Trade effects of WTO | VOX, CEPR Policy Portal
Mario Larch, José-Antonio Monteiro, Roberta Piermartini, Yoto Yotov 20 November 2019
Over the past decades, international trade has grown considerably. A broad range of factors are behind this trend, including the widening and deepening of the multilateral trading system. The conclusion of the GATT in 1947 and, more recently, the creation of the WTO in 1995 established a set of trade rules covering agriculture and manufacturing goods, services, and intellectual property rights. Besides trade rules, the WTO also serves as forum to solve disputes, negotiate new rules, and promote transparency of trade policy.
As of 2019, the WTO has 164 members, with 22 other countries currently negotiating their accession. For many of these members, joining the GATT/WTO required important changes in their domestic and trade policy. Besides binding and reducing tariffs, WTO membership involves commitments related to non-tariff barriers and transparency. These benefits are all the more important because – even with the explosion of regional trade agreements (RTAs) in recent years – 75% of world merchandise trade still takes place on a non-discriminatory most-favoured nation (MFN) basis. And in most cases, this rate applies to non-Member countries as well. The core role of the multilateral system in underpinning and anchoring world trade growth is often underestimated or overlooked in journalistic and some academic circles.
Estimating the effect of GATT/WTO membership on trade is not straightforward
Part of the controversy on the role of trade policy might be linked to the debate about the empirical evidence of the impact of GATT/WTO membership on international trade. While economic theory suggests a positive impact of Membership on trade through a reduction in trade costs by reducing tariffs and non-tariff barriers and enhancing transparency and predictability, the empirical evidence is more mixed.
The effect of GATT/WTO membership is typically estimated in empirical gravity models by introducing a dummy variable that denotes whether both countries in a pair are GATT/WTO members. According to economic theory, (1) growth in trade volumes because of GATT/WTO accession is influenced by the depth of the acceding country’s commitments; and (2) GATT/WTO membership diverts, if any, trade away from other non-GATT/WTO member countries.
Following Rose’s (2004) seminal paper, in which he finds no significant effect of GATT/WTO membership on bilateral trade and on diverting trade from non-members to members, several studies have reassessed the impact of GATT/WTO membership by refining the gravity model. Some of these studies suggest that the effect of GATT/WTO membership can only be accurately estimated when the main following issues are taken into account in the estimation: (1) the need to distinguish between asymmetries in commitments across membership (Subramanian and Wei 2003); (2) contemporaneous existence of other trade agreements (Tomz et al. 2004, Eicher and Henn 2011); and (3) the omission of zero trade observations (Helpman et al. 2004).
More recently, using nonparametric methods, Chang et al. (2011) report a large and positive effect of GATT/WTO membership on trade (between 74% to 277% for countries both in GATT/WTO). Dutt et al. (2013) find that WTO membership increases the extensive margin of exports by 25% but has a negative impact on the intensive margin. Roy (2011) finds no statistically significant effect of WTO membership once country-time fixed effects are included in the gravity model. As noted by Cheong et al. (2014), the fragility of GATT/WTO membership estimates in the standard gravity model is mainly due to the multicollinearity between the multilateral resistance terms and the WTO membership variable.
Yet, gravity models so far have neglected the effect of GATT/WTO membership on trade when these effects occur to WTO and non-WTO members alike. The rule-based system that underpins WTO is largely MFN – that is, it does not discriminate between members and non-members. This is true de facto because in many cases countries apply the same tariff rate to member and non-members countries. However, it is also true de jure because the nature of some commitments is MFN – as, for example, in the case of rules on transparency provisions as well as prohibition of some forms of subsidies. In other words, the system of rules guarded by WTO commitments has a public good dimension. It helps to generate a predictable and transparent trading environment that has the potential to positively affect trade. Measuring the trade effects due to the public good nature of GATT/WTO entry is difficult in a standard gravity model specification for various reasons, including multicollinearity.
A novel (theory-consistent) approach makes the case for the missing trade effect of GATT/WTO membership on trade
In a recent paper (Larch et al. 2019), we capitalise on the latest developments in the empirical structural gravity literature (Yotov et al. 2016) to revisit the question of whether and how much GATT/WTO membership promotes international trade of member countries. The most significant departure from the existing work is that our estimation of the structural gravity model includes intra-national trade in addition to international trade flows. This enables us to take into account possible diversion from domestic sales toward international trade flows. We make two contributions to the existing literature.
