Cabinet Coalition Formation: A Game-Theoretic Analysis

Abstract

The use of N-person game theory solution concepts to predict cabinet coalitions in European parliaments has been fairly limited. This is because most concepts predict only the alternatives in the solution space and not the coalitions which achieve these alternatives. In order to make coalition predictions, additional ad-hoc assumptions are necessary. The Competitive Solution attaches a coalition structure to each alternative in the solution space. It is thus extremely applicable to this problem.

The cabinet formation game is formulated spatially with parties viewed as the active players. The parties, cohesive units, negotiate over possible policy positions for the potential governing coalitions. It is this bargaining that results in particular coalitions being able to form and others being unable to compete. The Competitive Solution for 11 cases from five countries is computed, and the results are compared to those of other prediction techniques. Of the solution concepts tested, the Competitive Solution and Axelrod’s connected coalition theories are the most successful.

The author wishes to thank Richard McKelvey and Peter Ordeshook for their helpful comments and assistance. Any errors are the responsibility of the author.