Macroeconomics and political institutions in Western European multi-party democracies

In this thesis, I study three aspects of the interaction between politics and economics in coalition systems. First, theoretical and empirical issues concerning the political bargaining process over cabinet formation are addressed. Theoretical predictions on the duration of the process and its outcome are tested for the period 1950-1995. It turns out that the formation delay is increasing in the degree of ideological heterogeneity of coalition partners and that the share of portfolios secured by the formateur is decreasing in the degree of complexity of the bargaining environment. A few factors affecting the degree to which the outcome of the negotiation process can be defined as balanced are also identified. Second, I investigate the politics and economics of cabinet survival. A Proportional Hazards Specification for cabinet duration data is estimated by mean of a flexible parametric approach. I find that the hazard rate is determined by the majority status, the degree of fragmentation and ideological homogeneity of the coalition, the polarisation of the legislature and the time horizon at the moment the cabinet is formed. Interesting innovative results concern the greater stability of cabinets supported by coalitions ideologically closer to the median party and/or left-oriented. The overall state of the economy also appears to play a role. Graphical evidence suggests that the underlying distribution of duration data might be a Gompertz or a Weibull distribution, but not certainly an exponential distribution, as instead it is often assumed in the literature. Third, I look at the political and institutional determinants of fiscal policy outcomes (deficit, spending and taxation). Panel estimates show that the ideological orientation of the policymaker, the degree of cabinet instability, cross-country differences in electoral and budgetary institutions and the dispersion of political power within the ruling coalition all significantly affect fiscal policy variables. Instead, the preferences of the median voter appear to have relatively little importance. The evidence also rejects the hypothesis of systematic pre-electoral manipulation of fiscal instruments. A full description of the data-set of political indicators I have constructed and used to estimate the models in this thesis is also given. Specific attention is devoted to some methodological instances concerning the theoretical underpinning of empirical proxies. As an overview of the contents of the data-set, I discuss the evolution of basic attributes of legislatures and governments in the thirteen countries that constitute the sample for my analysis