization,gradual structural reforms, and non-transparent privatization favouringpolitical insiders. This created a perfect environment for huge rents, low-costprivatization by the few, and the evolution of what is now widely recognizedas an oligarchic regime and a captured state.

There were some temporary exceptions to this where an early big-bangeffort was attempted, but in all these cases it was either soon aborted (Russia),or insufficiently sustained to prevent reversal (Armenia, Kyrgyz Republic,Moldova). In Russia, the tendency to the oligarchy direction was not onlydue to insufficient stabilization and liberalization in the 1992 big-bang toclose the major opportunities for rent-seeking, it was further exacerbated byan explicit privatization strategy of coopting the powerful opponents ofreform by giving them privileged access to state assets. The Loans-for-Sharearrangement in the mid-1990s was perhaps not intended by its reformistarchitects to create a group of oligarchs, but its aim to create a capitalist classthat would favour further liberalization and rule-of-law to protect their newproperty backfired somewhat; an oligarch class that is inimical to liberalityand transparency appears to be the result.

EU membership as a factor in these countries was essentially SherlockHolmes’ dog that didn’t bark. There was certainly no signal from the EUitself of an invitation or potential membership even in the long term, andvery few countries made expressions of membership desire at the beginning.Those that did were clearly ambivalent and not sincere enough to acceptthat membership could only come on the EU’s terms. Thus, for example,Ukraine’s authorities began after the mid-1990s to express such interest, butthis was not backed up by consistent political or economic demarches.

By this time, the grip of oligarchs whose interests are not best-served by theliberalization demands of the Acquis process steered policy, and while it waseasy enough to express the desire, it was much harder to follow through withthe sort of demonstration effort that the Baltics had undertaken in the early1990s. It will remain a question for academics whether a bolder initiative bythe EU to use not only the stick of refusal to Ukraine, but also the carrot ofan explicit (always conditional) offer – as was used in Slovakia and Romania –would have enticed Ukraine to much more substantial liberalization on botheconomic and political dimensions.

In the case of the Intermediate Regimes (where until the late 1990s one wouldinclude Bulgaria, Romania and perhaps Croatia) the three principal determi-nants of the navigation paradigm not surprisingly entail a mixed picture.Some countries undertook early and rapid reform efforts, but these were notsustained; Bulgaria soon slowed its structural reforms and, more important,allowed the stabilization to fail by continuing large subsidies to state firms;Albania suffered a financial confidence crisis because of the inability to prop-erly regulate pyramid-scheme activities. All the other countries (within South-East Europe) proceeded at a very gradual pace, partly due to civil conflict andpolitical instability in the territory of the former Yugoslav Federation. Thenature of internal commitment and its relation to the nature of the first post-communist governments was also mixed. Albania was racked by continuedstruggles among political groupings from all parts of the spectrum. Bulgariaand Romania saw a dominance of the old power structures, but not withoutsignificant roles for new, more democratic coalitions. The former YugoslavRepublic also had considerable mixtures, starting with a continuation of theformer socialist political elites, but moving to more mixed governments.

Finally, the role of EU membership was also mixed. As noted, Bulgaria andRomania were in the very early group for informal discussion, while theothers may have had some distant hope that their geographic proximitywould put them in line for eventual membership but realism led to officialsilence on the matter. The EU itself was never encouraging until the mostrecent declarations of 2002–03 hinting at the possibility; nevertheless, it wasalso not as explicit in discouraging the idea as it was in the case of CIScountries farther east.

In sum, none of these three determinants alone explains the final outcomeone sees for each country in 2004, but together they tell a story that is rela-tively uncomplicated but still compelling. In the body of the text, severaltestable hypotheses are derived from this navigation paradigm, and wherepossible simple statistical tests are undertaken. Some of them give verystrong results, others are weaker as measuring a variable such as prospect ofEU membership is difficult, but in all cases the results are consistent withthese hypotheses. Thus, the extent of state capture by 1999 as measuredby the World Bank is significantly correlated with the extent of delay in start-ing the stabilization and reform process. When EU membership prospects

270 A Summing Up


are measured either as the strength of the offer from the EU side or thestrength of the desire by the country itself, the higher this value the shorterthe period of delay in reforms. Most important for future implications is thefollowing statistical result: the higher the degree of state capture in 1999, thelower is the level of progress to the market as measured by the EBRD transi-tion index; this is consistent with the hypothesis that in captured states tran-sition is frozen at an equilibrium part-way to a true market economy.

Two additional observations should be made about this paradigm. Aneven simpler paradigm has been put forth by some with the claim that isoverwhelmingly powerful as an explanation: geographic proximity toWestern Europe. In a superficial statistical sense any good measure of geo-graphic proximity is indeed a very powerful explanation of transitionprogress. But apart from the many exceptions to this that have been elabo-rated in the book, geography fails to leave any room for policy choice effec-tively implying that the closer to Europe the more a country will choose theright, progressive reform policy. It also fails to elaborate the behaviouralmechanisms that lead to such results, surely a necessary element in anysocial science analysis purporting to explain different outcomes in differentcountries. Geography as an explanation is too deterministic and rigid in itspath-dependency storyline. Path-determinacy is a powerful and generallycompelling analytical paradigm, but surely there are moments in the historyof a country when a new environment presents itself allowing for a sharpdeparture from the forces of the past, and surely the downfall of the com-munist empire and its ideological vision in 1989 was just such a moment.3

In many countries, elites seized the moment to undertake radical changes:Poland, Czechoslovakia, the Baltics, even for a short time Russia andKyrgyzstan, as well as others. That fact alone strongly suggests that in coun-tries where elites did not opt for radical change did so by choice and not dueto the rigidity of geographic pre-determinacy. This is not to say that geogra-phy and history cannot be used to explain why they chose non-radicalstrategies, but the fact remains they had the choice to do differently.

Consider just one illustration: a comparison of Czechoslovakia andUkraine, both with a nomenklatura that was ideologically untainted byreformist ideas like those in Hungary or Poland. Prior to the regime change,the powerful democracy movements of the late-1980s that brought massesof people on the streets were in both countries composed of non-communistdissidents, intellectuals with a limited number of technical-professionaltypes. In Czechoslovakia, this loose movement did not hesitate to takepower from the communists when the latter, in disarray, were losing control.In Ukraine, the Rukh forces were uncertain of their ability to run a govern-ment, and were willing to compromise with the nomenklatura led byKravchuk in what has been called the Faustian Bargain, with the latterpromising to support independence and the former agreeing that economicreforms were secondary and could be postponed.

As a final observation on the relationship between strategy chosen andfinal outcome, it is notable that in practice none of the countries that chosea gradual strategy followed closely the theoretical recommendations of firstestablishing good market institutions and then liberalizing the economy. Infact, it is easily seen in the time-path of the transition index componentsthat those countries which moved most slowly on liberalization elementsmoved even more slowly on the institution-building elements. This impliesthat the proclamations of the political elites in these countries, that thesociety was not ready for the market and a gradual evolution was necessary,were not sincere proclamations but masked a hidden agenda. It is not hardto conclude what that was: retain and enhance personal positions of power.It is in that sense that the big-bang strategy, while inferior theoretically asdemonstrated by many mathematical models, was in practice superior – itdid not as easily allow for the duplicity of incumbents pursuing a hiddenagenda in the name of the social good. There was no theoretical error in themodels of gradualism, there was an error in the assumption that self-interested politicians would implement its recommendations faithfully andnot abuse the science. Was this erroneous assumption due to the tendency ofgradualist arguments to be formalized mathematically? Not at all. A morerecent generation of mathematical transition models applied to capturedstates include self-interest of ‘oligarchs’ as part of the mathematics, and sen-sibly conclude that these oligarchs may very well choose the high-rent statusquo to a more liberalized economy which will ensure the security of theirproperty rights. That is to say, the abstraction of mathematics does notpreclude reaching the same conclusion as a practical interpretation.

Future prospects and policy implications

The 27 post-communist countries analysed in this volume were first groupedinto five categories using the TPI measure as a proxy of transition progress,this was then collapsed into four political-economy regime types observablein 2004. For a look forward, three groups will suffice to discuss generaltendencies: the first group comprises Central Europe and the Baltics, wheretransition is effectively over, plus most of South-East Europe where it is notbut the future path for the next 5–10 years is almost certain to be determinedby integration into the EU; a second group comprises the three laggingreformers where transition has barely begun; and the third comprises thecaptured states of the CIS. Policy implications for the future are simplest andeasiest to derive in the first two groups, but much more complex andincidentally more interesting in the third group.

For the countries now tied by membership or advanced accession negotia-tions to the EU, future policy is less related to transition from socialism tothe market and more related to the problems typical of middle-incomeemerging markets, with perhaps some institutional inadequacies still to

272 A Summing Up


be resolved. These are more the subject of development economics, publicfinances, monetary and exchange rate policies. There are some additionalspecificities related to EU membership, such as the optimal timing of euro-adoption, but these too are not transition issues any longer. For countriesless advanced in the transformation but highly likely to be put on a track toEU membership, the relevant transition issues (liberalization, privatization,institutional development) will be guided by the requisites of the negotia-tions of Acquis chapters, with limited room for deviation except perhaps ontiming. The analysis of this book bears two simple lessons for them:structural liberalization was done most successfully by those that movedearly and fast; and privatization gave the best results when it was open andtransparent. Direct sales under close supervision of committed reform gov-ernments has been more successful than other methods, but only because itis easier to ensure transparency under this method.

Once the laggards which are much closer to a socialist planned regimethan to a market economy are in a position to begin significant transforma-tions, the experiences of the other 24 countries can provide many usefulimplications. First, while it is important to develop good institutions, waitingfor these to be in place is costly, and risks yet again creating the rent-seekingenvironment that has been so harmful in the region. It is important to movefast enough on liberalization to close up the various loopholes of price dis-tortions and government regulations to minimize the rent opportunities.Big-bang, when sustained, has worked far better than gradual liberalization.Even the fears that it would cause greater social costs for the population haveproven unfounded, though of course separate mechanisms to compensatethose displaced by the inevitable job losses should be provided as well.Second, the process of large-scale privatization can come early or late, butmost important is that it be an open and transparent process that precludesthose on the inside and/or well-connected individuals to access large stateassets at bargain-basement prices. Such insider phenomena hurt theprospects of transition in two ways: they cause huge resentment in the pop-ulation and erode popular consensus for market reforms, and they may leadto the creation of powerful financial interests who can capture the policies ofthe state. Third, an underestimated aspect of a good transformation is thecreation of a market and institutional environment that encourages newsmall business. This requires dramatic simplification of new business licens-ing, taxation and regulation; it is not enough to put in place the simplifiedlegislation as some room for bureaucratic discretion will always remain – taxinspectors do have a legitimate purpose in the most open of economies –there will need to be sincere efforts to deal with bribery as well.Demonstration from the top has been the most powerful tool in this regard.

The captured states pose the greatest dilemma for future policy. To beginwith recommending policy to members of a government that is captured bythe financial benefits that ‘oligarchs’ bestow upon them may be about as

effective as showing the fox alternative routes out of the chicken-coop. Atthe extreme, captured governments and their oligarch captors are indistin-guishable. It is conceivable that some of these individuals will begin to desirelegitimation and historical glory, and will be inclined to accept advice onhow to overcome the power of oligarchs.

In principle, it is reasonable to argue that once opportunities for acquiringstate assets are exhausted, oligarchs will want to seek ways of protecting theirproperty rights. There is a powerful argument made by some analysts thatthis point will inevitably be reached and the political demands of the oli-garchs themselves will lead to liberalization and reduction of their power.But this ‘transition inevitable’ view is not consistent with the historical evi-dence of the behaviour of powerful vested interests, nor with the evolutionreached in captured states where opportunities for huge rents from existingassets will remain even after all privatization is done. The theory thatoligarchs want protection of property rights is correct, but it has beenamply demonstrated that these property rights can be purchased individu-ally and that their value may be far less than the rents to be gained fromnon-transparent lobbying. While there may indeed come a time that the oli-garchs become proponents of more open markets, in practice this may bevery far for the CIS countries finding themselves so entrapped; their transi-tion is likely to be stalled or frozen for the near future, though a radicalchange in government as in Georgia, Ukraine and perhaps Kyrgyz Republicmay provide a new liberalization opportunity.

But few, if any, of the countries are in this extreme form of state capture,so there is room for seeking out policy-makers and well-meaning oligarchswho may be able to implement policies that slowly undermine the power ofvested interests. Where a revolution of colour has occurred, and a moredemocratically-elected leadership has come into power, it is reasonable toassume that a new commitment to liberalism exists and policy-makers con-front the issue of reversing state capture directly. There are three types ofpossible policy actions to reverse the capture of the state:

● Reverse privatization and reduce the power of oligarchs.● Restrict the power of oligarchs by anti-monopoly regulations and elimi-

nation of any subsidies, and licences.● Facilitate development of new and small–medium-sized enterprises as a

political counterweight to the large economic entities.

The first of these is the most risky and the least advisable of the measures. Itis likely to be extremely popular politically in many of these countrieswhere, with some justice, the oligarchs are seen as nothing more nor lessthan thieves. But unless there is an extremely clear case of illegal privatiza-tion, this approach is likely in the long run to do more harm than good. Butsuch cases are few, given the general murkiness that pervaded this process,

274 A Summing Up


hence the overall impact of clean reversals may be very limited. Taking awayfrom one oligarch risks being a simply political rather than an objectiveaction, and sends the signal that the government is always more powerfulthan the private sector, exactly the spirit that transformation meant to over-come in this region. Even small entrepreneurs will read this less as renderingof justice and more as confirmation of the omnipotence of the state. Thegreatest danger is to have the reversal flawed by a re-privatization to anotherfavoured insider.

In sum, to undo some of the worst cases of insider privatization effectively,several conditions must be met. First, the number of cases must be very lim-ited. Second, they have to be selected very judiciously, focusing on thosethat are most egregious, and that have the greatest chance of being a rever-sal by an open judicial process. And, third, one must avoid an outcomewhere powerful people close to the new government are seen to benefit fromthe action.

Restricting the power of oligarchs by anti-monopoly regulations is not asimple matter of having the legislation and institutions; in fact, these arenow commonplace in the relevant countries. The problem is the classicalone that powerful monopolists influence or even ‘buy-out’ the regulatorsand the dilemma becomes ‘quis custodiet custodies’, who regulates the regula-tors. This, as well as elimination of programmes that provide subsidies, gov-ernment procurement favours and licences, means the ability to resist theinfluence of the oligarchs. If there is some will by politicians to do so, evenmarginally, that is all to the good, though it may not amount to much for avery long time and any sincere efforts of this sort may not be easily distin-guishable from political demonstrations of anti-corruption efforts. In short,governments with some degree of independence and desire to work for thegeneral good should go as far as they can in effective and sincere restrictionsof oligarch power, even if it is slow-going.

The third policy, facilitating the small business sector, will also be slow togive effects, but it is the most powerful of all the approaches. The word ‘facil-itating’ rather than ‘promoting’ is used to emphasize that it is not a matterof throwing government resources at small business, but easing the greatbureaucratic and taxation burdens that this sector faces in most of thesecountries. These burdens are the reason the sector is so little developed com-pared even to the newly transformed Central European countries, and itslimited development goes a long way to explain the weakness of civil soci-ety. The importance of having a strong small-enterprise sector lies in the factthat it not only tends to be the most innovative and dynamic part of anyeconomy, but its members are the strongest proponents of transparency,even-handed treatment by the state and democracy – in a word, proper rule-of-law. There is good reason to believe that this sector is slowly growing onits own even with the impediments it faces; but any effort to reduce theseimpediments will speed up the process and create perhaps the most effective

counterweight to the political influence of oligarchs. This approach iscentral to the future of the captured states where transition is stalled,where capitalism thrives only for the select few who function in a non-competitive and restricted market economy. Opening up the markets to newentrants and creating a level playing field of competition between big andsmall will be the most powerful lever to transform these states into realmarket economies with capitalism for all.

As of this writing, three special cases of captured states appear to havemoved sharply to reverse state capture: Georgia with its Rose Revolution,Ukraine with an Orange one, and Kgyrgyz Republic with Tulips. It is far tooearly to assess the prospects for a successful return to the dual liberal vision,beyond the obvious uniqueness of Ukraine: its much larger size and fargreater strategic position result in a much higher probability of being put onthe EU membership track. The evidence of 15 years of post-communismclearly shows how strong a discipline this can be towards felicitous policychange. As a final word, two points will be briefly noted here. First, each ofthese revolutions of colour had some unique aspect, but the overwhelmingcommon factor which drove the tremendous force of people power was theresentment of the populace at the egregious abuse of political and economicpower by the ‘oligarch’ clique. Second, the probabilities of analogous people-power revolutions in neighbouring countries are hard to judge; it is notablethat despite some precedents, like Slovakia 1998 and Serbia 2000, no-oneappears to have imagined – never mind predicted – any of the three recentcases. However, the common element of mass resentment against abuse ofeconomic and political power is surely there aplenty in other countries.Predictions should not be attempted, but pleasant surprises should beawaited.

276 A Summing Up

Notes

277

Introduction and Overview1 I borrow here the East–West distinction of society’s attitudes emphasized by Gaidar

(2003).

1 Key Debates on Transition1 Stiglitz (1999) and Campos and Coricelli (2002) both aver to this problem. A coun-

terargument to the lack of historical precedents is, for example, Dornbusch inZecchini (1991) which draws lessons from the dissolution of the Austro-Hungarianempire, though largely on macro-policy.

