Does WTO Accession Raise Income

1
Does WTO Accession Raise Income?

• May 22, 2006
• M.K. Tang Shang-Jin Wei
• Personal views, not necessarily shared by the IMF

2
Motivation I

• Accessions often involve extensive and sometime
  politically difficult policy changes (in recent
  years)
• It took China 14 years to join
• The median time gap b/n application and accession
  was 71 months for countries that joined after 1995

3
Motivation II

• Diverse Views on the Consequences
• Beneficial (if sometimes bitter)?
• WTO accession provides a predictable business
  environment and gives a powerful guarantee to
  investors that there will be no policy
  reversals. An advisor to the Ethiopian
  government
• Irrelevant?
• Andrew Rose (2004) Do we really know that the
  WTO has promoted trade? American Economic Review
• Worse than irrelevant?
• Developing countries squeezed for the benefits of
  rich and powerful nations, Jawara and Kwa (2003)
  Behind the Scenes at the WTO the Real World of
  International Trade Negotiations

4

• Regime change for developing country applicants
• Before the Uruguay Round Very little policy
  change needed
• After the Uruguay Round Much more extensive
  policy changes typically demanded by existing
  members (except for countries qualified for GATT
  Article XXVI 5(c))

5

• Value of accession
• Market access in exchange for political support
  from reform-opposition
• Commitment against future policy reversal
• Previous work on trade volume
• Subramanian and Wei, 2004
• This paper Effects on growth and investment

6

• Methodology
• Difference-in-differences in spirit
• Change in growth before/after 43 accessions
• Change in growth for non-accession developing
  countries during the same period
• Results
• Higher annual growth and Inv/GDP than before,
  especially for those undergoing more rigorous
  accession procedures

7

• Addressing Endogeneity Concerns
• Accessions with and without reforms
• Application vs. actual accession
• Formal statistical test using Heckman selection
  equation
• Varying effects along commitment/ governance
  dimensions

8
Roadmap

• Brief Description of Data and Specification
• Statistical Results
• Extensions and Robustness Checks
• Conclusions

9
Data and Specifications

• Data
• Per capita GDP, gross fixed capital formation,
  trade from WEO
• Institution quality from WBs Governance
  Matters
• 1981-2003, exclude all industrial and OPEC
  countries

10

• Empirical Specifications
• i country, t year
• s Time-profile of accession, years from
  accession null, -2, -1, 0, 1, 2, 3, 4, 5,
  beyond
• Country and year fixed effects included
• Treatment group accession countries
• Control group other developing countries
• Beginning period 8 years before accession

11
Accessions 1990-2001

• Results for All Accessions
• Growth sped up from 1 year before accession to 3
  years after
• Little effect on Inv/GDP though
• Positive coef. est. but insignificant

12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15

• Article XXVI 5(c) of GATT 1947
• If any of the customs territories, in respect
  of which a contracting party has accepted this
  Agreement, possesses or acquires full autonomy in
  the conduct of its external commercial relations
  and of the other matters provided for in this
  Agreement, such territory shall, upon sponsorship
  through a declaration by the responsible
  contracting party establishing the
  above-mentioned fact, be deemed to be a
  contracting party.

16

• Article XXVI5c vs. Non-Article XXVI5c Cos
• Heterogeneity in accession procedures
• Article XXVI 5(c) 18 countries
• Upon becoming independent, former colonies of
  GATT members were entitled to formally join GATT
  almost automatically
• The option can be exercised any time (up to the
  founding of the WTO)
• Non-Article XXVI5(c) 25 countries
• Acceded by normal procedures (Article XII),
  meaning lengthy negotiations!

17

• Results for AXXVI5c vs. non-AXXVI5c
• Faster growth after accession than before 72
  non-Article XXVI5c
• vs.
• 47 Article XXVI5c
• Higher Inv/GDP after accession than before
• 59 non-Article XXVI5c
• vs.
• 38 Article XXVI5c

18

• Non-Article XXVI 5c countries
• Growth became stronger since 1 year before
  accession
• Inv/GDP became higher since 2 years after
  accession
• Article XXVI 5c countries
• Growth Weakly positive, insignificant effects on
  change in growth
• Inv/GDP Negative, insignificant effects on
  change in Inv/GDP

19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
Extensions and Robustness Checks

• Timing of Spurs
• Median interval b/w application and accession
  71 months after 1995,
• or 55 months since 1990
• Pickups in Growth and Investment after
  application die down gradually
• But Growth and Investment rise again around
  accession

23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26

• Testing Selection Bias Formally
• Does the endogenous nature of membership status
  bias our results?
• Employ 2-stage estimation a la Heckman
• 1st stage Probit estimation of membership status
  as a function of lagged GDP and Trade/GDP ratio
• 2nd stage Add in the inverse Mills ratio as an
  independent variable
• Coef. est. on the inverse Mills ratio are
  insignificant gt No evidence that selection biase
  is quantitatively important

27
(No Transcript)
28

• Effects of Policy Commitments
• Sample Accessions after WTO formation
• More standardized documents
• 15 accessions
• Commitments touch on general policies
• Wide-ranging e.g., TRIPs, transparency,
  competition policy, price controls, etc.
• Might spillover to other non-trade areas
• Clearly incorporated in Protocol enforceable
  thru WTO dispute settlement mechanism

29

• Measure of degree of policy commitments
• commitments explicitly stated in Protocol
• Transparent, non-subjective
• Not one-size-fits-all
• accessions to the WTO will be on terms to be
  agreed essentially a process of negotiation,
  WTO.org
• Alternative measure words in WP Report

30
(No Transcript)
31

• Differential effects of policy commitments
• First View Countries w/ weak governance
• May lack incentives to implement reforms
• May lack credibility to follow through
• Second View Accession commitments may be a
  partial remedy for weak governance
• So, effect of commitments is stronger or weaker
  for countries w/ weak governance?

32

• Differential Effects Results
• Interact the commitment variable w/ governance
  measure
• Interaction terms negative and significant in
  both growth and investment regressions
• Support the interpretation that accession-related
  commitment may partly offset the shortcomings of
  weak governance

33
(No Transcript)
34

• Discrete categorization
• Governance
• Above world median (3) Good governance
• Otherwise Poor governance
• Policy commitments
• Above sample median (27) Many com.
• Alternatively, words gt sample median Many com.
• Otherwise Few com.
• Basic results hold Policy commitments have
  stronger effects among poor-gov countries (Table
  13)

35
Conclusions

• Accessions entail wide-ranging reform commitments
  
• We yield affirmative results of accession
  benefits by
• Distinguishing countries by the degree of
  accession requirements
• Tracing out the trajectories of performance
  changes
• Considering the endogenous nature of membership
• Showing the differential effects of accession
  requirements on countries with different
  governance quality

36

• The End
• Thank You

37

• Stronger effects among poor-gov. countries on
  both growth and Inv/GDP
• Difference in year 2 after accession among
• Poor-gov. 7.3 perc. points (change in growth)
• 40 (change in Inv/GDP)
• Good-gov. -0.8 perc. points (change in growth)
• -52 (change in Inv/GDP)
• Basic results hold with words
• Caveat based on only small sample