LAUSR

How unemployment affects electoral participation is a archetypal question in political sociology and of particular relevance in economic crises; in the 1930s as well as during a pandemic. A frequent argument in the literature is that unemployment leads to political withdrawal as the unemployed have to focus on their economy and other personal matters. Some scholars, on the other hand, reason that unemployment triggers political mobilization through feelings of grievance and a strive to protest against leading politicians. However, existing empirical evidence is mixed and often suffers from limitations in data and research design. To make progress on this difficult empirical question, the present study leverages Swedish register data on turnout, spanning nine elections between 1970 and 2018. This extensive panel dataset enables us to more adequately address the causal status of the relationship by relying on the with-in individual variation in turnout. Our results report significant but modest negative effects of both unemployment in general and of losing a relatively well-paid job (job loss). The effects are driven by the young. We also find that the negative effect is less pronounced for those who have previously worked at a workplace where most colleagues participate, supporting a socialization mechanism.