2 Kornai (1994) explores the possibility of what he prefers to call a ‘system para-digm’; one key argument he makes is that transition by definition does not need aparadigm or theory – only the beginning and end-point systems do.

3 The issue of minimizing dislocation was addressed theoretically in an early articleof Aghion and Blanchard (1994); this seminal piece spawned a large number ofmathematical models of the dynamics of reallocation, providing at least a theoret-ical basis for determining optimum speed of transition. A typical example isDewatripont and Roland (1992), which, as many of these articles tended to do,concluded gradual change is better, to allow time for new job-creation to absorbthe disemployed. The second section explains some of the debate on this, includ-ing counter-illustrations showing the model can just as easily conclude in favour ofbig-bang: Katz and Owen (2000).

4 A very successful practitioner of ‘shock therapy’, former Prime Minister Maat Laar(2002) of Estonia, chides Western critics for the misguided phrase ‘too much shocktoo little therapy’, and addresses directly their illustration of Russia by describingin detail the very limited reforms and the consequently limited restructuringwhich explains the failure there.

5 I agree fully with Stiglitz (1999) ‘that for a market economy to work well (to bePareto optimal) there must be both competition and private property’. Such a per-spective strongly informs a principal thesis of this volume.

6 The major exceptions which do propose a very different agenda are not by ana-lysts, but country governments like those in the CISL group who are lagging anddo proffer an alternative model: essentially socialism with a minor role for the mar-ket in Belarus and Turkmenistan, or an industrial policy approach as in Uzbekistan.

7 The present author has written two review works, Havrylyshyn and Nsouli (eds)(2001) a conference volume covering transition issues widely, and Havrylyshyn(2001) focusing on econometric studies of growth. These obviously inform myjudgment on what are the key issues, but I have attempted to derive the list objec-tively from the reviews by others as cited.

8 IMF (2000) shows that most transition countries have a far lower index of traderestrictiveness, using an IMF-constructed measure, than most other developingcountries.

9 EBRD (2003) chs 4 and 5 provide an up-to-date review.

10 Johnson, McMillan and Woodruff (2000).11 A question has arisen since 2000: why are the less reformed economies of the CIS

growing much faster than the advanced transition countries in Central Europe?This is considered in Chapter 2.

12 Roland and Verdier (1999), p. 40, while recognizing Central European countrieswhich reformed earliest are now growing again, nevertheless goes on to say thepolicies of the WC ‘led … to unexpected outcomes [such as] the important outputfall [which] was not predicted’.

13 Several chapters in Clague and Reuser (1992) address this; Frydman et al. (1999)give an extensive overview of the arguments, and World Bank (1996) provides anearly retrospective. Later reviews of privatization include Havrylyshyn andMcGettigan (2000), Nellis in Havrylyshyn and Nsouli (2001) and Megginson andNetter (2001).

14 The authors may be overly rhetorical in their claim that this result implies a ‘fail-ing grade’ for the Washington Consensus; even apart from the fact that one of theauthors, Sachs, had earlier been a strong proponent of rapid privatization in somecases – not Russia to his credit – the WC as explained nowhere pretends that it ismerely the transfer of ownership that is needed.

15 I will not cite the numerous recent statements from all corners about the impor-tance of institutions, the need for better ‘second-stage’ reforms (one example isSvejnar, 2002). A useful flavour of the agreement is in the narrower literature ofgrowth determinants and the increasing use of institutional development indicesas explanatory variables, reviewed in Havrylyshyn (2001) and Campos andCoricelli (2002). As two examples, consider Brunetti, Kisunko and Weder (1997)who find reform variables do not matter if institutions are included, v. Havrylyshynand van Rooden (2003) who argue that if properly specified such econometricsyield the result that both reform policies and institution matter. Two recentpapers provide broad-ranging reviews of the role of institutions and attemptto move towards a manageable framework: Djankov and Murrell (2002) andAcemoglu, Johnson and Robinson (2004).

16 Johnson and Kaufmann (2001) analyse how the rise of underground activity isrelated to institutions (ref. is ch. 10 in Havrylyshyn and Nsouli, 2001).

17 This early optimism of the World Bank analysts may be understandable as therealities of oligarch power were not yet evident. The Bank’s latest Country EconomicMemorandum on Russia, World Bank (2004), reflects a much less optimistic view ofthe demand for good law by capitalists.

18 Titling his article ‘Winners Take All’ nicely poignantly evokes this paradox.The sub-title ‘Politics of Partial Reform’ clearly signals his view that it was notrapid but gradual reform that is to blame. Another paradox is the contrastwith the expectations of strong opposition to reforms from ‘almost everyone inthe production hierarchy’: Holzmann (1992).

19 This section has been prepared jointly by the author and Karlo Basta, on the basisof Basta (2004a).

20 Guillermo O’Donnell and Philippe Schmitter, Transitions from Authoritarian Rule:Tentative Conclusions about Uncertain Democracies (Baltimore: Johns HopkinsUniversity Press, 1999).

21 Thomas Carothers, ‘The End of the Transition Paradigm’, in Journal of Democracyvol. 13, no. 1 (January 2002), pp. 5–21.

22 Fish (1999), p. 799.

278 Notes

Notes 279

23 For a review, see Fish (1999), p. 797. See also a review in Pridham (2000), pp. 5–8.However, Pridham calls this approach ‘functionalist’. Furthermore, the fact thathe lumps all theoretical approaches dealing with democratization under therubric of ‘transitology’ reveals the extent of the conceptual confusion in the studyof this phenomenon.

24 Philip Roeder’s chapter in Anderson, Fish, Hanson and Roeder (2001).25 For a brief summary of this view, especially as expounded by Herbert Kitschelt,

and an interesting critique, see Hanson and Kopstein (1997).26 Motyl (2004) also emphasizes this, noting that Croatia, Bulgaria, Romania and

Slovakia, which had indices of democracy much worse than Poland, Hungary andothers, saw a sharp move in the democratization direction, while many CISMcountries moved in the opposite direction approaching the magnitude of themost authoritarian states.

2 Measuring Progress in Transition1 Sachs and Warner (1996) is an early study using relatively simple cross-country

regressions; Fischer, Sahay and Vegh (1996) a more sophisticated one poolingtime-series and cross-country data. Two recent survey articles assess a large num-ber of econometric and other studies of growth in transition: Havrylyshyn (2001)and Campos and Coricelli (2002).

2 Roland and Verdier (1999) is an example of a large theoretical literature on thespeed of reforms, which often concludes that gradualism as in China would havebeen better for the European economies as well. Stiglitz (1999) is a key readingmaking the same case with ex post evidence of actual developments.

3 For a fuller explanation see EBRD Transition Report (2003). I have excluded infra-structure indicators in my calculations as it is not available in all earlier years forcomparison; arguably it may be as much a ‘results’ as an ‘input’ indicator.

4 I use the term CIS, Commonwealth of Independent States, for convenience as it iswidely, albeit improperly, used; formally not all the countries have fully acceptedmembership. There is equally a sensitivity to including the Baltics as a former repub-lic of the USSR, as their inclusion in 1945 was not internationally recognized by all.When I make such references, it will be obvious they refer to the objective function-ing of these economies within the USSR and not the willingness of the peoples to bethere. This last could be said about the populations of the other republics as well.

5 Ukraine’s prospects since the ‘Orange Revolution’ may make it an exception; thisis discussed in later chapters.

6 Kolodko (2000b) discusses opinion polls.7 See Buiter (2000) on some of these issues.8 The conclusion is mine, and though based on Feige and Urban (2004), they do

not necessarily agree nor state it explicitly.9 In the early 1990s, all CIS countries reached ratings as ‘Partly Free’ democracies.

By 2004 seven had fallen back to ‘NotFree’, and the other five all saw a slight dropin the Freedom House points system.

10 EBRD 2003 Transition Report also shows a strong positive correlation betweenmedia freedom and the TPI.

11 Those familiar with the writings on transition will have noted that critics of rapidreforms use the term shock-therapy, while proponents use big-bang or, even moreneutrally, rapid reforms.

12 There is little doubt that some of them, especially the Baltics, have greater marketliberality than many of the West European neighbours who have now joined,with more infrastructure privatized and trade barriers that will now have to go upto harmonize with EU levels.

13 The reader can quickly observe an illustration of this point by calculating theratios of INST to LIB in each group. For CEB where most rapid reformers are, thisratio is higher in 2003 than in 1994, reflecting a catching up. For CISM and CISL,dominated by slower reformers, the ratio in 2003 is lower, reflecting an evengreater lag of institutional reforms.

14 With the exception of the analysis of the growth surge in CIS countries since2000, this section is drawn in large part from Havrylyshyn (2001), and incorpo-rates several points from the survey of Campos and Coricelli (2002).

15 The use of synthetic indices for institutions may not be such a big shortcoming,because objective measures generally cannot capture the implementation qualitywhich is after all what matters. Most objective measures show the quantity of leg-islation, judicial resources devoted to commercial issues, etc. It is possible thatsome measures such as length of time taken for bankruptcy cases begin to capturequality, but so far the best measure of quality may indeed be ex post perceptionsof market actors.

16 The main exceptions can be mostly explained: Azerbaijan is due to oil bonusescoming in before exploitation; Armenia, Georgia and Tajikistan were hit hard bywar and conflicts, hence the usual sharp recovery.

3 Economic and Social Costs of Transition1 As a preview of conclusions, one could arguably say that China falls on the mar-

gin of the first and second, some Central European countries fall on the margin oftwo and three, while most of the rest fall clearly into the third category. ButChina’s outcome was due largely to felicitous circumstances and not necessarilysuperior policy, while the worst-case outcome for the last group was indeed due toinferior policy. On this debate see Brezis and Schnytzer (2003), Garner (2001) andJones and Owen (2003).

2 This last remains relevant even if it is purely a perception of ‘unfairness’ and doesnot entail any absolute loss of welfare. Such perceptions may have political econ-omy consequences, that is people who feel worse off may vote against a reformistgovernment, as in the Przeworski thesis. I am grateful to Gerald Helleiner for thispoint.

3 Micklewright (1992) contends it is because no formal mismeasurement problemcould be identified. For poverty and the Gini coefficient, several counter pointsare given later.

4 Dabrowski and Antzak in Kaminski (1996) point out the negative growth began atleast as early as 1990 in the USSR, and the cumulative decline through 1992 was29 per cent, that is an index of 71 before any reforms.

5 Though a counter-argument made by Milanovic is that the value of many prod-ucts like energy and machinery was underestimated.

6 The output loss is then calculated as the two triangles under the 1989 flat line ofAppendix Figure A3.2, including a loss in the period of decline, and a continuedloss in the period of recovery until actual output returns to its start level. This islabelled as the MAX loss.

7 The output loss in this case is only the ‘A’ triangle for the decline period shown inAppendix Figure A3.2.

280 Notes

Notes 281

8 Kolodko (2000b) is one of many insiders who argue the status quo could notpossibly have prevented a decline; indeed as a previous footnote points out, thenegative rates began well before transition.

9 Transmonee data is found at www.unicef-icdc.org The EBRD Report of 2000, ch. 5,provides an excellent overview of the statistics and analytical discussions on theissue. More recent data is found in the later reports.

10 Lehmann, Wadsworth and Acquiti (1999) has a detailed analysis of such under-employment in the case of Russia, but the process was very similar in other CIScountries.

11 It is not inconceivable, however, that properly measured underemployment inthe CIS countries could be well above 15 per cent, in which case the positive rela-tionship between performance and pace of reforms is maintained.

12 Critics of too-rapid reforms would be entirely justified to note that deterioration ofthe state’s ability to continue good data collection is itself a sign of social worsening.

13 Nor do they show any distributional effects. Even if HDI in Central Europe onaverage did not decline at any time, the very high unemployment rates must havecaused pain to a significant segment of the population.

14 Low-cost lunches, food and clothing for Gosplan employees is an illustration.15 Atkinson and Micklewright (1992) give useful comparisons, and for socialist

economies other earlier literature generally shows values in the same range.Replacing ‘inequality with dispersion’ raises a legitimate normative question: howbad was this increase of the Gini?

16 The importance of the difference between income and expenditure Gini is exem-plified by Keane and Prasad’s (2002) findings for Poland. Using wage earnings, thereis a sharp increase in Gini; using household expenditure data, there is a very littleor no increase. The explanation in this case is very generous transfers to the ‘losers’.

17 An exception is the extremely thorough and careful analysis of householdexpenditure data by Keane and Prasad (2002) for Poland, which comes to theconclusion that inequality did not rise between the late 1980s and the late 1990s,or at most the rise was de minimis.

18 See Eyal, Szelenyi and Townsley (1998) and the special issue of the Slavic Review,Winter 1999, articles by Roeder, Bunce and Fish.

19 This is shown more concretely in Chapter 6, especially in Table 6.2 of milestonesin transition.

20 Croatia is an outlier at 36.21 Russia is an outlier at 45.22 It is notable in the Benjamin et al. (2004) study that the Gini coefficient for

consumption is lower in both years (0.22 and 0.28) than that for income.23 An excellent review of the data and conceptual problems in measuring poverty in

any country are in Deaton (2004). He concludes current statistical procedurestend to overstate the magnitude of poverty.

24 Not to mention internal inconsistencies: the poverty ratio for Hungary in 1993 isshown in various issues of WDI as follows: 8.6, 14.5, 23.8 and 25.3. These are notminor discrepancies.

25 Both of the cited studies come to this conclusion as well; their findings affect itonly in pointing to a lesser degree of deterioration than indicated by the extremehigh-end estimates in other sources.

26 For perspective, this range includes countries like Bangladesh, Burkina Faso,Kenya and Philippines (WDI, 2003). In comparison, China has a national povertyline ratio of 6 per cent, but in an international comparison with 2$/day, thisjumps to 47 per cent, higher than most CIS countries.

27 Latvia shows an opposite trend: the values are in 1993, 2 per cent, and in 1998,8 per cent. There may be a country-specific explanation, but given the very simi-lar paths followed by all the Baltic countries one is more tempted to conclude thisis yet another reflection of the problems with which this variable is afflicted.

28 The meaning of this warning is illustrated by the nonsensical result of povertyratios in early 1990s for the Czech Republic and Slovenia of 55 and 85 per centrespectively; I noted earlier the anomaly of Latvia in the early 1990s having lessthan 2 per cent while other Baltics were 20–30 per cent. But apart from theseoutliers most of the values of WDI are more plausible. I take the risk.

29 Even in the cases of the worst declines in male life-expectancy in Russia and otherCIS countries, the trend for females was much less dramatic: in Russia this fell bythree years, in Ukraine and Kyrgyz republic less than two years, but in many otherCIS countries including Caucasus and Central Asia the female life-expectancyremained virtually constant at about 74–75 years, or even rose modestly (seeEBRD, 1997).

30 Public expenditures on healthcare do not seem to correlate well with the variationin the life-expectancy death rates. According to UNDP (1998), the Caucasian andCentral Asian economies with the exception of Kyrgyz saw sharp drops in thisratio, while the western states saw increases. One possible explanation is that bud-get data are a poor reflection of actual service delivery, or in some cases are notbelievable: for example Belarus is reported to have increased this share from 3.5 to7.0 per cent.

31 This is exemplified by writings on the subject by all the major IFIs which bear thebrunt of Reddaway and Glinski’s general criticism. The EBRD Transition Report forseveral years has made this point, and it is most recently thoroughly analysed inthe (2003) issue which concludes succinctly that ‘educational services havedeclined dramatically in quality in some countries’ (p. 6). Its chief economist,Willem Buiter, in a (2000) article assessing the state of affairs in Russia and itsneighbours, also points to the threat of the hollowing-out of human capital.

4 The SS Transition Navigation Model1 Modern reference is to Isaiah Berlin (1953), The Hedgehog and the Fox, who differ-

entiates those works which ‘pursue many ends [and] those who relate everythingto a single central vision’.

2 The focus on the post-communist era immediately distinguishes my analysis fromefforts to explain market reforms in Asian economies, such as China and Vietnam,where communism has remained the ruling political if not economic paradigm.

3 The KB paradigm I describe in Chapter 1 is a model of what changes are needed toachieve a transformation; here, I seek a paradigm that can, ex post, explain whythese changes occurred faster in some countries and have not yet occurred in afew laggards.

4 Selected examples of the former are the study of Russia by Reddaway and Glinski(2001), Ukraine by Banaian (1999), and Shen (1996), Romania by Shen (1997),Poland by Kolodko (2000a) and numerous others shown in Bibliography. Thecomparative approach is often found in conference volumes such as Kaminski(1996) or Havrylyshyn and Nsouli (2001) with chapters covering many countriesor different cross-country issues.

5 In Havrylyshyn (1997) the author has argued that overemphasis on building stateinstitutions in Ukraine, whether well-meaning or a tactic of old vested interests,

282 Notes

Notes 283

resulted in reform delays which allowed rent-seeking to flourish and create a newclass of oligarchs. A similar interpretation is given in Kuzio and Wilson (1994) andHarasymiw (2002) Motyl and Kravchenbo in Bremmer and Taras (1997) argueexactly the opposite: too rapid reforms caused oligarch development.

6 This was also true for Hungary and even more so in former Yugoslav Republics. Asofter version of this advantage is noted by Laar (2002) for Estonia which sawmore Nordic visitors in the Soviet period than other republics, and – given thesimilarities of Estonian and Finnish languages – had meaningful access to Finnishtelevision.

7 Which for econometricians implies the coefficient on the independent variable isof different value for each country, or in the worst case of a different sign for eachcountry, making any consistent quantitative analysis meaningless. Recall the caseof nationalism discussed earlier.

