This is the second article of a three-part series about two recent decisions by federal courts in Connecticut and California: Viens v. America Empire Surplus Lines Ins. Co., No. 3:14cv952 (D. Conn. June 23, 2015), and Jones v. Travelers Cas. Ins. Co. of Am., No. CV 5:13-02390 (N.D. Cal. May 7 2015). The first article discussed the legal and procedural background of the two cases. This article will discuss the issues these cases raise about the intended scope of the federal Fair Housing Act and the state laws that are based on it. A final article will discuss the more fundamental issues these cases raise about the application of federal law to underwriting decisions, and it will suggest steps that prudent insurers should be taking.
Discriminatory Housing Practices
The Fair Housing Act, 42 U.S.C §§ 3601 et seq. (“FHA”), prohibits certain forms of discrimination on the basis of race, color, religion, sex, familial status, national origin or disability. The FHA prohibits the following actions (among others) as “discriminatory housing practices“:
- “[M]ak[ing] [a dwelling] unavailable” to a person because the person is a member of a protected class. (42 U.S.C. § 3604(a).)
- “[D]iscriminat[ing] … in the provision of services … in connection [with]” the “sale or rental of a dwelling.” (§ 3604(b).)
- “[D]iscriminat[ing] against any person in making available” a “real estate-related transaction,” or “in theterms or conditions of such a transaction.” (§ 3605(a).)
- “[C]oerc[ion] … or interfere[nce] with any person … on account of his having aided or encouraged any other person in the exercise or enjoyment of, any right granted or protected by [the FHA].” (§ 3617.)
As discussed in part 1, there is general consensus that discrimination in the sale of property insurance can violate § 3604. See, e.g., 24 C.F.R. § 100.70(d)(4); 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“).
Some courts have also held that insurers may be liable under § 3605 or § 3617. E.g., NFHA at 58 (§ 3605); Nevels v. Western World Ins. Co., Inc., 359 F.Supp.2d 1110, 1122 (W.D. Wash. 2004) (“Nevels“) (§ 3617). But see NAACP at 297 (no claim under § 3605).
The New Disparate Impact Landscape
As a result of the Supreme Court’s June 2015 decision in Texas Dept. of Housing v. Inclusive Communities, No. 13–1371 (U.S. June 25, 2015), defendants can commit “discriminatory housing practices” without making any prohibited distinctions—by acting in a way that causes a “disproportionately adverse effect” on the housing rights of protected class members. These “disparate impact” claims must now proceed under the guidance of two recent, authoritative statements.
One is the “Discriminatory Effects Rule” issued by the U.S. Department of Housing and Urban Development (“HUD”) in 2013. Application of this regulation to insurance companies has been successfully challenged in recent litigation.Property Casualty Insurers Assoc. of Am. v. Donovan, No. 1:13-cv-08654 (N.D. Ill. Sept. 3, 2014) (“PCIA“), andAmerican Ins. Assoc. v. U.S. Dep’t of Housing & Urban Dev., No. 13-00966 (D.D.C. Nov. 7, 2014) (“AIA“). (PCIAand AIA were discussed in part 1 of this series.) Nevertheless, the outlines of the rule were later cited with approval by the majority in Inclusive Communities.
Under the rule, the FHA prohibits certain conduct that “predictably will cause a discriminatory effect.” 24 CFR § 100.500(c)(1). Thus, as discussed in part 1, the rule effectively imposes a duty on defendants, not just to refrain from intentional acts of discrimination, but also to calculate the likely consequences that their non-discriminatoryactions will have on the housing rights of protected classes.
