Classic studies in political sociology argued that differences in countries' patterns of mobility are responsible for variations in levels of class voting in these countries. In this paper the tenability of this classic claim is examined using survey data from 14 long-standing Western democracies over the period 1964–1990 (N=54 478). When testing this claim, we improved over earlier studies by employing detailed measures of class, by using multivariate statistical models that measure relative levels of class voting and by employing diagonal reference models that deal adequately with the effects of individual intergenerational class mobility. Our results show significant effects of individual intergenerational mobility. The longer mobile persons are members of a certain class the more they are apt to show the typical voting behaviour of their destination class. However, our results do not support the claim for the effects of intergenerational mobility at the macro level. Variation in levels of mobility can only to a very modest extent – if at all – be held responsible for cross-national and over-time variation in levels of class voting.