Deep Integration

Report
Deep Integration:
Free trade agreements heterogeneity and its
impact on bilateral trade
Séminaire DIAL
Jaime Ahcar Olmos
Under the supervision of :
Jean-Marc Siroën
PSL Université Paris-Dauphine
Laboratoire LEDa-DIAL
24 avril, 2014
Contents
1) Introduction
2) Methodology and Data
3) Results
4) Conclusions
5) References
1
Introduction
• International trade economists have apparently
successfully dealt with the question : does free
trade agreements increase trade flows?
Rose (2004), Baier & Bergstrand (2007), Martinez et al. (2009)
• This has been done neglecting thorny issues like
heterogeneity and the concept of deep
integration.
2
Introduction : Questions
• So we now try to answer new questions :
• Are all free trade agreements equal? What
happens if not ?
• Does a deeper free trade agreement really
increase trade more than a shallow agreement ?
• What would be a good indicator of deep
integration like ?
• Are other provisions related to trade but out of
the traditional WTO framework of negotiation
really important to expand trade ?
3
Introduction : Importance
• Preferential Trade Agreements PTA
heterogeneity little explored
• Indicators of heterogeneity
• Implications for the interpretation of PTA
coefficients
• Implications for our understanding of the limits
of trade liberalization
4
Literature review
• Bourgeois, Dawar & Evnenett (2007). A Comparative
Analysis of Selected Provisions in Free Trade
Agreements.
• Magee (2008) controlled for some levels of trade
integration.
• Vicard (2009) on the depth of the agreements found
that their trade creation effect does not statistically
differ according to the depth of the RTA.
• Vicard, V (2011) By using interactions terms found that
large, similar and close countriesn benefit more from
RTA. This approach doesn’t care about their dose or
their design.
5
Literature Review
• Horn, Mavroidis & Sapir (2010) and nontraditional WTO provisions in the analysis and
explored legal enforcement effects, but
ramained on the descriptive approach.
• Orefice & Rocha (2013) carried out empirical
analysis from Horn et al. observarions and
found positive evidence.
• Dür, Baccini & Elsig (2014) enlarged the
database and introduced latent trade analysis.
6
Database
• Two different classification of agreements
databases:
1) WTO (2011) Research division for the World Trade Report
2) Design of Trade Agreements DESTA-WTI (2014)
Then, we introduce this information in our gravity
model unbalanced panel data set which
encompasses 153 countries and 613.030 bilateral
trade flows from 1980 to 2012.
7
Databases and methodology :
• Following (Horn et al. 2010) approach:
• 1) WTO+ provisions: topics under the
competence of the WTO agreements.
• 2) WTO-X provisions: topics outside the current
mandate of the WTO.”
• This database is exploited by Orefice & Roche
(2013)
• Their regressions account for 66 PTA from 1980–
2007. We include in our calculations 103 PTA
including agreements for Peru & Colombia from
1980-2012.
8
Database variables
WTO+ Policy Areas Negociated in PTAs
Anti-dumping
Countervailing Measures
GATS General Agreement on Trade in Services
Public Procurement
Sanitary and Phytosanitary Measures
State Aid (Subventions)
State Trading Enterprises
Technical Barriers to Trade
Trade Related Aspects of Intellectual Property Rights TRIPs
TRIMs Trade-Related Investment Measures
Source: Orefice & Rocha (2013), Horn et al. (2010). WTO ( 2011) Research division for
the World Trade Report
9
Database variables
WTO-X Provisions Negociated in PTAs
Agriculture
Innovation policies
Anti-corruption
Investment measures
Approximation of Legislation
Audiovisual
Competition policy
Consumer protection
Cultural cooperation
Data protection
Economic policy dialogue
Education and training
Energy
Environmental laws
Financial assistance
Health
Human rights
Illegal immigration
Illicit drugs
Industrial cooperation
Information society
IPR
Labour market regulation
Mining
Money laundering
Movement of capital
Nuclear safety
Political dialogue
Public administration
Regional cooperation
Research and technology
SMEs
Social matters
Statistics
Taxation
Terrorism
Visa and asylum
Source: Orefice & Rocha (2013), Horn et al. (2010). WTO ( 2011) Research division for
the World Trade Report
10
Methodology
• Probit approach
• We put forward two main kinds of indicators
to try to capture deep integration.
• They depend on a set of variables (provisions)
that characterize PTA texts.
• 1) Additive indicators
• 2) Distilled indicators taken from PCA related
methodologies.
11
Methodology Contributions :
Additive indicators
• Does an additive indicator tell us something
about the depth of the agreement?