8 The cases illustrated are drawn from CISH for convenience, though the points aremade in many other writings on these countries.

9 The effect of strong nationalism may have changed after the ‘Orange Revolution’to that experienced in the Baltics at the start.

10 Basta (2004a) and Gray, ch. 9 in Heenan and Lamontagne (1999a), discuss theeasy-going but steady pace of transformation in Slovenia, attributing it to thestrong consensus in society on two points: Slovenia had done very well during theSocialist period and did not need to make any quantum leap; there was no risk ofgoing too slow because there was a strong commitment to the final goal of a full-fledged market economy, democracy, and eventual EU membership. One mightadd that the need to move fast was less urgent because the extent of market-orientation in 1989 (TPI � 1.7) was, together with Croatia, far more advancedthan in Poland (1.3) or even Hungary (1.5). Chapter 6 describes the starting pointsfor each of the transition countries.

11 I do simplify Kolodko’s views for purposes of exposition here: he also argued therewas insufficient systematic planning for a social safety net and institutional devel-opments, and disagrees with Balcerowicz (2002) that the period of ‘special poli-tics’ had to be used for a big-bang, which gave additional benefits of ensuringirreversibility. In the event, what is clear is that Poland was successful in the tran-sition, despite some costs, which may or may not have been avoided, and this isbecause fundamentally both of these major players representing different groupson the political spectrum had their sights firmly fixed on the same long-term target.

12 This point is made in Rzonca and Cizkowicz (2003) in trying to explain why theTPI as an explanatory variable for growth, becomes gradually less significant as itreaches values closer to the limit.

13 As I write this in Toronto, Ontario, I am starkly reminded that expecting ideolog-ical consistency or correspondence between election platforms and policy actionsis folly even in the most advanced democracies. The new Liberal Party Governmentof Ontario, which came into power in 2003 on a strong platform against any taxincreases, has gone ahead and done precisely that in 2004, reimposing a health-care premium that was abolished many years ago. Polls suggest the move is highlyunpopular, but the same polls suggest the public buys into the explanation that‘the finances were much worse than we realized and something had to be done’.I would nevertheless not care to predict the outcome of the next elections.

14 One does in fact come across similar lists of difficulties faced by the eventuallymore successful countries. Laar (2002) states: ‘The first years of independencewere difficult for Estonia. The markets in Russia were gone and it was necessary to

restructure the economy totally. The young country had to confront the urgentissue of land reform, reorganization of industrial production and markets, and themobilization of ethnic minorities.’ However, this was recognized as ‘nothingunique’ and taken as a reason for moving fast rather than a complaint of how dif-ficult it was. Similarly, Kolodko (2000b) notes many difficult initial conditions ofall countries but concludes that this was a reason to work even harder and notblame history.

15 Being at the historical crossroads of trade routes, or caught between competingempires, was not only Bulgaria’s or Ukraine’s inheritance, (as described in the rel-evant Handbook chapters) but can be legitimately claimed by Serbia, Slovenia, theBaltics, Kyrgyz Republic and many others – or in theory by any country sur-rounded by a number of other countries, which is to say most countries. Havingan apathetic population unprepared to throw the rascals out, when relatively freeelections allowed, could apply not only to Belarus (again in CIS Handbook), but toUkraine, Moldova and others or not: the Slovak, Serbian, Georgian and Ukrainianapathy very suddenly turned to Ghandian resolve when abuses by those in powerbecame overly egregious.

16 The reader is welcome to substitute Wyoming, southern Siberia, or, where itmay still exist, a Western European equivalent. I assume that at least the literaryallusion of not seeing the forest for the trees is clear.

17 Anderson and Boettke (1997) argue that is precisely what they did; more on thisin Chapter 6.

18 Bergloff and Roland (1997).19 This is a possible explanation for the early efforts of very radical reforms in Russia

and Kyrgyz Republic; but if so one require a good explanation of why this wasreversed, subsequent chapters elaborate on these and analogous cases.

20 This characterization applies before the Rose and Orange Revolutions in Georgiaand Ukraine. It is too early to judge whether they are moving out of their capturedstatus, but later chapters will discuss how it was that the oligarchies there lost elec-tions to democratic forces, and what are the prospects for reversing the oligarchpower.

21 I am not saying by this that the process of liberalization and democratization iscomplete in these countries, nor that former nomenklatura members had no abil-ity to use influence and become important new capitalists. I am saying that theextent of concentration of ownership is far lower than in the first group, and farfrom what it would need to be to control elections and policies.

22 Many text books on Comparative Economic Systems give Nazi Germany as theprime example of this form, and also note the softer versions thereof in manywar-period economies.

23 With one possible exception reflecting both their hubris and their human frail-ties: they have recently recognized publicly the extent of their political power.One of the first to do this, Berezovsky in Russia, may have paid the price of exile.

24 Hungary and Estonia did move much faster than Poland and were not later thanRussia, but the process was much more transparent, based mostly on direct sales, andthe governments much more committed to a clean privatization. This is explored inChapter 6 as an example of lesser vulnerability to oligarchic development. Thereader who wishes to see in this the continued role of old communist elite, isreminded that while in Estonia it did not continue, in Hungary it did much more.

25 Eyal, Szelenyi and Townsley (1999) review these predictions of ‘political capital-ism’, that is how the communist elite would conspire to become the capitalist

284 Notes

Notes 285

elite in the new regime. The main part of their study is to show that in CentralEurope this did not happen, and then explain why.

26 See Kraft (2000) for an application to transition economies measuring the timefrom independence to the achievement of this threshold, and showing the lagwas longer the weaker was national sentiment or commitment.

27 Updates in Kaufman, Kraay and Mastruzzi (2003) compute an analogous variablecalled ‘crony bias’, but the earliest data are most relevant here. Later data will beused to test the TF hypothesis in Chapter 8.

28 And it certainly does not come from the goodwill of capitalists as Smith’s succinctcomment on the baker emphasizes: ‘it is not because of the benevolence of thebaker that we eat fresh bread every morning but because of his desire to makemoney’.

29 I wish to emphasize that the TF criticism of TI is not a critique of the WCexcept in the limited ex post sense of recognizing that privatizations whichresulted in overly concentrated ownership created strong vested interest inmonopolistic and rent-seeking behaviour, which in turn means strong oppositionto full liberalization. This is not the place to discuss critiques which argue itshould have been foreseen and the negative outcome is therefore the fault of WCand its proponents; (Reddaway and Glinski (2001) is only one of many. Note thatin the model of the present work, delayed and incomplete reforms are the causeof the oligarchic formation, rather than the substance of the proposed reformsthemselves.

30 One of my modest objectives here is to bring back into focus the forgotten adjec-tive modifying ‘market’ in the neo-classical welfare maximizing arguments:Competitive.

5 The Search for a Navigation Chart1 The author was fortunate enough to be close to both events, first at an academic

conference in Budapest in mid-November, watching on television with CentralEuropean colleagues the Velvet Revolution events in Prague, then in late Auguston an interrupted vacation with his spouse in Yalta not far distant from Gorbachev’stemporary incarceration in Foros. The rollercoaster of emotions in that short weekincluded the fearful frenzy of crowds seeking safe return to their homes at theSimferopol airport, and the joyful celebration of Ukrainian independence on thestreets of Kyiv on 24 August 1991.

2 China of course had began such a transformation but focused first on the eco-nomic dimension.

3 Chapter 1 has discussed how this quest triggered considerable efforts of econo-mists to model alternative paths often coming to the mathematical conclusionthat gradual was better.

4 The concept of ‘uniqueness’ of each country also had considerable appeal andcontributed to the rationale for a longer debate. That this was a misguided viewcoopted by reform opponents has been argued in Chapter 4.

5 Thomas (1991) summarizes the reform experiences of developing market economies.6 The term was coined by John Williamson (1993), in a volume reviewing the expe-

riences of stabilization and liberalization in developing countries, but includingalready some early discussions on the lessons for transition.

7 Yalcin (2002). Its implementation may have been partial and its success not yetevident, as Chapter 9 will outline, but the degree of consistency of strategy is

relatively unique in the region. Turkmenistan cannot be considered as analogous,for its strategy starts from the primacy of the wisdom of the leader, goes on topolitical social issues and only last gives an economic recipe.

8 Schroeder (1996).9 Some cases shown in brackets in the figure are on the margin between two

categories; this is discussed more fully in the next section.10 This is elaborated in Chapter 9. Akayev was not a politician but an academic who

at first was inclined to be more experimental, and Yeltsin though a high rankingmember of the nomenklatura certainly supported a more radical approach for atleast a short period.

11 A brief history of this is in Bristow (1996) and Bristow’s chapter in Heenan andLamontagne (1999a).

12 No descriptions are given for 3� or 3.3. It is to be emphasized that the EBRD doesnot to my knowledge define any analogous milestones or benchmarks and all theinterpretations here are the author’s alone.

13 Hungary and then Czechoslovakia started with much less macro instability anddid not attain such a high level of inflation at any time.

14 TPI values in years 0 and �4 are not shown in Tables 5.1 and 5.2. Source is thesame as in the tables.

15 This is explored in Kolodko (2000b).16 These are not exact equivalents as in the second phase are also included large-

scale privatization; however all the other elements in the second phase areinstitutional in nature.

17 A third, commitment to EU membership, is part of this or parallel to it; it is elab-orated in Chapter 7. In the formalized Navigation Model of Chapter 4 Appendix,the first two forces combine to define the variable LIBVIS, while the third iscaptured by the variable EUM.

18 Bunce (1999), p. 280.19 See Demidkina, Chapter 4 in Heenan and Lamontaigne (1999b).20 The fact that a party with the name Communist won a minority and was not in

government should not mislead one; the mainstream pragmatist nomenklaturarenamed itself as the Democratic Party, while ideological communists and pro-USSR members stayed in a Communist party. This arrangement also exists to thepresent day in Russia, Moldova and the Caucasus.

21 Kubicek, ch. 5 in Heenan and Lamontagne (1999a).22 In Yugoslavia, the dominant party had been named ‘Socialist’ rather than

communist since Tito’s break from Moscow.23 This is the variable LEADER noted in Appendix of Chapter 5 describing the

Navigation Model. It is narrowly defined as cases where a leader was far morevisionary than the rest of the polity, but this is not to suggest that Havel orWaleson, or Baltic personalities, were not as visionary as Yeltsin and Akagev. Theysurely were and even more but in a context of a society showing and supportingthis vision.

24 The return of a communist government in Moldova in 2000 does not conform toPrzeworski’s thesis either. It came much later when the economy was in fact turn-ing around. It is also very different from the other cases as these were not therenamed and perhaps somewhat reformed-pragmatic former communists whohad been in power since 1991, but the much more ideological traditionalist com-munists who had been consistently against liberalization and privatization and

286 Notes

Notes 287

were victorious as much for the public revulsion against oligarchic power, as thepain of transformation. Not surprisingly they did indeed reverse both dimensionsof the transformation.

25 Kolodko (2000a) argues forcefully that the changes needed to overcome ‘thedisastrous experience with central planning … were not simply adjustments tothe existing system but the establishment of systemic new foundations’.

26 Shulman (2003) explores the opposite causation for Ukraine: does economicperformance affect the strength of national identity? He finds it does not.

27 But numerous minor ones do continue. As the author writes this in Toronto,several unresolved cases of border conflicts for Canada are written about on backpages. Tiny Hans Island in the strait between Greenland and Ellesmere Island wasrecently visited by the Navy of Denmark which reportedly laid claim to it in June2004. A much larger issue is the dispute with the USA on a vast part of the BeaufortSea, revolving around the angle of the line drawn from the mainland borderbetween Alaska and the Yukon Territory; oil resources there may be considerable.Toronto Star, 25 March 2004, p. A7, ‘Canada–Danish spat erupts over island’.

28 Economist, 28 August 2004, p. 38.29 Economist, 9 October 2004, p. 48.30 Gleason in Chapter 16 in Bremmert and Taras (1997) also notes that while in

Central Asia nationalism did not exist – this is perhaps too strong given the refer-ences that follow in the same work to the national myths of the Kahantes in Khiva,Samarkand, Tashkent – the settled peoples of the river plains in Uzbekistan were farmore authoritarian than the nomadic ones with traditions of tribal democracy.

6 From Rent-Seeking to Oligarchy to State Capture

1 Interested observers will not take long to come up with 5–10 names or more of oli-garchs, but almost all of them will be from Russia, perhaps one or two fromUkraine, maybe another from Central Asia where the family name of a presidentsuffices to make the association. Any from Poland, Hungary, Latvia? Highlyunlikely to be named; but is this because their oligarchs are less well-known? Theanswer given later is that it’s because their big-capitalists are not of the varietywhich can be labelled oligarch, both in size and political influence.

2 An exception was the capital accumulated by the Soviet era mafia based histori-cally on black-market trading in a shortage economy, but later including drugsand other criminal activity. Handelman (1994) describes this group of new capi-talists, but it may not have been the major source underpinning the future oli-garchs. It is notable that in the Index of his book, one cannot find a single one ofthe names that now comprise Russia’s oligarchs listed methodically in Barnes(2003b) and World Bank (2004).

3 The vast literature on post-communist privatization has unfortunately enshrinedthe nuance of ‘insider’ to mean the managers and owners of the enterprise; here Iwill emphasize its much broader meaning as in ‘insider-trading’ regulations: oper-ations undertaken by those who have special knowledge not available to the restof the market.

4 The literature on these activities is too vast to be cited; I mention here only afew good writings which give numerous examples, most of them focusingunfortunately on Russia. Handelman (1994), Aslund (1995), Brady (1999), Barnes

(2003b), Goldman (2003), Freeland (2000a,b); for Ukraine see Havrylyshyn(1997), Aslund (2003), Balmaceda (1998), Puglisi (2003) and Zimmer (2004).

5 Aslund (1999).6 As late as 1998, after considerable closing of these price gaps and loopholes, the

young reformist Prime Minister of Moldova, Jan Sturc, remarked in a personal con-versation: ‘give me 15 minutes sometime and I can tell you about sixteen differentways in which it is still possible to make a lot of money on energy operations’.

7 Handelman (1994) provides good evidence of their importance, citing on p. 63 astatement of a businessman: ‘Many of us [in the Komsomol] have been involvedin some kind of business since the late eighties’.

8 Hellman, Jones and Kaufmann (2003). It is notable that these authors who areresponsible for the measurement of a state-capture index (SC99) used by many todistinguish oligarchic from non-oligarchic cases, emphasize in this passage thatthe very low levels of SC99 for the lagging reformers are not meaningful.

9 This was in contrast to the delay in Poland which many eventually praised, as itallowed development of stronger market and competitive institutions. But delayalone was not the determinant of success, as Ukraine’s example attests; the differ-ence was far greater commitment in Poland which assured progress on institutions(see the figure in Chapter 2 Appendix which show the much slower institutionaldevelopments in Ukraine), and transparency in the privatization process.

10 Feige (1997) is an example; Eyal’s Szelenyi and Townsley (1999) explore theprocess of former communists leveraging political power into economic power inthe transition, and show this was far more effective in the USSR.

11 Alexeev and Pyll (2003).12 I rely in particular on Brady (1999), Freeland (2000b) and Goldman (2003) for

Russia; Aslund (1999), Puglisi (2003) and Zimmer (2004) on Ukraine. Forbes(2005) also provides useful input. On Central Asia, no focused writings are avail-able and the information must be gleaned from a large number of general studiesof these countries.

13 The details on the background of some new additions to the Billionaires’ Club isgiven in Forbes (2005) and illustrates the paths followed.

14 Kuzio (2005) argues that Akhmetov’s support of Yanukovich in Ukraine’s presi-dential election illustrates how oligarchs effectively block liberalism while advo-cating publicly democratic forms.

15 The head of a large Latvian bank with many rich Russian clients quite proudlyrelated to the author how his connections in the Komsomol were critical to busi-ness success, and how the organizational training the Komsomol provided ‘wasthe best MBA School in the USSR’.

16 A precise measure of GDP in transition countries remains elusive.17 The fact that Canada and Sweden have a ratio slightly above one may seem

surprising as both have a reputation as ‘welfare-states’ with a large public sector,virtually free education and health services, and – apart from the few billionaires – anincome distribution measured by the Gini coefficient that is more equitable thanin many other industrial countries. But for students of ‘capitalist entrenchment’,as reviewed in Morck, Wolfenson and Yeung (2005) and summarized in theAppendix to this chapter, the existence of a small but extremely wealthy group inthese two countries has been known for some time.

18 More recent values are available and will be used in later chapters. The earliestavailable value, 1999, is shown here because it is the best indicator of how the

288 Notes

Notes 289

dynamics of slow, partial reforms, plus non-transparent privatization led to verydifferent results in different countries.

19 Lecture of Prof. J. Mazesnikov, Slovak Institute of Policy Analysis, University ofToronto, 15 September 2004.

20 Motyl and Kravchenko in Bremmer and Taras (1997), p. 253. In fairness, theauthors argue in the event that Rukh probably did not have much choice andthat they made the right decision to not press for rapid economic reforms whichwould have threatened the stability of Ukraine. This is not a view shared by thepresent author.

21 Recall the argument of Rzonca and Cizkowicz (2003) in the discussion of growthdeterminants, Chapter 2.

22 Something similar was done in the 2002–03 programme with Brazil, with ananalogous result of the opposition winning the election.

23 True, land grants for latifundia in South America or railway-building are closer tothe privileged privatization of Soviet assets, but in fact without the activityof building railways or cultivating the land, the assets granted had little if anyvalue. In the case of railways, the grants were at least conditional on a completedrailway.

