In Fujita (1988) and Krugman (1991), the interaction of increasing returns to scale and transport costs creates a self-reinforcing process of industry agglomeration. Firms desire to concentrate production near large consumer markets since this allows them to economize both on transports costs and fixed production costs. As more firms locate in an industry center, the center becomes more attractive as a production site. Immobile factors of production, land rents, or other sources of congestion costs work against agglomeration. As long as agglomerative forces are sufficiently strong, firms will geographically concentrate in one or more industry centers. Where industry centers happen to form may only be partially determined by the exogenous features of regions, such as climate, access to waterways, or supplies of natural resources. Thus, historical factors or policy decisions, intentioned or not, may influence which regions develop an industrial base.

Free trade agreements influence the location of economic activity by expanding the set of markets that firms serve. As a country lowers its trade barriers, there is an increase in foreign demand for domestically-produced goods. All else equal, free trade gives domestic firms an incentive to move production to regions with relatively good access to foreign markets, such as border areas or port cities. The incentive to relocate is likely to be stronger in small economies, such as Canada and Mexico, where exports are a large fraction of sales, and weaker in large economies, such as the United States, where exports are a small fraction of sales. As firms begin to move towards specific regions, a self-reinforcing process is created, which may cause entire industry centers to relocate. A fall in trade barriers may thus cause some regions to expand and other regions to contract. loans

To examine the relationship between trade policy and industry location, I begin by reviewing patterns of regional development in North America. I then discuss evidence on how North American economic integration has influenced the location of economic activity in the three countries and conclude by considering the impact of future initiatives towards hemispheric integration on regional economies. I make no attempt at a comprehensive survey of the literature and instead choose papers to illustrate conceptual issues.