[The paper is forthcoming in Political Behavior] For any questions regarding the replication of the models, please contact Twan Huijsmans at t.m.huijsmans@uva.nl ABSTRACT: We investigated whether income gaps in voting turnout vary with country-level economic inequality, and whether patterns differ between wealthier and less-wealthy countries. Moreover, we investigated whether the prevalence of clientelism was the underlying mechanism that accounts for the presumed negative interaction between relative income and economic inequality at lower levels of national wealth per capita. The harmonised POLPART dataset, combining cross-national surveys including 66 countries, 292 country-years, and 510,184 individuals, was analysed using multilevel logistic regression models. In general, we found that the positive effect of relative income on voting was weaker at higher levels of economic inequality. This could only partly be explained by taking into account the moderating effect of clientelism. However, clientelism seems to decrease turnout of higher income groups, rather than increase turnout for lower income groups. Additional analyses in less-wealthy countries showed that clientelism is not the explanatory mechanism for the association between economic inequality and the income gap in voting in the countries where we expected this especially to be the case. Although clientelism partially explains why economic inequality reduces the income gap in voter turnout, it does not do so in the way we expected, and in the countries where we expected it most. This suggests that there must also be other mechanisms that explain the relationship between economic inequality and political inequality.