ECONOMIC INTEGRATION AND INDUSTRY LOCATION: Regional Economic Growth in Mexico Before and After Trade Reform

We expect trade liberalization to induce greater specialization in production and, as a consequence, a sectoral reallocation of resources. There are several reasons why falling trade barriers might also contribute to a spatial reallocation of resources. First, if individual industries are geographically concentrated, then regions that specialize in import-competing goods are likely to contract and regions that specialize in export industries are likely to expand. Second, the existence of transport costs gives firms that export a substantial fraction of their output or that use imported intermediate inputs intensively an incentive to locate their activities in regions with low-cost access to foreign markets. The existence of spatial agglomeration effects may compound these factors, as the size or mix of industries in a region also affects how it adjusts to trade. Venables (1996), for instance, shows that in the presence of scale economies vertical linkages between industries help make the location decisions of buyers and suppliers interdependent. Expansion in one industry may contribute to the expansion of upstream and downstream industries.

In Hanson (1998b), I examine regional industry employment growth in Mexico before and after trade reform in an attempt to disentangle the various effects that falling trade barriers have on industry location. I estimate state-level employment growth in individual manufacturing industries for two time periods, 1980-1985, the period immediately preceding trade reform, and 1985-1993, the period immediately succeeding trade reform, as a function of the following factors: (1) distance to the Mexico-U.S. border, which proxies for transport costs to foreign markets, (2) the initial regional concentration of upstream and downstream industries, which captures the effects of backward-forward linkages, and (3) the initial regional concentration of activity in the same industry, which captures the effects of within-industry agglomeration economies. comments