Posted: 2/1/2005

NAMIC Concurs With Texas Commissioner on Accuracy of Insurance Scoring in Predicting Risk

'Credit Scoring, if continued, is not unfairly discriminatory'

INDIANAPOLIS (Feb. 1, 2005)--On the controversial subject of credit-based insurance scoring, Texas Insurance Commissioner Jose Montemayor firmly agrees with the long-held positions of the National Association of Mutual Insurance Companies (NAMIC): Scoring accurately predicts risk and does not unfairly discriminate. That was made abundantly clear in a letter the commissioner sent to Gov. Rick Perry on Monday summarizing the conclusions of a new study by the Texas Department of Insurance.

"This has to be the final nail in the coffin of opponents to credit-based insurance scoring," said NAMIC State Affairs Director Neil Alldredge. "The data have convinced the Texas insurance commissioner, a self-admitted skeptic, that this legitimate underwriting tool is what the industry and other studies have said: Credit Scoring accurately predicts the likelihood that a person will file a claim for both auto and homeowners insurance. Moreover, credit scoring is not unfairly discriminatory because credit scoring is not based on race."

One of the few remaining questions in the insurance scoring debate was whether the relationship between an individual's credit score and claim experience provides significant additional information, over and above other rating variables, that serve to improve underwriting accuracy. Some critics had speculated that credit scores are merely a proxy for other factors; hence the correlation between credit scores and claim filing is merely coincidental.

To settle the matter, the TDI performed a multivariate analysis that considered the relative impact of other rating variables in addition to credit score. In the words of Commissioner Montemayor, the multivariate analysis proved "that credit scoring significantly improves pricing accuracy when combined with other rating variables in predicting risk." (Emphasis added.)

Addressing the question of whether credit-based insurance scoring unfairly discriminates on the basis of race and ethnicity, Commissioner Montemayor wrote: "Credit scoring, if continued, is not unfairly discriminatory as defined in current law because credit scoring is not based on race, nor is it a precise indicator of one's race. Recall that not all minorities are in the worst credit score categories. Further, its use is justified actuarially and it adds value to the insurance transaction."

A ban on credit scoring, added Mr. Montemayor, "would be a set-back to all Texans, of all races, especially those of moderate to lower income whose risk remains low." Here Commissioner Montemayor echoed a NAMIC op-ed that recently appeared in the Houston Chronicle, as well as a NAMIC policy paper released last year, The Legal Theory of Disparate Impact Does Not Apply to the Regulation of Credit-Based Insurance Scoring.

Key findings of the TDI report include:

• For both personal auto liability and homeowners, credit score was related to claim experience even after considering other commonly used rating variables. Credit score provides insurers with additional predictive information distinct from other rating variables. Using credit score, insurers can better classify and rate risks based on differences in claim experience.
• For personal auto liability, credit score was related to the probability of filing a claim or claim frequency. However, the Department found very little or no statistical evidence that credit score was related to the amount of a claim or claim severity.
• For homeowners, credit score was related to the probability of filing a claim. The Department found some statistical evidence that credit score was also related to the amount of a claim but was unable to draw a definite conclusion on this matter due to uncertainties caused by the unusual influences in the data (e.g., mold claims, large claims, and storm events).
• For personal auto liability, credit score varies in importance depending on the model the insurer is using, but was generally comparable in importance to territory and driving record for predicting claim experience. Only class (which reflects the age, gender and marital status of the driver combined with usage of the vehicle) was consistently a more important rating variable than credit score, driving record, or territory.
• For homeowners, credit score was one of several important rating variables for predicting claim experience. However, the Department was unable to draw definite conclusions regarding the relative ranking of rating variables due to the uncertainties mentioned above.
• For both personal auto liability and homeowners, the difference in claims experience by credit score was substantial. Typically, the claim experience for the 10 percent of policyholders with the worst credit scores was 1.5 to 2 times greater than that of the 10 percent of policyholders with the best credit scores. The magnitude of the variation noted in the earlier report remains unchanged even after considering other commonly used rating variables.

The commissioner's cover letter and the supplemental report can be found online.