trade, fdi, and global value chains - world...
TRANSCRIPT
Trade, FDI, and Global Value Chains
Hiau Looi Kee
The World Bank
September 2016
Kee (WB) OTRI, DVAR & Brexit 09/16 1 / 37
Disclaimer
The results and opinions present in this talk are my own, and do notnecessarily represent the views of the World Bank, its Executive Directors,or the countries they represent.
Kee (WB) OTRI, DVAR & Brexit 09/16 2 / 37
On June 23rd, 2016
UK voters decided to leave EU
British Pound nose dived to 30 year low
World stock markets plummeted by 8%
Prime Minister David Cameron resigned
As many as 52 trade deals to be re-negotiated
Most economists think that Brexit is a very bad idea (Baldwin, Bown,Dhingra, Ottaviano, Sampson, Van Reenen)
Some predicted big short term drop in trade, and income per capitadecrease as much as 2.6% post Brexit
Kee (WB) OTRI, DVAR & Brexit 09/16 3 / 37
Three Months Later
O¢ cial talks between UK and the EU haven�t started
Bank of England lowered its interest rate to 0.25%, the lowest level inthe bank�s 322 years, to fence o¤ �Brecession�
Pound has stabilized and stock markets have recovered
Teresa May is now the Prime Minister
There are signs that consumer spending is healthy, and retail sales areup
Credits to business and new regular job vacancies seem low
Kee (WB) OTRI, DVAR & Brexit 09/16 4 / 37
How about Trade and Competitiveness in a World withGlobal Value Chains?
Weaker currency should boost exports, but for global value chains,much of the exports contain imports, which is now more expensive
The net impact of a devalued pound on exports is not clear
Post Brexit: In short run, how do UK trade with EU and its othertrading partners such as the US and China?
Post Brexit: In long run, will UK remain competitive in the worldmarket?
Kee (WB) OTRI, DVAR & Brexit 09/16 5 / 37
What This Talk is All About?
This talk will analyze the potential short term trade in goods fall outsfrom Brexit, through potential changes in the trade policies of itsmain trading partners
The Overall Trade Restrictiveness Index (Kee, Nicita, Olarreaga,ReStat, 2008 and EJ, 2009)
Bilateral trade elasticities and the ad valorem equivalent of non-tari¤measures (Kee and Nicita, 2016)
This talk will discuss the potential long run impact from Brexit,through FDI and its participation in global value chains
FDI promotes local intermediate input industries which leads to risingdomestic value added in exports and long term competitiveness (Kee,JDE, 2014; Kee and Tang, AER, 2016)
Kee (WB) OTRI, DVAR & Brexit 09/16 6 / 37
What This Talk is NOT About?
This talk will not talk about services trade, �passporting�and the roleof London as a �nancial center post Brexit
This talk will not talk about immigration and the social aspect ofBrexit
This talk will not speculate about the optimal timing to invoke Article50 and �soft Brexit�
Kee (WB) OTRI, DVAR & Brexit 09/16 7 / 37
The Overall Trade Restrictiveness Index
OTRIc =∑n mn,c εn,cTn,c
∑n mn,c εn,c
OTRI summarizes the trade policies stance of a country, taking intoaccount the responsiveness of trade (εn,c ) with respect to trade policy(Tn,c ) , weighted by the share of each product in total trade (mn,c )Trade policy include both tari¤ and the ad valorem equivalent (AVE)of non-tari¤ measures (NTM)NTM include Sanitary and Phytosanitary Measures, such asrequirements on labelling, hygiene, maximum pesticide residue limitsand testing; as well as Technical Barriers to Trade, such asrequirements on labelling, product quality, packaging, andcerti�cationsTheoretical foundation: Anderson and Neary (1992, 1994, 1996, 2003and 2007), Feenstra (1995)Basic ingredients for OTRI: trade elasticities and AVE of NTMs
