Violence in conflict settings is seldom random, making its effects indistinguishable from the intentions of the perpetrator. We leverage on the quasi-randomness of accidental landmine explosions to study how violence shapes electoral outcomes in Colombia. We combine the geolocation of landmine blasts with the coordinates of voting polls in a regression discontinuity design that compares polls close to which a landmine exploded just before the election to those close to which it did just afterward. Blasts within a month from election day depress turnout by 23%. In addition, those who do vote penalize the democratic left for the explosions and are more likely to support political parties with ties with illegal paramilitary groups.

This paper examines new data on public sector employees from eighteen Latin American countries to shed light on the role of trust in the performance of government agencies. We developed an original survey taken during the first wave of the coronavirus pandemic that includes randomized experiments with pandemic-related treatments. We document that individual-level trust in coworkers, other public employees, and citizens is positively related to performance-enhancing behaviors, such as cooperation and information sharing, and policy attitudes, such as openness to technological innovations in public service delivery. Trust is more strongly linked to positive behaviors and attitudes in non-merit-based civil service systems. High-trust and low-trust respondents report different assessments of their main work constraints. Also, they draw different inferences and prefer different policy responses when exposed to data-based framing treatments about social distancing outcomes in their countries. Low-trust public employees are more likely to assign responsibility for a negative outcome to the government and to prefer stricter enforcement of social distancing.

Criminal groups use violence strategically to manipulate the behavior of victims and bystanders. At the same time, violence is a stimulus that causes fear, which also shapes people's reactions. Taking advantage of the randomness in the timing of antipersonnel landmine accidents in Colombia, as well as their coordinates relative to those of voting polls, we identify the effect of violence-induced fear (independent from intentions) on electoral behavior. Fortuitous landmine explosions reduce political participation. We further disentangle whether the type of fear caused by landmine explosions responds to an information channel (whereby people learn about the risk of future victimization) or by the salience of the explosion (which causes individuals to make impulsive decisions, driven by survival considerations), and show evidence in favor of the latter. While the turnout reduction takes place across the ideological spectrum, we document that the explosions induce a shift in the political preferences of individuals who do vote. These findings point to worrisome potential consequences for the consolidation of democracies in places affected by conflict.