Theory has two important predictions for the regional wage structure. First, nominal wages will be relatively low in regions that have relatively high transport costs to either Mexico City or the United States. Second, trade reform, by increasing U.S. demand for Mexican goods, will weaken the pull of the closed-economy industry center in Mexico City and strengthen the pull of the U.S. market. To examine these predictions, Table 4 shows regional average nominal manufacturing wages relative to national average nominal manufacturing wages from 1965 to 1988. Consistent with the theory, Mexico City has the highest wages, followed by the border states and central states, with northern and southern states lagging far behind.

To more formally examine the relationship between wages and proximity to industry centers, in Hanson (1997a) I estimate state manufacturing wages relative to national manufacturing wages as a function of distance to Mexico City and distance to the United States. Regional relative wages are strongly negatively correlated with distance to Mexico City and distance to the Mexico-U.S. border. A 10% increase in distance from Mexico City is associated with a 1.9% decrease in the relative state nominal wage, and a 10% increase in distance from the Mexico-U.S. border is associated with a 1.3% decrease in the relative state nominal wage. The results suggest that differential access to industry centers helps create regional wage differentials. Also consistent with the theory, in Hanson (1996a), 1 find that the effect of distance from Mexico City on state manufacturing wages has weakened following the liberalization of trade. comments

RegionBorder 19651.05 19701.06 19751.02 19801.04 19850.97 19881.00
North 0.60 0.58 0.63 0.70 0.72 0.71
Center 0.84 0.85 0.93 0.92 0.94 0.99
Mexico City 1.19 1.17 1.13 1.10 1.16 1.12
South 0.45 0.45 0.51 0.63 0.68 0.69

The strong correlation between regional wage differentials and proximity to industry centers suggests that trade policy plays an important role in regional economic development. However unintended it may have been, import substitution appears to have contributed in an important way to industry agglomeration in Mexico City. In a perhaps equally unintended manner, the opening of the economy to foreign trade and investment appears to undermine the economic rationale of such megacities.