A long-standing argument in political theory is that the use of a majoritarian electoral system, based on single member districts, is associated with stable government. Conversely, the use of proportional representation (PR) is associated with fragmented policies and `weak` coalition governments. Macro-political analyses to justify these arguments are somewhat inconclusive, because of the wide variety of PR policies. This research addresses this question by micro-political methods focusing on the nature of party and voter strategies in twelve polities. For nine of these polities (Belgium, Britain, Denmark, France, Germany, Ireland, Italy, Luxembourg, and the Netherlands) a `multinomial probit` electoral model has been devised, based on 1979 survey and elite data. In Israel, a conditional logit model has been constructed based on 1992 and 1996 surveys and manifesto data. These models will be extended to analyze elections in these countries to the present day, by incorporating party responses to electoral and economic disturbances. These data are augmented by new information on Poland and Russia. For Poland, a combination of opinion poll data, elite interviews and party manifestos will be used to study strategic behavior by parties and voters, in the context of a mixed majoritarian/PR electoral system. In Russia, elite interviews and study of party manifestos will be used to structure the nature of the policy space of relevance to political factions. This research focuses on the incentives for factions to coalesce in a political situation which is highly fragmented and uncertain. The theoretical purpose of the research is to understand the nature of the formation of factions and parties in the context of differing electoral systems. By `endogenizing` parties in this fashion, it is hoped that the research will lead, in the long run, to an account of the obvious differences, in party structure and coalition behavior, displayed by these various polities.
