Une approche comportementale de l'évaluation contingente.

Public economics proposed various models that intend to determine the optimal provision of public goods based on individual preferences. To provide decision makers with empirical recommendations, economists thus need to elicit individual preferences, and more precisely the marginal rate of substitution between private and public goods. Contingent valuation has proved a useful, and successful, tool to gather information on individual preferences. However, contingent valuation has been proved sensitive to various biases. In other words, variables that are not expected to have any influence do so in practice. In this paper, we propose a methodology, based on social psychology, which allows the identification of individuals that are proved immune to biases. This allows designing more powerfull, bias free, estimation of individual preferences. Two distinct applications are provided.