24 Another similarity is the source of an oligarch joke (anekdot) in Russia andUkraine: ‘Our oligarchs are just like the American Robber Barons – they bothinvested their money in America.’

25 In the 1970s and 1980s several multi-country projects investigated this problemunder the general editorship of top trade and development economists such asAnne Krueger, Jagdish Bhagwati and Bela Balassa. The last of these mega-projectscovered 18 countries: Papageorgiou, Choksi and Michaely (1991).

26 Sandbrook (2000).27 Viktor Pynzenyk, Deputy Prime Minister of Ukraine in the early 1990s, was put

the question at an international conference: ‘couldn’t one just get rid of all theold nomenklatura people and start over again as in Poland?’ His answer: ‘You cantake the pigs away from the trough, but then you just get another group of pigsjump in; the only solution is to take away the pig-trough.’

7 Safe Havens for Market Reforms1 As cited in Papadimitriou (2002). This book, plus the ones by Schimmelfennig

(2003) and Palankai (2003), provide excellent and comprehensive accounts of theEastern Enlargement of the EU; the following section draws much of its informa-tion from these three sources.

2 Papadimitriou, p. 35.3 Laar (2002) asserts unequivocally that for Estonia, at the very beginning the peo-

ple and political elites overwhelmingly saw EU membership as the central part ofthe transformation. He judges this was also the case in the other Baltic countries,and Central Europe.

4 Palankai (2003) p. 429.5 Although the pattern over time is for a shorter delay between the stages for late-

comers. Thus, the first four applied in 1994, three years after the Europe agreementsand had negotiations until 1999, while the Baltics and Bulgaria and Romaniamoved quickly from the Agreement to application to start of negotiations. Thisrecord gives some hope to Turkey (and Ukraine?) that the actual accession date

could be earlier than first noted. But late comers did not necessarily catch-up aswill be shown below.

6 An alternative model is posited by Plumper, Schneider and Troeger (2004); theyseparate the decision by candidates to apply as being determined by the level ofdemocracy and market reforms, from the EU decision to accept which is deter-mined by compliance with AC criteria. The historical process seems closer to asimultaneity.

7 Cooley’s chapter in Karatnycky et al. (2003) discusses the EU anchor and is rightto argue that the internal commitment in Central Europe was very strong withoutEU carrots. I argue later that sticks may have been even more important thancarrots to keep the laggards on track.

8 Lecture given by Bela Kadar on the EU enlargement, University of Toronto,16 September 2004.

9 An exception was Slovakia in the period 1994–98 with Méciar as Prime Minister;the EC gave negative reports and threatened delays in accession. This is exploredbelow as one example of the EU using membership as a stick.

10 Palankai (2003), p. 433; he goes on to note that despite having the highest wagecosts of all the post-communist countries, the Czech Republic, Hungary andPoland received the lion’s share of FDI, to a large extent because of the favourablesignals about EU accession.

11 Bleiere (1999).12 The same proposal also put Yugoslavia and the Soviet Union (not yet dissolved)

on the shelf for later examination. It is best to think of this not as a stick but awithholding of an offer for lack of clear prospects on reform progress.

13 A recent article on the series of second-wave democratic revolutions (Slovakia,Serbia, Georgia, Ukraine) reemphasizes this in a somewhat popular but funda-mentally sound conclusion: ‘Slovakia’s opposition … had two things workingin their favor: one was the immense lure of joining the EU and NATO. The otherwas the Internet.’ In ‘Belarus and Moldova, Hopes for Democracy’, New York Times,24 February 2005.

14 Indeed, as the Financial Times Survey of 22 June 2004 described it, the Slovenianleaders – and populace – always felt they were good managers of the economyunder socialist Yugoslavia, never suffered a financial crisis, were not far fromhaving a functioning market economy, and were confident they would get to EUmembership in due time.

15 Note in the Appendix table of Chapter 2 that the 2002 TPI measure for Ukraine,2.8, had nearly caught up to Romania’s value of 3.0; furthermore, this wasfor Ukraine forward momentum while for Romania it had been stalled at2.9–3.0 since 1997. This lies behind a judgment such as the Economist made ofgenerosity by the EU; such a long stall might instead have been cause for apply-ing the stick and delaying Romania instead of putting on the same pace asBulgaria.

16 Many countries are still not WTO members: Bosnia, Macedonia, Serbia andMontenegro in the Balkan region, and Azerbaijan, Belarus, Kazakhstan, Russia,Tajikistan, Turkmenistan, Ukraine and Uzbekistan in the CIS.

17 On the large EU effects and the potential superiority of this approach to WTO, seeTolchitskaya and Vinhas de Souza (2004). The reason EU free trade may have alarger impact though it includes only EU countries is that the extent of trade-barrier reduction is almost certainly larger than that initially obtained upon WTOmembership.

290 Notes

Notes 291

18 Argentina is not the only case of political influence by a powerful patron, but hasbecome one of the most thoroughly analysed, see for example Mussa (2002).

19 Ul-Hague and Khan (1998) show in a broad sample how difficult it is to observein econometric analysis any positive impact of IMF programme because it is notgenerally the most important determinant of success; but when the analysis isdone carefully, the effects are found to be positive if not always large. Mercer-Blackman and Unigovskaya (2002) find stronger results for transition countries’growth in the 1990s by measuring very carefully the effectiveness of actual imple-mentation; that raises the question whether countries that implement IMF pro-grammes well would have done all the right things without the programme.

20 For example, central bank operations, and its regulation of the banking system,continued to be steadily improved in Belarus. Personal communication fromfinancial sector specialists at the IMF.

21 There may be some dispute as to whether any ‘offer’ to Croatia existed at all; it wascertainly weaker and less explicit than to the other countries in the modest group,but informally it was at all times understood that given the advanced economictransformation, once the impediment of an awkward conflict situation andrelated questions of democracy was removed, Croatia was a good candidate. Inthe event the uncertainties on democracy remained for a long time after any con-flict was ended, but as noted earlier, with election of a government deemed moredemocratic, the EU moved fast to upgrade the offer.

22 A minor adjustment is made: Hungary and Czechoslovakia at no time had infla-tion as high as 5% monthly, but instead of showing zero months delay, the charttakes the values from regime change to the start of stabilization. Using zero clearlywould strengthen the pattern observed.

23 Estonia’s President Meri, for instance, stated in November of 1993 that ‘the Balticstates would “integrate into Europe economically, politically and militarily” ’(quoted in Park (1995), p. 33). In February 1994, he stated that ‘Estonian securitypolicy should be based on economic integration, first with the Baltic states, thenwith the Baltic sea states and the European Union’ (ibid.).

24 See the Commission Opinion on Estonia’s Application for Membership of theEuropean Union, p. 112. Accessed at http://europa.eu.int/comm/enlargement/dwn/opinions/estonia/es-op-en.pdf.

25 See the ‘Chronology of Estonia’s Accession to the EU’ at http://www.riigikogu.ee/?id�12974.

26 See Commission Opinion on Estonia’s Application for Membership of theEuropean Union, p. 116. Accessed at http://europa.eu.int/comm/enlargement/dwn/opinions/estonia/es-op-en.pdf.

27 The European Council outlined a set of broad membership criteria to CentralEuropean countries at its June 1993 meeting in Copenhagen. The criteria were ‘thestability of institutions guaranteeing democracy, the rule of law, human rights andrespect for and protection of minorities; the existence of a functioning market econ-omy; the capacity to cope with competitive pressures and market forces within theUnion; the ability to take on the obligations of membership, including adherenceto the aims of political, economic and monetary union’ (Mayhew, 1998, p. 162).

28 The social-democratic government of the reformed ex-communist Brazauskas wascontinuously attacked by the right wing politicians for being too friendly withRussia and betraying national interests (Lane, 2002, p. 201).

29 Lane is explicit in emphasizing that both the left and the right in Lithuania’spolitics wanted membership in the Western political, economic and security

organizations. However, ‘the question at issue was how to treat Russia; nationalistswere basically Russophobic and doctrinaire, the Left was pragmatic and flexible’(Lane, p. 202).

30 See Commission Opinion on Latvia’s Application for Membership of the EuropeanUnion, pp. 111–12, accessed at http://europa.eu.int/comm/enlargement/dwn/opinions/latvia/la-op-en.pdf. For Lithuania, see Commission Opinion onLithuania’s Application for Membership of the European Union, pp. 104–5, accessedat http://europa.eu.int/comm/enlargement/dwn/opinions/lithuania/li-op-en.pdf.

31 The progress in compliance with the EU criteria is illustrated in the regular annualreports of the EU from 1998 onward. For Latvia, reports can be found at http://europa.eu.int/comm/enlargement/latvia/index.htm#Overview%20of%20key%20documents%20related%20to%20enlargement. Lithuania’s reports can be foundat http://europa.eu.int/comm/enlargement/lithuania/index.htm#Overview%20of%20key%20documents%20related%20to%20enlargement.

8 Future Prospects for Captured States1 Somewhat less forthright arguments that oligarchs will eventually promote liber-

alization because it is in their own interest are found in Boone and Rodionov(2001), Cottrell (2002) and Shleifer and Treisman (2004) inter alia.

2 References to individual analysts are not meant to imply categorical adherence toeither the TI or TF views. Such pigeon-holing is neither necessary nor in factwould it be objective, because many analysts explore seriously both sides of thedebate in the same or different writings. Thus Gaidar and Buiter, as noted, recog-nize the validity of some arguments in both views and, as cited later, Aslundand Johnson (2004) also recognizes the primary role of small rather than largebusiness in pushing institutional development.

3 As cited in Financial Times, 7 June 2004.4 World Bank (2004) Country Economic Memorandum, Russian Federation.5 Titling his article ‘Winners Take All’ nicely and poignantly evokes this paradox.

The sub-title ‘Politics of Partial Reform’ clearly signals his view that it was notrapid but gradual reform that is to blame. Another paradox is the contrast withthe expectations of strong opposition to reforms from ‘almost everyone in theproduction hierarchy’: Holzmann (1992).

6 Rajan and Zingales (2003a and b) also emphasize how powerful capitalists cleverlyspin their entreaties to governments for protection against competition as beingfor the good of the workers, the nation.

7 Mr Poroshenko, who supported the opposition candidate, Mr Yuschenko, andMs Tymoshenko, remain for now far short of the billion-plus class of major oli-garchs, and should be regarded as ‘lesser-oligarchs’, or ‘mini-garchs’, with meretens or hundreds of millions of dollar assets. The lessons of the 2004 election and‘Orange Revolution’ in Ukraine are explored further late in the chapter.

8 I note only a few: Nellis in Havrylyshyn and Nsouli (2001), Megginson and Netter(2002), Barnes (2003a) and Frye (2003).

9 Many recent analyses show concentration has increased in all countries and thatthe share of major owners in Russia is not much higher than in Central Europe at35 per cent plus most recently. This is a measure of firm-level and not economy-wide concentration which is the relevant measure of oligarchic concentration.Hashi, Kozarzewski and Radygin (2004).

292 Notes

Notes 293

10 I cite only a few examples: Fidrmuc (2000), Frye (2003) and Jackson, Klich andPoznanska (2003).

11 Hellman and Schankermann (2000).12 One can imagine an inverted U-curve with an optimum SME share of

20–24 per cent.13 An almost surreal example of this was the action of the Uzbekistan government in

mid-2003. To meet the IMF conditions of a maximum gap between official andcurb rates for the som, they undertook a sweep of informal markets and imposednew licensing requirements for minor traders. This temporarily reduced suchactivities to near-zero levels; foreign currency demand dried up and the formaltarget was met; to the chagrin of IMF critics, the latter did not buy the trick.

14 That the architects of the loans-for-shares privatization believed in such aDamascene conversion of the new capitalists is confirmed in Freeland’s (2000b)fascinating account quoting Chubais: ‘Let them steal and take their property.They will then become owners and decent administrators of their property.’

15 Isajiw (2004).16 Johnson, MacMillan and Woodruff (2000).17 The large number of such writings is reviewed in the following three articles:

Havrylyshyn and McGettigan (2000), Djankov and Murrell (2002) andHaltiwanger et al. (2003).

18 Krugman (1994).19 Bugajski (2004) is an excellent source on the question of Russia’s future ambitions

on these dimensions.20 Plumper, Schneider and Troeger (2004) conclude with virtually the same words:

‘The EU requires severe reductions in market distortions.’ Thus, special intereststhat profit from these [distortions] oppose EU membership.

21 Many good colours remain. Imagine such a popular revolution in Moldova occurringin early summer of 200X – it might be called the Cherry Revolution for Moldova’sbountiful orchards, Mr Lukashenko’s own words may suggest a colour in Belarus:‘There will be no revolution of any kind here, not even a banana revolution.’

22 Her very successful actions as Minister of Energy in 1999–2001 cut rents in thissector by one-third or more (Aslund, 2002), suggesting she was not using her posi-tion to improve personal future opportunities. Coincidentally, and without prej-udice, her actions are said to reinforce the old adage ‘it takes a thief to catch athief’. Her own oligarch days were as an energy rent-seeker in partnership with thelater disfavoured Prime Minister Pavlo Lazarenko.

23 Judah (2005).24 The earlier references, Carothers (2002) and McFaul (2002), cover these ‘managed

democracies’ of some post-Soviet states.25 As Kuzio (2005) suggests: ‘The attempted poisoning of Yuschenko during the cam-

paign shows how far his opponents were willing to go to stop his election.’26 Forbes, 28 March 2005 annual ‘Billionaires’ issue estimates his worth at $2.3 billion.27 Michael McFaul, in an interview on American National Public Radio (28 March

2005), noted that in Kyrgyz Republic – which had neither the charismatic leaderof Georgia and Ukraine, nor the allure of EU membership in Serbia – the angerof the crowds was first and foremost directed at the large commercial outletsknown to be owned by members of the President’s family: a more focused expres-sion of ‘enough of the abuse’ cannot be imagined. It is notable that an Egyptianopposition movement has been named ‘Enough’.

9 Diverse Outcomes1 The reader is reminded of the analysis in Chapter 3 on the potential bias in official

GDP data and alternative estimates correcting for these problems; the measure hereis the average between official and Aslund-adjusted estimates as described inChapter 3.

2 One of the first to do these, Berezovsky in Russia, may have paid the price in exile.Potanin’s plea for social acceptance cited earlier is in the same category.

3 The second wave of ‘colour’ revolutions against oligarchies may also be such amoment.

294 Notes

Bibliography

295

Acemoglu, D. and J. Robinson (2000) ‘Political Losers as a Barrier to EconomicDevelopment’, AEA Papers and Proceeding, vol. 90, no. 2, pp. 126–30.

Acemoglu, D. (2003) ‘The Form of Property Rights: Oligarchic vs. DemocraticSocieties’, Mimeo, Department of Economics, MIT, Cambridge, MA, 46p.

Acemoglu, D., J. Robinson and T. Verdier (2003) ‘Kleptocracy and Divide-and-Rule:A Model of Personnel Rule’, paper presented at the European EconomicAssociation’s Annual Meeting, Stockholm, Aug. 2003.

Acemoglu, D., S. Johnson and J. Robinson (2004) ‘Institutions as the FundamentalCauses of Long-Run Growth’, NBER Working Paper no. 10481, May 2004.

Aghion, P. and O. Blanchard (1994) ‘On the Speed of Transition in Central Europe’,National Bureau of Economic Research Macroeconomic Annual, S. Fischer andJ. Rotemberg, eds. (Cambridge: MIT Press, pp. 283–320).

Aghion, P. and S. Commander (1999) ‘On the Dynamics of Inequality inthe Transition’, The Economics of Transition, vol. 7(2), pp. 275–91.

Akaev, A. (2001) An Economy in Transition (Canberra: Asia Pacific Press at the AustralianNational University).

Alexeev, M. and W. Pyll (2003) ‘A Note on Measuring the Unofficial Economy in theFormer Soviet Republics’, The Economics of Transition, vol. 11(1), pp. 153–75.

Anderson, G. M. and P. J. Boettke (1997) ‘Soviet Venality: A Rent-Seeking Model of theCommunist State’, Public Choice, vol. 93, pp. 37–53.

Anderson, J. (1999) Kyrgyzstan: Central Asia’s Island of Democracy? (Amsterdam:Harwood Academic Publishers).

Anderson, R., M. Steven Fish, S. Hanson and P. Roeder (2001) Post-communism and theTheory of Democracy (Princeton and Oxford: Princeton University Press).

Aron, L. (2003) ‘Crime and Punishment in Russia’, The New York Times, 6 October2003.

Arvedlund, E. (2004) ‘The Kremlin versus The Tycoon: A Battle for Russia’s Future’, TheNew York Times, 6 July 2004.

Aslund, A. (1995) How Russia Became a Market Economy (Washington, DC:The Brookings Institution).

Aslund, A. (1997) ‘The Russian Economy: Where is it Headed?’, Sturc MemorialLecture, November 1996 (Washington, DC: Johns Hopkins, SAIS).

Aslund, A. (1999) ‘Problems with Economic Transformation in Ukraine’, Paperpresented at the Fifth Dubrovnik Conference on Transition Economies, June 1999,www.ceip. org.

Aslund, A. (2001) ‘The Myth of Output Collapse after Communism’, Washington,DC: Carnegie Endowment for International Peace, Working Paper no. 18, March 2001.

Aslund, A. (2002) Building Capitalism: The Transformation of the Former Soviet Bloc(Cambridge: Cambridge University Press).

Aslund, A. (2003) ‘Left Behind: Ukraine’s Uncertain Transformation’, The NationalInterest, Fall, pp. 107–16.

