How should future utilities be discounted? Economists and philosophers have agonized over this question for almost a century. The choice of the pure rate of time preference has a major impact on governments’ evaluations of long-run public projects, from climate policy to infrastructure investments. Yet despite the vigor, longevity, and policy importance of this debate, no consensus has emerged as to `the’ appropriate value of this parameter. This paper adopts an approach which recognizes these persistent disagreements about the rate of time preference as good-faith differences of opinion that cannot be resolved by empirical analysis alone. The problem then is to find a political process that aggregates a heterogeneous set of opinions into a single representative rate that can be used for public decision making. We examine two such mechanisms: one based on efficiency, which leads to a representative rate of time preference that declines with time, and one based on voting over optimal consumption plans, which leads to an equilibrium in which the median rate of time preference is chosen. We analyze each of these mechanisms’ strengths and weaknesses, and then consider a `meta vote’ between the two approaches, and find conditions under which a majority of people will prefer one to the other.

Joint with Geoffrey Heal.