Kee (WB) OTRI, DVAR & Brexit 09/16 8 / 37
Import Demand Elasticities and Trade Distortions (Kee,Nicita and Olarreaga, ReStat, 2008)
Based on a GDP function approach: Overall 377,000 import demandelasticities are estimated for 117 countries across 4900 HS 6-digitproducts
Sample mean is around 3, which is in line with the estimates providedin the literature
Wide range of estimates suggesting product heterogeneity � tradedistortion could be higher
Import-weighted average tari¤s underestimate the restrictiveness of acountry�s tari¤ regime when tari¤ and demand elasticity is positivelycorrelated
Canada and the US, that impose high tari¤s on more elastic productswill have a larger trade and welfare distortions
Kee (WB) OTRI, DVAR & Brexit 09/16 9 / 37
Wide Range of the import demand elasticity estimates atHS 6-digit level Implies Large Trade Distortions
Mean=3.120
.2.4
.6.8
1Ke
rnel
Den
sity
4 3 2 1 0Import Demand Elasticity
Kee (WB) OTRI, DVAR & Brexit 09/16 10 / 37
Estimating Trade Restrictiveness Indices (Kee, Nicita andOlarreaga, EJ, 2009)
Estimate the AVE of NTM at importer-product level based on gravityregressions
The simple average ad-valorem equivalent in the entire sample forcore NTBs is 12%
Tari¤ and AVE are negatively correlated � suggesting trade policysubstitutions and the importance of including AVEs in analysis
Combining the elasticity and AVE estimates, OTRI of 78 countries areconstructed
NTMs have a signi�cant contribution to the level of traderestrictiveness
Developing countries tend to have higher trade barriers thandeveloped countries
Kee (WB) OTRI, DVAR & Brexit 09/16 11 / 37
OTRI Decreases with GDP per Capita
ALB
ARG
AUS
BFA
BGD
BLRBOL
BRA
BRN
CAN CHE
CHL
CHN
CIV
CMR
COL
CRICZE
DZA
EGY
EST
ETH
EUN
GAB
GHAGTM
HKG
HND
HUN
IDN
IND
ISL
JOR
JPN
KAZ
KEN
LBN
LKA LTU
LVA
MAR
MDA
MDG
MEX
MLI
MUS
MWI
MYS
NGA
NIC
NOR
NZL
OMN
PER
PHL
PNG
POL
PRY
ROM
RUS
RWASAU
SDN
SEN
SLV
SVN
THA
TTO
TUN
TUR
TZA
UGA
UKR
URY
USA
VEN
ZAF
ZMBALB
ARG
AUS
BFA
BGD
BLR
BOL
BRA
BRN
CAN CHE
CHL
CHN
CIV
CMR
COL
CRI CZE
DZAEGY
EST
ETH
EUN
GABGHA
GTM
HKG
HNDHUN
IDN
IND
ISL
JOR
JPN
KAZKEN
LBN
LKA
LTU
LVA
MAR
MDAMDG
MEX
MLI
MUS
MWI
MYS
NGA
NIC
NOR
NZL
OMN
PER
PHLPNG
POL
PRYROM
RUS
RWA
SAU
SDN
SEN
SLV
SVN
THA
TTO
TUN
TUR
TZA
UGA
UKR
URY
USA
VEN
ZAF
ZMB
0.2
.4.6
OTR
I
4 6 8 10 12Log of GDP per capita
Kee (WB) OTRI, DVAR & Brexit 09/16 12 / 37
Impacts of OTRI, AVEs of NTMs and Trade Elasticities
Within the Bank:
OTRI have been included in Global Monitoring Report from 2003 to2009, to keep track of trade policy reforms of member statesOTRI is part of the World Trade Indicators of the BankOTRI and the AVE estimates were included in many regional �agships,including MNA, LAC, SSA and SAR, as well as many country reports,such as Malaysia, Sri Lanka and others
Outside the Bank:
OTRI and trade elasticity estimates have been closely followed by manygovernments, and were used by the EU and the US trade negotiators intheir World Trade Organization (WTO) trade disputesMany leading academics have used these estimates for their own work(eg. Bob Staiger, Karl Bagwell, James Anderson, Peter Neary, DouglasIrwin and Peter Petri)Jointly these two papers generated more than 1,100 citations accordingto Google Scholar
Kee (WB) OTRI, DVAR & Brexit 09/16 13 / 37
Some Glossy Covers
Kee (WB) OTRI, DVAR & Brexit 09/16 14 / 37
Is Protectionism on the Rise? Assessing national tradepolicies during the crisis of 2008 (Kee, Neagu and Nicita,ReStat, 2013)
