Let's Make A Deal: Oil, Revenue Sharing, and Ethnic Group Rebellion in Civil War

Abstract

A principal finding in the cross-national literature on civil war links oil production with heightened risk of onset. Rebellion at the subnational level has been linked to oil, but the mechanisms remain underspecified or not tested systematically. The paper argues that the effect of oil on ethnic group rebellion in civil war is conditional on the quality of the ethnic group’s political representation in the central government. Specifically, when the group enjoys co-ethnic representation, it should receive a more favorable revenue sharing agreement, which should discourage rebellion. Using original data on the involvement of African ethnic groups in civil war between 1980 and 2006, statistical analysis provides some support for this argument: oil’s effect on the probability of rebellion is mitigated substantially by co-ethnic political representation in the presidency. The civil wars in Angola are discussed as an illustration of the argument that co-ethnic representation might shorten or even avoid civil war fueled by oil wealth. The argument has implications for the study of political representation and its simultaneous exacerbating and mitigating effect on conflict risk by relevant subnational groups. It also has implications for the bargaining literature on civil war and how likely the international community’s recent financial and budgeting transparency-promoting initiatives will reduce the risk of conflict. Finally, it is a useful addition to the literature on the effect of nontax revenue on regime stability, as it outlines a specific mechanism through which fiscal windfalls may affect conflict risk.

Keywords: civil war, oil, bargaining, ethnicity, political representation