• Reasonably yes. One can expect that more
comprehensive agreements are deeper.
• Isn’t the magnitude of the reduction on tariffs
the main indicator of depth? Maybe not, as
nontariff barriers can play a major role to
block trade.
12
Methodology Contributions :
Distilled indicators
• We calculate a Multiple Correspondence
Analysis MCA indicator for deep integration:
• MCA is a technique that can be applied to
obtain an indicator that summarizes in just
one dimension the variability of a set of
variables related with the content and design
of PTA, and then with the depth of the
integration.
13
Limits of the analysis
• “Europe’s single market is probably the best example
globally of successful deep integration” Claar S. &
Nölke A. (2010).
• Nevertheless, because of methodology consistency
requirements, Dür et al. (2014) additive indicator give
UE 1992 single market agreement a 5 while Colombia USA receive a 7
• Orefice & Rocha gives 6-11 to EU27 but 9-27 to EUChile.
• Enforceability of the agreements. Too subjective, it
implies going into the nuances of the language in PTA
texts.
14
Methodological Limits
• Due to the large size of our data set, we can not
completely control for multilateral resistance in
all specifications. Anderson and Van Wincoop (2003); Baldwin
& Taglioni (2006)
• We try to minimize this setback by introducing
time fixed effects and time invariant fixed
effects for importers and exporters countries.
15
Theoretical Model
xij is export from region i to region j, σ is the elasticity of substitution between all goods, tij is a trade
cost factor between i and j; yi and yj are GDP in region i and region j; yW is world nominal income; k is a
constant, (1 – σ) and (1 – σ)ρ are coefficients.
16
Probit model with single EV for the median rta coeffient on
some WTO+ provisions
(1)
brtamed brtamed brtamed brtamed brtamed brtamed brtamed brtamed brtamed
sps
0.071
(0.252)
tbt
0.206
(0.282)
ste
-0.268
(0.248)
ad
0.096
(0.319)
cvm
0.077
(0.275)
publicprocurement
-0.226
(0.250)
trims
-0.074
(0.247)
gats
-0.219
(0.273)
trips
-0.132
(0.253)
Constant
Observations
r2_p
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
-0.031
-0.140
0.145
-0.066
-0.043
0.140
0.050
0.168
0.092
(0.196)
(0.242)
(0.174)
(0.288)
(0.233)
(0.187)
(0.177)
(0.230)
(0.196)
103
103
103
103
103
103
103
103
103
0.000555
0.00375
0.00823
0.000634
0.000552
0.00576
0.000625
0.00454
0.00192
17
Probit model with single EV for the median
rta coeffieient on WTO-X provisions
brtamed brtamed brtamed
consumerprotection
-0.599*
(0.330)
-0.520**
environmentallaws
(0.252)
-0.496*
labourmarketregulation
(0.276)
0.120
0.244
0.155
(0.137)
(0.168)
(0.147)
103
Observations
0.0238
r2_p
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1
103
0.0302
103
0.0228
Constant
18
Results
• On a probit model regression with single EV. No
evidence that the presence of any particular
provision under WTO+ would increase trade
above the median level, which PTAs are supposed
to do.
• Only 3 out of 36 WTO-X provisions show
significant estimators : environmental laws,
labour market regulation and consumer
protection. All three show a negative coefficient.
• It could open the door to the possibility of PTA
that deters trade rather than promote it.
19
Results
• Robustness. Possible bias in the estimation of
the individual PTA coefficients and colinearity
between provisions.
• A better way could be capture the whole
variability of the provisions in just one
indicator.
• This indicator would give us a clue about the
depth of the integration.
20
Specification to estimate with PPML :
Control for large zero value flow observations in GM data sets and gives
consistent estimates.
(Silva & Tenreyro, 2006; 2011)
21
Deep integration: Latent trait analysis (Rasch) indicator Dur and
al (2014) and additive indicator on levels and logs. DESTA.
xij
dp_D_ra0
xij
xij
xij
0.129***
(0.006)
lndp_D_ra00001
0.027***
(0.001)
deep_DE_ad
0.085***
(0.003)
ln_dp_d_ad
0.033***
(0.001)
lnPIBi
0.730***
0.729***
0.746***
0.729***
(0.024)
(0.024)
(0.024)
(0.024)
0.647***
0.645***
0.662***
0.642***
(0.023)
(0.023)
(0.023)
(0.022)
-0.803***
-0.792***
-0.775***
-0.759***
(0.009)
(0.009)
(0.009)
(0.009)
-6.577***
-5.617***
-6.215***
-5.424***
(0.864)
(0.863)
(0.845)
(0.850)
587,654
587,654
587,654
587,654
R-squared
0.891
0.892
0.900
0.900
Exporter FE
YES
YES
YES
YES
Importer FE
YES
YES
YES
YES
NO
NO
NO
NO
YES
YES
YES
YES
lnPIBj
lndist
Constant
Observations
Country-pair FE
Time FE
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
22
Deep Integration: Additive indicators and its log
from WTO+ and WTO-X. PPML estimator.