The Great Trade Collapse 2008: Many economists fearedprotectionism , the trade collapseUse OTRI to quanti�es trade policy changes and the associated tradeimpacts for about 100 countries between 2008 and 2009
4OTRIc =∑j ∑n mn,c ,j εn,c ,j (4Tn,c ,j )
∑j ∑n mn,c εn,c ,j
While the rise in tari¤s and antidumping duties may have jointlycaused global trade to drop by US$43 billion, it explains less than 2%of the collapse in world trade during the crisis periodAlbie Award of Foreign Policy for the best papers on politicaleconomy in 2013; Cited in Economist; Media interview by ReutersReprinted in International Trade and the New Global Economy (2016)
Kee (WB) OTRI, DVAR & Brexit 09/16 15 / 37
Kee (WB) OTRI, DVAR & Brexit 09/16 16 / 37
Kee (WB) OTRI, DVAR & Brexit 09/16 17 / 37
Application: Brexit
Based on the newly collected NTM database (UNCTAD and WorldBank)
Bilateral import demand elasticities and AVEs are estimated in Keeand Nicita (2016)
Cover trade data of 35 importing countries with 96 exportingcountries
Kee (WB) OTRI, DVAR & Brexit 09/16 18 / 37
Export Pro�le of UK �by Destination Country
Kee (WB) OTRI, DVAR & Brexit 09/16 19 / 37
Export Pro�le of UK �by Sector
Kee (WB) OTRI, DVAR & Brexit 09/16 20 / 37
Trade Policies of Main Partners
With the exception of EU, most trading partners do not give tradepreference to UK�s exports
EU China USAMaximum MFN tari¤ (%) 74.9 65 350Minimum MFN tari¤ (%) 0 0 0
Product subjected to positive tari¤ (%) 75 93 58Maximum AVE (%) 1725 1627 1900Minimum AVE (%) 0 0 0
Product subjected to positive AVE (%) 33 48 25
Kee (WB) OTRI, DVAR & Brexit 09/16 21 / 37
Current OTRI of Main Partners
EU China USAImport Weighted Average MFN Tari¤ (%) 3.4 6.73 1.54
Import Weighted AVE (%) 3.43 0.74 3.18Import Weighted Trade Elasticity -6.4 -4.4 -0.18
Current OTRI (%) 0 0.76 1.36
Kee (WB) OTRI, DVAR & Brexit 09/16 22 / 37
Potential Trade Impact from Main Partners due to Brexit
EU China USAPotential Change in OTRI post Brexit (%) 0.26 0 0Potential Change in Trade post Brexit (%) -1.7 0 0
Potential Change in Trade post Brexit ($Billion) -3.43 0 0
Short term disruptive trade due to trade policy seems to be small �OTRI of main partners are low and UK has been complying withNTMs of current partners
Majority of UK�s exports to EU are either heterogeneous productswith low elasticities or homogeneous products with low MFN tari¤s
Without any trade deals, UK exports may drop by less than 2%
Kee (WB) OTRI, DVAR & Brexit 09/16 23 / 37
Why is EU�s OTRI Low?
Kee (WB) OTRI, DVAR & Brexit 09/16 24 / 37
Domestic Value Added in Exports: Theory and FirmEvidence from China (Kee and Tang, AER, 2016)
One way to measure long term competitiveness is to measuredomestic value added (DVA) in exports, which captures thecontribution of exports to GDP
Kee and Tang (AER, 2016) pioneers a �rm level approach to calculateDVA in exports based on customs transaction level data of China,instead of using conventional Input-Output Tables
Firm level approach embraces �rm heterogeneity and allow di¤erent�rms to have di¤erent production technology
DVA of Chinese exports are higher and are rising, from 65 to 70percent in 2000-2007
Kee (WB) OTRI, DVAR & Brexit 09/16 25 / 37
DVAR of China�s Aggregate (Processing + Ordinary)Exports
.45
.5.5
5.6
.65
.7.7
5
2000 2001 2002 2003 2004 2005 2006 2007year
DVAR (Processing + Ordinary Exp) DVAR (Processing Exp)
All firms with materials < imp & exp < imp, and processing firms with DVAR>DVAR(25 percentile of Ord Exporters) are excluded.