Aslund, A. (2004) ‘Liberalism’s Economic Triumph’, Moscow Times, 26 April 2004.Aslund, A. and S. Johnson (2004) ‘Small Enterprises and Economic Policy’, Carnegie

Endowment for International Peace, Working Paper no. 43, May 2004.

Aslund, A., P. Boone and S. Johnson (1996) ‘How to Stabilize: Lessons from Post-Communist Countries’, Brookings Papers on Economic Activity, 1, BrookingsInstitution, pp. 217–313.

Atkinson, A.B. and J. Micklewright (1992) Economic Transformation in Eastern Europeand The Distribution of Income (Cambridge, MA: Cambridge University Press).

Babetskaia-Kukharchuk, O. and M. Mathilde (2004) ‘Russia’s Accession to WTO: WhatPotential for Trade Increase?’ unpublished, CNRS-ROSES, University of Paris.

Balcerowicz, L. (1993) ‘Economic Transition in Central and Eastern Europe:Comparisons and Lessons’, Second Annual IFC Lecture (Washington, DC: IFC/WorldBank).

Balcerowicz, L. (2002) Post-Communist Transition: Some Lessons, 31st Wincott Lecture(London: The Institute of Economic Affairs).

Balmaceda, M. M. (1998) ‘Gas, Oil and the Linkages Between Domestic and ForeignPolicies: The Case of Ukraine’, Europe-Asia Studies, vol. 50, no. 2 (March), pp. 257–86.

Banaian, K. (1999) The Ukrainian Economy since Independence (Cheltenham: ElgarPublishers).

Barnes, A. (2003a) ‘Cognitive Theft: Context and Choice in the Hungarian, Czech andRussian Transformations, 1989–2000’, East European Politics and Societies, vol. 17,no. 3, pp. 533–65.

Barnes, A. (2003b) ‘Russia’s New Business Groups and State Power’, Post-Soviet Affairs,vol. 19, no. 2, pp. 154–86.

Barro, R.J. and X. Sala-i-Martin (1995) Economic Growth (New York: McGraw-Hill).Basta, K. (2003) ‘Tracing the Roots of Individual Policy in Post-Communist Slovenia

and Estonia’, Unpublished Paper, University of Toronto, Centre for Russian and EastEuropean Studies.

Basta, K. (2004a) ‘Post-Communist Democratization: A Survey of Recent Theories’,Unpublished Paper, University of Toronto, Centre for Russian and East EuropeanStudies.

Basta, K. (2004b) ‘Prospect of EU Membership as a Factor in Political and EconomicReform: The Baltic Republics’, Mimeo, University of Toronto, Centre for Russian andEast European Studies.

Beck, T. and L. Laeven (2005) ‘Institutions and Growth in Transition Economies’.Paper presented at the Dubrovnik Economic Conference, 29–30 June, National Bankof Croatia, www.hnb.hr.

Bedi, A. S. and A. Cielik (2002) ‘Wages and Wage Growth in Poland: The Role ofForeign Direct Investment’, The Economics of Transition, vol. 10(1), pp. 1–12.

Benjamin, D., L. Brandt and J. Giles (2004) ‘The Evolution of Income Inequality inRural China’, The William Davidson Institute of the University of MichiganBusiness School, Working Paper no. 654.

Bennett, J., S. Estrin, J. W. Maw and G. Urga (2004) ‘Privatization Methods andEconomic Growth in transition Economies’, CEPR Discussion paper no. 4291.

Berengaut, J., E. de Vrijer, K. Elborgh-Woytek, M. Lewis and B. Lissovolik (2002) ‘AnInterim Assessment of Output Developments in Ukraine, 2000–01’, IMF WorkingPaper 02/97 (Washington: The International Monetary Fund).

Berg, A., E. Borensztein, R. Sahay and J. Zettelmeyer (1991) ‘The Evolution of OutputTransition Economies: Explaining the Differences’, IMF Working Paper 91/73(Washington: The International Monetary Fund).

Bergloff, E. and G. Roland (1997) The EU as an Outside Anchor, paper presented forSwedish Governmental Commission on the Consequences of Eastern Enlargementof the European Union.

Berlin, I. (1953) The Hedgehog and the Fox (New York: Simon and Schuster).

296 Bibliography

Bibliography 297

Berry, A. and S. Karin ( July, 2003) ‘Economic Collapse, Poverty and Inequality DuringUkraine’s Difficult Transition’, Development Alternatives, Inc. – Boston Institute forDeveloping Economies, 71p.

Beselaere, M. (1999) ‘The Reform of the Economic Legislation in the Baltic States inthe Framework of their Integration into the European Union’, in K. Magliet andM. Keygnaert, eds. The Baltic States in an Enlarging European Union (Leuven: Institutefor European Policy).

Blanchard, O. and M. Kremer (1997) ‘Disorganization’, The Quarterly Journal ofEconomics, November, 1997, pp. 1091–126.

Blanchard, O. (1997) The Economics of Post-Communist Transition (Oxford: ClarendonPress).

Blais, J., M. Krumova and D. Kruse (1997) Kremlin Capitalism: Privatizing The RussianEconomy (Ithaca: Cornell University Press).

Bleiere, D. (1999) ‘Implementation of Copenhagen Political Criteria for EUMembership by Latvia’, in K. Malfliet and M. Keygnaert, eds. The Baltic States in anEnlarging, European Union (Leuven: Institute for European Policy).

Boeri, T. (2000) ‘Structural Change, Welfare Systems, and Labour Reallocation’, TheTransition of Former Planning Economies (Oxford: Oxford Press).

Bojcun, M. (2001) ‘Ukraine and European Integration’, Journal of Ukrainian Studies,vol. 26, no. 1–2, pp. 271–86.

Boone, P. and D. Rodionov, ‘Rent Seeking in Russia and the CIS’ (2001) Paper for EBRD10th Anniversary Conference, London, December, 2001.

Boycko, M., A. Shleifer and R. Vishny (1995) Privatizing Russia (Cambridge, MA: MITPress).

Brady, R. (1999) Kapitalizm: Russia’s Struggle to Free its Economy (New Haven: YaleUniversity Press).

Bremmer, I. and R. Taras (1997) New States, New Politics: Building the post-Soviet Nations(Cambridge: Cambridge University Press).

Brezis, E. S. and A. Schnytzer (2003) ‘Why are the Transition Paths in China andEastern Europe Different?’, The Economics of Transition, vol. 11 (1), pp. 3–20.

Bristow, J. A. (1996) The Bulgarian Economy in Transition (Cheltenham U.K.: EdwardElgar Publishing).

Broadman, H. G. (2000) ‘Reducing Structural Dominance and Entry Barriers in RussianIndustry’, The World Bank, Policy Research Working Paper, no. 2330.

Brown, D. and J. S. Earle (2004) ‘Economic Reforms and Productivity EnhancingReallocation in Post-Soviet Transition’, Bonn: D.p. no. 1044, Institute for the Studyof IZA.

Brunetti, A. G. Kisunko and B. Weder (1997) ‘Institutional Obstacles to DoingBusiness’ (Washington: World Bank). Policy Research Working Paper no. 1759.

Bruno, M. (1993) ‘Stabilization and the Macroeconomics of Transition – HowDifferent is Eastern Europe’, Economics of Transition, vol. 1 (1), pp. 5–19.

Brzezinski, Z. (1993) ‘The Great Transformation’, The National Interest, no. 33, Fall1993, pp. 3–13.

Bugajski, J. (2004) Cold Peace: Russia’s New Imperialism (Portsmouth, NH: GreenwoodPublishers).

Buiter, W. H. (2000) ‘From Production to Accumulation’, Economics of Transition,vol. 8, no. 3, pp. 603–22.

Bunce, V. (1999) ‘The Political Economy of Post-Socialism’, Slavic Review, vol. 58, no. 4,winter 1999, pp. 756–93.

Bungs, D. (1998) The Baltic States: Problems and Prospects of Membership in the EuropeanUnion (Baden-Bden: Normas Verlagsgeselschaft).

Calvo G. and F. Coricelli (1993) ‘Output Collapse in Eastern Europe: The Role ofCredit’, IMF Staff Papers, vol. 40, no. 1, pp. 32–52.

Campos, N. and F. Coricelli (2002) ‘Growth in Transition: What We Know, WhatWe Don’t and What We Should’, Journal of Economic Literature, vol. XL, Sept. 2002,pp. 793–836.

Carrere d’Encausse, H. (1981) The Soviet Socialist Republics in Revolt (New York: HarperPress, Colophon Books).

Carothers, T. (2002) ‘The End of the Transition Paradigm’, Journal of Democracy, vol.13, no. 1, pp. 5–21.

Che, J. and G. Facchini (2004) ‘Dual Track Liberalization: With or Without Losers’,William Davidson Institute, University of Michigan, Working Paper.

Clague, C. and E. Reuser (eds) (1992) The Emergence of Market Economies in EasternEurope (Cambridge: Blackwell Publishers).

Clunies-Ross, A. and P. Sudar (1998) Albania’s Economy in Transition and Turmoil,1990–1997 (Aldershot, U.K.: Ashgate Publishers).

Chowdhury, A. (2003) ‘WTO Accession: What is in it for Russia’, William DavidsonInstitute, University of Michigan, Working Paper no. 595, July 2003.

Commander, S. and F. Coricelli (eds) (1995) Restructuring and The Labor MarketingEastern Europe and Russia (Washington, DC: World Bank).

Commander, S., A. Tolstopiatenko and R. Yemtson (1999) ‘Channels of Redistribution:Inequality and Poverty in Russian Transition’, The Economics of Transition, vol. 7,no. 2, p. 411ff.

Cottrell, R. and A. Ostrovsky (2001) ‘After the Oligarchs’, The Financial Times, 16 April2001.

Cottrell, R. (2003) ‘Putin’s Trap’, The New York Review, 4 Dec., p. 26–29.Czaba, L. (2004) ‘Transition in and Towards Europe: Economic Development and EU

Accession of Post-Communist States’, Zeitschrift fuer Staats und Europawissenschaften,vol. 2, no. 3.

Dabrowski, M. and R. Antzak (1996) ‘Economic Reform in Kyrgyzstan’, Warsaw: CASEStudies and Analyses no. 28.

De Loecker, J. and J. Konings (2003) ‘Creative Destruction Productivity and Growth inan Emerging Economy: Evidence for Slovenian Manufacturing’, Bonn: IZADiscussion Paper no. 971.

De Melo, M., C. Denizer and A. Gelb (1997) ‘From Plan to Market: Patterns ofTransition’, Chapter 1 in Macroeconomic Stabilization in Transition Economics, ed., byM. Blejer and M. Skereb (Cambridge: Cambridge University Press).

Deaton, A. (2004) ‘Measuring Poverty in Growing World’, Mimeo, Minnesota:Princeton University, Woodrow Wilson School, February 2004.

Dewatripont, M. and G. Roland (1992) ‘Economic Reform and Dynamic PoliticalConstrains’, Review of Economic Studies, no. 59, pp. 703–30.

Djankov, S. and P. Murrell (2002) ‘Enterprise Restructuring in Transition: AQuantitative Survey (2002)’, Journal of Economic Literature, vol. 40, no. 3, pp. 739–92.

Djankov, S. et al (2003) ‘The New Comparative Economics’, Journal of ComparativeEconomics, vol. 31, pp. 595–619.

Doyle, O. and J. Fidrmuc (2004) ‘Who is in Favour of EU Enlargement? Determiningof Support for EU Membership in the Candidate Countries Referenda’, CEPRDiscussion Paper no. 4273.

Durnev, L., K. Li, K. Morck and B. Yeung (2004) ‘Capital Markets and Capital Allocation:Implications for Economies in Transition’, Economics of Transition vol. 12(4), pp. 593–634.

Earl, J. and S. Estrin (1997) ‘After Voucher Privatization: The Structure of CorporateOwnership in Russian Manufacturing Industry’, Working Paper, London BusinessSchool.

298 Bibliography

Bibliography 299

Easterly, M. and S. Fischer (1995) ‘The Soviet Economic Decline’, World Bank EconomicReview, vol. 9, no. 3, pp. 341–71.

Elborgh Woytek (2003) ‘Of Openness and Distance: Trade Developments in theCommmonwealth of Independent States’, IMF Working Paper 03/207 (Washington:The International Monetary Fund).

Engerman, S. and K. Sokoloff (2003) ‘Institutional and Non-Institutional Explanationsof Economic Differences’, National Bureau of Economic Research, Working Paper,September 2003.

Everaert, Greetje M.M. (2004) ‘The Political Economy of Restructuring andSubsidisation: An International Perspective’, unpublished monography, LICOSCentre for Transition Economics, K.U. Leuven, Belgium, February, 49p.

European Bank for Reconstruction and Development (Annual 1994–2004) TransitionReport (London: EBRD).

Eyal, G., T. Szelenyi and E. Townsley (1999) Marking Capitalism without Capitalists(London: Verso Books).

Faccio, M. (2003) ‘Politically Connected Firms’, University of Notre Dame, MendozaCollege of Business, Mimeo.

Feige, E. (1997) ‘Underground Activity and Institutional Change: Productive,Protective, and Predatory Behaviour in Transition Economies’, in Nelson and Tillyand John Walker, eds., Post-Communist Political Economics (Washington, DC:National Academy Press).

Feige, E. and I. Urban (2004) ‘Estimating the Size and Growth of UnrecordedEconomic Activity in Transition Countries’, WPI paper.

Ferreira, Francisco H.G. (1999) ‘Economic Transition and the Distribution of Incomeand Wealth’, The Economics of Transition, vol. 7 (2), pp. 337ff.

Fidrmuc, J. (2000) ‘Political Support for Reforms: Economics of Voting in TransitionCountries’, European Economic Review, vol. 44, pp. 1491–513.

Fish, M. Steven (1999) ‘Post-communist Subversion: Social Science andDemocratization in East Europe and Eurasia’, Slavic Review, vol. 58, no. 4 (Winter1999) p. 799.

Fischer, S. and A. Gelb (1991) ‘The Process of Socialist Economic Transformation’,Journal of Economic Perspectives, vol. 5, no. 4, pp. 91–101.

Fischer, S., R. Sahay and C. Vegh (1996) ‘Stabilization and Growth in TransitionEconomies: The Early Experience’, Journal of Economic Prospectives, vol. 10, Spring,pp. 45–66.

Flemming, J. and J. Micklewright (2000) ‘Income Distribution, Economic Systems andTransition’, UNICEF Innocent Occasional Paper 70.

Forbes (2005) ‘Special Issue: Billionaires’, 28 March, 2005.Frankel, J. and D. Romer (1999) ‘Does Trade Cause Growth’, American Economic Review,

vol. 89, no. 3, pp. 379–95.Freeland, C. (2000a) ‘Get-Rich-Quick is Now the Big Idea Shaping Russia’, Financial

Times, 28 May, 2000.Freeland, C. (2000b) Sale of the Century: Russia’s Wild Ride from Communism to

Capitalism (New York: Crown Business).Frydman, R., C. Gray, M. Hessel and A. Rapaczyshi (1999) ‘When does Privatization

Work?’ Quarterly Journal of Economics, vol. 114., no. 4, pp. 1153–91.Frye, T. (2003) ‘Markets, Democracy and New Private Business in Russia’, Post-Soviet

Affairs, vol. 19, no. 1, pp. 24–45.Frye, T. and A. Shleifer (1997) ‘The Invisible Hand and the Grabbing Hand’, American

Economic Review, vol. 97, no. 2.Gaidar, Y. (2003) State and Evolution: Russia’s Search for Free Market (Seattle: University

of Washington Press).

Garner, T. (2001) ‘Symposium on Economic Performance and Income Distribution,Introduction’, Economics of Transition, vol. 9(2), pp. 355–8.

Gerth, J. (2004) ‘Lukoil Begins Efforrt to Build Ties Globally’, New York Times,2 October 2004.

Gillingham J. (2003) European Integration. 1950–2003: Super state or New MarketEconomy (Cambridge: Cambridge University Press).

Gleason, G. (1997) ‘Uzbekistan: The Politics of National Independence’, in Bremmerand Taras (1997).

Goldman, M. (2003) The Privatization of Russia: Russian Reform Goes Awry (New York:Routledge).

Grun, C. and S. Klasen (2001) ‘Growth, Income Distribution and Well-Being inTransition Countries’, Economics of Transition, vol. 9(2), pp. 359–94.

Gustafsson, B. and L. Shi (2001) ‘The Effects of Transition on the Distribution ofIncome in China’, Economics of Transition, vol. 9(13), pp. 593–617.

Halpern, L. (1998) Hungary – Towards a Market Economy (Cambridge: CambridgeUniversity Press).

Haltiwanger, S. and Vodopivec, M. (2002) ‘Gross Worker and Job Flows in a TransitionEconomy: An Analysis of Estonia’, Labour Economics, vol. 9, no. 5, pp. 601–30.

Haltiwanger, J., H. Lehmann and K. Terrell (2003) ‘Job Creation and Job Destributionin Transition Countries’, The Economics of Transition, vol. 11(2), pp. 205–19.

Handelman, S. (1994) Comrade Criminal (London: Penguin Group).Hanson, E. S. and S. K. Jeffrey, ‘The Weimar/Russia Comparison’, Post Soviet Affairs,

vol. 13, no. 3 (1997) pp. 252–83.Harasymiw, B. (2002) Post-Communist Ukraine (Edmonton: Canadian Institute of

Ukrainian Studies Press).Hashi, I., Kozarzewski, P. and A. Radygin (2004) ‘The Legal Frameworks for Effective

Corporate Governance and Evolving Ownership Structure in Privatized Companiesin Poland and Russia’, Staffordshire University Business School, Mimeo, October2004.

