PERPETUATING SEGREGATION OR TURNING DISCRIMINATION ON ITS HEAD? AFFORDABLE HOUSING RESIDENCY PREFERENCES AS ANTI-DISPLACEMENT MEASURES

PERPETUATING SEGREGATION OR TURNING DISCRIMINATION ON ITS HEAD? AFFORDABLE HOUSING RESIDENCY PREFERENCES AS ANTI-DISPLACEMENT MEASURES

Affordable housing residency preferences give residents of a specific geographic “preference area” prioritized access to affordable housing units within that geographic area. Historically, majority-white munici­palities have sometimes used affordable housing residency preferences to systematically exclude racial minorities who reside in sur­rounding com­munities. Courts have invalidated such residency pref­erences, usually on the grounds that they perpetuate residential segregation in violation of the Fair Housing Act.

More recently, as gentrification spurs rising housing costs in many formerly majority-minority urban neighborhoods, cities including New York and San Francisco have implemented intramunicipal residency preferences as a mechanism for mitigating gentrification-induced dis­placement. These cities’ policies offer residents preferred access to affordable housing units in their own neighborhoods, relative to both nonresidents and to city residents living in other neighborhoods. Pro­ponents of these policies contend that their use on an intracity level preserves rather than excludes minority communities, thereby inverting the traditional discriminatory application of such preferences. Opponents of the policies argue that any residency preference imple­mented in a racially segregated area necessarily perpetuates segregation and violates the law.

This Note examines how neighborhood-level, anti-displacement res­idency preferences should be understood under the relevant law. It observes that the neighborhood-level residency preference is a potent anti-displacement tool that suffers from an emerging mismatch between fair housing goals and fair housing law. Neighborhood-level anti-displacement residency preferences likely suffer from the same legal defects as intercity preferences used to exclude minority applicants, and may even be at heightened risk because they are more likely to be ex­pressly race-conscious. Despite the fact that these preferences aim to promote accessible affordable housing for low-income and minority residents, they do so in response to displacement pressures that the Fair Housing Act does not contemplate and in a manner that arguably clashes with its anti-segregationist objective. If neighborhood-level resi­dency preference policies are to be effectively and legally utilized to address issues of urban displacement, either courts’ approaches to such policies or the policies themselves must evolve.

Introduction

In an era of gentrification-induced displacement, it is uncertain whether efforts to preserve existing neighborhood demographics should be understood as extending or subverting fair housing practices. Munici­palities use residency preference policies to restrict access to affordable housing units on the basis of applicants’ place of residence. Historically, residency preference policies have been challenged and invalidated when they exclude minority applicants from affordable housing in major­ity-white suburbs. 1 See infra section I.B (detailing judicial decisions regarding residency preferences). As gentrification elevates housing prices, many low-income and minority residents are displaced from their neighborhoods or even from their cities entirely. 2 See infra notes 126–139 and accompanying text (describing the displacement pressures created by gentrification). Cities, including San Francisco and New York, have offered residents preferred access to affordable housing in their own neighborhoods in an effort to mitigate population displacement. 3 See infra sections I.C–.D (describing gentrification-related preference policies in New York City and San Francisco).

Proponents of these policies contend that their use on an intracity level preserves rather than excludes minority communities. San Francisco’s City Attorney, for example, maintained that the city’s res­idency pref­erence plan takes a formerly exclusionary tool and “flips it on its head.” 4 Letter from Dennis J. Herrera, City Att’y, S.F., to Honorable Julián Castro, Sec’y, U.S. Dep’t of Hous. & Urban Dev. & Helen R. Kanovsky, Gen. Counsel, U.S. Dep’t of Hous. & Urban Dev. (Aug. 25, 2016) [hereinafter Letter from Herrera], http://www.
sfcityattorney.org/wp-content/uploads/2016/08/Herrera-to-HUD-2016-08-25.pdf [http://
perma.cc/QK8E-R69J].
New York City officials have similarly warned that invalidating their policy would “turn the [Fair Housing Act] on its head.” 5 Defendant’s Memorandum of Law in Support of Its Motion for Dismissal of the First Amended Complaint at 3, Winfield v. City of New York, No. 15-CV-05236-LTS (S.D.N.Y. Oct. 2, 2015), 2015 WL 12862506 [hereinafter Winfield, Motion to Dismiss]. Not every­one agrees, however, that neighborhood-level residency preferences amount to an in­clusionary headstand. New York’s community preference policy is the subject of a federal lawsuit, 6 See Winfield v. City of New York, No. 15CV5236-LTS-DCF, 2016 WL 6208564 (S.D.N.Y. Oct. 24, 2016); see also infra section I.C.2 (discussing the Winfield lawsuit). and San Francisco’s effort to use residency preferences as a lifeline for the city’s dwindling African American population was blocked by the Department of Housing and Urban Development (HUD). 7 See infra section I.D. In both cases, opponents claim that the policies violate the Fair Housing Act by perpetuating segregated housing patterns. 8 See infra section I.C.2 (discussing the Winfield lawsuit); see also infra section I.D.2 (discussing San Francisco’s clash with HUD over its resident preference policy).

Conventional legal analysis suggests that residency preferences are invalid in residentially segregated locales, regardless of whether the pref­erence favors primarily white or primarily nonwhite residents. 9 See infra section II.B. Moreover, affirmative-action-minded preferences may be at heightened legal risk because they are more likely to be expressly race conscious. 10 Classifications made on the basis of race, even when they seek to remediate past or continuing discrimination against racial minorities, are subject to strict scrutiny—the most rigorous form of judicial review. See, e.g., Regents of Univ. of Cal. v. Bakke, 438 U.S. 265, 291 (1978) (establishing that, even in the context of affirmative action programs, “[r]acial and ethnic distinctions of any sort are inherently suspect and thus call for the most exacting judicial examination”). Put simply, the neighborhood-level residency preference is a potent anti-displace­ment tool that suffers from an emerging mismatch between fair housing goals and fair housing law.

