When Do Allies Become Rivals?

Abstract

Explaining the formation and persistence of political and social coalitions is one of the most important problems in the social sciences. Solving this problem will help us understand which groups of people will cooperate and which political parties will form coalitions. However, in order to explain the formation of coalitions, we cannot simply examine the interests of their members. As Mancur Olson points out, a community of interests is not a sufficient condition for the emergence of social coalitions.' Extrapolating the same argument to political coalitions, we see that the mere existence of mutual interests among coalition partners is not enough to explain the existence of the coalition. coalition is a product of both cooperative and competitive tensions between the coalition partners. This means that the cohesion of a coalition is itself a variable to be explained. Game theory has provided a theoretical framework by which to begin studying political coalitions.2 However, the existing game theoretical literature, aside from being disputed on empirical grounds,3 focuses exclusively on the cooperative behavior of partners in a coalition. Thus, this literature addresses which coalition will form and not which coalition, once formed, is likely to dissolve. In a recent article published in Comparative Politics, Rochon and Pierce use the French Fifth Republic and, in particular, the two major partners of the Left-the Socialists and the Communists-to investigate both the cohesive and the competitive forces within coalitions.4 French Fifth Republic is the best empirical case for the study of the stability of coalitions. Under the French electoral system, parties have to stress both competition (in the first round) and cooperation (in the second round). In the first round of the run-off electoral system, strong criticism of the coalition partner may give a party the first place within the coalition and, therefore, the right to represent the coalition in the second round. In the second round, the cooperation of both partners is needed in order to win a parliamentary seat. But the French electoral system combines two unique institutional features, making it difficult for parties to pursue different strategies in the first and second rounds. First, the two electoral rounds are only a week apart and, therefore, there is no time to adopt two different strategies. Second, the strategies are electoral and, thus, visible as opposed to parliamentary. Rochon and Pierce's article is important and thought-provoking precisely because it focuses on an aspect of coalitions that has been neglected by the literature-the centrifugal forces that develop and operate inside an electoral coalition. Therefore, although it deals with a particular country, the article has more important theoretical implications for our understanding of political coalitions. Their analysis, however, comes to a surprising conclusion. The Socialist-Communist alliance appears (at least based on our data on the Socialist candidates) to be ruled by jealousy