Realignment and Generational Replacement in the Netherlands?

Since the 2002 elections, socio-cultural issues have become increasingly politicized, particularly the issues of European unification and immigration. In part, this politicization was a response to some critical events, such as the introduction of the Euro and the 9/11terrorist attacks of 2001. But to a large degree this change was the result of the surge of radical right parties, which mobilized support on the basis of these issues. As a consequence, these issues have become more important as predictors of party choice. In this paper we want to assess to which extent these processes are driven by generational replacements. We depart from the well-established fact that values and voting habits are acquired early in life and remain relatively stable afterwards. We test the hypothesis that the determinants of party choice will also be different among generations. Voters who, at a relatively young age learned to evaluate parties on the basis of socio-economic issues are expected to still evaluate parties on the basis of those considerations later in life. Those who grew up in a time when migration related topics were fiercely debated, are expected to evaluate parties more on the basis of those considerations. We study party choice by means of probability to support items (PTV’s). We distinguish groups of voters by age cohorts. We employ the Dutch National election studies since 1986, which was the first DNES in which these PTV-questions were included. The DNES has always included several batteries of issue-position questions. Because age, period and cohort have a perfect multicollinearity (i.e., age = period – cohort), APC models are not identified unless certain constraints are imposed. In this study, we disentangled age, period and cohort effects by imposing a theoretically informed functional form on our regression models.