What does the prospect of future hemispheric economic integration hold for industry location in North America? In considering this question, it is worthwhile to summarize the factors that have contributed to recent regional employment changes in the region. The most dramatic spatial movements in production have occurred in Mexico, with the breakup of the Mexico City manufacturing belt and the shift in economic activity to northern Mexico. The empirical literature I review suggests that increasing returns to scale and transport costs have contributed to these events. there

Several idiosyncratic factors have also shaped the emerging geography of production in North America. First, Mexico and the United States share a land border, and, as a result, over 80 percent of merchandise trade between the two countries is transported by truck or rail. This makes the Mexico-U.S. border an obvious production site for Mexican firms wishing to export to the U.S. market and U.S. firms wishing to export to the Mexican market. The recent formation of regional production networks on the Mexico-U.S. border mirrors the Canadian-U.S. production networks that formed earlier in the century.

Second, the U.S. economy is substantially larger than the Canadian or Mexican economies. As a result, the United States exerts a strong pull on industry location in the economies of its neighbors, but the effects in the reverse direction are relatively weak. In other words, Ross Perot got it wrong. The “giant sucking sound” he heard is the United States pulling Mexican industry north to the Mexico-U.S. border, more than Mexico pulling U.S. industry south. While employment in U.S. cities on the border with Mexico is growing much faster than in the United States as a whole, the U.S. border economy is still quite small relative to the rest of the United States. Mexico isn’t big enough yet to have a substantial impact on the spatial organization of production in the United States.