A recent opinion by the Texas Supreme Court (Ojo v. Farmers Group, Inc. No. 10-0245, issued May 27, 2011) has illuminated the crux of the conflict between state regulation of insurance and broader policy objectives of the federal government. Two laudable goals—Texas' desire to maintain a fair and equitable insurance market and the federal government's interest in preventing racial discrimination in housing—have collided. Which will yield? The Texas Supreme Court has said that, under the McCarran-Ferguson Act and Texas law, the state's regulation of insurance takes priority over the federal government's civil rights laws.
The Texas Supreme Court was responding to a certified question from the U.S. Court of Appeals for the Ninth Circuit, which was considering an appeal from a U.S. District Court ruling in a class action brought against Farmers Group for its use of a credit-scoring system. The class action was brought by an African-American resident of Texas on behalf of himself and other racial minorities whose homeowners insurance premiums were increased because of low credit scores. The plaintiffs claimed that the use of credit scoring had a disparate effect on minorities in violation of the federal Fair Housing Act (FHA), which prohibits discrimination "against any person in the terms, conditions, or privileges of sale or rental of a dwelling, or in the provision of services or facilities in connection therewith, because of race." 42 U.S.C. § 3604(b).
The District Court dismissed the case, on the ground that the FHA was reverse-preempted by the Texas Insurance Code under the federal McCarran-Ferguson Act (MFA), 15 U.S.C. §§ 1011 - 1015. On appeal, a divided three-judge panel of the Ninth Circuit reversed the District Court, but the case currently is under reconsideration en banc. Ojo v. Farmers Group, Inc., 600 F.3d 12 01 (9th Cir. 2 010). The Circuit Court certified to the Texas Supreme Court the question of whether Texas law permits an insurance company to use credit scoring even if it has a racially disparate impact. The Texas Supreme Court answered that "Texas law prohibits the use of race-based credit scoring, but permits race-neutral credit scoring even if it has a racially disparate impact."
The Texas Supreme Court cited the language in the MFA that "No Act of Congress shall be construed to invalidate, impair, or supersede any law enacted by any State for the purpose of regulating the business of insurance, . . . unless such Act specifically relates to the business of insurance." 15 U.S.C. § 1012(b). Since the Ninth Circuit had already determined that the FHA does not specifically relate to insurance and that the relevant Texas law was enacted for the purpose of regulating insurance, the only question for the Texas Supreme Court was whether the plaintiffs' claim under the FHA might "invalidate, impair, or supersede" Texas law.1 The court answered that question in the affirmative.
The court pointed out that the Texas Insurance Code prohibits "unfair discrimination," which, among other things, says that an insurer may not charge an individual a rate different from the rate charged to another individual because of that person's race, color, religion or national origin. The same provision (Tex. Ins. Code § 544.002(a)(2)) provides that it is not a violation of the prohibition against unfair discrimination if the refusal, limitation or charge is required or authorized by law or regulatory mandate. The court reasoned that since the statute prohibits charging a person a different rate for the same coverage because of the individual's race, the race-neutral application of credit scoring would accord with Texas law. Indeed, it would violate Texas law to treat persons with low credit scores more favorably than they otherwise would be treated because they belonged to a particular racial group.
Based on the answer provided by the unanimous Texas Supreme Court to the Ninth Circuit, it is likely that the District Court will be upheld and the case dismissed. Such a ruling would be in line with a similar case from the Eighth Circuit, also involving Farmers, Saunders v. Farmers Insurance Exchange, 537 F.3d 961 (8th Cir. 2008). In that case, the court upheld the dismissal of a class action claiming that residents of inner-city areas of Kansas City, MO—predominantly racial and ethnic minorities—were charged more for homeowners insurance than residents in suburban areas that are largely Caucasian. The court cited the reverse-preemption impact of the MFA, as well as the filed rate doctrine, which says that an insurer charging a rate that has been approved by the regulator cannot be sued for using that rate.
But not all courts that have considered this question have reached the same conclusion. The Fourth Circuit2, Fifth Circuit3, Sixth Circuit4, Seventh Circuit5 and the Eleventh Circuit6, all found in similar circumstances that the MFA did not reverse-preempt state insurance laws. As the Eleventh Circuit explained, because the applicable federal civil rights statutes did not "frustrate any declared state policy or interfere with the state's administrative regime in the insurance context," the MFA's reverse preemption did not apply. Moore, 267 F.3d at 1221-23. These courts found, in essence, that since state laws, including insurance laws, prohibited discrimination on racial grounds, the state laws were in harmony, not in conflict, with the federal civil rights statute at issue.