Havrylyshyn, O. (1994) ‘The Great Transition Robbery’, Financial Times, 22 July 1994.Havrylyshyn, O. (1995a) Economic Transformation: The Tasks Still Ahead, IMF,

Washington, D.C.: The Per Jacobson Lecture, October 1995.Havrylyshyn, O. (1995b) ‘How Patriarchs and Rent-Seekers Hijack The Transition to a

Market Economy’, Perspectives on Contemporary Ukraine, Harvard University,Ukrainian Research Institute, Issue of June–July 1995.

Havrylyshyn, O. (1997) ‘Economic Reform in Ukraine: Late is Better than Never, ButMore Difficult’, in M. Blejer and M. Skreb (eds), Macroeconomic Stabilization inTransition Economies (Cambridge: Cambridge University Press).

Havrylyshyn, O., T. Wolf, J. Berengaut, M. Castello-Branco, R. van Rooden, V. Mercer-Blackman (1999) Growth Experience in Transition Countries, 1990–98, IMF OccasionalPaper No. 184 (Washington: The International Monetary Fund).

Havrylyshyn, O. and D. McGettigan (2000) ‘Privatization in Transition Countries’,Post-Soviet Affairs, vol. 16, no. 3, pp. 257–86.

Havrylyshyn, O. (2001) ‘Recovery and Growth in Transition: A Decade of Evidence’,IMF Staff Papers, vol. 48, Special issue: ‘Transition Economies: How MuchProgress?’

Havrylyshyn, O. and S. Nsouli (eds) (2001) A Decade of Transition (Washington:International Monetary Fund).

Havrylyshyn, O. and R. van Rooden (2003) ‘Institutions Matter in Transition, But sodo Policies’, Comparative Economic Studies, vol. 45(1), pp. 2–24.

300 Bibliography

Bibliography 301

Havrylyshyn, O. (2004) ‘The Impact of EU Enlargement on Countries Beyond the NewFrontiers’, ch. 13 in Michael Landesmann and Dariusz Rosati eds, Shaping the NewEurope: Economic Policy Challenges of European Union Enlargement, London.

Heenan, P. and M. Lamontagne (1999a) The Central and Eastern Europe Handbook.London; Chicago: Fitzroy Dearborn, 1999.

Heenan, P. and M. Lamontagne (1999b) The CIS Handbook. London; Chicago: FitzroyDearborn, 1999.

Hellman, J. (1998) ‘Winners Take All: The Politics of Partial Reform in Post-Communist Nations’, World Politics, vol. 50, no. 2, pp. 203–34.

Hellman, J. and M. Schankermann (2000) ‘Intervention, Corruption and Capture’,Economics of Transition, vol. 8, no. 3, pp. 295–326.

Hellman, J., G. Jones and D. Kaufmann (2003) ‘Seize the State, Seize the Day: StateCapture and Influence in Transition economies’, Journal of Comparative Economics,vol. 91, pp. 751–73.

Hellman, J. and D. Kaufmann (2003) ‘The Inequality of Influence’, paper presented toStanford Corruption Workshop, 30–31 January 2003.

Heybey, B. and P. Murrell (1999) ‘The Relationship Between Economic Growth and theSpeed of Liberalization During Transition’, Policy Reform, vol. 3, pp. 121–37.

Holzman, F. (1992) ‘The Soviet Economy: Past, Present and Future’, Foreign PolicyAssociation Pamphlet no. 200, New York: Foreign Policy Association, September 1982.

International Finance Corporation (IFC) (1999) The Study of Small Business in Ukraine(Kyiv: IFC Mimeo).

IMF (1999) Republic of Belarus: Recent Economic Developement, IMF staff country reportno. 99/143, December 1999.

IMF (2000) World Economic Outlook. October 2000: Focus on Transition Economies(Washington, DC: IMF).

Irwin, D. A. (1995) ‘The GATT in Historical Perspective’, American Economic Review,vol. 85, 1995.

Isajiw, W. (2004) ‘Civil Society in Ukraine: Toward a Systematic Sociological ResearchAgenda’, paper persented at University of Ottawa conference, Understanding theTransformation of Ukraine, 15–16 October 2004.

Jack, W. (2002) ‘Institutional Design and the Closure of Public Facilities in TransitionEconomies’, Economics of Transition, vol. 10(3), pp. 619–35.

Jackson, J., J. Klich and K. Poznanska (2003) ‘Economic Transition and Elections inPoland’, Economics of Transition, vol. 11, no. 1, pp. 41–66.

Johnson, S. McMillan and C. Woodruff (2000) ‘Entrepreneurs and the Ordering ofInstitutional Reform: Poland, Slovakia, Romania, Russia and Ukraine’, Economics ofTransition, vol. 8(1), pp. 1–36.

Johnson, S. and A. Subramanian (2005) ‘Aid, Goverment and the Political Economy:Growth and Institutions’. Paper presented at Seminar on Foreign Aid andMacroeconomic Management, Maputo, 14–15 March 2005.

Jones, D., L. Cheng and A. Owen (2003) ‘Growth and Regional Inequality in ChinaDuring the Reform Era’, Working Paper no. 561, William Davidson Institute,University of Michigan, May, 31p.

Jowitt, K. (1996) ‘Dizzy with Democracy’, Problems of Post Communism, vol. 43, no. 1,January–February 1996.

Judah, T. (2005) ‘The Waiting Game in the Balkans’, New York Review of Books, vol. LII,no. 13, pp. 46–9.

Kaminski, B. (ed.) (1996) Economic Transformation in Russia and the New States of Eurasia(London: M.E. Sharpe).

Karatnycky, A., A. Motyl and B. Shor (1997) Nations in Transit: Civil Society, Democracyand Markets in East Central Europe and the Newly Independent States (Washington:Freedom House).

Karatnycky, A., Alexander Motyl and A. Schnetzer (2003) Nationalism in Transit.Democratization in East Central Europe and Eurasia (Washington: Freedom House,Transition Publishers).

Katz, B. and J. Owen (1993) ‘Privatization: Chossing the Optimal Time Path’, Journal ofComparative Economics, vol. 17, no. 4, pp. 715–36.

Katz, B. and J. Owen (2000) ‘Choosing Between Big Bang and Gradual Reform: AnOption Price Approach’, Journal of Comparative Economics, vol. 28, pp. 95–107.

Katz, B. and J. Owen (2002) ‘Voucher Privatization: A Detour The Road to Transition’,Economics of Transition, vol. 10(3), pp. 553–83.

Kaufmann, D., A. Kraay and p. Zoido-Lobaton (1999) ‘Government Matters’, WorldBank Policy Research Paper no. 2196 (Washington, DC: World Bank, 1999).

Kaufmann, D. and A. Kaliberda (1996) ‘Integrating the Unofficial Economy Into theDynamics of Post Socialist Economics’, World Bank Policy Research Working Paperno. 1691 (Washington, DC: World Bank).

Kaufmann, D., A. Kraay and M. Mastruzzi (2003) ‘Governance Matters III: GovernanceIndicators for 1996–2002’ (Washington, DC: The World Bank).

Keane, M. and E. Prasad (2002) ‘Inequality Transfer and Growth, New Evidence fromthe Economic Transition in Poland’, The Review of Economic and Statistics, vol. 8,no. 2, pp. 924–34.

Kemme D. and Claire E. Gordon (eds) (1990) The End of Central Planning?: SocialistEconomies in Transition: The Cases Of Czechoslovakia, Hungary, China, and the SovietUnion (New York: Institute for East-West Studies).

Kim, Bying-Yon and J. Pirtilla (2004) ‘Political Constrain of Economic Reform:Empirical Evidence from Post-Communist Transition in the 1990s’, Helsinki, BOFITWorking Paper.

King, Lawrence (2003) ‘Explaining Post-Communist Economic Performance’, WilliamWorking Paper no. 59, 2003.

Kitschelt, L. (1995) ‘Formation of Party Cleavages in Post-Communist Democracies’,Party Politics, vol. 1, no. 4, pp. 447–72.

Klaus, V. (1995) ‘The Ten Commandments of a Systematic Reform Revisited’,Washington, D.C., IFC Fourth Annual Lecture.

Kolodko, G. (1999) ‘Income Policy, Equity Issuers and poverty Reduction in TransitionEconomies’. Deutsche Stiftung fur Internationale Entwocklung (www.dsded.de).

Kolodko, Grzegorz W. (2000a) From Shock to Therapy: The Political Economy of Post-Communist Transformation (Oxford: Oxford University Press).

Kolodko, Grzegorz W. (2000b) Post-Communist Transition: The Thorny Road (Rochester:The University of Rochester Press).

Kolodko, Grzegorz W. (2004) ‘Institutionalism, Politicies and Growth’, Warsaw:Center for Transformation, Integration and Globalization Economic Research.(www.tiger.edu.pl).

Kopstein, J. and D. Reilly (2000) ‘Geographic Diffusion and Transformation of ThePost-Communist World’, World Politics, vol. 53, no. 1, October 2000, pp. 1–37.

Kornai, J. (1994) ‘Transformational Recession: The Main Causes’, Journal ofComparative Economics, vol. 19 (August 1994) pp. 34–63.

Kornai, J., E. Maskin and G. Roland (2003) ‘Understanding the Soft BudgetConstraint’, Journal of Economic Literature, vol. XLI, pp. 1095–136.

Knobl, A. and R. Haas (2003) ‘IMF and the Baltics: A Decade of Cooperation’, IMFWorking Paper, Washington, DC, December 2003.

302 Bibliography

Bibliography 303

Kraft, Evan (2000) ‘Independence and Macroeconomic Stabilization in Ex-Yugoslaviaand Former Soviet Republics’. Pittsburgh: University of Pittsburgh Centre forInternational Studies, Papers, no. 1502.

Krueger, A. (1974) ‘The Political Economy of Rent-Seeking Society’, The AmericanEconomic Review, vol. 64, no. 3, January 1974, pp. 291–303.

Krueger, A. (1992) ‘Institutions in the New Private Sector’, Ch. 12 in ChristopherClague and Gordon Rouser, eds., The Emergence of Market Economies in Eastern Europe.(Cambridge, MA: Blackwell Publishers), pp. 219–26.

Krueger, A. (1999) ‘Why Crony Capitalism is Bad for Economic Growth’, draft paper,Stanford Hoover Institute.

Krugman, P. (1994) ‘The Myth of Asia’s Miracle’, Foreign Affairs, vol. 73, no. 6,pp. 62–78.

Kubicek, P. (ed.) (2003) The European Union and Democratization (London: Routledge).Kuzio, T. and A. Wilson (1994) Ukraine: Perestroika to Independence (London: Macmillan

Press).Kuzio, T. (1997) Ukraine under Kuchma: Political Reform, Economic Transformation and

Security Policy in Independent Ukraine (London: Macmillan).Kuzio, T. (2003) ‘EU and Ukraine: A Turning Point in 2004?’, Occasional Paper no. 47,

European Union Institute for Security Studies, Paris, March 2003.Kuzio, T. (2005) ‘Ukraine’s Orange Revolution: The Opposition’s Road to Success’,

Journal of Democracy, vol. 16, no. 2, pp. 117–130.Laar, M. (2002) Little Country that Could (London: Centre for Research into Post-

Communist Economies).Lane, T. (2002) Lithuania: Stepping Westward (New York: Routledge).Lehmann, H., J. Wadsworth and A. Acquiti (1999) ‘Crime and Punishment:

Employment, Wages, and Wage Arrears in Russia’, Journal of Comparative Economics,vol. 27, no. 4, pp. 515–617.

Levits, E. (2001) ‘That Latvian and European Association Agreement and theTransformation of the Latvian Legal System’, in J. Talavs, ed., The Baltic States atHistorical Crossroads (Riga: Latvian Academy of Sciences).

Levitsky, S. and L. A. Way (2003) ‘Ties that Bind? International Linkage andCompetitive Authoritarian Regime Change In Africa, Latin America and Post-Communist Eurasia’, for Annual Meeting of the American Political ScienceAssociation, Philadelphia, PA.

Lieven, A. (1993) The Baltic Revolution (New Haven, CT: Yale University Press).Lipset, S.M. (1959) ‘Some Social Requisites of Democracy: Economic Development and

Political Legitimacy’, American Political Science Review, vol. 53, no. 1, pp. 69–105.Maniokas, K. (1999) ‘Management of European Union Affairs at the Executive

Level Lithuania, Latvia, and Estonia’, in Malfliet and Keygnaert and in Beselaere(2001).

Matthews, M. (1986) Poverty in the Soviet Union (Cambridge: Cambridge University Press).Mayhew, A. (1998) Recreating Europe: The European Union’s Policy Towards Central and

Eastern Europe (Cambridge: Cambridge University Press).McAuley, A.M. (1979) Economic Welfare in the Soviet Union (Madison: University of

Wisconsin Press).McCloskey, D. (1987) The Writing of Economics (New York: Macmillan Publishing).McFaul, M. (2002) ‘The Fourth World of Democracy and Dictatorship; Non-

Cooperative Transitions in the Postcommunist World’, World Politics, vol. 54,January 2002, pp. 212–44.

Megginson, W. and J. Netter (2001) ‘From State to Market: A Survey of EmpiricalStudies on Privatization’, Journal of Economic Literature, vol. 39, no. 2, pp. 321–89.

Mercer-Blackman, V. and A. Unigovskaya (2002) ‘Compliance with IMF ProgramIndicators and Growth in Transition Economies’, IMF, Working Paper WP/00/47(Washington: The International Monetary Fund).

Micklewright, J. (1992) ‘Education, Inequality and Transition’, The Economics ofTransition, vol. 7(2), pp. 343–76.

Milanovic, B. (1998) Income Inequality and Poverty During the Transition from Planned toMarket Economy (Washington, DC: The World Bank).

Milanovic, B. (1999) ‘Explaining the Increase in Inequality During Transition’, TheEconomics of Transition, vol. 7, no. 2, pp. 299–342.

Moers, L. (1999) ‘How important are Institutions for Growth in Transition Countries’,Tinbergen Institute, Amsterdam, Discussion Paper 99–804/2.

Morck, R., D. Wolfenzon and B. Yeung (2005) ‘Corporate Governance EconomicEntrenchment and Growth’, Journal of Economic Literature, vol. XLIII, Sept. 2005,pp. 657–722.

Motyl, A. (1997) ‘Structural Constraints and Starting Points: The Logic of System Changein Ukraine and Russia’, Comparative Politics, vol. 29, no. 4, July 1997, pp. 133–97.

Motyl, A. (2004) ‘Communist Legacies and Post-Communist Tragectories: Democracyand Dictatorship in the Former Soviet Union and Central Europe’, Ch. 2 in Brudny,Yitzhak M. eds., Restructuring Post Communist Russia (Cambridge: CambridgeUniversity Press), pp. 52–67.

Munich, D., J. Svejnar and K. Terrell (2002) ‘Returns to Human Capital Under theCommunist Wage Grid and During the Transition to a Market Economy’, Universityof Michigan, William Davidson Institute, Working Paper no. 272a, January 2002.

Murrell P. (1996) ‘How Far Has the Transition Progressed?’ Journal of EconomicPerspectives, vol. 10, no. 2, pp. 25–44.

Mussa, M. (2002) Argentina and the Fund: from Triumph to Tragedy (Washington, DC:Institute of International Economics).

Nissinen, M. (1999) Latvia’s Transition to a Market Economy: Political Determinants ofEconomic Reform Policy (London: MacMillan Press).

North, D. (1990) Institutions, Institutional Change, and Economic Performance (New York:Cambridge University Press).

North, D. C. (1993) ‘Institutions and Credible Commitment’, Journal of Institutionaland Theoretical Economics, vol. 149(1), pp. 11–23.

North, D. C. (1995) Institutions, Institutional Change and Economic Performance (NewYork: Cambridge University Press), p. 141.

Notten, G. (2004) ‘Poverty and Consumption Insurance in Russia’, paper presented atBank of Finland, workshop on Transition, 1–2 April 2004, Helsinki.

Odling-Smee, J. and P. Thomsen (2003) ‘Growth Without Reform Will Not HelpRussia’, Financial Times, 19 August.

Odling-Smee, J. (2004) ‘The IMF and Russia in the 1990’s’, IMF Working Paper 04/155(Washington: The International Monetary Fund).

O’Donnell, G. and P. Schmitter (1986) Transitions from Authoritarian Rule: TentativeConclusions about Uncertain Democracies (Baltimore: Johns Hopkins University Press).

O’Driscoll, G., E. Feulner and M. O’Grady (2000) 2003 Index of Economic Freedom,Heritage Foundation, Washington, DC.

O’Rourke, T. (1995) The Baltic Republics in Transition (Edinburgh: Scottish FinancialEnterprise).

Olson, M. (1993) ‘Why is Economic Performance Even Worse After Communism isAbandoned?’, Fairfax, Virginia (Ninth Annual Lecture in the Virginia PoliticalEconomy Lecture Series), 31p.

304 Bibliography

Bibliography 305

Olters, Jan-Peter (2001) ‘Modeling Politics with Economic Tools: A Critical Survey ofthe Literature’, IMF, Working Paper, 01/10, 61p.

Owen, D. and D. Robinson (2003) Russia Rebounds (Washington, DC: The InternationalMonetary Fund).

