Insurance Demand for Disaster-type Risks and the Separation of Attitudes toward Risk and Ambiguity: an Experimental Study
Cahier du LEF ; 2008-05Cet article a été présenté dans plusieurs conférences sous le titre : Ambiguity, Government Intervention and Inusrance Decision : an Experimental Study ; This article presents the results of an experiment designed to test theoretical predictions about the impact of public compensation schemes and ambiguity on insurance and self-insurance decisions. Consistent with theory, we find that government assistance significantly reduceswillingness to pay (WTP) for insurance and self-insurance (compared with a free insurance market). As expected, we also find significant differences between WTPs for insurance under different types of government compensation programs. For example, results from our experiment confirm the prediction that the WTP for insurance is smaller under a "Fixed Help" program thanunder a "Contingent Fixed Help" program where the government assistance is conditioned to the purchase of an insurance policy. Thirdly, we find that ambiguity, i.e., uncertainty about probability, significantly increases WTPs for insurance. This result, which indicates thatdecision-makers are ambiguity averse, is in line with previous results on the impact of ambiguity on insurance demand for low probability risks. Lastly, our experiment provides a clear support for the hypothesis that attitude to risk and attitude to ambiguity are two independent phenomena. In fact in this experiment, decision-makers are both risk-seekers (i.e., the mean WTP for insurance is on average smaller than the expected value of the loss) and ambiguity averse (i.e., the mean WTP for insurance is on average higher for an ambiguous risk than for a 'risky' risk) ; Cet article présente les résultats d'une expérience mise en place pour tester des prédictions théoriques concernant l'impact de programmes publics de compensations financières et de l'ambiguïté sur les décisions d'assurance et d'auto-assurance. En conformité avec la théorie, nous montrons que l'aide du gouvernement réduit significativement le consentement à payer (CAP) ...