Risk preferences may be time preferences: A comment on Andreoni and Sprenger (2012)

[Introduction] In an intensively discussed paper, Andreoni and Sprenger (2012), henceforth A&S, present an experiment where subjects can allocate money between two different points of time under the condition of risk. A&S claim that their results refute discounted expected utility (DEU) as well as prospect theory and other models relying on probability weighting. In this note I will show that the theoretical analysis of A&S is inappropriate and, therefore, that their claims are not valid. It turns out, that the experimental results of A&S are fully in line with DEU. The main problem of A&S's analysis is that is confounds income with consumption. There exist several other comments on A&S (Miao and Zhong, 2012; Epper and Fehr-Duda, 2014 and Cheung, 2014) which discuss interesting aspects of the analysis of A&S but have not identified the theoretical implications of equalizing consumption and income.