The Onset of Ethnic War as a Bargaining Process : Testing a Signaling Model

Abstract: Most theories of ethnic conflict explain ethnic war by reference to the factors that motivate and enable ethnic groups to rebel. To rebel is to rise up against or challenge government authority; but for war to be the outcome of a challenge the government must attempt to forcefully reassert its authority.In this dissertation I use a signaling model to explain both why ethnic groups challenge government authority, and under what circumstances this leads to war. The model is tested on a new data set covering 653 ethnic groups and 67 challenges to government authority, 1990-1998. The data includes all ethnic groups in the Minorities at Risk data set, plus an additional 370 groups to control for potential sampling bias.The signaling model depicts escalation as a bargaining process. The intuition is that through a process of self-selection less resolute actors tend to drop out earlier in the game, while more resolute actors stay in. The consequences are sometimes counterintuitive.The findings show that smaller ethnic groups are less likely to challenge their governments than are larger groups, but when smaller groups do challenge they are more likely to end up in war. The results also show that groups scoring higher on indicators of discontent are more likely to challenge, but surprisingly these groups are not more likely to end up in war when they challenge. These and similar findings which hold across a range of specifications clearly demonstrate the selection effects implied by the signaling model.