The WTO’s effect on trade: What you give is what you get

Abstract

Do countries trade more when they participate in theWorld Trade Organisation (WTO)? After Rose’s (Am Econ Rev 94:98-114, 2004) initial 'ʼnon-effect’', the literature has developed in several ways to re-examine this unexpected result. This paper gives a detailed overview of the developments and exposes the main biases that plague previous contributions. Using a dataset covering 181 countries for the period 1948-2007, we show that zero trade flows are best incorporated using (zero-inflated) negative binomial maximum likelihood estimation. We find that formal members gained more than non-member participants, and the level of WTO participants experienced gains go hand in hand with the extent of their multilateral liberalisation commitments. Developed nations gain more than developing or least developed countries, although poor countries do benefit from trading under the Generalised System of Preferences (GSP). We also correct for selection bias with respect to economic integration agreements and find, overall, that regionalism has a lower trade-promoting effect than WTO membership.