Evaluating a trade policy using a revealed preference approach.

Abstract

The traditional theory of international trade suggests that, in perfectly competitive markets, international trade always increases welfare. The theory of second best suggests that, in the presence of multiple distortions, a reduction in a distortion may actually reduce the welfare. With the welfare consequences of trade reform still being debated, there have been very sparse empirical studies to test these results. Moreover, the empirical studies done so far on this issue are mostly based on ex-ante approach, which looks for a set of policy prescriptions, which yield welfare improvement. Ju and Krishna's model built upon that of Dixit and Norman has shown that Ohyama's conditions under the assumption of many consumers and small country case are sufficient to ensure that a trade reform is a Pareto improvement. However, their model has not yet been tested empirically. This study, therefore, attempts to develop an empirical method to test Ohyama's and others' revealed preference approach, which looks for some indicator to determine if welfare has risen due to a trade reform. This study also applies the empirical method to test welfare effect of a trade reform. The study chose U.S.A. and Mexico for observation and considers the signing of NAFTA by the two countries as a form of trade reform. It then applies two empirical approaches: linear regression model approach and intervention model approach to test the hypothesis that U.S. welfare has increased due to liberalization of its trade with Mexico under the NAFTA agreement. The test results from both the linear regression model and the intervention model confirm the hypothesis that U.S. welfare has increased due to the liberalization of its trade with Mexico.