3
The central task facing societies emerging from the violence and destruction of civil war
is creating new governing institutions with the capacity to maintain peace among former
adversaries. While there are a number of existing statistical studies identifying factors that might
lead to civil war resolution through the different processes of victory by the government, the
rebels, or negotiated settlement (Mason and Fett 1996; Mason, Weingarten, Jr., and Fett 1999;
Gershenson and Grossman 2000; Enterline and Balch-Lindsay 2001), little effort has yet been
made to specify the conditions that might motivate decision-makers to design particular
institutions at war’s end.
In this paper we seek to address this gap in the established literature by examining factors
with the potential to shape the likelihood of adopting power-sharing or power-dividing
institutions following civil war. We employ a statistical test focusing on 106 civil wars ended
between 1945 and 1999 in order to identify factors that affect the likelihood of adopting any of
four power-sharing institutions at the conclusion of the conflict. Our findings suggest that
conditions that define both the nature of the conflict and the wartime bargaining environment
have an important role to play in determining the degree to which power-sharing and power-
dividing accommodations emerge as part of a settlement to end civil war.
This paper is divided into four sections. First, we identify and describe the conflict
management functions of the four power-sharing and power-dividing institutions that form the
dependent variable of this study. Second, we outline our theory and hypotheses concerning
factors with the potential to condition the likelihood of agreeing to power-sharing and power-
dividing arrangements among former civil war combatants. In the third section of the paper we
offer a test of our hypotheses and provide an interpretation of the statistical results. We conclude
with a discussion of the policy implications we derive from the empirical findings.