A new working class? : a cross-national and a longitudinal approach to class voting in post-industrial societies

Abstract

Post-industrial transformations in the occupational structure and new patterns of class-party alignments have fueled the debate on the relevance of social class as a determinant of political preferences and behavior. Although the growth of the service sector is one of the distinctive traits of post-industrial economies, low-skilled service workers have received limited attention in recent research on class politics. This dissertation analyzes the political implications of class in post-industrial societies, focusing specifically on the comparison between low-skilled production and service workers. Through a two-step analysis of class voting, this dissertation studies, first, the association between class and issue preferences and, second, the relationship between class and electoral behavior. This approach to class voting also allows me to theorize and analyze potential moderators and mechanisms of the individual-level association between class and political outcomes. To study these different aspects of class voting both cross-sectionally and longitudinally this thesis relies on multiple datasets like the European Social Survey, the Chapel Hill Expert Survey and the British Household Panel Survey, and on different estimation methods like multi-level, conditional logistic and panel data regression models. The results of a systematic comparison of production and service workers indicate that the two classes constitute a rather homogeneous electoral constituency both in terms of preferences on cultural and economic issues, as well as in their likelihood of voting for different party families. Thus, these two groups could constitute a new working class, characterized by its economically left-wing but culturally authoritarian political preferences, but also by its higher levels of electoral abstention. Other than revealing the similarity between production and service workers, this dissertation also contributes to the literature on class voting by studying moderators and mechanisms of the individual-level relationship between class location and political preferences. The analyses indicate that the politicization of policy issues by parties or the length of class tenure moderate this relationship. Moreover, I also consider how vertical and horizontal class mobility throughout an individuals’ career relates to differences in policy preferences. For this purpose, I implement a longitudinal approach, which has been rather infrequent in studies of class voting. The conclusion of this dissertation discusses the implications of these findings for the political representation of the working class and for aggregate levels of class voting. Overall, and in clear contrast with the dealignment thesis, this dissertation indicates that class is still a relevant determinant of political preferences in post-industrial societies.