| Abstract: | Government coalitions should be minimal winning. However, it is an empirical fact that oversized coalitions exist. Several theories have been offered to explain this phenomenon, but they are seldom put to a systematic empirical test. When empirical tests are performed, they are typically based on data on national government formations in post-war Europe. Since these are the data that gave rise to the theories in the first place, there is a risk of post hoc hypothesis reformulation. The purpose of this paper is to test explanations of oversized coalitions systematically in a new empirical setting and thus avoid this circularity problem. We focus on local governments in Denmark and have collected data by a survey sent to almost 3,000 local councillors. We draw hypotheses from three broad theoretical perspectives on oversized coalitions and test them in a logit regression analysis. The analysis shows that oversized coalitions cannot be explained by traditional coalition theories. Our results question the minimalist behavioural logic inherent in most coalition theories and suggest that parties may be motivated by norms. |
