Abstract
Since the start of the War on Drugs in 2006, Mexico has been living for years in a continuous state of high and extreme violence, that counts to date hundreds of thousands of deaths by homicides, massacres, and disappearances. Much of the literature has overtly focused on institutions and specific security policies to understand this puzzle. In this paper, I argue that high homicide rates happen in states that have some economic dynamism but low social cohesion, which leads to an increase in cartel recruitment and, therefore, to the rise in victims and perpetrators of violence. To prove this theory, in this paper, I execute multiple regressions with the homicide rate as a dependent variable. I divide my analysis into two sections: the relationship between a group of indicators of social cohesion and violence and the relationship between some structural variables and violence. Among my social cohesion variables, I found a negative relationship between the percentage of the population that speaks an indigenous language (my main variable of interest) and violence, a positive relationship between schooling years and violence, and no relationship between the presence of common land and violence. Also, among my structural variables, I found a negative relationship between poverty and violence, a positive relationship between economic activity and violence, and no relationship between distance to the US border and violence.
