<scp>The (1992) Bonus‐Malus System in Tunisia: An Empirical Evaluation</scp>

Abstract The objective of this study is to assess empirically what impact introduction of the bonus‐malus system (BMS) has had on road safety in Tunisia. The results of the Tunisian experiment are of particular importance since, during the last decade, many European countries decided to eliminate their mandatory bonus‐malus scheme. These results indicate that the BMS reduced the probability of reported accidents for good risks but had no effect on bad risks. Moreover, the reform's overall effect on reported accident rates is not statistically significant, but the exit variable is positive in explaining the number of reported accidents. To avoid any potential selectivity bias, we also made a joint estimate of the reported accident and selection equations. The reform has a positive effect on the exit variable but still does not affect the accidents reported. This indicates that policyholders who switch companies are those attempting to skirt the imposed incentive effects of the new rating policy. Some of the control variables are statistically significant in explaining the number of reported accidents: the vehicle's horsepower, the policyholder's place of residence, and the coverages for which policyholders are underwritten.