Willingness to pay for health protection: Inadequate sensitivity to probability?

Efficient investments in health protection require valid estimates of the p ublic's willingness to forgo consumption for diminished probabilities of de ath, injury, and disease. Stated valuations of risk reduction are not valid measures of economic preference if the valuations are insensitive to proba bility variation. This article reviews the existing literature on CV studie s of reductions in health risk and finds that most studies are poorly desig ned to assess the sensitivity of stated valuations to changes in risk magni tude. Replication of a recent study published in this journal by Johannesso n et al. (1997) demonstrates how serious the problem of insensitivity can b e, even for a study that reports plausible results. New empirical results a re presented from telephone surveys designed to provide internal and extern al tests of how WTP responds to size of risk reduction. The effect of varia tions in instrument design on estimated sensitivity to magnitude is examine d. Overall, estimated WTP for risk reduction is inadequately sensitive to t he difference in probability, that is, the magnitude of the difference in W TP for different reductions in risk is typically smaller than suggested by standard economic theory. Additional research to improve methods for commun icating changes in risk is needed, and future studies of stated WTP to redu ce risk should include rigorous validity checks.