One of the liveliest exchanges on my new blog was prompted early on by my Sept. 19 filing of Associate Editor Mark Ruquet's NU column on the credit scoring controversy, "Credit Scoring A Sore Point For Consumers." We received so much reader feedback that a few days later, on Sept. 22, Mr. Ruquet filed his own "Counterpoint" on this blog to respond to reader critiques, urging credit scoring proponents to "take off their blinders" and consider consumer concerns. With more responses pouring in, he penned a second column in the Nov. 20 edition of National Underwriter that is reprinted below. Feel free to continue the debate on this blog!
Readers Talk Back On Credit Score Debate
BY MARK E. RUQUET
One of the greatest fears of a reporter or any writer is not being read. So it was with some satisfaction to receive many responses—pro and con—to my column on credit scoring. I’ve received one or two comments on past columns, but in comparison, the reaction to my most recent piece—calling the practice into question—was an avalanche.
Readers had two chances to comment: once, when my views appeared in this space on Sept. 18 (“Insurers Shoot Messenger, Then Ignore The Message”), and then after my editor, Sam Friedman, posted my column on his blog (www.property-casualty.com) later that week.
Some agent readers agreed that the use of credit scores is controversial in the eyes of consumers, and raises suspicions. While unscientific, this might be a real indicator that among those who have to sit in their offices and explain the practice to customers, it’s not an easy sell.
“While reading your column…I was gratified to see that someone in the insurance industry is finally willing to admit and pronounce the fatal flaw in insurance scoring—the unreliability of the source,” wrote Len Stayton in an e-mail. “As an active agent working with clients on a daily basis, I see the damage that is done to consumers by this system of credit reporting that our industry has come to rely on so heavily.”
Another agent e-mailed to say he felt there was something to the opinion that the industry has a tendency to shoot the messenger instead of listening to the message. He also agreed that the industry is vulnerable to being inadvertently caught up in any alleged credit scoring abuses.
He felt that while “slow payers” increase expenses for insurers, the raw data for credit scoring does have defects. He concluded it would be wise for the industry to demonstrate some flexibility and listen to the criticism—to make the process better before the whole system is outlawed, “even where it can be actuarially supported.”
“I agree with you that this will blow up in someone's face,” said a third agent in a private e-mail, who requested anonymity. “Since I’m an agent, I know who the angry consumers will vent on.”
Others, however, remained steadfast defenders of credit scoring.
“This is just another example of how truth is being pushed out in favor of emotion more and more,” wrote Todd R. Bault, a senior analyst on non-life insurance for
Sanford C. Bernstein LLC, in New York, posting to Sam’s blog.
“Credit variables relate to insurance usage—i.e., how often one claims,” he said. “They do not penalize for credit crises, but they do penalize for using lots of credit because that is highly correlated to insurance usage. And compared to other data sources, credit is more accurate because people care more about it. That's part of why it works so well—the data’s good.”
“It seems to me the question to be answered is: Does the use of credit accurately predict losses?” wrote David Voteau on Sam’s blog. “If the answer is no, then why have so many insurers switched to using it? And why have their loss ratios improved? If the answer is yes, and the stats back it up, then let's quit having this debate.”
“The stats show there is a definite link between poor credit and poor driving habits,” he continued. “Basically, it is a life-style choice people make, and the way some people treat credit relates to the way they drive.”
Another agent wrote in an e-mail that I had “missed the proverbial ship on this one.” Though he had not read the Consumer Reports article casting doubt on the practice, he said insurers use credit scores in pricing personal lines because there is a “clear correlation between poor scores and poor drivers.” Databases, he noted, are never 100 percent accurate, but this one is far more accurate than inaccurate.
“We were buying a new car for cash recently, but the car dealer pulled out his chart showing loan rates. The chart listed credit score across the top and then rates that became progressively higher based on the scores. It’s interesting that there is not a public outcry about this like there is about insurers using the scores,” Stan Chraminski, with Learning and Organizational Development, noted on Sam’s blog. “The question is, how do we get like the banks, where the use of scores seems to be so readily accepted?”
No matter which side of the issue you are on, questions are still being raised and voters are being asked to answer them.
John Desmarteau, president of the Professional Insurance Agents of Oregon/Idaho, noted on Sam’s blog that Oregon’s referendum on whether to ban credit scoring was taking place on Election Day. Measure 42, as it was called, went down to defeat, by 65 to 35 percent.
Current law in Oregon, he noted, allows for the use of credit scores only on initial applications, and prohibits its use to raise or drop rates on existing customers. The association, he added, supports their use.
“Eliminating the use of credit information would be unfair to the great majority of insurance customers who carefully manage their business and personal finances,” he wrote. “Responsible individuals and well-run businesses end up subsidizing the insurance costs of those that aren’t as careful. Without credit scoring, rates are higher for everyone.”
He added that “removing tools that have proven to be accurate will only result in unfair cross-subsidies between risk groups.”
