Credit Scoring In The Underwriting Of Personal Property And Liability Insurance

Read Complete Research Material



Credit Scoring in the Underwriting of Personal Property and Liability Insurance

Introduction and Background

Credit scoring is a process of calculating the credit risk in underwriting of insurance policies. By means of statistical techniques and historical data, credit scoring attempts to evaluate the effects of applicant characteristics on defaults and wrongdoing. The method generates a "score" that Insurer can use to rank its applicant in terms of risk.

Credit scoring can be termed as the most important new development in the past two decades in insurance industry to calculate risk and predict potential losses. To this end, a formula is constructed using approximately 10-50 of more than 450 variables obtainable from an individual's credit file to derive a summary numerical score for predicting insurance losses (as opposed, e.g., to banks using credit records to predict default (credit risk) or its use in commercial insurance). The positive statistical relationship between credit scores and insured losses has been verified by multiple studies and multiple methods, and no study reporting a lack of statistical relationship has been published in the refereed literature. In automobile insurance, for example, Miller and Smith (, p. 97) found that of six possible automobile coverages, credit scores are always in the top three most important loss predicting variables, and often the most important variable. Credit score is the first variable considered in personal injury protection and medical payments coverage, second in bodily injury and property damage coverage (behind age and gender), and third in comprehensive and collision coverage (age and gender being second, with make and model of the car dictating costs more).

Credit history has, in fact, been used for decades in commercial lines of insurance and life insurance. Although it has been known since at least 1949 that credit history is related to driving accidents, the advent of high capacity, high-speed computers has made massive personal credit files available, and has made it feasible to routinely use this credit information for predicting insurance losses in personal lines of property and casualty insurance. Tillman and Hobbs (p. 106) show that drivers with bad credit history have repeated crashes at a rate six times higher than those with good credit history. Moreover, a 1968 study of Washington state drivers (Insurance Institute for Highway Safety, p. 44) showed that within the group of drivers who had a history of no automobile accidents, 64 percent had good credit while 35 percent had bad credit.

On the other hand, among drivers who had two or more automobile accidents, 35 percent had bad credit while only three percent had good credit. Tillman and Hobbs, who examined other lifestyle variables in addition to credit scores, put it quite succinctly: "... a man drives as he lives." (p. 329).

In spite of the success of credit scores as an underwriting tool and the clear and consistent demonstration of their association with losses, cross validated over multiple methods and multiple studies (c.f., Monaghan, p. 79; Kellison et al., 2003, p. 64; Miller and Smith, p. 106; Wu ...
Related Ads