While traditional political economic theory predicts that higher income inequality leads to more extensive redistribution policy, empirical studies often show the opposite relation. Many alternative theories have been proposed to reconcile these. We suggest a test of these alternative theories. Two steps are articulated in the prediction of the traditional theory: 1) the effect of income distribution on people’s preferences for redistribution, and 2) the translation of the given preferences into an actual policy. As alternative theories change either of the two steps of the traditional theory, we can test these alternative theories by checking the viability of each step. While our test supports the theory emphasizing the role of social mobility, the correlation between the measure of social mobility and the redistribution policy does not support it. A new theory is required.