Evaluating the validity of stated-preference estimates of health values

Economists have long been faced with the challenge of valuing goods and services that are not traded in markets. A variety of multiple-response stated-preference (SP) methods, which evolved independently in the area of consumer market research, also have been applied to nonmarket-valuation problems such as health-state value measurements. Unfortunately, distortions of the essential supply-demand relationships in such markets make it difficult to verify the validity of SP surveys using market data. However, because SP data include multiple observations for each respondent, it is possible to test a variety of hypotheses about consistency with welfare-theoretic principles and the use of judgment heuristics and learning and fatigue effects. This paper provides some results from two SP surveys eliciting health-state valuations. It focuses on the economic consistency of stated preferences. We show that with a few exceptions, subjects' stated preferences generally agree with welfare-theoretic requirements.