That calculation must now take account of a second statement: Justice Kennedy’s majority opinion in Inclusive Communities. For a defendant to be liable under the FHA, the opinion emphasized that the causal connection between a defendant’s conduct and its alleged discriminatory effect must be firmly established; Justice Kennedy called this a “robust causality requirement“:
[A] disparate-impact claim that relies on a statistical disparity must fail if the plaintiff cannot point to a defendant’s policy … causing that disparity. … A robust causality requirement ensures that ‘[r]acial imbalance … does not, without more, establish a prima facie case …’ and thus protects defendants from being held liable for racial disparities they did not create. …
The court’s discussion of this requirement suggests several ways in which defendants might challenge assertions about what effects their actions “predictably will cause.” One would be a showing that the defendant’s action merely contributes to a situation that can be traced to more than one cause. Another might show that the defendant’s choices are constrained by statute or regulation:
It may … be difficult to establish causation [in a disparate impact case] because of the multiple factors that go into decisions [affecting the housing market]…. . And … if the [plaintiff] cannot show a causal connection between the [defendant’s] policy and a disparate impact—for instance, because [governing] law substantially limits the [defendant’s] discretion—that should result in dismissal … .
HUD’s Discriminatory Effects Rule also provides that disparate impact will not result in liability, if the defendant’s conduct is “necessary to achieve … substantial, legitimate, nondiscriminatory interests” that could not “be served by another practice that has a less discriminatory effect.” 24 CFR §§ 100.500(c)(2) and (3). Thus, defendants must be prepared to show both that their action had a non-discriminatory rationale, and that the rationale has a solid basis in fact. The second showing is important, because (as discussed in part 1) HUD’s regulation effectively imposes an affirmative duty to select the least discriminatory alternative among all possible approaches to a business goal. To defend a disparate impact claim, therefore, companies must be prepared to offer evidence that the conduct they adopted is the most efficacious way to pursue the relevant nondiscriminatory objective.
It is not clear that the Supreme Court has gone quite so far. Quoting Griggs v. Duke Power Co., 401 U.S. 424, 431 (1971), the majority opinion in Inclusive Communities stated that the FHA forbids housing “barriers” that are not only “unnecessary,” but also “artificial” and “arbitrary“—language that implies at least a residue of discriminatory intent. Moreover, the court admonished:
Courts should avoid interpreting disparate-impact liability to be so expansive as to inject racial considerations into every housing decision.
Clearly, facially neutral policies that are mere pretexts for discrimination are prohibited. But the court’s language suggests that defendants acting in good faith might be shown some leeway on the issues of whether their policies are “necessary,” and whether they should have chosen a different option with a “less discriminatory effect.”
Disparate Impact Claims Based on the Intervening Acts of Others
As explained in part 1, the Viens and Jones cases assert a novel theory of FHA liability. The plaintiffs in Viens do notallege either that the insurer intended to discriminate against minority tenants, or that its underwriting guidelines make distinctions that are prohibited by the FHA; they contend, rather, that the insurer’s otherwise lawful act has a discriminatory effect on members of protected classes. (The plaintiffs in Jones asserted both disparate impact anddisparate treatment.) But unlike previous disparate impact suits against insurers (such as Dehoyos v. Allstate, 345 F.3d 290 (5th Cir. 2003) (“Dehoyos“)), Viens and Jones involve claims that the alleged discriminatory effect will be felt by someone other than the insurer’s own customers—specifically, by tenants of insured landlords who receive housing assistance under the Section 8 program, and who will be displaced if the landlords choose to stop participating in that program. They allege, that is, that the insurers will produce a disparate impact through theintervening acts of their customers—acts the insurers will allegedly “cause.”
Cases of this new variety will test the “robust causality requirement” that Inclusive Communities announced shortly after Viens and Jones were decided. As a general rule, “FHA plaintiffs’ injuries must be proximately caused by the defendant’s discriminatory acts.” Pacific Shores Properties, LLC v. City of Newport Beach, 730 F.3d 1142, 1168 n.32 (9th Cir. 2013). Proximate causation is usually absent where “an intervening act of a third party … actively operates to produce harm after the first person’s wrongful act has been committed.” E.g., Egervary v. Young, 366 F.3d 238, 246 (3d Cir. 2004). The exception is the rare case in which the intervening act was a “foreseeable” consequence of the first one. E.g., S.E.C. v. Apuzzo, 689 F.3d 204, 215 (2d Cir. 2012).
