In order to prevent the occurrence or recurrence of armed conflict in societies at risk, scholars and practitioners commonly recommend power-sharing arrangements that include all previous parties to the conflict and any potential spoilers. Power-sharing institutions include grand coalitions, federal institutions, and proportional allocations of government positions and benefits. Existing scholarship on power-sharing institutions, including cross-national quantitative studies as well as historical case studies, recognize that power-sharing comes in many forms, but argue that such institutions generically have beneficial effects. Yet, existing studies of power-sharing have tended to use fairly simple institutional measures and not to differentiate between arrangements that serve to include factional representatives in government (power-sharing institutions) and those that exclude them (power-dividing institutions). Moreover, there has been little rigorous theoretical recognition of the ways in which power-sharing may affect government decision making or the relationships between politicians and their constituents. Finally, the empirical literature on power-sharing has suffered from selection bias, as studies have focused much more on societies that have actually experienced civil conflict than on societies that have faced similar challenges but avoided overt conflict.
The intellectual merit of this project lies in the fact that it addresses each of these issues. It improves the theoretical understanding of both power-sharing and power-dividing institutions by identifying and distinguishing their most important features. It develops game-theoretic models to examine the short- and longer-term effects of these institutions on conflict and public goods provision. The project will create a quantitative data set that covers all major states from 1946 to the present and reports specifically on the mandate and implementation of power-sharing and power-dividing institutions. The data set will include both societies that have undergone civil conflict and societies with no such history but having characteristics that put them at risk for civil war. Such data help avoid the selection bias of many previous studies and help determine whether post-conflict situations are in fact more difficult than other situations in which there is a credible threat of internal violence.
The broader impact of this project is that it will give scholars and policy makers a better understanding of the strengths and weaknesses of different institutional strategies for reducing the civil conflict. Civil conflict is by far the most common type of armed conflict in the world today, and it is responsible for the great majority of military and civilian casualties. The project will assess the performance of political institutions often adopted in societies that have undergone civil conflict, and it will for the first time allow us to compare the performance of such societies with others that are at risk of war but have not yet experienced civil conflict. Finally, the project will generate an extensive data base on transitional aspects of political institutions, data that are currently unavailable and that will provide substantial benefit to other researchers in this very important field of study.
