2
Abstract
In this paper we develop a statistical model designed to predict whether or not institutions
intended to share or divide state power among former combatants will be created at the
conclusion of civil war. We categorize the independent variables we include in our study as
factors that define either the nature of the conflict or the wartime bargaining environment.
Testing our statistical model using data related to 106 civil wars ended between the years 1945
and 1999, we find that factors associated with both categories of influences play an important
role in determining the number of power-sharing and power-dividing arrangements that develop
at the end of hostilities.