xij
ad_ch1
xij
xij
xij
0.058***
(0.002)
lnad_ch00001
0.036***
(0.001)
ad_x1
0.023***
(0.001)
lnad_x100001
0.035***
(0.001)
lnPIBi
0.765***
0.734***
0.756***
0.738***
(0.024)
(0.024)
(0.024)
(0.024)
0.679***
0.647***
0.671***
0.650***
(0.023)
(0.023)
(0.024)
(0.023)
-0.784***
-0.746***
-0.819***
-0.747***
(0.009)
(0.009)
(0.008)
(0.009)
-5.860***
-5.607***
-5.810***
-5.623***
(0.886)
(0.889)
(0.904)
(0.886)
572,657
572,657
575,383
575,383
R-squared
0.900
0.903
0.894
0.903
Exporter FE
YES
YES
YES
YES
Importer FE
YES
YES
YES
YES
NO
NO
NO
NO
lnPIBj
lndist
Constant
Observations
Country-pair FE
Time FE
YES
YES
YES
YES
Time Varying Exporter FE
NO
NO
NO
NO
Time Varying Importer FE
NO
NO
NO
NO
Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1
23
Conclusions
• We are now more aware that not all PTA are
equal. This would imply that much of previous
estimations on the average effect of PTA on trade
could be biased.
• Through different indicators we have confirmed
that deeper PTAs increase trade more than
shallow ones. However these indicators also
present limits to their interpretation. A better
definition of deep integration continues to be a
challenge.
24
Conclusions
• A better deep integration indicator should show EC
model as the deepest. Meanwhile, these imperfect
indicators present enough power to give us sufficient
clues about the direction of the impact of
heterogeneity of the agreements on trade.
• Other provisions out of the traditional WTO scope
present significant positive impact on trade. Yet, some
of them could individually bear a negative impact.
• More research is needed specially to try to identify
which provisions or combination of provisions presents
a higher impact.
25
Data Set : Sources
PTA provisionas database : WTO ( 2011) Research division for the World Trade Report.
http://www.wto.org/english/res_e/publications_e/wtr11_dataset_e.htm
World Trade Institute (2014) . The Design of trade agreements database DESTA.
Bilateral Exports: International Monetary Fund (IMF) Direction of Trade Statistics Database DOTS (2013).
Current GDP and population: World Development Indicators (WDI) database, World Bank, (2013)
Area, Island and Landlocked, constructed by the author based on the World Factbook from the Central Intelligence
Agency of the United States of America (CIA)
Weighted distance, contiguity, col45 and comlang_eth9 : CEPII (2013): Head, K., Mayer, T. & Ries, J. (2010), Gravity
dataset, obs. till 2006.
Regional Trade Agreements: constructed by the author, based on the Regional Trade Agreements Information
System (RTA-IS), World Trade Organization WTO (2013)
GATT membership: constructed by the author based on the World Trade Organization WTO information (2013).
CEPII: Head, K., Mayer, T. & Ries, J. (2010), Gravity dataset, obs. Till 2006.
OCDE membership: constructed by the author based on the Organisation de coopération et de développement
économiques OCDE (2013) information.
26
References
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Horn, H., Mavroidis, P. C. and Sapir, A. (2010), “Beyond the WTO? an anatomy of EU and US preferential trade agreements”, The
World Economy 33(11): 1565-1588
Bourgeois, J., K. Dawar and S.J. evenett. 2007. A Comparative Analysis of Selected Provisions in Free Trade Agreements, Study
commissioned by DG Trade, Brussels: European Commission, mimeo
Claar S. & Nölke A. (2010) Cross-Border Cooperation: Deep integration. http://www.dandc.eu/en/article/even-though-deepintegration-could-serve-developing-countries-strategy-politically
Dür, A., Baccini, L.& Elsig, M. (2014) The Design of International Trade Agreements: Introducing a New Dataset. The Review of
International Organizations.
