The Construction of Automobile Rating Territories in Massachusetts

In Massachusetts, the past ten years have witnessed the evolution of an increasingly sophisticated system of methodologies for determining the definitions of rating territories for private passenger automobile insurance. In contrast to territory schemes in other states, which tend to group geographically contiguous towns, these Massachusetts methodologies have had as their goal the grouping of towns with similar expected losses per exposure, regardless of the geographic contiguity or non-contiguity of the grouped towns. This paper describes the evolving Massachusetts methodologies during that then year period. The paper includes the latest methodology, which was employed to establish rating territories for use in Massachusetts in 1986. That methodology evaluates by-town claim frequency and by-town claim severity separately and then combines the results. The claim frequency approach is to compile detailed insurance data by town, and to compare those actual observations to an a priori model of the expected insurance losses in each town. The model and the actual observations are blended using empirical Bayesian credibility procedures. The claim severity analysis uses a two layer hierarchical empirical Bayesian method in which countrywide and statewide severity data supplement less-than-fully-credible town severity data. The combined results of the frequency and severity analyses serve as the basis for ranking the towns according to expected losses per exposure and for placing the towns into rating groups. Class Rating, Territorial Rating