Late in June, in Texas Dept. of Housing v. Inclusive Communities, No. 13–1371 (U.S. June 25, 2015), the U.S. Supreme Court ended years of debate by embracing a “disparate impact” claim against a housing authority under the federal Fair Housing Act (“FHA”). For insurers, that case was just one of a series of battles over what kinds of information insurers must consider when making basic underwriting decisions. In 2014, the industry enjoyed some success in challenges to a regulation that exposes them to disparate impact liability if they fail to take account of certain demographic information. But in Viens v. America Empire Surplus Lines Ins. Co., No. 3:14cv952 (D. Conn. June 23, 2015), and in Jones v. Travelers Cas. Ins. Co. of Am., No. CV 5:13-02390 (N.D. Cal. May 7 2015), federal courts ruled that an insurer may be liable under the FHA, if its decisions “predictably will cause” someone else to commit an action with a disparate impact on protected classes—and that this is so, even if the second-order action is also non-discriminatory on its face. Because the issues these cases present appear to be a long way from being settled, now would be a good time for insurers to review their underwriting guidelines and data protocols for potential exposure to claims of this kind.

This article will discuss the legal and procedural background of the Viens and Jones cases. A second article will discuss the issues these cases raise about the intended scope of the FHA. A final article will discuss the more fundamental issues they raise about the application of federal law to underwriting decisions, and it will suggest steps that prudent insurers should be taking.

Fair Housing Laws

Following the assassination of Dr. Martin Luther King, Jr., Congress enacted the Fair Housing Act, 42 U.S.C §§ 3601 et seq., to outlaw housing discrimination on the basis of “race, color, religion, sex, familial status, or national origin.” Specific provisions of the law prohibit: “mak[ing] [a dwelling] unavailable” to any person “because of” membership in a protected class (§ 3604(a)); “discriminat[ion]” against protected classes in the “sale or rental” of a dwelling, “or in the provision of services … in connection therewith” (§ 3604(b)); and discrimination in making available “residential real estate-related transactions,” or “in the terms or conditions of such a transaction” (§ 3605). The FHA also makes it unlawful to “coerce, intimidate, threaten, or interfere with any person” in the exercise of housing rights, or “on account of” the person’s having “aided or encouraged” another to exercise rights granted by the FHA. (§ 3617.) Similar provisions separately prohibit discrimination “because of a handicap.” (§ 3604(f).)

Any person who either claims to have been injured or believes she will be injured by a violation of the FHA may bring an action for compensatory and punitive damages, or for injunctive relief, in state or federal court. (§§ 3602, 3613.) Alternatively, an “aggrieved person” may seek relief in an administrative proceeding before the U.S. Department of Housing and Urban Development (“HUD”). (§§ 3610-12.)

State fair housing laws are generally modeled on the federal statute, but they can be broader in important respects. Some, for example, prohibit additional forms of discrimination—including discrimination on the basis of “source of income,” which can target participants in government anti-poverty programs or recipients of government housing subsidies. The Connecticut and California statutes at issue in Viens and Jones both bar discrimination of all the types that are covered by the FHA, but Connecticut also prohibits distinctions based on creed, ancestry, gender identity or expression, lawful source of income and marital status (C.G.S.A. § 46a-64c), while the California law adds protections based on gender, gender identity, gender expression, sexual orientation, marital status, ancestry, familial status, source of income—and even genetic information. (Cal. Gov. Code § 12955.) In California, the definition of these categories “includes a perception that [a] person has any of those characteristics or that the person is associated with a person who has, or is perceived to have, any of those characteristics.” (Id.)

Fair Housing Claims Against Insurers

In Mackey v. Nationwide Ins., 724 F.2d 419 (4th Cir. 1984), the U.S. Court of Appeals for the Fourth Circuit held that discriminatory practices in the sale of homeowners’ insurance do not violate the FHA. A few years later, however, HUD adopted a regulation that prohibits discrimination in connection with “property or hazard insurance.” 24 C.F.R. § 100.70(d)(4). Since that time, and in deference to that regulation, courts (including some within the Fourth Circuit) have generally held that property and liability insurance are “services” connected to the sale or rental of a dwelling (under FHA § 3604(b)), and that withholding such insurance makes the dwelling “unavailable” (under FHA § 3604(a).) E.g., N.A.A.C.P. v. American Family Mut. Ins. Co., 978 F.2d 287 (7th Cir. 1992) (“NAACP“); Fuller v. Teachers Ins. Co., No. 5:06-CV-00438-F (E.D.N.C., Sept. 19, 2007); National Fair Housing Alliance v. Prudential Ins. Co., 208 F.Supp.2d 46, 57 (D.D.C. 2002) (“NFHA“); Strange v. Nationwide Mut. Ins. Co., 867 F.Supp. 1209 (E.D. Pa. 1994).