Palánkai, T. (2003) The Economics of European Integration. Budapest: Akadémiai Kiadó.Pabriks, A. and A. Purs (2002) Latvia: The Challenges of Change (New York: Routledge).Papadimitriou, D. (2002) Negotiating the New Europe. The European Union and Eastern

Europe (Aldershot: Ashgate Publishing).Park, A. (1995) ‘Russia and the Estonian Security Dilemas’, Europe-Asia Studies, vol. 47,

no. 1, pp. 27–45.Papageorgiou, D., A. Choksi and M. Michaely (eds) (1991) Liberalizing Foreign Trade in

Developing Countries (London: Basil Blackwell).Pisuke, H. (2001) ‘The Process of Bringing Estonia’s Legal System into Conformity with

the European Union’, in J. Talavs, ed., as in Levits (2001).Pirttilá, J. (2001) ‘Fiscal Policy and Structural Reforms in Transition Economies:

An Empirical Analyses’, Economies in Transition, vol. 9(1), pp. 29–52.Plúmper, T., C. Schneider and U.F. Troeger (2004) ‘The Politics of EU Eastern

Enlargement Evidence from the Heckann Selective Model’, Public Choice and PoliticalEconomy Working Paper Series, vol. 3, no. 16, 16 July 2004.

Polischuk, L. and A. Savvateev (2004) ‘Spontaneous (non) emergence of propertyrights’, Economics of Transition, vol. 12, no. 1, pp. 103–27.

Pomfret, R. (1997) ‘Growth and Transition: Why Has China’s Performance Been SoDifferent?’, Journal of Comparative Economics, no. 25, pp. 422–40.

Pomfret, R. (2000) ‘The Uzbek Model of Economic Development, 1991–1999’,Economics of Transition, vol. 8(3), pp. 733–48.

Popova, M. (2004) ‘Income Inequality and Poverty of Economies in Transition’, paperpresented at Bank of Finland, BOFIT Workshop on Transition, 2–3 April 2004,Helsinki.

Pridham, G. (2000) The Dynamics of Democratization: A Comparative Approach (Londonand New York: Continuum), 2000.

Przeworski, A. (1991) Democracy and the Market: Political and Economic Reforms inEastern Europe and Latin America (Cambridge: Cambridge University Press).

Puglisi, R. (2003) ‘The Rise of the Ukrainian Oligarchs’, Democratization, vol. 10, no. 3(Autumn), pp. 99–123.

Qian, Y. and C. Xu (1993) ‘Why China’s Economic Reforms Differ: The M-formHierarchy and Entry Expansion of Non-state Sector’, Economics of Transition,vol. 1(2), pp. 135–70.

Radulescu, R. and D. Barlow (2002) ‘The Relationship Between Policies and Growth inTransition Countries’, Economics of Transition, vol. 10(3), pp. 719–45.

Rajan, R. and L. Zingales (2003a) ‘The Great Reversals: The Politics of Financial Devel-opment in the Twentieth Century’, Journal of Financial Economics, vol. 69, pp. 5–50.

Rajan, R. and L. Zingales (2003b) Saving Capitalism From the Capitalists: Unleashing thePower of Financial Markets to Create Wealth and Spread Opportunity. Crown Business,2003, 369p.

Reddaway, P. and D. Glinski (2001) The Tragedy of Russia’s Reforms: Market BolshevismAgainst Democracy (Washington, DC: United States Institute of Peace Studies).

Resmini, L. (2000) ‘The Determinants of Foreign Direct Investment in the CEECs: NewEvidence From Sectoral Patterns’, Economics of Transition, vol. 8(3), pp. 665–89.

Roland, G. and T. Verdier (1999) ‘Transition and the Output Fall’, Economics ofTransition, vol. 7(1), 1999, pp. 1–28.

Ronnas, P. (1996) ‘Private Entrepreneurship in the Nascent Market Economy ofVietnam: Markets and Linkages’ in Reforming Asian Socialistm, The Growth of MarketInstitutions, B. Naughton and S. Haggard (eds.) (Ann Arbor: University of MichiganPress).

Rose, A. (2002a) ‘Do We Really Know that the WTO Increases Trade?’, NBER WorkingPaper no. 9273, October 2002.

Rose, A. (2002b) ‘Do WTO Members Have a More Liberal Trade Policy?’, NBERWorking Paper no. 9347, November 2002.

Rose, A. K. (2004) ‘Does the WTO Make Trade More Stable?’, NBER Working Paperno. 10207, January, Cambridge, 2p.

Rzonca, A. and P. Cizkowicz (2003) ‘A Comment on the Relationship BetweenPolicies and Growth in Transition Countries’, Economics of Transition, vol. 11(4),pp. 743–48.

Sachs, J. (1994) Poland’s Jump to a Market Economy (Cambridge, Mass: MIT Press).Sachs, J. and W. Thye Woo (1994) ‘Structural Factors in the Economic Reforms of

China, Eastern Europe, and the Former Soviet Union’, Economic Policy, April,pp. 102–44.

Sachs, J. and A. Warner (1996) ‘Achieving Rapid Growth in the Transition Economies ofCentral Europe’, HIID Development Discussion Paper no. 544 (Cambridge, Harvard:Institute for International Development), pp. 1–58.

Sandbrook, R. (2000) Closing The Circle: Democratization and Development in Africa.(Toronto: Between The Lines).

Schimmelfennig, F. (2003) The EU, NATO, and the Integration of Europe: Rules andRhetoric (Cambridge: Cambridge University Press).

Schneider, F. and R. Klingalmair (2004) ‘Shadow Economies Around the World: WhatDo We Know?’ IZA, Discussion Paper no. 1043, Bonn: March 2004.

Schroeder, G. (1996) ‘Economic Transformation in the Post Soviet Republics’, ch. 2 inB. Kaminski ed., Economic Transition in Russia and the New States (London:M.E. Sharpe).

Sen, A. (1999) Development as Freedom (New York: Anchor Books).Shen, R. (1997) The Restructuring of Romania’s Economy. A Paradigm of Flexibility and

Adaptability (Westport, CT: Praeger Publishers).Shen, R. (1996) Ukraine’s Economic Reform: Obstacles, Errors, Lessons (Westport, CT:

Praeger Publishers).Shen, R. (1994) Restructuring the Baltic Economies: Disengaging Fifty Years of Integration

With the USSR (Westport, CT: Praeger Publishers).Shen, R. (1993) Economic Reform in Poland and Czechoslovakia. Lessons in Systemic

Transformations (Westport, CT: Praeger Publishers).Shevtsova, L. (1999) Yeltsin’s Russia: Myths and Realities (Washington, DC: Carnegie

Endowment for International Peace).Shleifer, A. (1997) ‘Establishing Property Rights’, Proceedings of the World Bank Annual

Conference on Development Economics of 1996 (Washington, DC).Shleifer, A. and R. Vishny (1998) The Grabbing Hand: Government Pathologies and Their

Cures (Cambridge: Harvard University Press).Shleifer, A. and D. Treisman (2004) ‘A Normal Country’, Foreign Affairs, May, pp. 20–39.Shulman, S. (2003) ‘The Role of Economic Performance in Ukrainian Nationalism’,

Europe-Asia Studies, vol. 55, no. 2., pp. 217–39.Sonin, K. (2003) ‘Why the Rich May Favor Poor Protection of Property Rights’, Journal

of Comparative Economics, vol. 31(4), pp. 718–31.Spagat, M. (2002) ‘Human Capital, Growth and Inequality in Transition Economies’,

Journal of Comparative Economics, San Diego: Dec 1999, vol. 27(4), p. 618.

306 Bibliography

Bibliography 307

Stern, N. (1997) ‘The Transition in Eastern Europe’, Ch. 2 in Salvatore Zecchini, ed.,Lessons From the Economic Transition (Paris: OECD).

Stiglitz, J. (1999) ‘Winter Reform: Ten Years of Transition’, The World Bank EconomicReview, September 1999.

Stillman, S. and D. Thomas (2004) ‘The Effect of Economic Crises on NutritionalStatus: Evidence from Russia’, IZA Discussion Paper no. 1092 (Bonn: Institute forthe Study of Labor, IZA).

Subramanian, A. and Shang-Jin Wei (2003) ‘The WTO Promotes Trade, Strongly butUnevenly’, NBER Working Paper no. 10024, October 2003.

Sukiassyn, G. (2004) ‘Inequality and Growth: What Does the Transition EconomyData Say?’, paper presented at Bank of Finalnd Workship on Transition, 2–3 April2004, Helsinki.

Svejnar, J. (2002) ‘Transition Economies’ Performance and Challenges’, Journal ofEconomic Perspectives, vol. 16, no. 2.

Székely, István P. and D. Newman (1992) Hungary: An Economy in Transition(Cambridge: Cambridge University Press).

Thomas, V. (1991) Restructuring Economies in Distress: Policy Reform and the World Bank(Washington, DC: The World Bank).

Tikhomirov, Vladimir I. (2000) The Political Economy of Post Soviet Russia (London:Macmillan Press).

Tolchitskaya, I. and L. Vinhas de Souza (2004) ‘Trade Integration in Eastern Europe:Russia and The European Union’, Paper presented at UN Economic CommissionSpring Conference 2005, Draft August 2004.

Tsygankov, A. (2002) Pathways After Empire: National Identity and Foreign EconomicPolicy in the Post Soviet World (Lanham: Rowman & Littlefield Publishers).

Ul Hague, N. and M. S. Khan (1998) ‘Do IMF-supported Programs Work? A Survey ofthe Cross-Country Empirical Evidence’, IMF, Working Paper, WP/98/169(Washington, DC: The International Monetary Fund).

UNICEF (1999) After the Fall: The Human Impact of Ten Years of Transition (Florence:UNICEF).

United Nations Development Programme (1998) Poverty in Transition (New York:UNDP Regional Bureau for Europe and the CIS).

Vinod, T., (ed.) (1991) Restructuring Economics in Distress: Policy Reform and the WorldBank (New York: Oxford University Press).

Van Zon, H. (1998) The Political Economy of Independent Ukraine (Houndsmills: PalgraveMacmillan Publishers).

Wan, G.H. (2002) ‘Income Inequality and Growth in Transition Economies’, WIDERDiscussion Paper no. 2002/14 (Helsinki: UNU/WIDER).

Way, L. (2005) ‘Ukraine’s Orange Revolution: Kuchma’s Failed Authoritarianism’,Journal of Democracy, vol. 16, no. 2, pp. 131–45.

Weder, B. (2001) ‘Institutional Reform in Transition Economics: How Far Have TheyCome?’ IMF Working Paper, WP/01/114 (Washington, DC: The InternationalMonetary Fund).

Williamson, J. (ed.) (1993) The Political Economy of Policy Reform (Washington, DC:Institute for International Economics).

Wolczuk, K. (2004) ‘Ukraine’s European Choice’, Policy Brief, Centre for EuropeanReform, London.

Wolf, H. (1997) ‘Transition Strategies Choices and Outcomes’, Working Paper, NewYork University Stern School of Business.

World Bank (1996) World Development Report 1996: From Plan to Market(Washington, DC).

World Bank (2000) Transition: The First Ten Years. Analysis and Lessons For Eastern Europeand The Former Soviet Union (Washington, DC: World Bank).

World Bank (2002) Building Institutions for Markets (Washington, DC: World Bank).World Bank (2004) From Transition to Development: A Country Economic Memorandum for

the Russian Federation, Draft Document of the World Bank, Washington, DC: April(2004).

Wöhrmann, F. (2000) Economic Discourse in Uzbekistan: The Perception of EconomicChange Between Market Principles and Social Traditions (Verlag fúr Entwicklungspolitik).

Yalcin, R. (2002) The Rebirth of Uzbekistan: Politics, Economy and Society in the Post-SovietEra (New York: Ithaca Press), 2002.

Yavlinsky, G. (2003) ‘The De-Modernization of Russia: Economic and PoliticalChallenges and Prospects’ (Moscow: unpublished paper).

Zakaria, F. (2003) The Future of Freedom: Illiberal Democracy at Home and Abroad (NewYork: W.W. Norton).

Zecchini, S. (ed.) (1991) The Transition To a Market Economy (Paris: OECD).Zimmer, K. (2004) ‘From ‘Roving’ to Stationary Bandits? Financial Industrial Groups

in the Post-Soviet Space – An Eastern Ukrainian Case Study’, Paper prepared forWarsaw Special Convention, Association for the Study of Nationalities, Warsaw,18–22 July 2004.

Zinnes, C. and J. Sachs (2001) ‘Bench Marking Comparativeness in TransitionEconomies’, Economics of Transition, vol. 9(2), pp. 315–53.

Zinnes, C., Y. Eilat and J. Sachs (2001) ‘The Gains from Privatization in TransitionEconomies: Is Change of Ownership Enough?’ IMF Staff Papers, vol. 48 (SpecialIssue), pp. 146–70.

308 Bibliography

Index

309

Accession to EU, 203–4Acquis Communautaire, see European

UnionAgricultural sector, 7, 79Akaev, Askar, 186–7, 195, 197, 225Albania, 29, 51, 66, 68, 72, 89, 128, 130,

137, 144, 163–5, 171, 206, 216,222–4, 263, 270

Anti-trust legislation, 148, 199–201, 249Armenia, 36, 66, 68, 87, 103, 108, 110,

112, 115, 128, 130, 135, 137, 144,156, 169, 175, 193, 195, 197, 240,262, 269

Aslund, Anders, 23, 29, 39, 52–3, 68,82–6, 120, 148, 235, 241–4, 287 n.4,288 nn.4, 5, 12, 292 n.2, 293 n.22

Association Agreements, 203–5see also Europe Agreements

Azerbaijan, 51, 54, 68, 71, 87–8, 98, 108,112, 115, 137, 168, 171, 175, 188,197, 262

Balcerowicz, Leszek, 19, 27, 45, 128,151–2, 172, 283 n.11

Baltics, 2, 10economic performance, 53–4, 69, 71,

81–2, 88EU membership, 3, 6, 29, 173–4,

209–10, 221–2, 225–30, 268institutions, 36–7, 57, 60, 65, 128,

186, 284 n.15nationalist sentiment, 125–6, 128,

173–5social costs, 93, 97, 100, 102, 105–6,

108, 112, 114, 116–17transition progress, 2, 9–10, 20, 24,

50–1, 73–4, 90, 128, 137, 145,158, 162, 165, 195, 258, 260, 272,280 n.12

see also Estonia; Latvia; LithuaniaBarnes, Andrew, 29, 32, 41, 134, 181,

236, 241, 287 n.2, 288 n.4, 292 n.8Belarus, 8, 28, 44, 47, 51, 53–5, 61, 84,

87, 94, 98, 109, 112, 123, 125, 128,138, 156, 158, l63, 166, 167–9, 171,

175, 182, 192, 197, 207, 212–13,215–16, 218, 234, 251–2, 260–1,263, 267

Berlin, Isaiah, 123, 282 n.1Berlin Wall, 1, 2, 5, 151, 203

as date of transition start, 8Big-Bang, see reform strategyBosnia and Hercegovina, 50, 60, 89, 111,

174, 206, 216, 263see also Yugoslav Federation

Blanchard, Olivier, 16–17, 23–4, 27, 63,277 n.3

Buiter, Willem, 21, 40–1, 133, 235, 241,246, 279 n.6, 292 n.1

Bulgaria, 25, 39, 44, 53, 58, 63, 66, 82,89, 100, 107, 109, 110, 132, 137,144–5, 163–5, 174, 179–80, 185,193, 195, 198, 205–6, 209, 211, 214,222–3, 233, 247, 254, 260–3, 268–70

Campos, Nauro, 21, 23–4, 26–7, 62–3,89–91, 278 n.15, 279 n.1, 280 n.14

Captured states, see State captureCarothers, Thomas, 42–3, 57Central Europe, 2, 10, 27, 33, 97, 100,

102, 105–6, 109, 112, 117, 135, 145,173, 177, 183, 279 n.4

economic performance, 52–4, 63, 69,81–6, 186, 281 n.11

EU membership, 3, 6, 7, 155, 203–4,208–9, 220–3, 233, 272–3, 289 n.3

institutions, 36–7, 42, 47, 55–9, 178,192, 196, 241, 280 n.13

transition progress, 2, 8, 9, 30, 35, 50, 72–3, 158–64, 255–62, 272,282 n.29

China, 7–8, 20, 22, 26, 38, 49, 79–80, 95, 101, 103, 117, 154–6, 174, 279 n.2, 280 n.1, 281 n.26, 282 n.2,285 n.2

Chubais, Anatolyi, 187, 293 n.14CISL, see Lagging reformersClientelism in Africa, 201–2Coase theorem, 40, 235, 241

see also Transition inevitable thesis

Collective action theory, 241Commonwealth of Independent States

(CIS), 3, 10, 33, 131, 135, 145, 156,177, 190, 196, 198, 210, 216, 233–4,273–4

economic performance, 52–4, 63,81–6, 186

growth surge after 1998, 67–9, 245–6institutions, 36–7, 40, 42, 55–9, 192,

235, 241, 274–5, 280 n.13social costs, 88, 93–4, 97–102, 104–8,

112, 114, 117transition progress, 36, 47, 50, 72–3,

158–64, 179, 255–62, 272Comparisons with non-transition

economies, 95, 102, 153, 156,173–4, 178, 191, 199–202, 226, 239,247, 249, 251–3, 267, 281, 291

Competitive or open, markets, 146–7,183–4, 191–3, 262

Copenhagen Criteria, 205, 291 n.27,see also European Union

Coricelli, Fabrizio, 21, 22, 23, 26–7,62–3, 89–91, 277 n.1, 278 n.15, 279 n.1, 280 n.14

