Preferential Trade Agreements and Services

Ms Amélie Guillin, Ph.D student in International Trade (University of Paris I)


(solamente en inglés)

Trade in services is growing faster than trade in goods (Koncz and Flatness, 2007). As a consequence, multilateral and bilateral trade negotiations have increasingly included services issues. Though negotiations can be tough in general, it has been particularly difficult for services. It might come from the characteristics of services. First, there are various modes of delivery of services which could present an impediment to quickly concluding an agreement. The four modes are: cross-border trade in services (mode 1), consumption abroad, (mode 2), commercial presence (mode 3), temporary movements of natural persons (mode 4). Furthermore, the barriers in services are different in nature involving regulatory barriers to market access and national treatment for foreign service providers.

When two (or more) countries negotiate an agreement, they generally expect an increase in trade flows. Although it seems to be evident that trade between partners would increase in response to lower tariff and/or non-tariff barriers, some studies conclude that the impact of a trade agreement on intra-trade is not significant. Nevertheless, Baier and Bergstrand (2002) find a positive impact of RTAs on trade, highlighting that the common specification of the gravity equation underestimates the RTA coefficient (roughly by 75% with OLS), as in Trefler (1993) due to an endogeneity bias. Using a panel with bilateral-pair fixed effects and accounting for multilateral resistance terms, Baier and Bergstrand (2007) conclude that free trade agreements (FTAs) increase trade by 58%.

Following the literature, I use a simple gravity model with panel data including bilateral fixed effects and country-and-time fixed effects, and first-differenced panel estimates to analyse the role played by RTAs on trade in services. Accounting for the potential endogeneity issue, we can observe that RTAs increase intra-trade in services by 11%-17% according to the methodology, as the coefficient of RTAs is positive and significant. Nevertheless, all the RTAs do not increase trade in services equally : only Economic Integration Agreements (EIAs), which covers services only, make trade grow (by 32%), while non EIAs do not boost intra-trade. Even distinguishing amongst non EIAs, as Preferential Trade Agreements (PTAs), Custom Unions and Free Trade Agreements, no other trade agreements impact significantly trade in services. Consequently, signing a PTA only does not lead to sufficient linkages between trading partners to promote trade in services effectively.

Concluding an agreement on goods and subsequently negotiating an agreement in services is, nevertheless, not useless (case of European Union, ANZCERTA, or ASEAN-China as examples) because it does increase trade : by 34% instead of 32% without any previous negotiations. Creating strong commercial relationships with the other signatories, encourages those countries to choose the members of the agreement as trading partner for services.

Amongst the RTAs, commitments can differ significantly. As EIA is the only type of agreement available when countries negotiate on services, it may be important to account for a potential heterogeneity amongst EIAs. To distinguish the EIAs according to the depth of liberalization, I collected information on four modes of services delivery accessible in the legal texts of the RTAs. I focused my analysis on the presence or not of 30 articles, giving a weight of zero if the article was not included in the agreement. From these scores by mode I calculated, I do a hierarchical clustering to classify and divide EIAs into three groups according to their openness. (The cluster 3, EIAhigh, corresponds to the deepest agreement, the cluster 2, EIAmedium, to a less deep agreement and so on).

Considering the three groups obtained from the hierarchical clustering, we can say that even amongst EIAs, not all have a similar impact on trade in services. Deeper liberalization leads to a higher intra-trade between the members of a same EIA. There is no impact at all of EIAlow, while the coefficients of EIAmedium and EIAhigh are positive and significant. The depth does matter for trade in services.

Services have become an issue that no-one can ignore. More and more, goods and services are included in agreements as countries sign a FTA and an EIA.  In order to have  greater coverage, PTA signatories now follow this tendency, combining an agreement on goods with a deep agreement on services, sequencing or not the liberalization of the economies.



Baier, S. L. and J. H. Bergstrand (2002). On the Endogeneity of International Trade Flows and Free Trade Agreements.  Manuscript (August).

Baier, S. L. and J. H. Bergstrand (2007). Do Free Trade Agreements Actually Increase Members International Trade?, Journal of International Economics, 71, 1, 72.95.

Guillin, A. (2010).  Trade in Services and Regional Trade Agreements: Do Negotiations on Services Have to Be Specific., Mimeo.

Koncz, J. and A. Flatness (2007). U.S. International Services: Cross-Border Trade in 2006 and Sales Through Affiliates in 2005. Survey of Current Business, October.

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


Ms Amélie Guillin Ph.D student in International Trade (University of Paris I), and early stage researcher fellow at University College Dublin in the Globalization, Investment and Services Trade (GIST-CEPR) network.Her research mainly focuses on trade in services and regional trade agreements.

> Guide to downloading files