Lets Give Credit Where Credit Is Due

The link between credit score and loss propensity is often found to be stronger than most of the variables traditionally used in auto insurance, like driving record, age, miles driven, or type of vehicle.

Most people have good credit, so using the credit score lowers insurance costs for about two-thirds of consumers.

For consumers with poor credit, its use increases availability of coverage by allowing the insurer to charge a price that reflects loss potential.

It is almost impossible for regulators to ban the use of credit information.

Those of us whose work involves government affairs like to say that legislators can repeal a lot of laws, but not the laws of economics. If credit is indeed a strong predictor of loss experience, a free market will find a way to use the information.

Assume for a moment that regulators ban or restrict insurers ability to use credit information in underwriting and rating. Certainly, insurers will comply with the law, and if forced to stop using credit, will do so. But what will stop banks from mailing solicitations for low auto rates from their allied insurer to their gold cardholders or preferred mortgagors?

A bank doing this would not be underwriting or rating based on credit, but the effect would be the same. It defies common sense to believe that a competitive market will ignore such powerful information and that regulators can effectively control it.

Using credit does not increase overall rate levels at the expense of the consumer.

Ratemaking is a two-part exercise. Overall rate levels are set to reflect overall costs. Then the costs are distributed among the population of potential insureds based on predictive rating variables. The use of credit–or any rating variable–affects an individual customers premium, but not the total amount of revenue collected.

Despite the calls from some critics for an explanation of "why" scoring works, such an explanation is not necessary for its use to be valid and fair in risk classification. This is codified in Actuarial Standard of Practice 12, Section 5.2, which clearly states that we do not need to show "causality" for a rating variable to be appropriate.

Using credit information does not increase overall premium levels, distributes costs more fairly, lowers rates for most consumers, makes it easier for those consumers with poor credit to get coverage (by pricing them at an appropriate rate), and is consistent with generally accepted actuarial standards for setting rates.

Beyond that, even ignoring actuarial precepts and statistical evidence for a moment, the notion that poor credit makes one a poorer insurance risk is pretty easy to understand.

Some would argue, for example, that people who are responsible about managing their finances are probably responsible behind the wheel. More importantly, so much of the insurance claims process depends on the propensity of a claimant to demand compensation for an event. After all, the injured party self-reports how bad his or her neck hurts after an accident. Is it that hard to imagine that someone threatened by bill collectors might be seduced into filing or inflating a claim?

So why have so many consumer advocates vehemently opposed the use of credit information in insurance underwriting and pricing?

First, not all consumer advocates represent all consumers. As with any group of advocates, some represent various subsets of the population.

One should not assume that because someone calls himself or herself a "consumer advocate" that the individual represents the views of all, or even most, consumers. We have a diverse population, and the adage that the "squeaky wheel gets the grease" needs to be remembered. It is understandable that the minority of people adversely affected by credit scoring will complain the loudest and draw the lions share of attention.

Second, public views of the insurance industry generally range from skeptical to downright hostile. This negative attitude is partly the fault of insurers, which have historically done a poor job of promoting their value to society, and partly an inherent problem with a product that requires a regular payment for an infrequent "reward" (claim payment).

Third, people allege all sorts of things about the use of credit without strong supporting data. Examples are claims of discrimination against low-income and minority consumers, and allegations that the practice is meant to fatten insurer profits.

Fourth, the press has generally been very hostile to the use of credit. Anecdotal examples of consumers with clean driving records whose premiums have ballooned due to poor credit history make great news stories.

However, stories about the millions of consumers–the silent majority–who have benefited from the use of scoring are less likely to sell newspapers. Thus, press coverage has naturally gravitated towards human-interest stories on the problems some consumers face with their credit, and ignored the reality that most consumers are better off by its use.

Fifth, many of these anecdotal examples involve errors on credit reports.

Of course, given the millions of credit transactions filtered through credit systems daily, some erroneous information slips through, but at a tiny fraction of the rate for motor vehicle records. A recent Insurance Research Council study showed that 22 percent of MVR records were inaccurate in key states. By comparison to some other insurance rating variables, the accuracy of credit reports is exemplary.

Finally, and perhaps most disappointing to insurance executives, is that some insurance agents have loudly opposed the use of credit information.

The opposition probably has much to do with the way credit was previously used. Since insurers in the past lacked a way to explicitly rate for poor credit, risks with poor credit became candidates for close scrutiny by underwriters.

But the same thing would happen if classification plans lacked a charge for, say, youthful operators. If the insurer knew that youthful operators got into more accidents than average, yet had to charge them the same rate as adult drivers, the insurer would expect a bad loss ratio on youthful operators. The logical response would be to adjust its underwriting rules to make it more difficult to accept youthful operators.

However, once the companys actuaries implemented a proper classification and rating system, the insurer would expect the same loss ratio on drivers of all ages, and would tell its underwriters to ignore this factor in their decision making.

Insurers are now working toward building credit information into their classification and rating systems in the same way they have age, driving record, or other variables shown to predict loss potential. Once this happens, insurers should become indifferent to accepting risks with varying credit histories, opening up markets for agents, and (hopefully) earning their enthusiastic support for the practice.

Insurers bear a lot of the blame for the credit controversy due to numerous blunders that were made in the early implementation of credit scoring. These include:

A failure to clearly explain to the public why using credit is good for consumers.

A failure to implement credit programs in a way that minimizes the disruptive effects on existing customers.

A failure to make reasonable exceptions for hardship cases, such as erroneous reports or severe medical collection records.

A failure to periodically reevaluate a customers credit status.

A failure to inform consumers about how to improve their credit status and therefore insurance rates.

Insurers clearly need to do a better job of addressing the legitimate issues noted earlier regarding the use of credit in insurance. For example, systems must contain provisions to recognize specifically defined extenuating circumstances, and override the standard score-based rating algorithm. However, insurers must continue to fight to use all available and valid information to properly classify risks.

Incorporating credit into rating and underwriting plans will improve availability of coverage and lower insurance costs for the majority of consumers without increasing overall rates–a win-win situation.

Winning the public policy debate on this issue will require insurers to think outside the box and make common sense adjustments to address the legitimate concerns of regulators and consumers. If the industry prevails, it will have done a great service to consumers, even if some of their self-appointed advocates disagree.

Rade T. Musulin is vice president-operations, public affairs and reinsurance for the Florida Farm Bureau Insurance Companies in Gainesville, Fla.

Reproduced from National Underwriter Property & Casualty/Risk & Benefits Management Edition, August 26, 2002. Copyright 2002 by The National Underwriter Company in the serial publication. All rights reserved.Copyright in this article as an independent work may be held by the author.