Mexico : impact evaluation of sme programs using panel firm data

Unlike social programs targeting individuals, few enterprise support programs have been rigorously evaluated, and existing evaluations have mostly been done in high-income countries such as the United States and Europe. Mexico spends a large share of government resources on small and medium enterprise programs each year. How effective these programs have been in achieving their objectives is unclear. In Mexico, impact evaluations of small and medium enterprise programs are rare, and most are qualitative in nature. This is the first paper evaluating these programs in Mexico using firm-level panel data. The continuous and ten-year panel data -- from the 1994-2005 period -- allow the authors to address selectivity bias and unobserved firm heterogeneity by applying a generalization of differences-in-differences models combined with propensity score matching methods. This study finds evidence that participation in small and medium enterprise programs is associated with improvements in key variables such as value added, gross production, and wages. Furthermore, the study finds evidence that some of the positive effects can take several years to realize. The results also call for streamlining and greater efficiency in Mexico's small and medium enterprise programs.