Discounting in decision making: the consistency argument revisited empirically

Discounting is one of the prominent topics of debate in health economics. While the standard practice in economic evaluation is to discount costs and effects alike with a 3-5% discount rate, many have raised questions about this practice. The debate sometimes seems trapped in Weinstein and Stason's consistency argument. In this paper, we use a set of health care programs--resembling Weinstein and Stason's hypothetical programs--to test whether appointed societal decision makers are consistent in their preferences over present and future costs and health effects, and whether they discount costs and effects at the same rate. Our results demonstrate these appointed decision makers to be fairly inconsistent on both issues, susceptible to the framing of problems and in part myopic. In other words, our respondents appear to be incapable of providing reasonable and consistent preferences between present and future costs, and health effects for use in economic evaluations. There is some support for the idea that rather than using constant and identical rates for costs and effects, real differences in health endowment over time (the growth rate for health) could serve as a basis for discount rates. Our respondents seem to relate their discount rate for health to their expectations about future life expectancy, but this also is dependent on the elicitation method.