Type of Parliamentarism and the Political Coherence of Coalition Governments

The objective of this paper is to specify the role of the type of parliamentarism in the functioning of representative democracy by focussing on the relative political coherence of coalition governments. In a representative democracy, with a chain of accountability running from the electorate to the government (Strom et al. 2003), we assume that parties are predominantly policy motivated. Therefore, we could expect that governing coalitions are “congenial in policy terms” with parties seeking to “minimise the policy range between themselves and their partners” (Strom and Müller 1999, 7). Considering that positive parliamentarism implies securing a majority in the parliament during coalition bargaining and thus more clearly differentiates between the coalition and the opposition (as opposed to minority governments in their relationship to the opposition), the relative coherence of coalitions can be hypothesised to be greater in systems with positive parliamentarism. Testing whether this is true would give us information if this basic difference in type of parliamentarism is associated with the realisation of some of the core principles of representation and accountability. The analysis operationalises the relative coherence of government coalitions as the ratio of the weighted average election manifesto overlap in the coalition to the opposition. The analysis will test if type of parliamentarism is associated with relatively more coherent coalitions, controlling for factors that can be hypothesised to have an impact on the political make-up of the coalitions like party system fragmentation, cabinet type, electoral volatility and possibly others suggested in previous research (e.g. Müller and Strom 2000, Strom et al. 2008). Information about governments will be drawn from the European Representative Democracy Data Archive and supplemented from other sources. The indicator of policy coherence will be calculated on the basis of the Manifesto Project unaggregated content analysis data.