In Inclusive Communities, Justice Kennedy made a point of observing that “multiple factors” go into real estate decisions. His remark was made in a different context, but it still could reasonably be applied to a landlord’s decision about whether to participate in the Section 8 program. That being the case, it might be hard to support a claim that the landlord’s decision is a “foreseeable” result of decisions about the pricing of property insurance—or, in the language of the FHA, that it is the landlord’s insurer who will make housing “unavailable” to minority tenants.
On the other hand, the insurance industry is currently investing substantial resources in systems and analyses whose express purpose is to predict how pricing structures will affect the decision-making of insureds. The term “price optimization” is generally applied to a number of different processes of this kind, and they are the subject of considerable controversy and regulatory attention. For the most part, these systems address consumers’ decisions to purchase or renew insurance policies—not decisions about whether (for example) to change their own business models by turning away certain classes of tenants. But by using these tools, insurers endorse the validity of at leastsome beliefs that attribute the behavior of policyholders directly to underwriting decisions. It is also possible that these tools will generate records that will be used as evidence in support of arguments over causation.
Prohibited Conduct Issues
Cases like Viens and Jones also raise questions about whether the FHA should be applied to every kind of conduct that can somehow be connected to a “disparate impact.” As discussed in part 1, the burden-shifting framework set out in HUD’s Discriminatory Effects Rule implies that insurers (and all other defendants) have a duty to identify and adopt the course of conduct that is least likely to have a disparate impact on the housing rights of protected classes. The theory underlying Viens and Jones is that this duty requires insurers analyze both (i) how their actions will affect the decision-making of their customers, and (ii) what unintended discriminatory effects those customers’ decisions might have. It is at least debatable whether an insurer that fails to carry out that analysis can rightfully be charged with creating an “artificial” and “arbitrary … barrier” to minority housing. That being the case, one can also fairly question whether Congress intended the FHA to regulate decision-making on this level.
The plaintiffs in AIA, one of the cases that challenged the Discriminatory Effects Rule, made a related argument. They contended that the only way for insurers to avoid committing “discrimination” through disparate impact might be to collect and analyze data about protected classes, and to take account of that data when making underwriting decisions. They pointed out, however, that several states prohibit insurers from using race and other protected categories in at least some ratemaking decisions. E.g., 215 ILCS 5/424(3). In some states, insurers may not even collect the relevant data in connection with any application for insurance. E.g. Md. Code, Insurance, § 27-501(c)(1). Given these state laws, it is at least problematic for the FHA to impose an additional regulatory overlay on the non-discriminatory conduct of insurers.
(These statutes are also relevant to arguments about causation, especially in light of Justice Kennedy’s observation about situations in which the “law substantially limits the [defendant’s] discretion.” They are also relevant to arguments about the McCarran-Ferguson Act, which will be discussed in part 3 of this series.)
The majority opinion in Inclusive Communities lends some support to these arguments, because Justice Kennedy warned that disparate impact liability should not “be so expansive as to inject racial considerations into every housing decision.” Unfortunately, Inclusive Communities does not provide much practical guidance about which decisions the court meant to exempt.
Furthermore, it cannot be denied that insurers are in the process of voluntarily introducing many new types of data as “considerations” in their decision-making. In addition to the “price optimization” process noted above, insurers are developing uses for new information resources. For example: Acxiom, one of the country’s largest data brokers, reported to a Senate Committee in 2013 that its customers included “five of the top 10 life/health insurance providers” and “nine of the top 10 property and casualty insurers.”As the underwriting process achieves new levels of granularity, it is likely that calls for subjecting the process to antidiscrimination safeguards will gather strength.
Statutory Language Issues
Disparate impact liability based on the actions of third parties might also force the language of the FHA to carry more than it can bear. It is true, of course, that the FHA contains terms that go beyond the “sale or rental” of a “dwelling.” But even if insurance is a “service” within the meaning of § 3604(b), or a “real estate-related transaction” within the meaning of § 3605, it does not follow that an insurer “discriminates” in either the “provision” or the “terms or conditions” of insurance, if it makes the insurance available on equal terms to all its customers, and if none of those customers suffers a discriminatory effect.
