In response to the article "Credit Scoring: Tough to explain, hard to beat" by Bart Anderson (AA&B 3/10), I don't believe that rates should be based on our clients' spending habits.
People grow up and their driving habits change. We only charge for minor violations for 3 years and major violations for 10 years (here in California). It can take longer than that to repair a client's credit. These are the same people who last year had excellent credit. Their driving habits have not changed, the economy has.
This is the time to rate on actual figures that have a direct correlation with the risk. Driving records, claims experience, age, geography, mileage, use, vehicle type and marital status are excellent rating factors. Most of these are choices that give the client an opportunity to change in the short term if they cannot afford insurance. Rating on credit in these times is not as accurate an interpretation of the client.
As you said, "Removing credit from the underwriting/price mix" will not make insurance more expensive. Look at New Jersey and Michigan, where auto insurance rates are through the roof and they use insurance scores. States like California and Hawaii have reasonable rates and those states do not use insurance scores. I don't believe for a second that taking away insurance scores will cause mass rate increases.
Jeff Apperson
Apperson Insurance Services
Carmichael, Calif.
The article by Bart Anderson really interests me. I have heard over and over that "credit scoring is a fair and accurate predictor of loss." Where is the actual proof? The reasons given are "because of all these studies," but there is never an explanation of what data was collected, the length of time it was collected, and where the information can be examined by the general public. I haven't seen any documentation of such research relating to credit scoring.
A "study" is nothing but a snapshot of a selected historical sample--10,000 auto claims. Then common factors are looked at. Will you look at this--the vast majority of the claims are submitted by people with low credit scores. Now comes the magic--an assumption is made that people with low credit scores are more likely to file claims (have losses). Based on what principle of cause and effect? Coincidence of a snapshot?
Real research involves a large random sample of people being surveyed for many years. All kinds of data would be collected, including but not limited to insurance claims, credit score, etc. If the data revealed a direct relationship confirming that as their credit scores decreased or remained low and their claim frequency increased, I would be a believer. Until I can access some real research, I say the king has no clothes and credit scoring has no relevance to claims potential or frequency. If the data exists from true research and not a snapshot study, show it to me or give me the source and I'll examine it for myself.
Until such time, it hasn't been proven and no matter how many experts tout "because we say so," it makes no difference to thinking people like myself. This is my 39th year in the insurance business writing personal lines and I maintain the use of credit profiling to predict loss frequency is pure bunk. I am forced to use it through our comparative rater, but I don't like it or agree with it.
Darald l. Novak, AAI
company withheld
Altamont, N.Y.
Bart Anderson responds: The Federal Trade Commission study, available at www.ftc.gov, contained more than 2.7 million randomly generated records and concluded that "credit-based insurance scores are effective predictors of risk under automobile policies. They are predictive of the number of claims consumers file and the total cost of those claims. The use of scores is therefore likely to make the price of insurance
better match the risk of loss posed by the consumer. Thus, on average, higher-risk consumers will pay higher premiums and lower-risk consumers will pay lower premiums."
EPIC Actuaries LLC completed its research, "The relationship of credit-based insurance scores to private passenger automobile insurance loss propensity" in 2003 (www.ask-epic.com), and concluded that "insurance scores were found to be correlated with the propensity for loss" and that "insurance scores significantly increase the accuracy of the risk assessment process."
In 2005, the Texas Dept. of Insurance released its "Report to the 79th legislature: Use of credit information by insurers in Texas" (www.tdi.state.tx.us) based on its examination of more than two million policies. The conclusion: "There appears to be a strong relationship between credit scores and
claims experience."
One statement in these reports best sums up the findings and provides us all with the great mystery that fuels the debate. "A consistent finding of prior research and the FTC's analysis is that credit information, specifically credit-based insurance scores, is predictive of the claims made under automobile policies. However, it is not clear what causes scores to be effective predictors of risk."
It is the "why does it work" question and our attempts to find reasoning or simple logic in all that science that fuels the debate and controversy. However, the answers to the questions about "Does it work?" or "Is it a fair and accurate risk assessment tool?" are clear. At some point we need to stop debating whether it works or not--it just does.
