Bargaining in n-party legislatures over government formation

This paper questions results that claim to extend non-cooperative models of bargaining in legislatures from the highly atypical three-party case to a generic n-party setting. It identifies problems both with the derivation of theoretical results and the empirical evaluation of these. No empirically robust formateur advantage can be observed in field data on bargaining over government formation. The paper concludes with a modeling agenda of uncontroversial empirical statements about the government formation process and argues that these should form the premises of a more compelling new model of this crucial political process