National Housing Law Project Housing Law Bulletin

Recently the United States District Court in Seattle decided that a landlord's plan to end its participation in the Section 8 certificate and voucher programs violated the federal Fair Housing Act. Green v. Sunpointe Associates, Ltd., No C96-1542C (W.D. Wash. Order Granting Summary Judgment filed May 12, 1997).¹ The litigation involved an African American woman with two children who was living in a large apartment complex with her rent subsidized by the local housing authority under the Section 8 program. The court concluded that the landlord's plan to withdraw from the Section 8 program would disparately impact African Americans, women and families with children, and that this particular landlord could not show any business necessity for getting out of Section 8. Thus the court ordered the landlord to remain in the Section 8 program.

This problem of landlords refusing to participate in the Section 8 program is one that has received widespread legislative attention during the past 20 years, at least. In 1987, Congress passed a statute, known as Section 8(t), that prohibited any landlord who has entered into a Section 8 contract from refusing to rent to certificate or voucher holders merely because they were participating in those programs.² That provision came under severe attack in 1994 when the Multi-Housing Council released a report claiming, among many other things, that that statute discouraged landlords from participating in Section 8 in the first place.³ The claim was that the landlords were afraid that if they took one Section 8 tenant, they would have to take them all. Responding to the label "take one, take all," Congress suspended that statutory provision in 1996.⁴

There are, however, other federal statutes that prohibit some landlords from refusing to rent to certificate and voucher holders. For example, the same 1987 legislation that enacted the non-discrimination provision discussed above included another provision that bars subsidized landlords, i.e., those with Section 221(d)(3) BMIR (below-market interest rate) mortgages, Section 236 interest reduction payments, rent supplements, Section 202 loans or Section 8 contracts other than certificates or vouchers, from refusing to rent to certificate and voucher holders because they hold such subsidies.⁵ Again, landlords who purchase buildings from HUD as part of its property disposition program are barred from refusing to rent to certificate holders because of their status as such.⁶

Beyond the federal statutes, some states have laws that prohibit landlords from discriminating against applicants on the basis of their source of income.⁷ Their language may be construed to mean that landlords may not discriminate against applicants who hold Section 8 certificates or vouchers. Some state statutes are more explicit and prohibit discrimination on the basis of receipt of a housing subsidy.⁸

An alternative prohibition against refusal to rent to certificate and voucher holders arises from the Fair Housing Act.⁹ That act prohibits discrimination in housing on the basis of race, national origin, sex and the presence of children in the family. If a landlord were to refuse to rent to certificate and voucher holders as a subterfuge for refusing to rent to members of one of those protected classes, that refusal would violate the Fair Housing Act. Even if the landlord's refusal were not intentionally designed to keep protected class members out of his buildings, but it had that effect, such refusal also would violate the Fair Housing Act if there were no business necessity for the policy.

The Green case involved such a refusal. The landlord had built a large apartment complex in a suburb outside Seattle with financing that had been raised by the Washington State Housing Financing Commission (State HFC) through tax-exempt bonds. Initially the landlord rented to certificate and voucher holders. In fact, it had signed a regulatory agreement with the State HFC under which it was barred from discriminating on the basis of receipt of public assistance or housing assistance. Eventually, however, the landlord decided not to participate in the Section 8 program anymore. Apparently without re-reading the regulatory agreement, the landlord implemented a plan to get out of Section 8. One step was to terminate the tenancies of the certificate and voucher holders already living in the buildings.

The plaintiff, Angela Green, brought this litigation in federal court. She had received a notice that her lease would not be renewed and that she would have to move out by September 30, 1996. Her claims were that the landlord's actions discriminated against her on the basis of race, sex and family status, in violation of the Fair Housing Act, the Civil Rights Act of 1866 and Washington state law against discrimination. She also included a claim that the defendants had breached their regulatory agreement with the State HFC.

The plaintiff's Fair Housing Act claim had two parts: first, that the defendants were acting intentionally and, second, that the defendants' policy had a disparate impact. The court declined to rule on the intentional discrimination claim on the summary judgment motions because of factual disputes. However, it did reach the disparate impact claim and ruled for the plaintiff.

With regard to the disparate impact case, the primary arguments related to what the relevant comparisons should be. The plaintiff demonstrated that, of the 22 families already living in the buildings with Section 8, all were headed by women, all had children, and 82 percent (18 out of 22) were African American. Similarly, of the housing authority's certificate and voucher holders, 85 percent were families headed by women, 81 percent were families with children, and 49 percent were African American. The court concluded that those figures demonstrated that the landlord's plan to withdraw from Section 8 would fall most heavily, and in some cases exclusively, on people protected by the Fair Housing Act.

