A central question in Western European electoral research is whether electoral changes over the past decades should be interpreted in terms of dealignment or realignment. Although many scholars study this question, they have not paid much attention to the role of generational replacement. This ‘age-period-cohort’ (APC) study fills this void by examining the last 10 national elections in the Netherlands (1986–2017). The hypothesis is tested that the determinants of party choice differ systematically across generations. With regard to period effects, it is found that the association between party preference and its predictors has mainly weakened. Over generations, contrarily no decreasing associations are found. The effects of religion, social class, partisanship, left–right and redistribution have not structurally decreased with each successive generation, whereas the effects of education, immigration and European unification are stronger for younger generations. Taken together, these findings reveal how ...