The Impact of Insurance Prices on Decision-Making Biases: An Experimental Analysis

Abstract

This paper tests whether the use of endogenous risk categorization by insurers enables consumers to make better-informed decisions even if they do not choose to purchase insurance. We do so by adding a simple insurance market to an experimental test of optimal (Bayesian) updating. In some sessions, no insurance is offered. In others, actuarially fair insurance prices are posted, and a subset of subjects is allowed to purchase this insurance. We find significant differences in the decision rules used depending on whether or not one observes insurance prices. Although the majority of choices correspond to Bayesian updating, the incidence of optimal decisions is higher in sessions with an insurance option. Most subjects given the option to purchase actuarially fair insurance choose to do so, however fewer subjects purchase insurance when the probability of a loss is higher.