Plaintiff also demonstrated that members of protected classes were present in the class affected by the landlord's "no Section 8" policy to a much greater extent than they were in the population of the suburb where the project was located or the Seattle metropolitan area. In the suburb, African Americans made up 5 percent of the population, families headed by women constituted 9 percent and only 28 percent of the households had children. The comparable figures for the metropolitan area as a whole were 4 percent, 9 percent and 32 percent, respectively.

Plaintiff also showed that protected class members in the buildings were affected to a greater extent than members of other classes. The "no Section 8" policy affected 13.5 percent of the African American families in the buildings, but only 6.7 percent of the Caucasian households, i.e., the percentage of African American families in the building affected by the policy was twice as great as the percentage of white families affected. Similarly, that policy affected 17.5 percent of the families with children, but none of the households without children.

Plaintiff had thus shown that, among existing and potential tenants, almost all of the households affected by the "no Section 8" policy would be members of protected classes. In addition, protected classes would be adversely affected by the policy at a greater rate than classes not protected by the Fair Housing Act.

The defendants' response was that the plaintiff had made the wrong comparisons. First, the landlord pointed out that even though most of the people affected by the policy were protected class members, there were so many other protected class members without Section 8 that the overall impact on protected class members would be slight. For example, because 141 of the families in the buildings were African American, evicting the Section 8 families would reduce the African American population in the buildings by only 2 percent.

The court, however, considered that fact to be irrelevant, because the Fair Housing Act protects individuals, not a minority group as a whole. The fact that many protected class members would not be affected by the policy did not make irrelevant the fact that most of the people adversely affected were protected class members. Just as the intentional rejection of an African American applicant because of her race cannot be excused because many other apartments are rented to African Americans, following a policy that adversely affects only African Americans cannot be excused because many African Americans are admitted despite the policy.

The landlord's other statistical argument was that the court should not compare the percentages of protected class members among the class affected by the policy with their percentages in the general population. Instead, it maintained that the comparison should be with the protected class members' presence among those renters in the area who would be qualified to rent the apartments in question. The court rejected that argument as well. In the court's view, in housing cases, where a policy falls more heavily on protected class members than their presence in the local community, a showing of disparate impact is made. In this case, there was the added showing that the policy affected protected classes at a greater rate than people not in the protected classes.

On the business necessity defense, the plaintiff had an easier case than might be true in some other cases. For purposes of summary judgment, the defendants claimed that it cost them about $25,000 per year to participate in the Section 8 program. However, in order to secure the bond financing from the State HFC, the landlord had agreed not to discriminate against Section 8 applicants. As a result, the court concluded that the landlord could not rely on the costs of participating in Section 8 as a business necessity defense. That determination quickly lead the court to impose an injunction against discriminating against Section 8 applicants and tenants, as long as the regulatory agreement with the State HFC was in place.

An alternative argument presented by the defendants was that the Fair Housing Act could not be interpreted to ban landlords from withdrawing from Section 8, if withdrawing would have a disparate impact on protected classes. Otherwise, in the defendants' view, Congress' suspension of Section 8(t) would have no effect because, in most cases, Section 8 includes a high percentage of class members protected by the Fair Housing Act. In more general terms, defendants' argument was that, by suspending Section 8(t), Congress granted landlords the freedom not to participate in Section 8, and the Fair Housing Act could not take that freedom away if the landlord was not intentionally discriminating on grounds of race, sex or familial status.

The argument was based upon a case from the Seventh Circuit, Knapp v. Eagle Property Management, 54 F.3d 1272 (7th Cir. 1995). In Knapp, the plaintiff had applied for an apartment and been rejected, she alleged, because of her race and because she was using Section 8. Her primary claim was that the landlord had violated Section 8(t); the jury agreed, awarding her $95,000 damages. However, the trial court reduced that award to $1 on the grounds that contractual damages, not tort damages, were the only remedy for a violation of Section 8(t), and she had shown no damages compensable under a contract theory. The Court of Appeals had affirmed that trial court ruling.

