Essays on the political economy of international trade

This dissertation examines the theoretical and empirical issues of the political economy of international trade. Chapter 1 investigates the founding principle of GATT, the Most Favored Nation Clause (MFN), and why the United States and other industrial countries which once supported unconditional MFN now routinely adopt policies that circumvent MFN. This chapter develops a model of multilateral trade negotiations to illustrate some of the trade-offs large countries face in adopting MFN and how these trade-offs may change over time. It is shown that, as capital accumulates in the export sectors of the South, the cost of MFN to the North rises relative to the benefits. Thus, the very success of MFN in promoting comparative advantage may cause the North to switch from being a net beneficiary to a net loser from MFN.; Chapter 2 develops a model to explain the relationship between R&D and lobbying in an international trade framework. First we show that both types of expenditures are increasing with the level of investment effectiveness or the level of development of the country. Next we compare the welfare under free trade with the welfare under endogenous tariff formation. It is found that when the policy making country is more developed than its trading partner, free trade will never be the preferred choice when there are domestic special interest groups. When the policy making country is less developed, the outcome depends on the investment effectiveness of its trading partner and other political parameters.; Chapter 3 considers an empirical application of the gravity model with non-homothetic preferences developed by Deardorff (1998). In contrast to existing research, non-homotheticity is introduced in the form of an index which looks at the deviation of trading partners' consumption and production patterns from the world patterns. Even though the initial results for trade within the OECD and between OECD and non-OECD countries are as predicted by the model, they are not robust to different datasets. While the results of the Tobit regression for the non-OECD countries appear to be consistent with the model's prediction, the results of the fixed effects regressions are not.