How does trade respond to Preferential Trade Agreements?

Neil Foster and Robert Stehrer — The Vienna Institute for International Economic Studies (WIIW)


Beyond Trade Creation

Most studies considering the effects of Preferential Trade Agreements (PTAs) on international trade concentrate on the trade creating and diverting effects of PTAs. Trade creation arises from the elimination in distortions between the relative prices of domestic goods and those of other members, while trade distortion arises from the introduction of distortions between the relative prices of members and non-members. Empirical research looks to quantify these two effects — usually using the gravity model of international trade — with results generally supporting the trade creation argument and being mixed for the trade diversion hypothesis.

In two recent papers we move beyond the relationship between PTAs and the volume of trade to consider how PTAs affect trade structures. In particular, we examine whether the presence of a PTA between two countries impacts upon the structure of trade (i.e. whether it affects the level of intra-industry trade) and whether PTAs affect trade through the extensive (i.e. the variety of traded goods) or intensive (i.e. the volume of trade) margin of trade. We make use of publicly available trade data from Feenstra et al (2005) over the period 1962-2000 for up to 172 countries.

PTAs and Intra-Industry Trade

With one or two recent exceptions the impact of PTA membership on the structure of trade has largely been ignored. The issue of whether PTAs stimulate gains due to specialisation (i.e. inter-industry trade) or to gains from scale economies and product differentiation (i.e. intra-industry trade, IIT) is an important one however, particularly when considering the question of which countries should form a PTA. If the trade-creating effects of PTAs work through an increase in IIT we may expect the resource reallocation effects in the short- to medium-run to be relatively low as little change in inter-industry factor movements would be required.

We examine how trade responds to PTA presence using the gravity model and the Grubel-Lloyd index of IIT. Being careful to take account of endogeneity through the use of a variety of fixed effects (as suggested by Baldwin and Taglioni, 2006) we regress both the level of trade and the Grubel-Lloyd index on standard gravity determinants and a dummy for the presence of a PTA. Our results indicate that the presence of a PTA between two countries is trade-creating and that a good deal of this additional trade is due to an increase in IIT. Despite this general conclusion we also show that there is a great deal of heterogeneity in results when we consider specific PTAs, where in some cases we find that PTAs work primarily through increased specialisation. We consider these results further by introducing the possibility of non-linear effects of PTAs. In particular, we test a number of hypotheses developed by Bergstrand (1990) and include interactions between the PTA variable and the difference in per capita GDP and the product of per capita GDP. Our results indicate that the impact of PTAs on IIT is increasing in the product of a bilateral-pair’s GDP per capita and decreasing in the difference between GDP per capita. Such results are in line with the theoretical predictions of Bergstrand (1990) and suggest that PTAs affect IIT to a greater extent for trade involving at least one developed country.

PTAs and the Variety of Exports

In a second paper, we again adopt a gravity approach as well as a matching procedure to examine the impact of PTAs on export variety. To date there has been very little literature considering the impact of PTAs — or liberalisation more generally — on the variety of trade. This is despite many theoretical models of trade and growth emphasising the importance of access to new products and varieties from abroad in raising productivity. Recently, Feenstra and Kee (2008) have shown that the variety of exports is also related to country productivity in a sample of 48 countries. Their theoretical model relates to the recent literature on heterogeneous firms, with firms self-selecting into exporting markets. Since more productive firms self-select into export markets and are thus more productive than the average domestic firm, an increase in the number of firms exporting and therefore an increase in export variety is associated with rising productivity.

We use a similar dataset to above, but concentrate on the contemporaneous effects of PTAs on both export variety and volumes (i.e. the extensive and intensive margins). The results indicate, once again, that PTAs are trade-creating, with the contemporaneous effect of PTAs be to increase exports by around 12 percent (depending upon the specification).  Our results also indicate that much of the trade creating effects of PTAs occurs along the extensive margin, i.e. by increasing the variety of products exported. In particular, we find that the extensive margin accounts for between 78 and 106 percent of the increase in exports following the formation of a PTA. We once again consider the possibility of non-linearities, since Frankel et al (1995) show that the formation of a PTA between two large partners creates trade in more varieties than a PTA between two small partners. Our results indicate that the trade creating effects of PTA formation are larger in smaller exporters and for smaller country pairs, but that PTAs have a significantly larger effect on the extensive margin in larger country pairs, consistent with the hypothesis of Frankel et al (1995).


PTAs have effects beyond trade creation and diversion, increasing the extent of IIT and trade in variety. To the extent that the former is enhanced the resource reallocation costs of PTAs are likely to be muted, while the increase in the latter may lead to an additional productivity effect of PTAs, as emphasised in the theoretical literature on trade and growth. Such benefits have until recently rarely been discussed in the literature on PTAs. Our results indicate that these benefits are likely to be unevenly spread across country-pairs however, being stronger for economically larger country-pairs and for more similar country-pairs.


Baldwin, R. and D. Taglioni, 2006. Gravity for dummies and dummies for gravity equations. NBER Working Paper no. 12516, National Bureau of Economic Research.
Bergstrand, J.H., 1990. The Heckscher-Ohlin-Samuelson model, the Linder hypothesis, and the determinants of bilateral intra-industry trade. Economic Journal, 100, 1216–1229.

Feenstra, R.C. and H.L. Kee, 2008. Export variety and country productivity: Estimating the monopolistic competition model with endogenous productivity. Journal of International Economics, 74, 500-514.

Frankel, J., Stein, E. and S-J Wei, 1995. Trading blocs and the Americas: The natural, the unnatural, and the super-natural. Journal of Development Economics, 47, 61-95.

Foster N. and Stehrer R., 2010, Preferential Trade Agreements and the Structure of International Trade, The Vienna Institute for International Economic Studies Working Paper Series no. 61.

Foster N., Poeschl J. and Stehrer R. ,2010, The Impact of Preferential Trade Agreements on the Margins of International Trade, The Vienna Institute for International Economic Studies Working Paper Series no. 70.

The research summarised was prepared as part of the EU FP7 programme project ‘World Input-Output Database: Construction and Applications’ (WIOD). This project is funded by the European Commission, Research Directorate General as part of the Seventh Framework Programme, Theme 8: Socio-Economic Sciences and Humanities. Grant Agreement no. 225 281.



Robert Stehrer is Deputy Director of Research at the Vienna Institute for International Economic Studies and Lecturer at the University of Vienna

Neil Foster is Research Economist at the Vienna Institute for International Economic Studies and Lecturer at the University of Vienna


> Guide to downloading files