Wednesday, February 27, 2013

This paper investigates, theoretically and empirically, the relationship between municipal-level ethnic composition and the spread of conflict. Cross-country literature on conflict finds that ethnic diversity, and ethnic polarization in particular, are associated with greater incidence of conflict. However, the question of where within ethnically diverse countries conflict begins and where it spreads has been unexplored. My work suggests that conflict arises in ethnically diverse areas only after it is initiated in neighboring ethnically homogenous areas. After an ethnic group is successful at establishing ethnic stronghold(s), then co-ethnics in neighboring ethnically diverse areas may decide to attack the other group in their local community. I present a parsimonious model in which the decision to attack depends on only three observable variables: 1) ethnic population shares, 2) ethnic groups' weapons ratio 3) the share of co-ethnic successes in the battles that took place in the previous period. I use the model of conflict spread to simulate conflict. I use data from the Bosnian civil war to test two main predictions of the model: first, that conflict starts later in more diverse areas and, second, to test predictions of individual municipal attacks. In conclusion, I show how my findings are relevant for the design of peacekeeping efforts.