Although it avoids the negative externalities associated with the damages caused by uninsured

individuals, compulsory insurance raises the issue of insurance crowding out prevention.

Interestingly, Pannequin and Corcos ((2020)) show that on a theoretical level, although

compulsory insurance and self-insurance (prevention investments dedicated to loss reduction)

are substitutes for risk averters, they are complementary for risk lovers. The present

contribution aims to test, in the Lab, these surprising results using a model-based

experimental design. Our experimental results support the theoretical predictions:

compulsory insurance and self-insurance are complementary for risk lovers and substitutes

for risk averters. This contribution fully supports public policies that aim to implement

mandatory insurance. Far from deterring prevention activities and providing that its level is

high enough, mandatory insurance increases prevention levels.

Keywords: compulsory insurance; self-insurance; experiment; risk-attitudes; substitutability;

complementarity.