Violence in Mexico and its effects on labor productivity

Abstract

This paper examines the evolution of Mexico’s labor productivity (GDP per worker) across its 32 sub-national entities from 2003 to 2013, during a period of rising drug-related crimes. Using quarterly data and economic controls, fixed effects models suggest the effects of crime are small and differ depending on whether such crimes are prosecuted by state/local or federal authorities. However, results from System Generalized Methods of Moments regressions generate stronger responses for (endogenous) wages and labor productivity. First, crime has negative effects on Mexican labor productivity across states during the “war on drugs” period. Second, increases in expenditures on public security lead to falling labor productivity, which can be interpreted as indirect effects of crime. Third, federal authorities are found to be more effective (in not causing lower productivity) than state/local authorities.