Abstract
How much should society be willing to pay for reducing inequality? The standard approach to this normative question relates inequality aversion to risk aversion by treating inequality as an outcome of a lottery. However, in the presence of heterogeneous risk preferences, it is unclear whose preferences should be used for evaluating this lottery. This paper derives a social welfare function as a limit of an iterative procedure, in which each iteration constructs a lottery based on the certainty equivalents from the previous iteration. The limit of this procedure can be interpreted as the equally distributed equivalent of the initial allocation.