by Walter C. Wright
Following is an excerpt from Matthew Rodermund's November 1979 editorial focusing on a session at the CAS Spring Meeting that year. Change the words "age, sex, and marital status," to "credit history" and the debate is as fresh now as it was then.
Cost Based Classification Criteria
We believe it is agreed by the technicians that a proper risk classification needs more than to be cost based. But many segments of the industry have resisted current efforts to eliminate age, sex, and marital status as classification definitions largely on the ground that such distinctions can be statistically proved to be valid. They are cost based. What has happened is that in my quarters the importance of the cost based concept has been broadened to embraceand thus minimizeother considerations that the public regards as vital.
Certainly age, sex, and marital status meet Mr. [Michael] Fusco's standards of definiteness and practicality, but the insurance business has not yet established conclusively that those classifications meet the homogeneity criterion. There is a suspicion, moreover, that the agreed validity of the cost based criterion for age, sex, and marital status has discouraged the incentive for actuarial innovation that Ms. Bogue [Linda A. Bogue of the Stanford Research Institute] called for, and tends to sidestep her question on asymmetrical fairness, i.e., are good risks unreasonably overcharged?
The cost based concept is fundamental, but it ought not to be used as club to beat down other considerations.