At least in Oregon, instead of simply assuming the issue is not worth arguing over, members of the industry put up a case to consumers that there is some benefit to credit scoring for them, and won the day. (See “Oregon Voters Okay Credit Scoring For Insurance Rating,” Nov. 13, page 33.)
Meanwhile, in Michigan, the state Appeals Court is considering whether to reinstate the insurance department’s regulations banning the use of credit scores.
Pro or con, this issue is not settled. If insurers really believe it is, they could be in for a rude awakening one day from a consuming public that still views the industry’s use of credit scores with a sense of suspicion—one I can’t help but share because of the unreliable nature of the source data itself.
To comment further on this issue, contact Associate Editor Mark E. Ruquet at mruquet@nuco.com, or right here on Sam's blog.
Comments (9)
The credit-score predictor of claim frequencies (claims per 100 insured car
years) is simply a function of Demand Law reaction to charging insurance as a cost of car owning rather than car operating.
When finances get tight, to save on car insurance drivers cut down on cars proportionally more than they cut down on miles. Hence the average miles per car year on shared cars rises.
With the increase in average miles per car, claims per 100 car years also must rise.
See the short and long explanations on our Web site: www.centspermilenow.org
See especially the explanation in CARTOON form for Texas legislators, item #736 (index page, fifth bullet down on the left).
Keep up the good reporting.
Patrick Butler, Ph.D.
Insurance Project Director
National Organization for Women
Washington, D.C.
Posted by Patrick Butler | November 27, 2006 5:45 PM
Posted on November 27, 2006 17:45
Sooner or later, when insurers do something that is legal but viewed widely by the public and policyholders as illegitimate, the public will succeed in asserting its position. It may do this through the auto line, or another line of insurance. Which means that auto insurer use of credit scores puts insurers of all personal lines at risk of retaliation.
One reason why the use of credit scores can appear illegitimate is that, from what I can see, no consumer can reasonably understand how she or he is gaining from it. This is not too far off from the skepticism held by many employers about the legitimacy of their workers' compensation premium. The pricing protocol is just too complicated to reasonably expect a customer to understand it.
Auto insurers should find a way to use credit scores to help make safer drivers, to reduce claims (and losses retained by the customer) to the ultimate benefit of the customer.
I don’t know how to do this. But I bet that many, if not most, risk management innovations of merit succeed only if the policyholder community comes to view them as serving their interests, not just the insurers’.
Posted by Peter Rousmaniere | November 27, 2006 10:50 PM
Posted on November 27, 2006 22:50
Missing Data Element:
The answer to credit scoring lies in the middle of the two extreme arguments.
Credit scoring is based on actuarial statistics, but these statistics leave off a major criteria-insurance fraud losses. Insurance companies often site figures ranging from 10-to-20 percent of insured losses being due to fraud. In essence, these losses should be removed from the data and spread evenly to all insureds.
The current approach in essence is saying people with lower credit scores are more apt to commit fraud, or a crime, which is unfair discrimination, which most states specifically prohibit. The result of a recalibration is a result in the middle.
Posted by Stacey Parsons | November 28, 2006 8:12 AM
Posted on November 28, 2006 08:12
A further thought, pertaining to the problem of legitimacy as perceived by the public.
The credit scores are not being correlated to losses. No, they are correlated to "claimed" losses. Every car owner knows the difference.
You might say that we all make economic decisions on whether to file a claim, eat the repair cost, or simply not repair. At the end of the day, they say, we are all economists.
But this leaves hanging, unanswered, a question: What is fair in the public mind about a system of insurance which induces the policyholder not to claim? And implicitly rewards policyholders who have the funds to self-pay repairs?
In the workers' comp realm, this gets close to, and crosses over, into criminal fraud. The situation is much different in auto insurance, but there remains an odor.
Posted by Peter Rousmaniere | November 28, 2006 10:04 AM
Posted on November 28, 2006 10:04
In Oregon, like many states, the voters turned down removing credit scoring because the ads say, if you have good credit and you remove credit scoring your rates will go up.
I don't think this is the problem with credit scoring. I think the problem is the abuse of the insurer.
They are working on a bell curve when they use credit scoring. It isn't worth it to them to correct problems that affect a small percentage of consumers on the outside edges of the curve.
Thus the people with a wrong report, for whatever the reason, do not have easy ways to correct it--or sometimes no way to correct it.
As an example, if a person with good credit moves into a new state, he will upset his credit score while he is getting established. The insurer does have a way to see this.
Since it falls out of the bell curve range of acceptable credit risk, they aren't willing to spend time or money to correct the problem for this small percentage of consumers. It just doesn't make them any money.
This is never explained to the voter when these issues come up, and even if it was, most of the voters would say that doesn't affect me and will still vote regulations down.
Posted by Rodger Lance | November 28, 2006 1:59 PM
Posted on November 28, 2006 13:59
The problem with credit scoring is that it is not just about credit. The age someone is when they first obtain credit, the number of revolving cards with no balance, the ownership of an oil company credit card and credit inquiries don't relate in most consumer's minds as "credit" scoring.
