A View From The Press Box

Am I the only one who thinks that all of these people are asking the wrong questions?

The main point is this: Just because an actuarial relationship can be shown between credit scores and risk does NOT mean that it should be used.

In other words, the simple fact that a relationship exists does NOT justify its use. Whether or not the relationship is valid is entirely beside the point!

What if we could show that criminal record or medical background are good predictors of claims risk? Should we then grant insurers access to all of those records, too?

I don't know what the truth is behind the use of credit scores. Insurers argue that good credit consumers are subsidizing the premiums of low credit score consumers--if only we would let them use credit scores, then premiums would drop for the majority of consumers.

What? Premiums would drop for the majority of consumers? Are you asking us to believe that insurers are fighting tooth and nail for new laws that would lower their revenue base?

Please, thoughtful readers of this blog, set me straight.

In almost 30 years as an underwriter, I have never seen a correlation between a credit score and the quality of a person's driving record or their homeowners loss experience.

As with all exposures, there are going to be insureds who present a moral hazard with inflated or fraudulent claims, but these are as likely to be well-to-do as they are to have a low credit score.

It appears this issue is never going to go away.

With all the statistics showing the use of credit to predict losses is accurate and reliable, the question posed by the congressman is a pretty good one and should be addressed.

Being in the industry, I have customers who are the very worst from a credit standpoint and have never had a loss in five years, along with some of the best who have had three losses in a year.

Of course, there will always be some who go agianst the statistics. It is very hard to look at the data in whole and dispute that credit does matter.

But I do wonder what would happen if some improved their score in, say, three months...Would it improve their driving habits?

There have been some articles published about how people can change their score by being added to someone else's credit cards, and their score can go up by as much as 50 to 100 points. Once again, does this make them a better driver or a better risk for their home insurance?

I do think the government is going to keep on this, and if a Democrat is elected to the White House and the House and Senate stay in Democrat control, the industry had better be prepared to deal with the possiblilty that the use of credit scores in underwriting and pricing will be outlawed.

I have been in the insurance business for the past 24 years. My credit is NOT impecable, and the four drivers in my household have excellent driving records, yet we are being penalized with higher prices because of credit scores.

I think it's ridiculous and unfair to all consumers. The tiering programs these companies are using are shafting the general public.

I believe the underwriter's comments to be right on the money. However, I am sure we will hear from other underwriters who will contradict her.

In analyzing my almost 50 years of driving, I would say the times I did get into an accident, where it was my fault, were more directly connected to my emotional state of mind than my financial situation or credit score.

Being distracted, not concentrating and not planning properly were the primary or direct reasons for my accidents, not my below-average credit score.

I do not believe people with above-average or high scores are better drivers than those with below-average or low scores.

The industry has picked the wrong criterion or been sold a bill of goods by these credit scoring firms, and they have once again shot themselves in the foot from a PR standpoint.

I don't believe that credit scoring is a reasonable predictor of accident-prone behavior. If this were true, I would likely not be seeing all the BMW, Mercedes and high-priced vehicle owners in wrecks here in California that are directly attributable to their own poor driving.

It would always be the other fellow with the lower credit score who could afford only the cheaper vehicle!

That said, with tongue-in-cheek, credit scoring might be of some value only if there were serious deficiencies that would be a predictor of present or upcoming
severe financial distress, such as bankruptcy or other impending financial problems likely to impact an individual's emotional state.

Yet that, too, is so individualistic that it varies widely from one person to another.

Maybe I've been involved in the law enforcement side of the game all too long, but looking at a criminal history, as mentioned by Informed Consumer, really does make sense to me.

That would likely be a greater predictor of real risk than how you do or don't control your money. It would, perhaps, be indicative of how you control yourself, and isn't that a major cause of loss behind the wheel?

Another question to ponder...

The insurance industry never claimed a low insurance score makes you a worse driver. They said that it makes you more likely to file a claim.

Most auto claims are small and costly to investigate. Those people who have high credit scores are likely to pay for the repairs out of pocket so their rates don't go up. People with low scores will file a small claim because they don't have the option not to.

Besides, if people with low scores were truly irresponsible, they wouldn't carry insurance in the first place.

I am in complete agreement with the others, both as a consumer and as a 25-plus-year insurance veteran.

