Do we really know that trade agreements increase trade?

This study surveys the empirical literature in which the gravity equation has been used to study the effect of economic integration agreements (EIAs) on international trade flows. We show that most studies either focus on improving the methodology to assess regionalism’s overall impact, or on a small set of well-known agreements without necessarily adopting new methodological improvements. We bridge this gap by providing individual estimates for EIAs on world trade, while employing first-differencing techniques to correct for endogeneity bias and account for phase-in effects. Overall, EIAs promote trade by at most 50 %. Surprisingly, more than half of the EIAs investigated have had no discernible impact on trade at all, while only about one quarter of the agreements are trade promoting. Characteristics of these agreements, such as their institutional quality, design, and their members’ involvement in the World Trade Organisation, shed more light on how this variation can be understood.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic €32.70 /Month

Buy Now

Price includes VAT (France)

Instant access to the full article PDF.

Rent this article via DeepDyve

Similar content being viewed by others

The WTO’s Effect on Trade: What You Give is What You Get

Chapter © 2017

Cliometric Approaches to International Trade

Chapter © 2019

Cliometric Approaches to International Trade

Chapter © 2019

Notes

According to Kohl et al. (2013), the EEA incorporates a total of 16 policy areas. In contrast, US-Israel FTA covers just slightly more than half that amount.

The user-written Stata packages reg2hdfe and reg3hdfe were used to estimate our linear regression models with high-dimensional fixed effects for the entire dataset. See Carneiro et al. (2012) and Head and Mayer (2014).

The appendix shows that our findings are similar to those of B&B when the dataset is restricted to their sample of 96 countries and 1950–2000.

Foster and Stehrer (2011) provide an interesting exception by investigating both the aggregate and individual impact of several major EIAs on intra-industry trade.

Abbreviations are explained in Table 5.

In related work, Park and Park (2011) find that agreements established under GATT Article XXIV are more effective than those established under the Enabling Clause.

Group means were compared using ANOVA and a Tukey HSD pairwise test. “Mixed” and “All in WTO” are both significantly different to “None in WTO”, but there is no significant difference between the “Mixed” and “All in WTO” groups.

The nine indicators convey information about whether the signatories have agreed to how consultations are to proceed, whether technical issues have been defined in the agreement, whether details about a dispute settlement mechanism are provided, whether an evolutionary clause has been specified, whether the institutional framework has been laid out, the objectives of the agreements formulated, if a plan and/or liberalisation schedule has been specified and whether information is provided as to how the signatories intend for the institutional arrangement to be transparent. See Kohl et al. (2013) for full details.

The 17 provisions are on policy areas such as agriculture, anti-dumping and countervailing measures, capital mobility, competition, customs administration, environmental issues, export restrictions, labour issues, import restrictions, intellectual property rights, investment, public procurement, sanitary and phytosanitary measures, services, state aid, state trading enterprises, and technical barriers to trade. See Kohl et al. (2013) and the online dataset at http://tristankohl.org for more detail about the dataset and for an application of the gravity equation that supports the findings in this section.

This finding raises the idea that EIAs serve a network function, connecting a broad range of countries and commodities. As the global value chain is continually sliced into smaller pieces, it is not the separate parts comprised of EIAs or regions that matter as much for international trade as the sum of these parts in an intertwined, interconnected network of economic integration. This avenue is further explored in Alba et al. (2010).

References

Acknowledgments

The author thanks Rob Alessie, Jeffrey Bergstrand, Steven Brakman, Harry Garretsen, Will Martin, Peter van Bergeijk, participants at the Spring Meeting of Young Economists (Groningen), International Trade and Finance Association’s Annual Conference (Eilat) and the Conference on Globalisation, Strategies and Effects (Koldingfjord), three anonymous referees and the editor for valuable comments.

Author information

Authors and Affiliations

  1. Faculty of Economics and Business, University of Groningen, P.O. Box. 800, 9700 AV, Groningen, The Netherlands Tristan Kohl
  1. Tristan Kohl