The Overconfidence Problem in Insurance Markets

Adverse selection has long been recognized as a rationale for government intervention in insurance markets and for the adoption of public compulsory insurance. A different rationale for compulsory insurance is that overconfident individuals may underinsure because they underestimate the relevant risks. We show that government intervention is not a Pareto improvement in an adverse selection model with a significant fraction of overconfident agents. We underline that behavioral biases need not be the basis for government intervention. In fact, behavioral biases may overturn existing compelling reasons for intervention in the economy. Our model also delivers novel positive implications on aggregate variables that have been at the center of recent empirical investigation