Similarly, if the insurer raises the price of coverage for buildings with Section 8 tenants, it still might not be accurate to say that the insurer has “coerced” or “interfered with” the landlord, in violation of § 3617. The statute prohibits “coercion” or “interference” that is committed “on account of” the landlord’s “having aided or encouraged” tenants to enjoy rights protected by the FHA. Some courts have held that “intent to discriminate” is an element of a § 3617 claim,Bloch v. Frischholz, 587 F.3d 771, 783 (7th Cir. 2009), and others require at least “a causal link … between the protected activity and the adverse action.” Walker v. City of Lakewood, 272 F.3d 1114, 1128 (9th Cir. 2001).
There is also the question of whether the FHA applies, under any circumstances, to conduct the occurs after a dwelling has been sold or rented—an issue on which there is currently a split in authority. Compare Halprin v. Prairie Single Family Homes of Dearborn Park Ass’n, 388 F.3d 327 (7th Cir. 2004) (no claim under § 3604 for alleged religious harassment of plaintiffs who had already rented apartment), with The Comm. Concerning Community Improvement v. City of Modesto, 583 F.3d 690 (9th Cir. 2009) (upholding claims under § 3604 alleging racial discrimination in the provision of municipal services to plaintiffs who were already residents of community). If an insurer cancels a policy because Section 8 participants are already renting apartments in the insured dwelling, at least some cases suggest that the FHA does not apply at all.
Finally, the type of claim asserted in Viens and Jones present unusual circumstances relating to standing. Under the FHA, a civil action or administrative proceeding may be filed by an “aggrieved person.” 42 U.SC. §§ 3602, 3610, 3613. An “aggrieved person” can include someone who “believes … [she] will be injured” by an FHA violation “that is about to occur.” 42 U.S.C. § 3602(i). The Supreme Court has held that these provisions “extend to the full limits of Art[icle] III” of the Constitution. Havens Realty Corp. v. Coleman, 455 U.S. 363, 372 (1982). That is, a plaintiff has standing if she “ha[s] suffered or [is] imminently threatened with a concrete and particularized ‘injury in fact‘ that is fairly traceable to the challenged action.” Lujan v. Defenders of Wildlife, 504 U.S. 555, 560, (1992). The “fairly traceable” standard is less stringent than “proximate cause,” which is an element of the substantive claim. Lexmark Int’l, Inc. v. Static Control Components, Inc., 134 S.Ct. 1337, 1391 and n. 6 (2014).
In Viens and Jones, the courts had to apply these rules to plaintiffs who were not victims of the defendants’ alleged acts of discrimination. The landlords relied on two propositions. First, a plaintiff need not be a member of a protected class to be an “aggrieved person” within the meaning of the FHA. Trafficante v. Metropolitan Life Insurance Co., 409 U.S. 205, 209-10 (1972). Second, “paying a higher price for insurance” has been found to be “an injury” that confers standing. NAACP, 978 F.3d at 293. The other plaintiffs—nonprofit civil rights organizations—asserted “association standing,” on the ground that they had to expend resources investigating and advocating against the defendants’ policies. See Havens Realty Corp., 455 U.S. at 378–79; Olsen v. Stark Homes, Inc., 759 F.3d 140, 158 (2d Cir. 2014).
The landlords in Viens argued, further, that their “injury” would give them standing, even if they could not establish injury to any member of a protected class. The court rejected that argument: without any injury to tenants, it held, the plaintiffs could not explain how the defendant had engaged in “discrimination” based on disparate impact. In other words, failure to show that tenants were injured would amount to failure to state a claim. Thus, although standing can usually be established under a “fairly traceable” standard, the circumstances in Viens appear to require an additional showing of proximate causation—one that is not only “plausible,” Bell Atl. Corp. v. Twombly, 548 U.S. 903 (2007), but also “robust.” Inclusive Communities.
The Rulings in Viens and Jones
The courts in Viens and Jones faced many of the issues discussed above. Although they decided virtually all of these issues against the insurers, neither case is likely to be the last word on any of them. Moreover, details of the two cases bring out some of the conceptual difficulties of using a disparate impact theory to make insurers responsible for the intervening acts of others.
