By examining the Belgian case, this article aims to show that methodological nationalism is strongly present in the literature on nationalization of party offer and voting behaviour. In nationalization studies, Belgium is often presented as a typical example of a denationalized country. This is true for the party offer, as it is de facto split between the two language groups since the 1980s, and therefore also voter response at the national level. However, voter response within each separate subnational party system is very homogeneous and shows interesting differences between these party systems that inform us about important electoral dynamics. We argue, on the basis of our results, that rather than stretching the concept of nationalization, it is preferable and justified to treat the concepts of nationalization of the party offer and homogenization of voter response as analytically distinct and not as two sides of the same coin.

Taking Lijphart’s work on consensus democracies as our point of departure, we signal a major shortcoming in Lijphart’s focus being almost exclusively on the political hardware of the state structure, leaving little attention for the administrative and bureaucratic characteristics of governance systems. We propose to expand the Lijphart’s model which overviews structural aspects of the executive and the state with seven additional features of the bureaucratic system. We argue that these features are critical for understanding the processes of policymaking and service delivery. Next, in order to better understand the functioning of the Netherlands and Belgium as consensus democracies, we provide a short analysis of the historical context and current characteristics of the political-administrative systems in both countries.

This article takes up Lijphart’s claim that consensus democracy is a ‘kinder, gentler’ form of democracy than majoritarian democracy. We zoom in on contemporary welfare state change, particularly the shift towards social investment, and argue that the kinder, gentler hypothesis remains relevant. Consensus democracies stand out in regard to the extent to which their political institutions help to overcome the politically delicate intricacies of governing for the long term. We theorize the features that can help to solve the problem of temporal commitment in democracy through processual mechanisms and illustrate these with short case studies of the contrasting welfare state reform experiences in the Netherlands and Belgium.