Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Spatial analyses focus to a large extent on the ‘bright side of proximity’, namely voluntary (positive-sum) interactions such as, e.g., in trade and innovation. In contrast, the violent ‘dark side of proximity’ has often been overlooked. To address this gap, we study the role of spatial proximity in ethnic conflict, developing a structural model of spatial violence in which ethnic groups recruit fighters strategically across space. The spatial decay of violence determines the equilibrium placement of fighters and drives specific spatial patterns of conflict. The structural parameters of the model are estimated using fine-grained data on ethnic groups and violence from twenty-four ethnically divided countries. We find that in more than half of these, spatial decay is substantial: half of all ethnic violence dissipates after 350 km. Violence is asymmetric, is higher near ethnic borders and typically originates from outside a location. Counterfactual estimates suggest that setting up barriers would reduce violence, but pacifying groups suffering from grievances would often be more effective.