Viens v. American Empire
Because the plaintiffs in Viens do not allege that the insurer’s actions caused them to deny housing to any member of a protected class, their claim is based on the theory that the defendant’s policy “predictably will cause” unlawful discrimination by other landlords. 24 CFR § 100.500(c). In denying the defendant’s motion to dismiss, the court accepted that theory: it said the insurer’s policy imposed a “financial penalty [on policyholders] for renting to Section 8 tenants,” and it inferred that policyholders who are subjected to this “penalty” will “be less likely to participate in the program.” This, the court reasoned, “would result in less housing being available to Section 8 participants”—a disproportionate number of whom are members of racial minorities.
But the court also acknowledged (in a footnote) that the Connecticut Fair Housing Act (“CFHA”) differs from federal law, in that it prohibits landlords from declining to rent to Section 8 tenants. See C.G.S.A. § 46a-64c (prohibiting discrimination on the basis of “lawful source of income”). Thus, the court’s holding was actually based on a finding that the insurer’s policy “creates an incentive for [policyholders] to not comply with this requirement.”
This finding raises two immediate questions. The first has to do with the facts of the case. The court made its finding in the context of a motion under Federal Rule 12(b)(6), which requires that the facts alleged by the plaintiff provide a “plausible” basis for each element of the plaintiffs’ claim—including proximate cause. Bell Atl. Corp. v. Twombly,supra. The court therefore found it “plausible” that the insurer’s underwriting policy “predictably will cause” landlords to violate the CFHA—thereby exposing themselves to potential liability for both compensatory and punitive damages.
One such landlord—one of the plaintiffs in the case—alleged that his replacement policy cost $162 more per year(with a higher deductible) than the policy the defendant had canceled. The court did not specifically discuss whether this amount was a plausible “incentive” for unlawful conduct, but other courts have clearly taken a narrower view of what is necessary to establish proximate causation under the FHA. E.g., City of Miami v. Bank of America Corp., No. 13–24506–CIV (S.D. Fla. July 9, 2014) (where city alleged that the defendant bank’s original lending practices caused subsequent foreclosures by the bank’s customers in minority neighborhoods, the court ruled that “[t]he independent actions of [a] multitude of non-parties break the causal chain”).
The second question is more fundamental: if a defendant has not discriminated on the basis of any of the categories identified in the FHA, are the statute’s objectives actually advanced by making the defendant liable for “causing” others to act in a way for which the law already provides a remedy? The court’s opinion did not reach that question.
Jones v. Travelers
The decision in Jones presents the other side of this anomaly. As in Viens, the landlord plaintiffs in Jones did not claim to have turned away any Section 8 tenants; the court ruled, instead, that a reasonable jury could find it “predictabl[e]” that minority tenants will suffer from the defendant’s policies. But those California landlords, unlike their Connecticut counterparts in Viens, are not legally required to accept Section 8 tenants. Because the Section 8 program is voluntary under both federal and California law, a California landlord who discriminates against Section 8 participants does not make housing “unavailable” in a way that is prohibited by either the FHA or California’s Fair Employment and Housing Act (“FEHA”). E.g., Doe v. WCP I, LLC, No. 05-438885 (Cal. Super. Ct. June 4, 2009).
As discussed in part 1 of this series, the Jones court ruled that insurers do not enjoy the same discretion: it found “no cases where … landlords and insurers are similarly situated under Section 8.” What that means, however, is that an insurer can “discriminate” under the FHA by causing a landlord to commit a second act that does not itself count as discrimination. Under the reasoning in Jones, therefore, Section 8 tenants have no remedy when landlords opt out of the Section 8 program because of business considerations in general, but they do have a remedy—potentially including punitive damages—if the landlords happen to be trying to save on insurance.
That is not how the law usually works. See, e.g., Pagan v. Calderon, 448 F.3d 16, 30 (1st Cir. 2006) (“the standing inquiry turns on the plaintiff’s injury, not the defendant’s motive”). But, because the case settled, the court in Jones did not get a chance to look any deeper into the question of how an insurer can violate the FHA by “causing” someone else to act in a way that the FHA itself permits.