Joumal of Risk and Uncertainty, 3:369-379 (1990) © 1990 Kluwer Academic Publishers Discounting and the Evaluation of Lifesaving Programs MAUREEN L. CROPPER Department of Economics, University of Maryland, Resources for the Future, Washington, D.C. 20036 PAUL R. PORTNEY* Resources for the Future, Washington, D.C. 20036 Key words: discounting, latency, risk valuation Abstract The evaluation of lifesaving programs whose benefits extend into the future involves two discounting issues. The intragenerational discounting problem is how to express, in age-; dollars, reductions in an individual's conditional probability of dying at some future age k. Having discounted future lifesaving benefits to the beginning of each individual's life, one is faced with the problem of discounting these benefits to the present—the intergenerational discounting problem. We discuss both problems from the perspectives of cost-benefit and cost- effectiveness analyses. These principles are then applied to lifesaving programs that involve a latency period. In evaluating a proposed regulation or making a public investment decision, it is stan- dard practice to compare the discounted present value of costs and benefits of the project, i.e., to apply a benefit-cost criterion. Application of this criterion, however, often meets with resistance when benefits or costs take the form of lives saved. This is especially tme when lives are saved or lost in the future, thus raising the question of whether these lives, or the monetary value of the corresponding risk reductions, should be discounted. The problem of discounting human lives arises frequently in the context of environ- mental policy. Perhaps the most striking example is nuclear waste disposal, which may impose risks on generations thousands of years into the future. The time pattern of risks to human life is, however, important even in the context of shorter planning horizons. Many environmental programs—for example, those concemed with asbestos—reduce exposure to carcinogens with long latency periods. This implies that, while the costs of reduced exposure may largely be bome today, the benefits do not occur until the end of the latency period. Compared with a program that reduces an individual's risk of death today, a program that reduces that same person's risk of death at the end of a 20-year latency period saves fewer expected life-years. This fact has often been ignored in valuing the benefits of environmental regulations. 'The authors are, respectively. Associate Professor of Economics, University of Maryland and Senior Fellow, Resources for the Future; and Senior Fellow and Vice President, Resources for the Future. We thank the National Science Foundation for their support under grant DIR-8711083.