Indecisiveness in Behavioral Welfare Economics

Abstract

When an individual's preferences depend on the time or 'frame' at which decisions are made, the preferences that appear at different frames must be aggregated in order to make social decisions. Suppose we aggregate each individual i's frame-based preferences with a 'behavioral welfare relation' that ranks x above y if, when both x and y are available, i sometimes choose x and not y and never chooses y and not x. The set of Pareto optima can then be large. In fact a small amount of preference diversity across frames can cause every allocation to be Pareto optimal. More generally, the set of Pareto optima will have the same dimension as the set of allocations. The Pareto criterion then will not be able to discriminate locally among policy options. A small distortion, for example, will call for no policy response.

Keywords: behavioral welfare economics, incomplete preferences, Pareto optimality

JEL Classification: D11, D60, C60