Optimal Capital Income Taxation and Redistribution

This paper studies the effects of agent heterogeneity on optimal capital income taxes, where taxes are collected for redistributional purposes. In a finite horizon model with an arbitrary number of heterogeneous agents, we ask what forms of taxation are optimal for different causes of inequality and how correlations, e.g. between labor incomes and wealth levels, affect optimal tax rates. While steady state analyses such as Judd (1985) have found that the optimal capital income tax rate in the steady state is zero, we show that off the steady state this is generally not the case. In a two period model with arbitrarily many heterogeneous households, we find that if households are heterogeneous with respect to productivities and endowments, capital income taxes generally increase welfare. If they are heterogeneous only with respect to productivities, it is optimal not to tax capital income. The extent of the inequality and the joint distribution of its different components (productivities and endowments in our model) are crucial for the size of the marginal welfare effects of taxation. A positive correlation between endowments and productivities, for example, increases the marginal welfare effects of capital income taxation. The paper emphasizes the analogy between commodity taxes and capital income taxes, which effectively tax consumption goods at different points in time. With the help of the literature on optimal commodity taxation based on Atkinson and Stiglitz (1976), we can check the robustness of our results. We find that zero capital income taxation is optimal only if endowments are homogeneous and production is weakly separable between labor and capital and utility functions are homothetic and identical across agents. When we link our model to the infinite horizon steady state analyses as described in Judd (1985), we find that the models are compatible under appropriate assumptions.