ECONOMIC INTEGRATION AND INDUSTRY LOCATION: TRADE, INCREASING RETURNS, AND REGIONAL ECONOMIES 2

To understand the relationship between international trade and industry location, we must first identify the underlying forces that shape the spatial distribution of economic activity. In this section, I briefly review empirical research on the relationship between increasing returns to scale and industry location and then examine the particular question of how trade shapes the location and organization of production in North America.

Is Geographic Concentration Evidence of Increasing Returns?

The geographic concentration of economic activity is often cited as evidence of increasing returns to scale. With capital and labor mobile in the long run, there would need to be an implausible concentration of immobile resources to produce cities the size of Chicago, Los Angeles, or New York. Yet, the time-invariant characteristics of regions, such as the supply of water, the existence of natural harbors, and weather, surely influence the formation of cities and the spatial distribution of resources.

One strategy to empirically distinguish the effects of increasing returns on industry location from those of site-specific factors is to examine the variation in industry growth rates across regions. While the concentration of natural resources in a particular location may contribute to industry agglomeration in that region, it will not, except under specific forms of technological change, contribute to the industry’s expansion over time. Using data on long-run changes in industry employment in U.S. cities, Glaeser et al. (1992) find that employment growth in a regional industry is positively correlated with the initial diversity of industrial activity in the region. Henderson, Kuncoro, and Turner (1995), using annual data for individual manufacturing industries in U.S. cities, discover that regional industries with larger initial employment levels tend to grow faster, but only over short horizons. Both studies interpret the positive correlation between regional industry employment growth and the initial concentration of economic activity as evidence of dynamic agglomeration economies. In a related branch of literature, Wheeler and Mody (1992) and Head, Ries, and Swenson (1995) find that, controlling for international differences in factor prices, multinational firms are more likely to locate in a region the larger is the local concentration of existing multinationals. These findings are also interpreted as evidence of agglomeration economies. direct lenders payday loans