There are a lot of components to credit scores. Having a poor score does not automatically mean a person has unpaid, defaulted debts. Scores are impacted by ratio of credit limit to balance, number of credit inquiries, late payments, etc.

From an underwriting perspective, I could almost buy using credit scores in pricing new business--where the insured is an unknown quantity and the Insurer has no first-hand information on past loss history to rely upon for risk assessment.

However, once an Insured has an established record with a company, there is no longer any reason to use credit scores in the pricing.

In my own case, I have been with the same carrier for 20 years with two claims (totalling less than $5,000). Yet I still receive a notice with each renewal offer that my premium has been affected by my less-than-spotless credit. Am I still considered a risk for fraudulent claims after 20 years? I think not.

The telling phrase I've seen in industry publications in discussions of the use of credit scores is "revenue leakage." I consider this to be the true motivator. As long as the regulatory bodies allow it, well, no one wants to leave premium on the table, right?

I would be curious to hear from other readers who are more informed than me about whether the regulations have any stipulations as to frequency of review, consistency of application, etc.

Is there any proof required that if scores have improved, the insured is receiving the benefit? If insureds' premiums are being debited for poor credit, is there an equivalent credit for insureds with excellent credit?

Insurance may be my livelihood, but at heart, I'm on the side of the consumer!

Although I haven't been in the insurance industry for a long time like some of the other posters, I have reviewed hundreds of credit reports when I worked in banking. I have also read quite a bit about what factors are used in calculating a FICO score.

Based on that experience, I am convinced that credit scoring is not a form of discrimination. One poster made the incorrect comparison of someone with poor credit to someone who is "well to do." Being weathly does not mean you automatically have good credit. In fact, I have seen very wealthy people with very poor credit scores.

While there are certainly exceptions to every "rule," I think there is a correlation between credit scores and personal responsibility. Why do I say that? Because one of the biggest factors in determining your credit score is how well you do at paying your bills on time. One late payment can reduce your credit score by 100 points, just like one careless moment can result in an accident that costs thousands of dollars in damages.

If that is true, why not allow people with good credit to pay less money for their auto insurance? It is relatively easy to check your credit score, and there is a lot of information out there about how to improve your score. Will it make you a better driver? Why not try it and find out!

Are you folks paying attention? It seems that after eight more comments I must ask again: What justifiable right to insurers have to use credit scores at all?

Credit scores are for the sole purpose of determining the liklihood of a consumer to repay a debt obligation. PERIOD. Insurers are not loaning money!

Whether or not there is a correlation is not the central issue.The issue is, where does it stop? Should we also grant auto insurers complete access to our medical history? Job history? School grades? IQ tests? if they can prove there is a coorelation? What about criminal background checks?

For goodness sakes, readers, a jury court is not even allowed to see your criminal history when you are being tried for a separate crime!

Hello? Wake up! The issue is NOT whether or not there is a correlation. Who cares if there is? The question is whether or not they have any reason accessing that information in the first place!

If we allow them to access this information "because they can prove a coorelation," then why not give them access to anything else they can prove a corelation with?

SAM RESPONDS:
Don't give carriers any ideas, Insured Consumer!

The 'common sense' questions posed, (i.e. if your credit score goes down, do you become a worse driver) do indeed make sense in an oversimplified way.

However, car insurance aside, I have no problem answering those questions relative to property insurance.

If some major financial setback causes your credit score to drop, you might indeed also be forced to put off needed maintenance on your home, leading to a greater risk of loss.

The opposite is also true if you had credit issues but get your life into good order, chances are you also might fix your roof or put in a new sidewalk to get rid of the cracks.

I am an agen,t and although I understand the math of credit scoring, the (oversimplified) connection that the congressman is trying to make here lends itself better to homeowners.

On credit scoring, what if the individual pays everything in cash and rents an apartment, thus has no credit history? Does that make that individual a poor driver?

Thank you, Mr. Biegen, for being one of the few here who actually posed a "common sense question."

The credit scoring basis does not have anything to do with someone being a "better" or "worse" driver. It has to do with financial responsibility.

For the respondent who answered, "we are not a lending institution," I would respond that in fact we are. If a cancellation goes out 30 days past due, and is not allowed to cancel until 10 days after that (per most state statutes), then we just afforded that person financial coverage.

