Do monetary incentives and chained questions affect the validity of risk estimates elicited via the Exchangeability Method? An experimental investigation

Using a laboratory experiment, we investigate the validity of stated risks elicited via the Exchangeability Method (EM) using an evaluation method based on de Finetti’s notion of coherence, under which probability estimates are valid if and only if they obey all axioms of probability theory. The validity of risk estimates elicited through the EM has been theoretically questioned because the chained structure of the game is thought to potentially undermine the incentive compatibility of the elicitation mechanism even when real monetary incentives are provided. We investigate this by designing and implementing four experimental treatments. Respondents are divided in two initial treatment groups: in the first, they are provided with real monetary incentives, and in the second, subjects are not. Each group is further sub-divided in two treatment groups, in the first, the chained structure of the experimental design made quite clear to the subjects, while, in the second, the chained structure is hidden by resorting the elicitation questions. The superiority of real monetary incentives is not evident when people are presented with chained experimental design.