Contract design and insurance fraud : an experimental investigation

Abstract

This paper investigates the impact of insurance contract design on the behavior of filing fraudulent claims in an experimental setup. We test how fraud behavior varies for insurance contracts with full coverage, a straight deductible or variable premiums (bonus-malus contract). In our experiment, filing fraudulent claims is a dominant strategy for selfish participants, with no psychological costs of committing fraud. While some people always commit fraud, a substantial share of people only occasionally or never defraud. In addition, we find that deductible contracts may be perceived as unfair and thus increase the extent of claim build-up compared to full coverage contracts. In contrast, bonus-malus contracts with variable insurance premiums significantly reduce the filing of fictitious claims compared to both full coverage and deductible contracts. This reduction cannot be explained by monetary incentives. Our results indicate that contract design significantly affects psychological costs and, consequently, the extent of fraudulent behavior of policyholders.