The collapse of the housing market and the severe downturn inthe economy could reignite insurance scoring battles in statelegislatures and in Congress, according to consumer groups, whichexpect lawmakers to aggressively take another look at the ratingfactor in the wake of the subprime mortgage crisis and subsequentcredit crunch. With millions seeing their credit standingthreatened and many at risk of foreclosure, penalizing them furtherwith higher insurance rates would be unfair, these groupscontend.

Legislative and regulatory disagreements regarding insurer useof credit information in determining homeowners and auto insurancerates are certainly not new--and in recent years, the controversyeven seemed to be dying down.

The battles over the issue between the insurance industry andconsumer groups had been particularly heated throughout 2002 and2003, as buyer advocates contended, among other arguments, thatinsurance scores adversely impact minorities and the poor, whileinsurers defended the rating tool as a race-blind,actuarially-sound way of measuring risk. (See related story, page14.)

By the end of 2004, most states had settled on adopting some orall components of the 2002 National Conference of InsuranceLegislators Model Act, which offered guidelines on how creditinformation should be used by carriers, and called for enoughconsumer protections to satisfy many concerned lawmakers.

Since then, bills seeking a complete ban on the use of insurancescoring have continued to sporadically pop up in legislativecommittees across the country, but insurer associations, armed witha well-honed message and reports from the Federal Trade Commissionand various state insurance departments, were increasinglyconfident about their ability to defeat the measures (seeaccompanying story).

Now, however, consumer advocates are vowing to press the issuewith more intensity given recent changes in the economy. BirnyBirnbaum, executive director of the Center for Economic Justice inAustin, Texas, described what he sees as the challenges forconsumers in today's environment.

"One of the things that's happening is because of the subprime[mortgage] crisis and the huge increase in the number offoreclosures, and the amount of financial stress on consumersthrough no fault of their own--through reckless and abusive lendingpractices--millions of consumers are going to see higher auto andhomeowners insurance rates because of credit scoring, even thoughclaims aren't increasing and even though consumers didn't really doanything wrong," he said.

These factors will likely lead state legislatures to view theissue with more scrutiny, according to J. Robert Hunter, directorof insurance for the Consumer Federation of America inWashington.

"I expect that if we do go into a recession, there'll be a hugepickup [in legislative activity]," he predicted. "How can youadversely impact all those people who may lose their jobs, or theirhomes [because of] foreclosures? You certainly shouldn't becharging them for what may be beyond their control."

Of course, any activity in the states may be rendered moot ifthe federal government decides to take action. A new bill, HR 5633,has been introduced in the House of Representatives that would barthe use of credit information where there is a government findingof discrimination, or if the credit information serves as a proxyfor race or ethnicity. (See this week's upfront "Top Stories"section for details.)

Also on the federal stage, Mr. Hunter said he will be workingwith Congress to perhaps get a second FTC study commissioned toonce again evaluate the fairness of credit scoring.

But barring significant federal action, insurers and consumergroups will continue to fight the information war on insurancescoring state-by-state. "Many, many states are going to bebattlegrounds," said Mr. Birnbaum, noting that not all legislatureswill be in session this year, and predicting that 2009 could seeeven more activity.

In addition, there might be more driving this renewed interestamong lawmakers than concerns about the credit market and theeconomy, some insurer groups said.

Neil Alldredge, vice president of state and regulatory affairsfor the National Association of Mutual Insurance Companies, hasanother explanation for what he sees as a noticeable increase ininsurance scoring legislative activity in the states: "I think alot of it has to do with the fact that we had a lot of change inthe makeup of state legislatures following the 2006 elections," hesaid, noting that 17 states have seen a change in the majority ofone or both legislative chambers.

He explained that many of these states had already passedinsurance scoring legislation consistent with the NCOIL model, butnow legislators who weren't able to address or throw significantweight behind ban proposals are revisiting the enacted laws.

"In many of the states where we're seeing activity, you havelegislators who were either not in the legislature, or were in theminority when their state enacted the law they have now...and so Ithink those folks now are having sort of their bite at the apple onthe issue," according to Mr. Alldredge. And while the level ofactivity is not what it was prior to NCOIL's model, he said it ishigher than in recent years.

But not all in the industry share that assessment. David Snyder,vice president and assistant general counsel for the AmericanInsurance Association, said, "I think we're seeing about the levelof activity we've seen in prior years. It doesn't go away, but it'snot unexpected." He added, "It's not an unusual volume of activity,and there are thousands of bills introduced every year that don'tgo anywhere, and so we don't see anything particularlyextraordinary."

Alex Hageli, manager of personal lines for the Property CasualtyInsurers Association of America, concurred with Mr. Snyder's view.He agreed both that activity is dying down, rather than picking up,and that, for the most part, the ban bills that are beingintroduced have little chance of passing. "Every year, there are anumber of ban bills introduced," he said. "A lot of them just sitin committee and don't go anywhere. But in some states, they move alittle bit."

Regarding why bills are continually introduced every year, Mr.Hageli said, "I think what usually happens with the automaticintroductions is that you have one or two members [of alegislature] who just flat out think [insurance scoring] is unfairto people. They think poor people necessarily have lower creditscores or insurance scores, which is not the case. And then fromthat basis they just assume that regardless of whether or notcompanies can justify its use, it should be banned."

