A second strategy to identify the effects of increasing returns on industry location is to use the spatial covariation in wages, employment, and income to estimate the magnitude of scale economies directly. Ciccone and Hall (1996) find that the correlation between labor productivity and the spatial density of employment across U.S. states is consistent with small but significant aggregate increasing returns to scale. To control for the potential endogeneity of spatial employment densities, they use historical data on population levels and transportation infrastructure as instrumental variables. In a similar vein, Rauch (1993) finds that wages for workers are positively correlated with the local concentration of economic activity. In Hanson (1998a), I use data on U.S. counties to estimate the structural equations of the Krugman (1991) model. Using a time-differenced specification to control for the unobserved site-specific characteristics, I find evidence of large transport costs and moderate scale economies. The magnitude of increasing returns that I estimate are sufficiently strong to support the spatial agglomeration of industry. online payday loan lenders
A growing body of empirical work suggests that increasing returns to scale contribute to the geographic concentration of economic activity. The literature helps answer questions such as why industrial firms tend to concentrate near large cities or, at the level of individual industries, why microelectronics firms tend to concentrate in a place like Silicon Valley. An important question for trade policy is how specific shocks, such as the formation of free trade areas, influence the spatial distribution of economic activity. Recent literature extends empirical work on increasing returns to address this question.