There is also authority holding that the sale of insurance can be a “real estate-related transaction,” within the meaning of FHA § 3605, but courts remain divided over that issue. Compare NAACP, 978 F.2d at 297 (“It would strain language past the breaking point” to apply § 3605 to the sale of insurance), with NFHA, 208 F.Supp.2d at 58 (holding that insurance redlining violates § 3605). Additionally, there is limited authority for the proposition that insurance practices can support an “interference” claim under § 3617. E.g. Nevels v. Western World Ins. Co., Inc., 359 F.Supp.2d 1110, 1122 (W.D. Wash. 2004) (“Nevels“).

Disparate Treatment and Disparate Impact

Both state and federal statutes expressly prohibit disparate treatment—i.e., intentional acts of discrimination, based on the characteristics listed in the statute. For many years, most federal courts have also recognized claims under the FHA based on disparate impact—actions that do not make prohibited distinctions, and which may not have been undertaken for discriminatory reasons, but which nonetheless have a “disproportionately adverse effect” on the housing rights of protected class members. Some state laws, such as California’s Fair Employment and Housing Act (“FEHA”), incorporate a disparate impact theory expressly. See Cal. Gov. Code § 12955.8(b).

In February 2013, HUD proposed a rule “to prohibit housing practices with a discriminatory effect.” 78 Fed. Reg. 11482. As applied to insurers, certain aspects of this “Discriminatory Effects Rule” were successfully challenged in two cases that are discussed below. But the basic structure of the rule was confirmed by last month’s Supreme Court decision in Inclusive Communities, which held definitively that the FHA prohibits “housing decisions with a disparate impact.”

The Duty To See The Future

HUD’s rule creates a burden-shifting framework for disparate impact claims. One way to look at that framework is as an elaboration of a new legal duty: a duty to consider the future effects of each person’s actions on the housing rights of others. That duty does not require a defendant to advance the rights of others at a cost to its own legitimate business interests, but it can impose an affirmative obligation to choose the least “discriminatory” among possible courses of action.

The duty is announced in the first part of the framework: in a case alleging that a defendant has violated the FHA, plaintiffs have the initial burden of showing that the defendant’s action “caused or predictably will cause a discriminatory effect.”   24 CFR § 100.500(c). In other words, a plaintiff can state a claim for “discrimination” under the FHA by showing the future effect of a defendant’s conduct on the housing rights of protected classes, regardless of the defendant’s intent.  In theory, at least, the plaintiff need show nothing more than that the defendant has overlooked or misjudged the future consequences for housing of conduct that is otherwise lawful.”

The second part of the framework declares that the duty is not absolute: a defendant must consider the housing rights of protected classes, but it is not required to place their interests before its own. Thus, if the plaintiff makes the required showing, the defendant can shift the burden back by “prov[ing] that the challenged practice is necessary to achieve one or more [of the defendant’s] substantial, legitimate, nondiscriminatory interests.” Id.

The third and final element, however, potentially extends the duty into a requirement to choose specific business options. Even if the defendant carries its burden of proving that its decision was necessary to achieve a legitimate goal, the plaintiff can still prevail, if it proves that the defendant’s interests “could be served by another practice that has a less discriminatory effect.” Id.

Thus, to defend a disparate impact claim, defendants must justify not only their motives (by demonstrating that their acts had a legitimate business goal), but also their business judgment (by proving that the approach they chose is objectively the least discriminatory one).

Disparate Impact Claims Against Insurers

Most cases against insurers have involved alleged acts of disparate treatment—intentional discrimination, such as “redlining” in the sale of property insurance to homeowners. E.g., NAACP, supra. Claims against insurers based on disparate impact are more controversial, but they have previously been accepted in some cases. E.g. Dehoyos v. Allstate, 345 F.3d 290 (5th Cir. 2003) (“Dehoyos“) (credit scoring allegedly had a disparate impact on customers for homeowners insurance who were members of protected classes); NFHA, supra (underwriting criteria for sale of individual homeowners policies allegedly had disparate impact).

