Abstract:
How does violence affect social capital? I argue that its impact depends on two factors: i) the ability to identify the perpetrating group, and ii) the intensity of the violence. These factors help to reconcile the seemingly contradictory effects of violence on social capital presented in the literature. I study this question in the context of Colombia by exploiting changes in violence attributed to cross-border shocks on coca markets in neighboring countries interacted with a novel index of suitability for coca cultivation. This index uses satellite data from ecological conditions for growing coca. I document that violence has a negative effect on social capital measures such as trust, participation in community organizations, and cooperation. Notably, this effect is stronger when it is not possible to identify the enemy. The results are robust to a large number of controls that account for potential confounders. In particular, I show evidence that this effect is not related to the presence of drug cartels in Colombia during the Escobar and Cali era.