While insurers may disagree on current and future legislativeappetite regarding insurance scoring, they all are confident intheir ability to defeat these measures as they arise. For one, Mr.Snyder said, the issue is not being driven by consumerdissatisfaction. He said insurance departments are not receiving ahigh level of complaints, even though, by law, consumers arenotified when adverse action is taken against them because ofcredit information.

Furthermore, Jeffrey Junkas, an AIA spokesman, said the one timea measure to ban insurance scoring was brought directly beforeconsumers for a vote--in Oregon in November 2006--it was defeatedby more than a two-to-one margin.

Mr. Snyder said while banning the use of credit information hasits appeal on an ideological level for some politicians, "it's justnot objectively an issue that is being raised by real, liveconsumers."

Insurers have also been emboldened over the years by studiesthat, they say, justify the use of credit information indetermining rates. Industry representatives have continuallypointed to these studies when testifying before legislators, andthe lawmakers, by and large, have accepted the findings assufficient evidence to allow the use of insurance scoring.

Perhaps the most cited of these studies is one conducted by theFTC, released in July 2007. On the effectiveness of insurancescores in evaluating risks, the reports states: "Credit-basedinsurance scores are effective predictors of risk under automobilepolicies. They are predictive of the number of claims consumersfile and the total cost of those claims....Thus, on average,higher-risk consumers will pay higher premiums and lower-riskconsumers will pay lower premiums."

This finding is virtually word-for-word what insurerrepresentatives have argued in the state legislatures, and itdovetails with the industry's own studies. Mr. Hageli said that a2003 study by Epic Actuaries, which is now a part of Tillinghast,concluded that "individuals with the lowest insurance scores werefound to incur 33 percent higher losses than average, while thosewith the highest scores incurred 19 percent lower losses thanaverage. And at 33 percent higher, that's considered one of themost accurate actuarial tools [for insurers]."

While it cannot be denied that there is a correlation betweencredit scores and insurance risk, the unanswered question asked bylegislators and opponents of the practice is, why? Even consumergroups do not doubt the existence of a correlation, but ratherobject to the absence of a plausible basis for the classificationscheme.

"For example, a plausible basis might be number of accidents foran auto policyholder," Mr. Hunter explained. "If you have moreaccidents, the thesis is that you're demonstrating an inability tobe a good driver....Then you test [the thesis] with data and youfind out that's true."

Lately, though, Mr. Hunter said the industry has begun usingdata-mining methods to find correlations that are not based on athesis. "In recent years, with all this data, people are coming upwith...correlations and saying, 'Oh, look what we found.'Obviously, you may find a correlation in certain data, but it oughtto be testing a thesis so that, for one thing, people canunderstand how to avoid the risk, and it should be risk-relatedsomehow," he said.

"If you talk to the industry about that, they're going to say,'Hunter's wrong--in the actuarial standards, you don't have to havea cause and effect relationship,'" he added. "I'm not saying a'cause and effect relationship.' Cause and effect doesn'thappen...I'm talking about a logical relationship."

Industry representatives are less interested in the why. "Wecan't explain it," Mr. Hageli admitted. "We don't know why thecorrelation exists, but it does."

Insurers have seen that insurance scoring can predict risk, andpredict it well--and that, essentially, is where the argument endsfor them. NAMIC's Mr. Alldredge summed this up while speakinggenerally about the effectiveness of using credit to predictrisk.

"Most studies have shown that, for some reason, insurancescoring tends to measure variables and factors that otherunderwriting tools don't catch," he said. Without insurancescoring, he added, "you do not get the same accuracy in terms ofpredicting the loss."

Mr. Hageli said this greater degree of accuracy has allowed fora "substantial increase of rating categories from prior years."Previously, he said, insurers had a small number of rating tiers,but "with insurance scoring, it's allowed for a much more precisemeasure of risk, which allows for much greater stratification ofrisk over many more tiers than what you previously had." Somecompanies, he noted, have gone from three-to-200 rating tiers, andthis has "allowed for much greater customization of pricing."

Up to this point, evidence presented by the industrydemonstrating that insurance scoring does predict risk--evenwithout a plausible basis for the correlation--has swayed themajority of legislators to leave the practice alone.

PCI's Mr. Hageli said that "when we go into these committeehearings and we show them the evidence and the studies, like theFTC study, that demonstrate that [insurance scoring] is actuariallyjustified--that for whatever reason insurance scoring is predictiveof future risk, and that shouldn't people who represent a greaterrisk pay more for their insurance--that's usually the clincher forconvincing people to continue to allow insurance scoring."

Mr. Birnbaum sees a different reason legislators are acceptingthe industry's arguments--threat-mongering. "I think the mainargument that is swaying legislators is insurer threat--that allthese consumers are going to get rate increases if you ban creditscoring."

He called this a "false threat," noting that insurers would notraise the rates on customers they consider most favorable "becausethere are going to be any number of insurance companies that wouldjump to get these so-called best and least-risky customers."

As the insurers and consumer groups continue to make theirarguments to state legislators, to Congress and even in the courts,industry representatives remain confident in their message, even ifsome express concern about increased legislative activity.

"A couple of years ago, I predicted that we'd seen the highwater mark on the issue, and I thought the issue would die down,"said NAMIC's Mr. Alldredge. "I turned out to be wrong about thatone, [but] I feel pretty confident about the industry's ability todefend the use [of insurance scores] in state legislatures."

For his part, Mr. Hunter has vowed to continue to make his caseto legislators as well. "There are times when we've had somesuccess, and there are times when we get beat," he said, "but we'renot going away. It's not like it's going to be over some day. It'snot over until it's gone."