Reforming democracy : institutional engineering in Western Europe, 1990-2010

Defence date: 29 April 2014Examining Board: Professor Sven Steinmo, European University Institute (Supervisor), for Professor Peter Mair (†), EUI Professor Nicolas Sauger, Sciences Po Paris (External Co-supervisor) Professor Stefano Bartolini, European University Institute Professor Jean-Benoit Pilet, Université Libre de Bruxelles.First made available online 27 May 2019The modification of the formal rules regulating the functioning of democracy has become an increasingly important field of research in political science, all the more so in a context of decline of political support in consolidated democracies. This thesis focuses on reforms of the core democratic rules in Western European democracies during the two last decades, adopting an expansive focus in order to adopt a multidimensional and longitudinal approach to institutional engineering. It investigates the occurrence of reform, successively examining the description and the characterisation of reforms, their contextual determinants and the mechanisms of change, articulating mixed-methods including large-N statistical analyses and case studies. The main theoretical innovation of the thesis is the concept of bundle of reform, defined as institutional reforms linked to each other and relating to several dimensions of the institutional architecture. In the first part of the thesis, thanks to an original database including 6 categories of reform, it is demonstrated that institutional reforms in the last few decades have been frequent, have happened in bundles, and have been moving towards more inclusive institutions. When studying the long- and short-term determinants of the number of reforms, certain long-term factors foster a favourable structural environment for the occurrence of institutional reforms, in particular a lack of political support from citizens for their institutions, but also the disproportional electoral systems in majoritarian democracies. In the short-term, institutional reforms in a given legislature tend to be more numerous after political alternation, and in contexts of rising electoral uncertainty. The thesis also evidences the fact that neither constitutional rigidity nor the number of partisan veto players can be considered to be the main explanation, or even relevant at all in some cases. The second part of the thesis focuses on understanding why, in favourable contexts, some reforms succeed and others fail, investigating case studies of three bundles of reforms in France, Ireland and Italy. The main finding evidences that the final outcomes of reforms are dependent on the combination of the nature of the reforms at stake (divisive or consensual) and the process used to adopt them (majoritarian, supermajoritarian or externalised). Divisive reforms tend to obey a self-interested logic, and their success depends on the ability to build an agreement in relation to a package deal of reforms. During the processes of consensual reform, credit-claiming logics prevail. When the majority chooses to exclude opposition parties and interest groups from the elaboration of reforms in order to claim sole credit, the fate of such reforms depends on the attitude of the opposition, who may choose to oppose the reform in order to symbolically defeat the government