Underlying these cases is the law enacted in every U.S. jurisdiction that requires the state insurance regulator to ensure that rates charged for insurance "are not excessive, inadequate or unfairly discriminatory." This language comes from the model rating law of the National Association of Insurance Commissioners (NAIC). "Not unfairly discriminatory" in insurance terms means that persons are not charged different rates for the same risks or the same rates for different risks. If the risks are higher—because the house is made of wood instead of brick, or because multiple losses were filed in the past several years, or because the insured has a low credit rating (all factors that have been empirically demonstrated to be predictive of higher losses in the future)—then the insurer must charge higher rates to everyone who falls in that category. The fact that such uniform and race-blind application of rates from these rating categories might disparately impact racial minorities should be irrelevant. The words "unfairly discriminatory" in insurance, then, means something very different than "discrimination" in federal civil rights statutes, such as the FHA.
Note, however, that rating categories are themselves artificial constructs, designed by insurance companies to help them predict future losses and to charge appropriately for the anticipated risks. How those categories are defined, then, can determine what rates will be charged, regardless of the actual loss experience (if any) of the individual insured. That is because insurers charge rates based on expected future losses, not past claims. An insured's claims history can be relevant, of course, to predicting future claims, but many other factors can be predictive also. People accept some rating categories as reasonable (although there are always objectors). Teenage drivers, for example, are charged substantially higher rates for automobile insurance than older drivers because they have far more accidents—even though a particular 18-year-old may never have had an accident. But other rating categories are seen as unfair even though they may be just as predictive. For example, because women live longer than men should they therefore pay less for life insurance? Young women who drive have far fewer accidents than young men—should they be charged less for their auto insurance? Women have higher health expenses. So do older persons. Should they pay more for health insurance?
These are not easy questions to answer. Public policy determinations can and have been used to nullify the results of carefully analyzed and very accurate actuarial calculations. Insurance, after all, involves the spreading of risk. Inevitably that means that some people who are never going to file claims will pay for those who do. Some risk characteristics are outside of our control—age is one, so is gender and so is race. Other risk characteristics are at least theoretically under our control—obesity, for example, or smoking, or other kinds of "lifestyle" choices. Credit score is another, since the score is largely determined by whether the individual is scrupulous about paying incurred debt. Where we choose to build our house, and how strongly we build it, is also within our control, as is the kind of car we buy and how we drive it. Which of these risk categories should insurers be allowed to use to determine rates and which should be prohibited or homogenized by so-called community rating?
At present these choices are left largely up to insurance companies and to the state officials who regulate them. This means that overarching federal policy goals, such as equal treatment in access to housing, may be subordinated to the economic requirements of a vibrant and competitive insurance market. Much of the competitive success of an insurer is based on its ability to determine—and to charge appropriately for—the risk of loss inherent in the policies it sells. Competition among insurers demonstrably results in lower insurance rates for most persons. Insurance regulators know that and therefore allow wide latitude for companies to develop risk categories, as long as those categories are not explicitly based on prohibited factors, such as race and sometimes gender or even age.
But if the use of permitted factors has a disparate impact on a particular racial group, should such factors be prohibited? Or must rating factors used by insurers always reflect the exact racial makeup of their customers—a practical impossibility, since insurers rarely know explicitly the race of their customers? Should insurers be required to charge the same rate to all persons purchasing a particular product, such as a homeowners policy or an auto policy, regardless of the risk that any particular applicant or those like him presents? Should low-risk persons be charged more for their insurance so that higher-risk persons can pay less for theirs? Insurers always answer that question "no," arguing that persons should pay premiums based on the risk of loss that they represent. But of course no system will ensure complete fairness. The careful 18-year-old male driver still will be charged more because of the reckless behavior of his peers in the same rating group.
These fundamental questions about the nature of insurance are at the root of the dispute over whether MFA means that federal civil rights laws must yield to state insurance regulatory decisions. The use of credit scoring is but one example of where the ability of insurance companies to determine risk may result in unfair treatment of persons who fall into a rating category that may not (or may) be adequate predictors of their individual risk of loss.