This research project examines, both theoretically and empirically, the effects of non-discrimination commitments in international trade agreements. The focus is on the WTO rule of generalized, unconditional Most-Favored Nation (MFN) treatment. It has long been suspected that this rule leads some countries, mostly developing countries, to "free ride" on the trade negotiations of others, resulting in trade agreements with inefficiently high levels of protection. A formal theoretical model of this free-rider effect is developed and tested first with US tariff data. It is then extended to conduct a cross-country study of the MFN free-rider problem and several related issues.
The theoretical model offers two key predictions. First, there should be a negative relationship between a country's negotiated tariff on a given product and the degree to which the exports of that product to that country are concentrated among foreign countries, as measured by a Herfindahl-Hirschman index (HHI) of export shares. Using US MFN tariff rates for both 1983 and 1989-1999, we find strong evidence of this prediction, even after controlling for a host of political and economic factors.
The second prediction is that the severity of the MFN free rider effect should be more pronounced for markets in which the country has greater monopoly power in trade. This is addressed using measures of monopoly power across goods and across countries. The project builds a global trade, production, and tariff database and uses it to test the generality of the MFN free-rider effect and to add up the global impact of MFN.
The project also addresses two related issues. First, the data show that US tariffs are lower on products imported principally from major US export destinations. This suggests that tariff concessions reflect considerations of reciprocity. To model this, the project expands the dimensionality of the existing theoretical model to accommodate many countries with interdependent tariff schedules. Second, the theory predicts and US data show that unlike tariffs, non-tariff barriers (NTBs) are uncorrelated with the HHI. This provides a strategy for testing whether NTBs and tariffs are complements or substitutes, using HHI as an instrument for the tariff.
This project is the first of its kind to empirically identify the MFN free rider problem, and it contributes to a burgeoning literature on the goals and side effects of WTO rules. Its broader impact is to contribute to better institutional design of the WTO and possibly to reduce one of the main sources of tension between developed and developing countries in global trade negotiations.