HUD first proposed its Discriminatory Effects Rule in November 2011, and, on its face, the rule applied to the sale of “property or hazard insurance.” It drew strong opposition from the insurance industry, which argued that the disparate impact theory should not be applied to “risk-based business practices.” Despite that opposition, the rule was promulgated without substantial changes in February 2013. 78 Fed. Reg. 11482.

Industry groups challenged certain aspects of the new rule in two separate cases: Property Casualty Insurers Assoc. of Am. v. Donovan, No. 1:13-cv-08654 (N.D. Ill. Sept. 3, 2014) (“PCIA“), and American Ins. Assoc. v. U.S. Dep’t of Housing & Urban Dev., No. 13-00966 (D.D.C. Nov. 7, 2014) (“AIA“). Both challenges were at least partially successful, but neither of them resolved any of the larger disputes over disparate impact claims against insurers.

In PCIA, the court held that HUD’s adoption of the rule had been “arbitrary and capricious,” in violation of the Administrative Procedure Act (5 U.S.C. §§ 551 et seq.), because the agency had not given adequate consideration to certain arguments advanced by the insurance industry: that applying the rule to insurers would violate the McCarran-Ferguson Act, 15 U.S.C. § 1012; that it would violate the “filed rate” doctrine; and that it was inconsistent with the “fundamental nature of insurance.” (These arguments will be discussed in part 3 of this series.) The case was remanded to HUD for further proceedings on those issues. The court’s decision means that HUD’s regulation will not receive deference with respect to the issues the court identified, but this result does not prevent other courts (such as the courts in Viens and Jones) from deciding those issues in the same way that HUD did.

In AIA, the court held that the FHA does not authorize any claims based on disparate impact, and so that HUD had exceeded its authority by issuing the Discriminatory Effects Rule. Seven months later, that ruling was superseded by the Supreme Court’s decision in Inclusive Communities.

Thus, even after two successful challenges to the HUD rule, disparate impact claims against property insurers remained viable—enough so to survive a motion to dismiss in Viens and a motion for summary judgment in Jones.

A New Breed of Housing Claim

The complaints in Viens and Jones extend disparate impact theories into new territory.

   Viens v. American Empire

Section 8 of the Housing Act of 1937 (42 U.S.C. § 1437f) authorizes housing “assistance payments” to low-income families, which are distributed through programs operated by HUD. The plaintiffs in Viens include two building owners in Connecticut who rent apartments to participants in the Section 8 Housing Choice Voucher program. (The other plaintiff is a civil rights organization dedicated to promoting equal housing opportunities.) Under that program, tenants make payments towards their rent in an amount that is capped at a portion of their incomes; using funds from the federal government, the program pays the remaining amount of rent directly to landlords.

Both of the landlord plaintiffs were insured under policies issued by American Empire. In 2013 and 2014, one policy was canceled and the other nonrenewed, allegedly because the insurer’s underwriting guidelines excluded properties in which tenants receive housing assistance. The landlords were then forced to purchase substitute policies that offered less favorable terms at a higher price.

The Connecticut Fair Housing Act, C.G.S.A. §§ 46a-63 et seq. (“CFHA”), prohibits discrimination on the basis of “lawful source of income”; the federal FHA does not. Thus, while the plaintiffs asserted a disparate treatment claim under the CFHA, they could not do so under the FHA. Instead, plaintiffs alleged that the defendant’s underwriting guidelines “predictably will cause” landlords to make fewer units available to participants in the Section 8 program. They also cited evidence that African-American and Hispanic households in Connecticut are, respectively, 12.9 and 12.4 times more likely than White, non-Hispanic households to participate in that program. Consequently, they alleged that the burden of a reduction in the availability of Section 8 housing will fall disproportionately on members of protected classes.

Viens is not the first case in which an insurer allegedly violated the FHA by influencing the actions of others. But earlier examples—Nevels, supra; Wai v. Allstate Ins. Co., 75 F.Supp.2d 1 (D.D.C. 1999)— were disparate treatment cases. The insurers in those cases allegedly refused to insure properties with disabled residents, and persons with disabilities are protected by § 3604(f) of the FHA.

