Abstract

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 geo-location 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 percent. 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. We provide evidence that the reduction in turnout is driven by fear, and that the post-explosion voting patterns are not driven by changes in composition of voters but rather by changes in voting preferences that are consistent with blaming the democratic left for the actions of illegal left-wing insurgents. Finally, we provide suggestive evidence that demining campaigns increase turnout, this partially offsetting the effect of explosions.