Few social science relationships have spawned as much interest -- or as many elaborate theoretical models and arguments -- as that between democracy and income inequality. However, the empirical literature has generally employed statistical models based on problematic assumptions, and has produced quite mixed results. Hence, this paper makes an important empirical contribution by applying models that, for instance, account for endogeneity biases and control for country--specific effects. Despite being correlated, there is very little evidence of any effect of income inequality on level of democracy once employing appropriate model specifications. Furthermore, there is no robust evidence that inequality systematically affects either democratization prospects or democratic stability. In contrast, there is evidence that democracy reduces income inequality when inequality is proxied by share of income going to wages. However, also this effect is sensitive to the choice of inequality measure. Democracy does, for instance, not reduce inequalities in disposable household incomes.