First, we obtain unilateral estimates of the effects of GATT/WTO membership. These results are new in the literature. Although these effects tend to be country- and time-specific, they are often large, positive, and statistically significant at any conventional level. Specifically, our estimate implies that, on average, joining GATT or the WTO has led to about a 72% increase in the international trade of member countries relative to their domestic sales in our full dataset covering 178 trading partners over the period 1980-2016.
Second, we obtain positive, statistically significant and larger bilateral estimates of the effects of GATT/WTO membership on international trade between member countries, as compared to corresponding estimates from the related literature. Specifically, our results imply that GATT/WTO membership has increased trade by 171% between member countries on average. The explanation for this result is that our specification captures trade creation effects due to GATT/WTO membership that stem from the reduction of domestic sales. This in no way means that production is lowered. It simply means that more is exported and imported rather than sold domestically. This reallocation is one of the drivers of the gains from trade leading to larger benefits for consumers.
Third, the estimation specification that simultaneously includes the unilateral and the bilateral GATT/WTO variables enables us to draw an inference about the impact of GATT/WTO membership on trade between member states as well as trade between member and non-member countries. The estimates of these effects are also large, positive, and statistically significant. Specifically, our estimate implies that, on average, joining GATT or WTO has led to about an 88% increase in the international trade between member and non-member countries.
The creation of the GATT/WTO has generated significant trade gains for its members but also non-members. Our research shows that, on average, GATT/WTO membership increased trade between members by 171% and trade between member and non-member countries by about 88%. Part of this positive effect of GATT/WTO membership on trade between members as well as trade between member and non-member stems from the non-preferential nature of some of the GATT/WTO commitments that gravity models have so far neglected. Commitments to notify trade policy changes, to avoid unnecessarily restrictive technical regulations, or to remove export subsidies in agriculture are some examples of WTO rules that increase transparency and reduce the uncertainty of trade policy to the benefit of members and non-members.
Authors’ note: The views expressed in this column are those of the authors and do not necessarily reflect the position of the WTO. Any errors or omissions are the responsibility of the authors.
Chang, P-L, and M-J Lee (2011), “The WTO Trade Eﬀect,” Journal of International Economics 85(1): 53–71.
Cheong, J, D Won Kwak, and K Ki Tang (2014), “The WTO Puzzle, Multilateral Resistance Terms and Multicollinearity,” Applied Economics Letters 21(13): 928–933.
Dutt, P, I Mihov, and T Van Zandt (2013), “The Eﬀect of WTO on the Extensive and the Intensive Margins of Trade,” Journal of International Economics 91(2): 204–219.
Eicher, T S, and C Henn (2011), “In Search of WTO Trade Effects: Preferential Trade Agreements Promote Trade Strongly, but Unevenly,” Journal of International Economics 83(2): 137-153.
Helpman, E, M J Melitz, and S R Yeaple (2004), “Export versus FDI with Heterogenous Firms,” American Economic Review 94(1): 300–316.
Larch, M, J-A Monteiro, R Piermartini, and Y Yotov (2019), “On the Effects of GATT/WTO Membership on Trade: They are Positive and Large after All”, CESifo Working Paper no. 7721.
Rose, A K (2004), “Do We Really Know That the WTO Increases Trade?,” American Economic Review 94(1): 98–114.
Roy, J (2011), “Is the WTO Mystery Really Solved?,” Economics Letters 113(2): 127–130.
Subramanian, A, and S-J Wei (2007), “The WTO Promotes Trade, Strongly But Unevenly,” Journal of International Economics 72(1): 151–175.
Tomz, M, J Goldstein, and D Rivers (2007), “Do We Really Know That the WTO Increases Trade? Comment,” American Economic Review 97(5): 2005–2018.
Yotov, Y, M Larch, L Piermartini, and J-A Monteiro (2016), An Advanced Guide to Trade Policy Analysis: The Structural Gravity Model, WTO & UNCTAD.