Viens is also not the first disparate impact case against an insurer. But previous cases, such as Dehoyos and NFHA, asserted that the “impact” of the insurer’s decisions fell on the insurer’s own customers; they did not seek to hold the insurer responsible for the consequences of those customers’ intervening acts.

The novelty of Viens lies in the combination of these elements. The FHA claim in Viens asserts that the insurer engaged in an unlawful act of “discrimination,” by making non-discriminatory underwriting decisions that will cause someone else to act in a way that—although otherwise lawful—disproportionately affects the housing rights of protected class members. In other words, the plaintiffs in Viens assert that insurers have a duty under the FHA to consider both (i) how their underwriting decisions will affect the future behavior of customers, in those customers’ interactions with third parties, and (ii) whether those third-party interactions will produce a disparate impact.

   Jones v. Travelers

Jones was based on similar allegations and theories. The landlord plaintiffs (who sued together with a non-profit fair housing group) owned a building in California that was insured by Travelers; that policy was nonrenewed because of the residents of the building included Section 8 tenants; and the landlords ended up paying higher premiums for a replacement policy. The plaintiffs cited statistics showing that households receiving Section 8 assistance were more likely than others in the county to be headed by members of several different protected classes.

In opposing Travelers’s motion for summary judgment, the plaintiffs argued that circumstantial evidence supported a claim of disparate treatment—specifically, that Travelers’s decision not to insure subsidized housing was based on “stereotypes” about tenants who receive government assistance. But the plaintiffs also asserted that their claims could be pursued solely on disparate impact grounds—for the same reasons that were advanced in Viens.

An unusual aspect of the disparate impact claim in Jones has to do with the fact that the Section 8 program is voluntary under federal law. For this reason, courts in California have held that landlords who decline to rent to Section 8 participants are not liable under the FHA—even under a disparate impact theory. E.g., Doe v. WCP I, LLC, No. 05-438885 (Cal. Super. Ct. June 4, 2009). Plaintiffs’ claim in Jones, therefore, went beyond the one in Viens—that is, beyond the theory that an insurer can violate the FHA if its non-discriminatory decisions cause someone else to commit acts with a disparate impact on protected classes. In Jones, plaintiffs alleged that the insurer can be guilty of “discrimination,” even if the third-party actions that directly cause the alleged disparate impact were lawful under the FHA.

   Administrative Proceedings

Fair housing groups have recently asserted similar claims in recent administrative proceedings before HUD.  E.g., Brevard Neighborhood Development Coalition v. Int’l Catastrophe Ins. Managers, HUD file No. 04-14-0858-8. Complaints in those cases track the language of HUD’s Discriminatory Effects Rule, alleging that underwriting decisions based on the presence of Section 8 tenants “ha[ve] no legally sufficient justification,” and that, in any case, “there are reasonable alternatives that have no or a less discriminatory effect.”

Rulings to Date

On May 7, 2015, the U.S. District Court for the Northern District of California denied the insurer’s motion for summary judgment in Jones v. Travelers. It held that plaintiffs had proffered enough “circumstantial evidence of … discriminatory intent” to raise a triable issue of fact on the disparate treatment claim. On the issue of disparate impact, it also found sufficient evidence that the insurer’s policy “predictably falls more heavily on protected classes and results in discrimination.” The court acknowledged the voluntary nature of the Section 8 program, but it reasoned that there are “no cases where a court has held that landlords and insurers are similarly situated under Section 8.”

The Jones case will not proceed to trial, because the parties settled in late June. (The court’s views were expressed during a hearing on the motion for summary judgment; the case settled before the court could issue a written opinion.) For the same reason, the rulings in Jones will not be tested on appeal.

On June 23, 2015, the U.S. District Court for the District of Connecticut denied the defendant’s motion to dismiss the complaint in Viens v. American Empire. The court found that the plaintiffs had stated a claim under FHA §§ 3604 and 3605. Along the way, the court specifically ruled that the sale of insurance is a “real estate-related transaction,” within the meaning of § 3605; that claims under the FHA may be based on transactions that occur after the sale or rental of a dwelling; and that the plaintiffs’ claim was not “reverse preempted” by state law under the McCarran-Ferguson Act.

These rulings will be discussed in greater detail in part 2 and part 3 of this series.

Image source: CBS Television (Wikimedia)