Orefice & Rocha (2013) Deep Integration and Production Networks: An Empirical Analysis. The World Economy. John Wiley &
Sons Ltd
Le Roux, B. and H. Rouanet (2010). Multiple Correspondence Analysis. Series Quantitative Applications in Social Sciences. 163.
Sage Publications.
Anderson, J. E. (1979) “A Theoretical Foundation for the Gravity Equation,” American Economic Review 69, 106-16.
Anderson, J & Wincoop, E. (2003) “Gravity with Gravitas: A Solution to the Border Puzzle” American Economic Review, 93 (1), p
170 – 193.
Baier, S & Bergstrand, J. (2007) Do free trade agreements actually increase members’ international trade? Journal of
International Economics. 71, 72-95.
Baldwin, R. & D. Taglioni (2006). ¨Gravity for Dummies and Dummies for gravity equations¨ NBER working paper No.12516,
National Bureau of Economics Research, Inc.
Breusch, T. S., & Pagan, A. R. (1980). The Lagrange multiplier test and its applications to model specification in econometrics.
Review of Economic Studies, 47, 239−253.
Cipollina, M. & Pietrovito, F. (2011) The trade impact of European Union Preferential Policies: A Meta-Analysis if the Literature.
In De Benedictis, L. and Salvatini, L. (2011) The trade impact of European Union Preferential Policies. An Analysis Throug Gravity
Models. Springer. 91-109.
Cipollina, M. & Salvatici, L. (2011) Trade Impact of European Union Preferences. Chp. 6, in: De Benedictis, L. and Salvatini, L.
(2011) The trade impact of European Union Preferential Policies. An Analysis Throug Gravity Models. Springer. 111-125
De Benedictis, L & Vicarelli, C. (2005) Trade Potentials in Gravity Panel Data Models. The B.E. Journal of Economic Analysis & 27
Policy. Topics in Economic Analysis & Policy. V. 5, No. 1.
References
•
Feenstra, R. (2004b) Advanced International Trade. Theory and Evidence. Chap. 5. Increasing Returns and the gravity equation. Princeton University Press. 137 -173.
•
Hausman, J. (1978) Specification Tests in Econometrics. Econometrica, Vol. 46, No. 6 (Nov., 1978), pp. 1251-1271. Stable URL: http://www.jstor.org/stable/1913827.
•
Head, K. Mayer, T., & Ries, J. (2010) The Erosion of colonial linkages after Independence. Journal of International Economics. 81, 1-14.
•
Krugman, P. (1979) “Incresing Rerturns, Monopolistic Competition, and International Trade”, Journal of International Economics, 9, pp. 469-479.
•
Liu, X., Burridge P., and Sinclair, P. (2002) Relationships between economic growth, foreign direct investment and trade: evidence from China. Applied Economics,
2002, 34, 1433-1440.
•
McCallum, J. (1995) "National Borders Matter: Canada-U.S. Regional Trade Patterns."American Economic Review, 85(3), pp. 615-23.
•
Martínez-Zarzosoa, I. & Nowak-Lehmann, F. & Horsewood, N. (2009) Are regional trading agreements beneficial? Static and dynamic panel gravity models. North
American Journal of Economics and Finance 20 46–65.
•
Markusen, J. (2002) Multinational firms and the theory of international trade. Cambridge Etats-Unis : MIT Press.
•
Muchielli, J-L. et Mayer, T. (2011) Economie International. Dalloz. 2e Edition. Paris.
•
Rose, A (2004) Do we really know that the WTO increases trade? American Economic Review 13 (4), 682–698.
•
Rose, A. (2005) Which international institutions promote trade? Review of International Economics 13 (4), 682–698.
•
Rose, A., & Engel, C., (2002) Currency Unions and International Integration, Journal of Money, Credit and Banking, Blackwell Publishing, vol. 34(4), pp. 1067-89.
•
Santos Silva, J. & Tenreyro, S. (2006) The log of gravity, The Review of Economics and Statistics, 88, 641-58.
•
Santos Silva, J. & Tenreyro, S. (2011) Further simulation evidence on the performance of the Poisson pseudo-maximum likelihood estimator. Economics Letters.
Volume 112, Issue 2, August 2011, Pages 220–222.
•
Timbergen, J. (1962) Shaping the World Economy: Suggestions for an International Economic Policy, New York : Twentieth Century Fund.
•
Trefler, D. (1993) Trade liberalization and the theory of endogenous protection: an econometric study of U.S. import policy. Journal of Political Economy. 101 (1), 138–
160.
•
Subramanian, A., Wei, S. (2007) The WTO promotes trade strongly, but unevenly. Journal of International Economics 72 (1), 151–175.
28

similar documents