The other theory of the Knapp case had been that a refusal to accept Section 8 had an adverse effect on African Americans and thus violated the Fair Housing Act. The trial court had refused to allow the plaintiff to present the expert testimony regarding adverse impact that she had proffered. Plaintiff's theory was that a much larger percentage of the local African American population than of the white population participated in the Section 8 program. With that difference, the policy of not accepting Section 8 disproportionately affects the African American population.

The Seventh Circuit, however, concluded that the disparate impact test should not be available to a protected class member adversely affected by a refusal to accept Section 8. Its reasoning was complicated. First, it started with the premise that, in the Seventh Circuit, the disparate impact analysis is not appropriate in all Fair Housing Act cases. Next, it assumed that a landlord who was not already participating in the Section 8 program could refuse to participate even if doing so would have a disparate impact upon protected classes. Then it concluded that applying the disparate impact test only to landlords who had already agreed to participate in Section 8 but who had changed their policies would deter landlords from participating in Section 8 in the first place. As a result, the court concluded, landlords participating in Section 8 should not be subject to the disparate impact analysis if they should later decide to withdraw from the program.

Of course, the weakness of this reasoning is the court's erroneous assumption that landlords who have never participated in Section 8 are not subject to the Fair Housing Act's disparate impact analysis. If they are subject to the Act, using the disparate impact analysis on landlords who are withdrawing from participation will not create any deterrence to their initial participation.

The district court in Green decided that the fact that Congress has made participation in Section 8 voluntary does not preclude the plaintiff from making out a prima facie case of disparate impact. If that case is made, the burden then shifts to the defendant to prove business necessity. In the court's view, Congress' judgment to make participation voluntary is relevant when considering the landlord's claim of business necessity.

Given the facts of this case -- i.e., that the landlord had no business necessity defense because of its promise not to discriminate against Section 8 applicants -- it was easy for the court to consider Congress' suspension of Section 8(t) as relevant to the business necessity defense. However, making the suspension of Section 8t relevant to that defense is probably more than the ordinary rules of statutory interpretation allow. Congress may rewrite statutes, including the Fair Housing Act, to create new defenses. However, when it revises one statute without expressly modifying another, there is a strong presumption that the other law is not impliedly repealed or otherwise modified. If the suspension of Section 8(t) is read not only to do what it expressly does, but also to create a new defense for landlords who would otherwise be violating the Fair Housing Act, the rule disfavoring implied repeals is being violated.

The proper way to handle the suspension of Section 8(t) and the voluntary nature of Section 8 is to recognize that neither the suspension nor the voluntary nature includes the freedom to ignore the Fair Housing Act. If refusing to participate in Section 8 has a disparate impact upon protected classes and there is no business necessity not to participate, the landlord is not free to reject Section 8 applicants.

One also might wonder why the court did not decide this case on the ground that the landlord had violated his contract with the State HFC. That is particularly true when one realizes that the court found plaintiff to be a third-party beneficiary of the contract under Washington state law. However, the plaintiff was seeking summary judgment and the contract allowed the landlord 60 days to cure any breach from the date it learned of the breach. When the suit was filed, the landlord renewed the plaintiff's lease and offered to renew all other Section 8 leases and to accept new Section 8 tenants. The court considered that there was a factual dispute whether that action came within the 60-day cure period.

The landlord has now appealed the court's award of the permanent injunction on the disparate impact claim and the rest of the claims are proceeding to trial.

1. On file at the National Housing Law Project's Oakland office.
2. 42 U.S.C.A. § 1437f(t) (West 1994).
3. Abt Associates, Final Report on Recommendations on Ways to Make the Section 8 Program More Acceptable in the Private Rental Market, at 10 and 22-23 (Mar. 1994). This report is summarized in A Critique of Landlord-Commissioned Report on the Certificate and Voucher Programs, 24 HOUS. L. BULL. 35 (May/June 1994).
4. Pub. L. No. 104-134, § 101(e), enacting the HUD FY 1996 Appropriations Act, § 203, 110 Stat. 1321-281 (1996); Pub. L. No. 104-204, § 201(e), 110 Stat. 2893 (1996).
5. Pub. L. No. 100-242, § 183(c), 101 Stat. 1872 (1988), noted after 42 U.S.C.A. § 1437f (West 1994).
6. 12 U.S.C.A. § 1701z-12 (West 1989).
7. E.g., Wis. Stat. § 101.22(6).
8. See, e.g., Mass. Gen. L. ch. 151b, § 4(10); Attorney General v. Brown, 400 Mass. 826 (1987).
9. 42 U.S.C.A. §§ 3600-3617 (West 1994).