If that individual has an accident in that time frame (premium paid or not), it is but guaranteed that the company will be paying the claim.

For the people using individual experiences, just stop! In the words of Mr. Spock, "the needs of the many outweigh the needs of the few." This certainly holds true here.

There are 175 million drivers in the USA, and this system is designed to the industry as a whole, not on individual experiences. It's the best we have, and has shown to be pretty accurate up to this point.

So if I understand Mr. Burnell, who thanks Mr. Beigen because credit scoring is about financial responsibility.

Then it is okay to charge people more for their premiums because they do not have money to pay for claims out of their pocket--small claims? Who decides what is a small claim and what is big claim? This is why the consumer groups and goverment has a problem with the whole credit issue.

It should not be about how much money a person has to pay out of pocket. Credit scores can be a very poor indicator of how much cash on hand someone has.

I do believe credit is an indicator of potential claims frequency, but with that said, should there not be a way to take care of the person who defies the statistics?

If a low credit-score person who is supposed to have claims has none, why would that person still pay more in premium than a high credit-score person who has filed claims, whether auto or property?

I think that one thing some prior commenters are not remembering is that personal lines insurance is class-rated, not individual-rated. There are always exceptions to the class, but we take the class average to devise rates.

So, the individuals that say that credit scoring is not accurate because they don't file claims even though they don't have good credit are the exception to the class.

This is similar to a 17-year-old wondering why she pays more for auto insurance, with her spotless record, than her 40-year-old dad pays even though he just had an accident.

That said, insurance scoring cannot be the only measure for devising premiums and/or acceptability. All appropriate factors should be taken into consideration. In auto insurance, credit should be a small factor in rating, behind driving record and age of the driver.

I wonder why the debate always seems to be around auto insurance, without hardly mentioning homeowners? Is it because it is easier to defend the practice in homeowners insurance?

I personally believe the practice is beneficial in this line of business, but I am not so sure it is useful in auto insurance.

How often to the people with mysterious fire losses also have poor credit? This is often one of the first pieces of evidence an insurer gets when investigating suspicious fires.

A critical point that seems to be ignored here is that correlation does NOT imply causation.

There are many patterns in this world where things match up more or less, and we may not understand what commonalities they have. An example from Wikipedia:

Since the 1950s, both the atmospheric CO2 level and crime levels have increased sharply. Hence, atmospheric CO2 causes crime.

The bottom line is, nobody is quite sure WHY credit score correlates highly with insurance risk--we just know that it DOES. In the insurance industry, it is our business to associate premium with risk, therefore, the use of credit score is highly desirable.

The completely unrelated question that remains unanswered is this: Even IF the correlation exists between credit score and risk, is it socially responsible and ethical to use it?

If the insurance industry is using credit scores to help determine who will file a claim for a minor loss, then they really ought to look at what goes into the credit score.

As someone familiar with credit scores from the mortgage industry, you'll have plenty of high-income people with low credit scores--doctors, real estate investors & others. They have far more cash/income than my wife, who doesn't work (except caring for our children) and has a higher score than I do.

You'd be better off looking at cash and assets on hand (bank statements, etc.).

I had a heated argument with an actuary that I used to work with about credit scoring. He posited that a homeowner in financial trouble would be more likely to burn his home than someone with a better credit score!

I advised him that arson is fraud, the claim would not be paid, and this is not a good case for rating increases.

Statistics can make anything seem logical, like the fact that I have brown eyes definitely correlates with the fact that I'm a great driver. The premise, however, is illogical.

Lewis Carroll was an expert logician, but he also knew too well about falling through the looking glass into a fantasy world.

I have been in the industry 25 years, and can't remember a more reprehensible form of redlining.

Insurance companies pay on fraudulent arson claims all the time.

First of all, it can sometimes be difficult to prove fraud, therefore the claim is paid. Also, even if fraud can be proven, if the insured has a lien on the home, we have to pay the lien.

I almost forgot that some states also have innocent-spouse laws. If we can prove one of the insureds burned the home, but we can't prove the other insured knew that his or her spouse was going to do it, we have to pay that spouse.

The standard rating variables are gender, age and marital status. Why should insurers be allowed to discriminate based on those attributes?

A standard tiering variable is years continuously insured or time of lapse. Shouldn't that be thrown out as well?

On what attributes would you allow insurers to base a rate?