Assessing the impact of government alternation on party unity : the case of Italy, 1988-2008

Literature has for quite long time identified at least two different sources of the party unity in the legislative arena. In a famous article Diermeier and Feddersen (1998) argue that it is the confidence relationship, i.e., the threat of being voted out of office and losing agenda setting powers, that makes parties more united in parliamentary than in presidential systems. On the other hand, many scholars argue that in electoral systems (such as proportional with open list) where candidates compete for a personal vote, MPs are likely to pay attention to the demands of their constituency as well as to the demands of their party, while in situations where the candidate\u201fs chances depend only on the party leadership (such us under a proportional system with closed list) loyalty to the party should be the rule. In spite of the a-priori plausibility of both arguments, the empirical evidence in literature is mixed and contradictory. We argue that the present theoretical models and empirical analyses miss a crucial explanatory factor: the competitiveness of the party system or, in other words, the presence of the government alternation as a stable and expected feature of the political system. We test this very general idea by using a MPs\u201f index of party-dispersion derived from a Wnominate analysis applied to the roll call data of the Italian Chamber of Deputies in the last 20 years. The uncommon use of data at the individual level allows us to effectively control the role played by the original level of party affinity and by other idiosyncratic characteristics. Moreover, it also allows us to distinguish (empirically) between party cohesion and party discipline