Abstract
Whether respondents disclose their preferences truthfully in surveys that are used to assess the values of public goods remains a crucial question for the practical application of stated preference methods. The literature suggests that in order to elicit true preferences, respondents should see a valuation survey as consequential: they must believe in the actual consequences that may follow from the survey result. Drawing on recent empirical findings, we develop a model depicting the importance of the consequentiality requirement for truthful preference disclosure in a survey that evaluates a public policy project based on a referendum-format value elicitation question. First, we show that a respondent’s belief that his vote may influence the outcome of the referendum plays a central role for revealing his preferences truthfully. Second, we find that the subjectively perceived probabilities of the successful provision of the public good and of the collection of the payment related to the project implementation not only need to be positive but also to be in a particular relationship with each other. This relationship varies in respondents’ preferences towards risk.