Abstract
Heterogeneous risk preferences seem to play an important role in shaping the wealth distribution and help to explain wealth inequality at the top (Laurent et al. 2020). Surprisingly, tax theorists have not devoted much attention to this type of heterogeneity. This project aims to fill this gap by investigating the theoretical implications of heterogeneous risk preferences on optimal ex-post allocation and capital income tax theory. I consider a two-period model where agents make portfolio decisions that shape their choices of consumption and savings. Coherently with the literature, two types of government policies are investigated: lump-sum transfers and public good provision. The first best outcome can be obtained when the government sets state-contingent, type-specific, lump-sum taxes. When state-contingent taxes are not available, type-specific taxation yields "ex-ante efficiency" only. When the safe and excess returns are targeted separately to finance a public good, the optimal excess return tax is positive, while the safe return tax is different from zero. Therefore, a trade-off between insurance and efficiency arises.