Lethal violence and migration in mexico: An analysis of internal and international moves

Abstract

Linking municipal-level homicide rates from 1990 through 2018 with data from the Mexican Migration Project, we estimate a series of multinomial discrete-time event history models to assess the effect that exposure to lethal violence has on the likelihood of migration within Mexico and to the U.S. without documents. Statistical estimates indicate that the homicide rate negatively predicts the probability of taking a first undocumented trip to the U.S. but positively predicts the likelihood of taking a first trip within Mexico. Among those undocumented migrants, who have already taken a first U.S. trip, lethal violence also negatively predicts the likelihood of taking a second undocumented trip. Among returned internal migrants whose first trip was to a Mexican destination, the odds of taking a first undocumented trip were also negatively predicted by the municipal homicide rate. Violence in Mexico is not a driver of undocumented migration to the U.S.