Extending a Statistical Model of Multiparty Elections for Comparative Analysis Across Different Electoral Rules.

This paper proposes an extension of a statistical model of multiparty aggregate electoral data to conduct a truly comparative analysis of elections across different electoral rules. Such a fully comparative study requires the analysis of intra-party votes for electoral rules with intra-party competition (e.g., open-list PR) and the estimation of winning probabilities which are, unlike vote shares, directly comparable across SMDs and MMDs. The paper develops a model of intra-party vote shares and proposes a two-step estimation of inter- and intra-party vote shares to accommodate electoral rules with intra-party competition. It also uses a simulation of vote shares and seat distributions based on these models to estimate winning probabilities of seats and candidates. To demonstrate its utility, the paper applies the methodology to comparing incumbency advantage in UK and Finland which use, respectively, an archetype of electoral rules at the opposite end of the spectrum: the SMD without intra-party competition and the MMD with intra-party competition. The estimation results support the hypothesis that incumbency advantage is higher in Finland than in UK due to the difference in legislator’s incentives to pursue personal votes. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]