Recent years have seen a surge of literature examining how political institutions influence the risk of civil conflict. A comparatively neglected aspect of this debate has been the heterogeneous impact of different forms of authoritarianism. In this article, I theoretically and empirically unpack the authoritarian regime category. I argue that authoritarian regimes differ both in their capacity to forcefully control opposition and in their ability to co-opt their rivals through offers of power positions and rents. Authoritarian regimes thus exhibit predictable differences in their ability to avoid organized violent challenges to their authority. I examine the association between four types of authoritarian regimes—military, monarchy, single-party, and multi-party electoral autocracies—and the onset of civil conflict from 1973 to 2004. I find that military regimes and multi-party electoral autocracies run a higher risk of armed conflict than single-party authoritarian regimes, which on the other hand seem to have an institutional set-up that makes them particularly resilient to armed challenges to their authority. These findings suggest that the emerging view, that political institutions are not a significant determinant of civil conflict, results from treating a heterogeneous set of authoritarian regimes as homogenous.
Fjelde, Hanne (2010) Generals, Dictators, and Kings: Authoritarian Regimes and Civil Conflict, 1973—2004, Conflict Management and Peace Science 27 (3): 195–218.