ABSTRACT
If risks are not all equal in an insurance scheme, it seems fair (and perhaps essential) to require insured parties to contribute (in the form of premiums) roughly in proportion to their relative risk. The risk of a motor accident, which often gives rise to an insurance claim, varies from driver to driver. What are the factors that contribute most to high risks of claims in motor insurance? It would seem that the risk of a motor accident must be correlated with
the annual driving distance (mileage) of a driver, but this is often difficult to measure in an economical way. Some attempts have been made to use it directly (Sweden and Holland) or indirectly (using, for example, distance between home and work in the USA), but many of these measures are often underreported. In the USA there are indications (Lemaire [37]) that such underreporting results in more motor insurance fraud than faked accidents. The ability or skill of a driver is also related to the propensity to give rise to
a claim. This encompasses such factors as knowledge and appreciation of the rules of the road, care with respect to speed restrictions and driving conditions, and good judgement about driving when tired or under the influence of factors such as alcohol, tension and aggressiveness. However, these factors are also difficult if not impossible to measure cheaply and effectively. Some variables that are more accessible and commonly used in rating are the age, gender, marital status, occupation, residence and driving experience of the driver. The age, engine size, model, purpose and even color of the vehicle are also used. In his study, Coutts [17] suggested using type of cover, policyholder age, vehicle age and vehicle size as rating factors, while Brockman and Wright [10] suggested adding district and vehicle use to these. Having selected appropriate variables, each set of individuals with common values for these factors (or covariates) constitutes a rating or tariff group. The group is then charged an appropriate premium by the insurance company based on its own (and others’) experience. However, such groups can still be quite heterogeneous in their risk factors, and insurers are constantly trying to reduce heterogeneity in risk groups. Research has shown (e.g., Lemaire [35]) that if insurers are allowed to use
some form of Evidence suggests that the best indicator of future claims for a driver is the individual’s past claims history. In the 1950s the idea emerged to adjust one’s initial premium assessment following the actual observation of the individual’s claims experience with the use of a No Claim Discount (NCD) system for rating. No Claim Discount systems (sometimes also called Bonus-Malus systems)
are experience rating systems which are commonly used in motor insurance. NCD schemes (or systems) represent an attempt to categorize policyholders into relatively homogeneous risk groups who pay premiums relative to their claims experience. Those who have made few claims in recent years are rewarded with discounts on their initial premium, hence are enticed to stay with the company. Depending on the rules in the scheme, new policyholders may be required to pay the full premium initially and then obtain discounts in the future as a result of claim-free years. There are a wide variety of NCD schemes. The regulatory systems in the
UK and Ireland give insurance companies considerable latitude in designing their own schemes and tariff structures. The number of discount classes, their range, and the rules for moving between them can vary considerably, but typical features of such schemes are the following:
1. The number of classes (or discount levels) in the scheme varies between four and eight, with six being the most common.