If I don't want an oil credit card, it shouldn't affect my score. I can't change the age I was when I first got credit. How many revolvoing cards I have with a zero balance should not be of an auto insurer's concern.
Credit has always been used for underwriting, and if the criteria was limited to judgments, liens, bankruptcies, late payments and length of credit history, I'm not sure consumers would mind.
How the industry will ethically handle Katrina victims' insurability could speak volumes about public acceptance of this system.
I have placed two customers this year with credit scores in excess of 800 with DWI's for less money than they were paying prior to the DWI.
Somebody's scoring model thinks that financially responsible drunk drivers deserve a rate break, and that serves to make the system as it stand make no sense to consumers.
Posted by Gloria | November 29, 2006 10:42 AM
Posted on November 29, 2006 10:42
Mark, I appreciate your efforts to revisit this issue.
However, there is one point you keep asserting (and others concur) that is simply wrong.
Your biggest beef seems to be that credit scoring data is unreliable. But compared to what? You and others seem to think that there is some other magic data source out there that is better, and somehow insurers are suppressing this.
Based upon everything I have read (and I'm an actuary and stock analyst, so I read a lot), credit data is probably the BEST data available. Its statistical effectiveness is practically self-testiment to this point.
If you think there is better data source, name it and provide justification. I'm telling you, based upon firsthand knowledge, that most insurance data is poor quality, and there is little that can be done about that. Credit scoring is a huge improvement.
I have a bit more sympathy to a new point that seems to be arising, namely that consumers can't understand the technique. But only a little more sympathy.
This was the point I made last time about emotion running things. It really is bad policy to discard science (or things close to science, like statistics) in favor of "gut checks."
Unfortunately, Americans have a terrible anti-intellectual bias, to our discredit. The fact that too many of us are proud of this fact (i.e. they would likely call me "elitist") is even more reprehensible.
When we disregard more rigorous and objective methods like science and mathematics in favor of gut checks, we do real harm. The fact that experts can be wrong or abuse their power is not an argument--relying on non-expert gut checks is worse.
I know that many will not agree with this point, but frankly this is an issue I think believers in science and good-faith attempts at objectivity need to stand up for.
Currently, I realize my side of this issue is probably the loser (and I do agree that this is the major threat to credit scoring), but that is a poor statement about American anti-intellectual attitudes, frankly.
Posted by Todd Bault | November 29, 2006 10:48 AM
Posted on November 29, 2006 10:48
There are many factors that could be used to predict rates. Some of these the public finds unacceptable (race, for example). I do not think there is enough logic behind the link between loss and credit scoring to enable the public to understand the connection.
Some insurers are sending consumers to collection agencies when there is a minimal amount of premium due. (I saw one notice for less than $5.) How would that consumer feel if the insurer's action caused their credit rating to be reduced, and thus allowed them to charge higher rates? What about those cases where health insurance claims are not paid in full, the consumer chalks up a huge medical debt, and their credit rating is reduced?
Personally, I was able to explain a few anomolies in my credit file when I bought my home. The bank and realtor realized there are sometimes errors, unknown items, or other issues in the credit files.
Why don't the insurance companies allow consumers to ensure the accuracy of their credit reports before they are used to create an insurance score?
If the industry really wants to keep credit scoring as an option, perhaps they should consider removing some of the secrecy about their scoring. Consumers would need education to accept the practice, yet most companies choose not to proactively educate potential clients.
Finally, the use of credit scoring is only one of the controversial issues that will raise consumer ire. The use of education and profession are two additional factors that, once people figure out they are being utilized, could create a huge outcry.
If an individual chooses not to attend college, does that REALLY make them a higher risk? If so, how does that explain that some of the most educated (doctors and attorneys) are involved in more accidents?
Again, the industry has a responsibility to educate the public if they want to use these risk selection items.
Posted by Kay | November 29, 2006 4:35 PM
Posted on November 29, 2006 16:35
"... no consumer can reasonably understand how she or he is gaining from it."
I do not believe that is true. To the degree that it is true, it reflects poorly on insurers' and agents' ability to educate their insureds.
About half of all insureds (those with better than average scores) will gain.
It's pretty simple. No need for fancy charts or graphics.
On a separate issue, I'm surprised no one discusses the public policy arguments in favor of credit scoring.
Insurers, of course, aren't in the public policy business. From the insurer's standpoint, it's simply a matter of actuarial validity. Arguments that insurers are "punishing" people with bad credit reflect a lack of understanding on the part of consumers.
However, there is, to me, an incidental societal benefit to providing an additional incentive for people to: 1) PAY THEIR BILLS AS THEY COME DUE; and 2) monitor their credit history periodically for accuracy.
Posted by John | December 5, 2006 4:08 PM
Posted on December 5, 2006 16:08