Corruption, 36, 48, 64, 172, 185, 202,239, 244, 254

index, 202, 258; see also TransparencyInternational

tax bribes, 250, 273Croatia, 39, 44, 50, 55, 60, 63, 71, 82–3,

94, 98, 123, 128, 130, 137, 144–5,159–60, 161, 164, 168, 172, 174–5,185, 193, 195, 198, 205–6, 212, 214,222–4, 233, 240, 254, 260, 262–3,267–8, 270, 291 n.21

see also Yugoslav FederationCrony capitalists, 238

see also Robber baronsCzech Republic, 24, 29, 32, 35, 39, 44, 57,

65–6, 90, 115, 123, 128–30, 137, 165,167, 185, 205, 222, 259, 262, 268

Czechoslovakia, 44, 71, 141, 145, 186,204, 267, 271

see also Czech Republic; Slovakia

Delayed reformsand EU membership, 220–3and national first government, 168–72and nationalism, 173–5and social costs, 115

Democracy, 38, 45measures, 54–7relationship to markets, 11, 38, 42–6,

57–8see also Transitology paradigm

Distance from Brussels, see Geography

Economic elites in history, 148, 178,199–202, 236, 238–9

see also Oligarchs; Robber baronsEconomic performance, see Transition

progressEconomic rhetoric, 92Education, 113–15

see also Reform strategy; Social costsElite entrenchment, 199–201, 238–9

see also Economic elites; Transitionfrozen thesis

End of transition, see Transition is overEnergy sector, 179–80, 238Estonia, 29, 30, 39, 50, 60, 63, 89–90,

99, 107–8, 123, 137, 153, 155, 167,172, 204–5, 208–10, 212, 222–3,225–7, 229, 244, 259, 260, 262, 267, 269

Europe Agreements, 160, 203–5, 225,227, 229–30

see also Association AgreementsEuropean Bank for Reconstruction

and Development, 7, 10, 17, 21, 24, 45, 48–9, 51, 55, 58, 62, 87, 91, 104, 110, 114–15, 118, 137, 144, 162, 171, 174, 218, 242, 261,266, 270

see also Transition progressEuropean Neighbourhood Policy (ENP),

213, 215European Union (EU)

Acquis Communautaire, 35–6, 154–8,195, 205, 210–11, 228, 230

membership as reform discipline, 4,5–6, 10, 137, 141, 198, chapter 7,(203–25), 211–12, 225–30, 233,266–7

Faustian bargain, see Kravchuk, UkraineFinancial–Industrial Groups (FIGs),

241–2, 248First transition government, see Initial

conditions, continuity ofcommunists

310 Index

Index 311

Fischer, Stanley, 17, 19, 40, 90, 224Forbes billionaires list, 185, 191Foreign direct investment, 54, 209Freedom House, see Democracy,

measuressee also Transition progress, democracy

and freedom indicator

Gaidar, Egor, 4, 19, 160, 187, 235–6, 249,292 n.1

Geography, and EU membership, 7, 213,215, 224–5, 271

see also Initial conditions, locationGeorgia, 3, 9, 61, 66, 68, 82, 87, 98,

103–4, 107–8, 112, 115, 126, 137,156, 169, 193, 198, 216, 234, 248,252–3, 263, 274, 276

Rose Revolution, 3, 9, 252–4Gini coefficient, see Income

distributionGlinsky, Dimitry, see ReddawayGorbachev, Mikhail, 1, 151, 156, 179,

186, 237Government guarantees, supports, see

SubsidiesGradualism, see Reform strategyGrowth recovery, 2, 3, 54, 62–70, 86

prospects, 245–7see also Output decline

Handelman, Steven, 36, 139, 144, 189,287 nn.2, 4, 288 n.7

Hard budget, 16, 68Havrylyshyn, Oleh, 23, 29, 41, 62, 65,

182, 196, 209, 237Health indicators, see Life expectancyHeenan, Patrick, 109, 129, 132, 146,

160, 169, 172, 204, 283 n.10, 286 nn.19, 21

Hellman, Joel, 32, 41, 135, 144, 149,182, 192, 198, 237, 240, 243, 288 n.8, 293 n.14

Historical cross roads, 132, 284 n.15,see also Initial conditions, uniqueness

Historical inertia, see Initial conditionsHuman Development Index, 48, 92–5Hungary, 29, 32, 39, 50, 61, 89, 90, 104,

107–8, 123, 127–9, 137, 141, 145,165, 167–9, 171–2, 184–5, 193,204–5, 208, 222, 225, 259, 262, 267,269, 271

Import-substitution model, 156Income distribution, 95–103

see also Poverty ratiosInflation, 54, 63–4, 162–3, 180, 238

see also MacrostabilizationInformal sector, 70–2, 243

see also KaufmannInitial conditions, 65–7, 126

continuity of communists, 128–30,167–9, 190, 286 n.20

location, 127; see also Geographynationalist commitment, 128, 173–5overindustrialization, 127religion, 127–8uniqueness, 131–2years under communism, 144, 160,

167–8and policy choice, 196–7

Insiders, 179, 181, 197political insider v. firm insider, 181,

287 n.3Institutions, 18, 33–8, 59–62, 154,

156–7, 166, 280 n.15International Monetary Fund (IMF), 10,

19, 20, 21, 137, 198, 210, 216,218–19, 243, 266

International organizations and reform,216–20

technical assistance, 219–20see also EU, membership as reform

disciplineInvisible Hand, 146–7

Johnson, Simon, 23, 61, 65–6, 148, 241–5,278 nn.10, 15, 16, 292 n.2, 293 n.16

Kaufmann, Daniel, 72, 135, 192, 243,278 n.16, 285 n.27, 288 n.8

Kazakhstan, 54, 68, 87, 89, 112, 137,156, 167, 175, 178, 188, 191, 193,195, 202, 240, 262

Khodorkovsky, Mikhail, 39, 188–9, 235,248

Klaus, Vaclav, 19, 59, 156, 186Kolodko, Grzegorcz, 17, 19, 22, 27, 100,

126–8, 131, 153, 172, 204, 279 n.6,281 n.8, 282 n.4, 283 n.11, 284 n.14,286 n.15, 286 n.25

Komsomol, 181and new capitalists, 181, 187–8see also Oligarchs, origin

Kornai, Janos, 16–17, 24, 26, 63, 277 n.2see also Transition model; Output decline

Kravchuk, Leonid, 168, 196Krueger, Anne, 31–2, 40, 134, 201, 238,

289 n.25Kuchma, Leonid, 188, 197Kuzio, Taras, 196, 213, 247, 253, 283 n.5,

293 n.25Kyrgyz Republic, 3, 9, 33, 44, 61, 66, 82,

87, 94, 104, 106–8, 112, 126–8, 135,137, 145, 158, 161, 163–6, 170, 175,182, 186, 193, 195, 197, 217, 225,234, 252–3, 259, 263, 266, 269, 271,274, 276

Tulip Revolution, 3, 9, 252–4

Laar, Mart (former PM of Estonia), 153,167, 172, 204, 205, 272 n.4, 283 n. 14, 289 n.3

Labour markets, 89–90Lagging reformers, 10, 37, 50–1, 59–60,

69, 115–16, 138, 261, 263–4, 267, 273see also Belarus; Gradualism;

Transition progress;Turkmenistan; Uzbekistan

Lamontagne, Monique, see HeenanLatvia, 30, 35, 39, 50, 63, 66, 89, 123, 137,

167, 172, 185, 205, 208–12, 222–3,225–9, 259, 262, 267, 269, 292

Law on cooperatives, 238see also Perestroika

Liberal societies, 137, 262, 267–8see also Liberalism

Liberalism, 55, 167–71dual vision: markets and democracy,

11, 152, 162, 204, 208Life expectancy, 111–13, 282 n.29Limited reforms, see Lagging reformersLipset, Seymour, 58Lithuania, 29, 30, 35, 39, 63, 89, 107,

123, 137, 167, 205, 208–12, 222–3,225–9, 259, 262, 267, 269

Loans-for-shares exercise, 30, 187, 197see also Privatization

Lobbying, 178, 185, 199–201, 239difference from state capture, see State

capturesee also Rent-seeking

Lukoil, 180, 251

Macedonia, 51, 89, 91, 100, 159–61, 163,165, 206, 223, 261

Macrostabilization, 18, 22, 59–62, 159,162

Mafia, 36, 144, 186–7, 189–90, 287 n.2Market memory, see Initial conditions,

years under communismMcFaul, Michael, 43–6, 57, 254,

293 nn.24, 27Moldova, 55, 66, 82, 94, 107, 110,

112, 114–15, 125–8, 132, 137, 145,156, 158, 161, 164–5, 175, 180, 192,207, 212–13, 215–17, 224, 252, 262, 269

Montenegro, 263see also Serbia and Montenegro;

Yugoslav FederationMurrell, Peter, 21, 23–4, 30–1, 33–4, 36,

64, 157, 278 n.15, 293 n.17

Navigation charts, 152–8see also Reform strategy; Transition

modelNegative interest rates, 180New firms, or denovo firms, 28, 31, 42,

182North Atlantic Treaty Organization

(NATO), 10, 137, 207, 216–17, 220,266

Oil prices, 68Oligarchs, 2, 3, 7, 33, 40, 42, 70–1,

134–40, 142, 147–8, 177–91, 197–9,215, 234–9, 241–6, 248–51, 255–9,261–4, 266–76

origin, 7, 142, 177–85and robber barons, 148, 238and state capture, 3, 5–6, 10, 32–3,

135–6, 183and transformation frozen, 3, 11, 183,

233Orange revolution, see UkraineOrganisation for Economic Co-operation

and Development (OECD), 16, 70–1,101–2, 243

Output decline, 8, 26–8, 81–6, 118–20

Pareto optimality, 8, 79see also Reform strategy

Partial reforms, see Delayed reformsPartnership and Cooperation

Agreement, 213, 215People revolutions, 234, 255

see also Revolutions of colour

312 Index

Index 313

Perestroika, 11, 154, 156, 180, 237Poland, 19, 22, 26–7, 30, 39, 44, 57, 63,

82, 90, 100, 104, 109, 123, 127–8,141, 145, 153, 163, 165, 167, 171–2,179, 184–5, 191, 204–5, 222, 225,242, 259, 262, 267, 271, 281 n.16

Pope John Paul II, 1Poverty ratios, 103–10, 281 n.23,

see also Social indicators; Reformstrategy and social costs

Price controls or regulation, 178–80see also Price liberalization

Price liberalization, 17–19, 59–60, 178–9Privatization, 18, 24, 28–31, 59–60, 67,

162, 177, 178–9, 195, 197, 261, 278 n.13

see also New firms, or denovo firms;Privatization

Property rights, 35–6, 39–40, 136,199–201, 235–7

Przeworski, Adam, 44–5, 49, 55–7, 252,257–8, 286 n.24

Rapid reformsand EU membership, 220–3and nationalism, 173–5and social costs, 115, 259see also Reform strategy, big-bang

Reagan, Ronald, 1, 151Red Directors, 180–1, 186, 197Reddaway, Peter, 19, 22, 33, 113, 117,

126, 181, 197Reform strategy, 159, 164–7

big-bang, or rapid, 6, 19, 23–6, 153–5debate and delay, 5, 10, 11, 17–20,

136, 138–40, 152, 154, 158, 176,192–4, 220–22

gradualism, 6, 19, 23–6, 59–60, 156–7measuring success or failure, 8–9, 94shock-therapy, 6, 19and social costs, 94, 115–17, 259–60

Rent-seeking, 3, 10, 32, 134–5, 178–85,191–2, 233, 266–7

see also Krueger, AnneRe-privatization, see Reversing state captureReversing state capture, 248–54, 274–6

see also Revolutions of colourRevolutions of colour, 234, 252–4, 263,

293 n.21,see also Georgia, Rose Revolution;

Kyrgyz Republic, Tulip Revolution;Ukraine, Orange Revolution

Roadmaps for transition, see Navigationcharts; Reform strategy

Robber barons, 199–202see also Oligarchs

Roland, Gerard, 23–4, 277 n.3, 278 n.12,279 n.2, 284 n.18

Romania, 29, 39, 44, 53, 58, 66, 72, 82,89, 99–100, 107, 109–10, 125, 137,145, 159–61, 163, 165, 168, 174,179, 185, 198, 205–6, 209, 211, 214,222–3, 233, 247, 254, 260–3, 268–70

Rose Revolution, see GeorgiaRukh, 168, 196–7, 271Rule-of-law, 35, 40, 130–1, 136, 139,

199–201Russia, 20, 24, 29, 30, 32–3, 35, 39–40, 44,

54, 57, 63, 65, 66–8, 70, 82, 87, 89,98–103, 107–8, 110–13, 126–8, 135,137, 144, 148, 151, 155–6, 158,159–61, 164–6, 168, 170, 178, 180–2,184–93, 195, 197, 199, 202, 206, 209,219, 228, 235–6, 238, 241–2, 245,246–8, 250–1, 259–60, 262, 266, 269, 271–8

Sachs, Jeffrey, 23, 31, 34, 51–2, 54, 278 n.14, 279 n.1

Schengen curtain, 212–13Security issues, 216–17Serbia and Montenegro, 50, 60, 89, 90,

123, 127–8, 130, 145, 164, 167,174–5, 206, 216, 252–3, 263, 276

see also Yugoslav FederationShadow economy, see Informal sectorShleifer, Andrei, 39–40, 178, 197, 202,

235, 242, 246Shock therapy, see Reform strategySlovakia, 29, 35, 39, 44, 55, 65, 123, 128,

130, 133, 137, 165, 168–9, 172–5,193, 205, 212, 218, 222, 225, 259,262, 267, 269, 283 n.10, 290 n.13

Slovenia, 30, 39, 60, 89–90, 123, 127–8,130, 145, 164, 167

see also Yugoslav FederationSMEs (small and medium enterprises),

138–9, 241, 242–4, 275–6and oligarch influence, 243see also New firms, or denovo enterprises

Smith, Adam, 146see also Invisible Hand

Social indicators, 10, 91–2see also Reform strategy, and social costs

South-East Europe, 3, 9, 10, 79–81Soviet Mafia, see MafiaState capture, 3, 10, 135–6, 178, 191–3,

262, 269–70and reform delay, 192–6v. lobbying, 33, 178, 188, 202, 239,

262–3, 273–4see also Oligarchs; Rent-seeking;

Transition inevitable; Transitionfrozen

Stiglitz, Joseph, 17–22, 29, 31, 33, 34, 36,277 nn.1, 5, 279 n.2

Subsidies, 179–80Svejnar, Jan, 21, 29–30, 100–1, 115,

278 n.15

Tajikistan, 51, 61, 68, 87, 95, 108, 110,112, 138, 161, 263

Trade and exchange liberalization,17–19, 59–60

Transformational recession, 26see also Kornai; Output decline

Transition costs, 3, 78–81Transition frozen (TF) thesis, 40–1, 46,

134, 147, 183–5, 197–8, 236–8empirical evidence, 238–40

Transition inevitable thesis (TI), 39–40,46, 133–4, 181, 233–6

Transition is over, 3–4, 255, 260–1Transition model, 15–20, 124, 133–43,

149–50, 264–6debates on transition, 21–3,

138–40, 152see also Reform strategy

Transition progress, 47–8, 146, 255–6correlation among indicators, 10, 49,

51, 54, 58, 256democracy and freedom indicator,

54–7EBRD Transition Progress Indicator

(TPI), 10, 48–52, 73–7, 162–4,256–8

economic performance measures, 52–9historical explanations, 125–33see also Human Development Index;

Social indicatorsTransition start, 152, 158–64Transitology paradigm, 42–6Transparency International, 36, 48Tulip Revolution, see Kyrgyz Republic

Turkmenistan, 8, 28, 47, 51, 54, 61, 68,71, 87, 89, 94, 98–9, 109, 123, 138,166–7, 175, 182, 218, 234, 251, 257,260–1, 263

Twain, Mark, 200, 233

Ukraine, 3, 9, 20, 29–30, 32, 35, 44, 55, 65–6, 82, 87, 98, 100–1, 106–8,110, 112, 126–8, 130, 133, 135, 137, 152, 155–6, 168, 174, 177–8,180–2, 184–6, 188–91, 196–9, 202,207, 211–16, 233–4, 236, 239,242–4, 247–54, 262–3, 269–71, 274, 276

and EU membership efforts, 213–14,247

Orange Revolution, 3, 9, 215, 252–4timeline, 158–64see also Kravchuk; Kuchma; Rukh;

YuschenkoUnderground economy, see Informal

sectorUnemployment, 87–91United Nations Development

Programme (UNDP), 48, 78, 80,91–3, 95, 104, 106, 111, 114, 286

Uzbekistan, 8, 47, 51, 54, 61, 72, 84, 87,94, 109, 123, 138, 156, 158, 163,166, 167, 175, 182, 192, 234, 251,257, 260–3

Veblen, Thorsten, 200Visegrad countries, 8, 23–6, 34, 40, 63,

80, 92, 265

Washington Consensus, 17–20, 154–5,210

World Bank, 10, 15, 20–1, 29–30, 32, 34,36, 40–1, 48, 94, 104, 106, 110, 135,137, 144, 149, 188, 191–2, 195, 210,216, 218–19, 236, 240–3, 266, 270

World Trade Organization (WTO), 10,127, 137, 216–17, 220, 266

Yeltsin, Boris, 186–7, 197Yugoslav Federation, 8, 172, 184, 261,

270Yukos, see KhodorkovskyYuschenko, Viktor, President of Ukraine

(2005–), 68, 213–15, 249, 292 n.7

314 Index

divergent paths in post-communist transformation: capitalism for all or capitalism for the few?

Documents