Kee (WB) OTRI, DVAR & Brexit 09/16 26 / 37
DVAR of China�s Exports to its Top 5 Trading Partners
.45
.5.5
5.4
5.5
.55
2000 2002 2004 2006
2000 2002 2004 2006 2000 2002 2004 2006
Germany Hong Kong, SAR Japan
Korea United States
DVA
R o
f Exp
orts
yearGraphs by twodigit iso code
Kee (WB) OTRI, DVAR & Brexit 09/16 27 / 37
Factors behind China�s Rising DVAR: Trade and FDILiberalization which Promote GVC Participation
Rising DVAR was induced by the country�s trade and FDIliberalization, which deepened its engagement in global value chainsand led to a greater variety of domestic materials becoming availableat lower prices.
Increase presence of downstream FDI and decrease in upstream tari¤sboth help expand domestic variety of intermediate inputs, which leadsto lower domestic prices and higher DVAR
Kee (WB) OTRI, DVAR & Brexit 09/16 28 / 37
Trade and FDI Liberalization A¤ect DVAR
Kee (WB) OTRI, DVAR & Brexit 09/16 29 / 37
Local Intermediate Inputs and the Shared SupplierSpillovers of Foreign Direct Investment (Kee, JDE, 2014)
Cross support requested by the Government of Bangladesh: to studythe potential bene�ts to domestic garment �rms should thegovernment liberalize FDI restrictions after the end of the Multi-FiberAgreement (2005)
Conduct �rm level survey to collect information on garment �rms andtheir suppliers
Sharing local suppliers with domestic �rms, FDI �rms help promote1/4 of the product scope expansion and 1/3 of the productivity gainswithin domestic �rms
Paper was really well received by the Government and the privatesector and the country has since open up for more FDI
Currently Bangladesh is one of the top 5 garment exporters of theworld
Kee (WB) OTRI, DVAR & Brexit 09/16 30 / 37
Numbers of Garment FDI �rms and Local Suppliers InBangladesh (1984=1)
Kee (WB) OTRI, DVAR & Brexit 09/16 31 / 37
Long Term Competitiveness: FDI and Global Value Chains
FDI bene�ts local economy beyond the immediate sectors theyparticipate in
Presence of FDI helps promote local intermediate input industries ofChina and Bangladesh
Competitive local intermediate input industries allow countries tomove up the global production chains
Kee (WB) OTRI, DVAR & Brexit 09/16 32 / 37
Application: Brexit
UK as a FDI destination: #1 in EU and #4 in the World (behind US,China and India)
In 2014/2015, UK attracted 1/3 of FDI in EU, mostly in the �nancialservices sector, headquarter services and R&D
Largest FDI source countries: US, EU
Manufacturing industries: automotive, food, and machinery andequipment sectors
Kee (WB) OTRI, DVAR & Brexit 09/16 33 / 37
FDI Pro�le of UK �by Source Country
Kee (WB) OTRI, DVAR & Brexit 09/16 34 / 37
FDI Pro�le of UK �by Sector
Kee (WB) OTRI, DVAR & Brexit 09/16 35 / 37
Potential Long Term Implications
2015 UK attractiveness survey by Ernst and Young: 72% of investorsciting access to the European single market as important to the UK�sattractiveness
Brexit anxiety: uncertainty may hurt existing and future FDI
Brexit may not a¤ect manufacturing FDI and UK participation inglobal value chains
Business services may be signi�cantly a¤ected
Kee (WB) OTRI, DVAR & Brexit 09/16 36 / 37
Conclusions
Short term collapse in trade due to trade policy changes is unlikely
Long term competitiveness depends on whether UK can retain andattract FDI
Kee (WB) OTRI, DVAR & Brexit 09/16 37 / 37