Could trade policy be responsible for regional employment shifts in Canada, Mexico, and the United States? The sheer size of the U.S. economy makes it unlikely that foreign trade could have a substantial impact on industry location. From Table 1 it is evident that the regional reallocation of employment towards the west and southeast has been ongoing for nearly a century, during which the U.S. economy’s exposure to world markets has oscillated.

Factors such as the development of the interstate highway system and the advent of air conditioning likely have more to do with regional population changes in the United States than does trade with the rest of the world.

For Canada and Mexico, however, the relationship between international trade and industry location appears to be much stronger. Canada’s industry centers developed alongside those in the United States, forming an extension of U.S. industry agglomerations. Until recently, for instance, auto production in the two countries was geographically concentrated in the Michigan-Ontario industrial corridor. In Mexico, the demise of the Mexico City manufacturing belt and the rapid expansion of industry in northern Mexico followed the formation of closer economic ties with the rest of the world, and with the United States in particular. At first glance, at least, access to foreign markets seems to have played an important role in these countries’ regional economic development. credit

A correlation between proximity to the United States and regional economic development in Canada and Mexico in no way implies causality. The Canadian-U.S. border is the site of important waterways that link markets in interior Canada to markets in Europe and elsewhere. Similarly, northern Mexico has stocks of natural resources, including timber and iron ore, which supply Mexican industry. How do we know that it is international trade, and not these other factors, that is driving regional economic development in Canada and Mexico? A related question has been at the heart of much recent empirical literature on industry location and regional economic growth: how do we know that industry agglomeration, such as that which occurred in the U.S. manufacturing belt, is due to scale economies and not to exogenous factors, such as supplies of natural resources?