Trade liberalization and the geographic distribution of industry in the United States and Mexico

This dissertation consists of four chapters: (1) an introduction; (2) a history of the relationship between international trade theory and location theory; (3) an empirical study of recent changes in the geographic distribution of industry in the United States and Mexico; and (4) a model in the 'new economic geography' class intended to help explain the changes. Specifically, following the introduction, chapter 2 surveys the history of attempts throughout the twentieth century to find common ground for trade theory and location theory. The chapter suggests that the new economic geography can be read as a new attempt to achieve an old objective: to unify the two sub-fields within a general theoretical structure. Chapter 3 outlines the empirical changes in the geographic distribution of industry in the U.S. and Mexico from 1988 to 1993, a period of dramatic trade liberalization. Maps of the density of manufacturing employment in the U.S. and Mexico are presented to reflect the changes: the two nations are depicted as four locations on an axis running from north to south, with both nations' traditional manufacturing centers at the extremities. The extremities have experienced declines in their shares of national manufacturing employment in the period of trade liberalization, while the locations nearer the border have experienced growth in their shares. Chapter 4 develops a new economic geography model to study the causes of the stylized changes. The model features multiple regions in each of multiple nations, allowing analysis of changes in the distribution of industry both within and between nations. The international border is defined in terms of limitations to the mobility of goods and labor and the geographic distribution of tariff revenue. The border breaks the symmetry among locations that is typical of the new economic geography models: locations nearer the border have a 'natural advantage' in their closer average proximity to all other markets. Simulations of the model show how the stylized changes in the U.S. and Mexico may arise from trade liberalization. The natural advantage of border regions is enhanced by liberalization, outweighing the 'circumstantial advantage' of the pre-existing industry concentrations at the extremities.