Discounting of Life-Saving and Other Nonmonetary Effects

Abstract

Cost-effectiveness analysts generally assume that preferences over time are such that streams of monetary and nonmonetary program effects can be reduced to one discounted sum of monetary costs and another of effects. It is known that if the nonmonetary effects can be cashed out in a way that does not vary with time, then the rates of discount for monetary and nonmonetary effects have to be equal. This paper presents a more compelling argument for the equality of those rates when hard to monetize benefits such as life-saving are involved. It shows that if the ability to produce the nonmonetary effect does not diminish too quickly over time, failure to discount benefits implies that programs are always improved by delay. In general, discounting benefits and costs at different rates can lead to peculiar results.