The Longitudinal Trends of Personality Politics in Voting Behavior

One of the most debated features of the “personalization of politics” relates to the actual effect of political leaders on citizens’ voting behavior. Several decades of cross-sectional analyses have supported the idea that candidate evaluations exert a statistically significant impact in multivariate, fully specified models of voting in both presidential and parliamentary systems. But have leaders become more important in voters’ minds, or have they always been this relevant? Prominent party leaders are certainly not a product of the 21st century. Studies on the personalization of politics have long struggled to determine whether the personalization of voting behavior corresponds to a long-term process or if leaders are as important today as they have ever been. Against this background, the present contribution offers the most encompassing assessment of leader effects on vote choice in comparative and longitudinal perspective so far. To do so, I rely on a novel dataset pooling 265 national election study (NES) datasets collected in 20 established democracies worldwide across the last seven decades. The results show that leader effects increased over time as a function of the decline of party identifications among mass democratic publics. Furthermore, the results demonstrate a sizable growth in the incidence of negative party leader evaluations across time. This finding highlights the promising role of “negative personalization” as an epistemological tool to understand leader effects in a context ever more defined by negative politics and widespread sociopolitical polarization.