Pair-Wise Approach to Test the Regional Convergence Hypothesis

This paper assesses the gross domestic product (GDP) per capita convergence of the 32 Mexican States in the period 1940-2010 through the method proposed by Pesaran (2007), which is based on the convergence stochastic criterion. One of the main advantages of this method is not only on a model of leading economy, also on a pair-wise approach that considers all possible gap pairs of per capita logarithms of all the Mexican States analyzed in the sample. According to this method, all the differences or output gaps of the States must be stationary around a constant mean. Most results provide evidence against the hypothesis of convergence especially for the total sample from 1940 to 2010 and the first period from 1940 to 1985. However, mixed evidence of this hypothesis was observed in the second period from 1986 to 2010. Additionally, the test results applied to a set of States considered as the richest suggest these findings are not due to the unique behavior of these States.