Explaining the Length of Ministerial Tenure in Consensual Coalition Governments.

What are the determinants of ministerial duration in complex multiparty democracies? What are its main differences from majoritarian systems? The present study is an attempt to answer these questions. The research on election timing in consensual systems showed how electorally advantaged parties can extort non-electoral gains from disadvantaged coalition partners. In this paper, I argue that the demotion of a minister, or an alteration of the distribution of ministerial powers in a coalition cabinet, can be thought, under the lenses of bargaining theory, as a currency in the power game among incumbent parties. The present study attempts to compensate three main limitations of the literature on ministerial tenure. First, its general lack of interest for consensual systems: amost any effort was concentrated on Westminster systems, with an outstanding dominance of the United Kingdom. Second, its lack of focus on the effects of coalition bargaining, and how the threat of an anticipated cabinet termination can be exploited by electorally advantaged parties in order to achieve a more favourable distribution of ministerial prerogatives. Third, no previous empirical study employed electoral expectations as an explanatory variable of ministerial duration. I combined, for the first time in the literature, event history methodology with time-varying measures based on voting intention trends, that will allow to produce novel, untested assumptions. My argument is that large imbalances in incumbent parties' popularity are associated with more frequent redistributions of ministerial powers within the governing coalition. When the disproportion of electoral advantages is high, the pressure on unpopular parties for portfolio concessions increases, and more coalition members have the opportunity to ask for the demotion of a minister, or an alteration of his or her policymaking prerogatives. This and other corollary hypotheses have been successfully tested on an original dataset based on opinion polls from four consensual democracies (Austria, Italy, Denmark, the Netherlands), covering about sixty years of parliamentary politics. From voting intention data, a measure of "popularity turbulence" within the coalition was operationalized. Large changes in the incumbents' electoral expectations are shown to be associated with a greater probability of a ministerial layoff, and of a redistribution of policymaking prerogatives within the cabinet members. All the bargaining variables employed outperformed the measures of coalition heterogeneity traditionally employed in the literature. Additionally, this research, conducted on consensual parliamentary systems alones, provides some novel comparative insights on their differences from Westminster systems. From the empirical analysis it emerges that single ministers' demographic variables (such as gender, age, education) are not empirically relevant for explaining ministerial tenure. Previous researches, based on evidence from (completely, or at least partially) Westminster systems, where the coalition dynamics analyzed herein are not at play, showed their relevance instead. The insights of this research tell us that this difference depends from the peculiar nature of coalition, parliamentary governments, where ministerial duration is the outcome of a negotiation among coalition parties. This might represent a significant discovery regarding the differences between the two main models of parliamentary democracy.