Monetary values of increasing life expectancy: sensitivity to shifts of the survival curve

Harvard University, Cambridge MA

Investigators

Abstract

Individuals' risks of dying at different ages are affected by government policies including health, safety, and environmental regulations. To help evaluate whether new and existing regulations enhance national welfare, government policies require comparing estimates of the monetary value of the life-saving and other benefits with the costs of the regulations. The monetary value of reductions in mortality risk is usually estimated using one of two approaches. One approach, common in transportation and environmental regulation, estimates the annual monetary benefits as the product of a "value per statistical life" (VSL) and the expected number of "lives saved" per year. The second approach, common in food-safety and drug regulation, estimates the benefits as the product of a "value per statistical life year" (VSLY) and the average increase in life expectancy. Because life expectancy depends on mortality risk at all future ages, this second approach assumes that individuals assign the same monetary value to any time path of risk reduction that produces the same gain in life expectancy; e.g., a reduction in accident risk when young or heart-attack risk when older. This assumption is challenged by recent research that finds individuals have systematically different preferences over time paths of mortality-risk reduction; some people prefer near-term risk reductions, others prefer risk reductions at older ages. The goal of this research is to estimate the monetary value of mortality-risk reduction in the American population and to describe how it varies with the total gain in life expectancy and the time path of annual risk reductions. This knowledge should help in refining estimates of the monetary value of the benefits of certain government policies for comparison with their costs, and contribute to decision making in other domains such as health care where evaluation is less formal. The project will develop a stated-preference survey instrument, administer the survey over the internet to a sample of 1000 or more individuals representative of the adult US population, and analyze the results to determine how the monetary value of mortality-risk reduction depends on the gain in life expectancy and the time path of the risk reductions. The survey will present to each respondent a description and graphic illustrating his or her baseline annual chances of dying in or surviving each of the following decades through age 100 (conditional on age and gender). It will also present perturbations of this baseline hazard function, obtained by reducing mortality risk for only the first decade, and by reducing mortality risk in all future decades by subtracting a constant or multiplying by a constant. The three perturbations will be calibrated to produce the same increase in life expectancy. Respondents will be asked to report their preferences between each pair of the three perturbations (to test for consistency) and separately to report their willingness This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.