Accounting for risk preferences in Stated Preferences studies

Abstract

Environmental policy outcomes are often uncertain, but choosing the socially preferred policy can be improved with an understanding of how people value such policies. While uncertainty considerations are increasingly incorporated into discrete choice experiment stated preference studies, these studies rarely address the multiple dimensions of risk that respondents may care about: the direction and magnitude of risk at different reference points. Here, we assess preferences for uncertain policies to conserve the Iberian lynx, focusing on these risk dimensions. We measure respondents’ risk aversion, loss aversion, and probability sensitivity using lottery tasks, and then use this information to investigate preference heterogeneity regarding the conservation programmes. We find that lottery choices align with theory, indicating risk aversion, and loss aversion and probability weighting consistent with prospect theory. This is also reflected in preferences for risky conservation policies, where we again find evidence of risk aversion, loss aversion, and – to some extent – probability weighting. We show that preferences over the variance of policy outcomes can matter, and excluding these dimensions in preference studies may lead to erroneous policy recommendations.