Economic integration and regional convergence: effects of NAFTA on local convergence in Mexico, 1980–2008

Continental integration processes can alter traditional development axes. Paelinck and Polèse´s work in 1999 explains that in the case of Mexico there is a tension between the U.S. border and the rest of the country: Mexico´s integration in the NAFTA should have reduced the U.S. border attraction, extending the growth to territories located between this border and Mexico City. To test this hypothesis, we propose a spatial conditional β-convergence model that uses as regressors both the distances to the U.S. border and to Mexico City, together with other control variables. This model is applied to the period from 1980 to 2008 using GVA at the municipality level. Working with municipal-level data allows to observe convergence patterns across space and identify the effects of location. The time-span studied distinguishes between before and after NAFTA. Estimates based on Mundlak´s approach were obtained for time-invariant regressors. Results show that during the pre-NAFTA period there was a general process of convergence, but it is mainly explained by the faster growth of municipalities located near to the U.S. border. However, post-NAFTA, convergence in municipalities disappeared and the effect of distance to the U.S. border reversed its sign, as predicted in Paelinck and Polèse´s model.