This Note examines how neighborhood-level, anti-displacement res­idency preferences should be understood under the relevant law. Part I describes the legal history of exclusionary, intercity residency preferences and details New York’s and San Francisco’s efforts to implement and de­fend intracity preferences. Part II analyzes both the potential value and legal vulnerabilities of anti-displacement, intracity residency preferences and concludes that they are unlikely to withstand legal challenge. Part III proposes several solutions to this dilemma, suggesting alternative ap­proaches to the residency preference model and urging a more expan­sive understanding of fair housing goals in light of gentrification pressures.

I. Overview of Affordable Housing Residency Preferences

The Fair Housing Act (FHA) 11 42 U.S.C. §§ 3601–3631 (2012). and the federal regulations through which HUD enforces it endow local governments with the authority to govern applicant eligibility for affordable housing units. 12 Id. § 3616 (outlining the federal–state collaboration in handling housing discrim­ination); 24 C.F.R. §§ 5.655, 960.202 (2016) (regulating tenant-selection policies). Public Housing Authorities (PHAs) may, within certain limits, restrict eligibility or create a priority system for eligibility on the basis of any legally permissible cri­teria. 13 See 24 C.F.R. § 960.202 (regulating PHA tenant selection). Many local governments elect to restrict eligibility on the basis of applicants’ geographical residence in order to protect their own resi­dents from losing affordable housing opportunities to nonresidents. 14 See, e.g., Maya Srikrishnan, How National City Is Fighting Displacement of Poor Residents, Voice of San Diego (June 17, 2016), http://www.voiceofsandiego.org/topics/
land-use/how-national-city-is-fighting-resident-displacement/ [http://perma.cc/VA6G-EVEN] (describing a San Diego-area municipality’s residency preference policy designed “to ensure local residents would get the new homes”).

No court has held that residency preference policies are per se ille­gal, and HUD has tacitly endorsed the proper use of such policies. 15 Several HUD regulations specifically address residency preference policies, simultaneously affirming their potential legitimacy and limiting their application. See 24 C.F.R. §§ 5.655, 982.207.
One provision establishes that PHAs may utilize residency preferences but not residency requirements. See id. § 982.207. Valid residency preferences enacted by PHAs must treat employees who work in a “residency preference area” as qualifying residents and may not discriminate on the basis of the duration of applicants’ residency. This provision also requires that the “residency preference area” be no smaller than a munic­ipality or county. Id. Note, however, that because this provision specifically applies to tenant-assistance programs enacted by PHAs with federal funding, it does not amount to a total prohibition on intramunicipality preferences. An additional provision that regulates the administration of Section 8 project-based rental assistance imposes similar require­ments. Id. § 5.655.
The case law regarding residency preferences, however, suggests that residency preferences are often on tenuous legal ground. Courts have repeatedly found that residency preferences, when applied in racially segregated areas, facilitate or perpetuate segregation by limiting the opportunities for proximate nonresidents of color to procure affordable housing in predominantly white municipalities. 16 See infra section I.B (charting the landscape of judicial decisions on residency preference policies). Against this legal backdrop, cities have encountered resistance to the implementation of intracity residency preferences, even when they are enacted with the purported intention of supporting communities of racial minorities. 17 See infra sections I.C–.D.

This Part examines the trajectory of the legal controversies sur­rounding affordable housing residency preferences. Section I.A introduces residency preferences and residency requirements generally, as well as the relevant legal boundaries on affordable housing residency pref­erences. Section I.B charts the existing case law on residency prefer­ences in affordable housing. Sections I.C and I.D address recent contro­versies surrounding the preference policies of New York and San Francisco, respectively.

A. Background on Residency Preferences

1. Residency Requirements and Preferences. — The landscape of resi­dency requirements and preferences is extensive and varied. 18 See Robert C. Farrell, Classifications that Disadvantage Newcomers and the Problem of Equality, 28 U. Rich. L. Rev. 547, 548–50 (1994) (acknowledging the numer­ous ways in which states classify between residents and nonresidents and describing the Supreme Court’s jurisprudence on the subject as “inconsistent and inadequate”). Litigation over such policies often invokes the Article IV Privileges and Immunities Clause, 19 U.S. Const. art. IV, § 2, cl. 1. which is understood to prohibit governmental discrimination on the basis of state and municipal residency, 20 See Richard H. Seamon, Note, The Market Participant Test in Dormant Commerce Clause Analysis—Protecting Protectionism?, 1985 Duke L.J. 697, 724–25 (discussing the Supreme Court’s extension of the prohibition on residency-based discrim­ination to the municipal level). and the Dormant Commerce Clause doctrine, under which states are generally proscribed from im­plementing economic protectionism. 21 Id. at 702–04 (describing the Dormant Commerce Clause rule of per se invalidity). Nonetheless, these doc­trines have notable exceptions, 22 See, e.g., Dan T. Coenen, Untangling the Market-Participant Exemption to the Dormant Commerce Clause, 88 Mich. L. Rev. 395, 401 (1989) (discussing the evolution of the market-participant exception). and states frequently enact laws and regulations that endow state residents with preferential access to jobs, social services, and other opportunities—or that foreclose nonresidents from accessing those opportunities entirely. 23 See, e.g., Treg A. Julander, State Resident Preference Statutes and the Market Participant Exception to the Dormant Commerce Clause, 24 Whittier L. Rev. 541, 541 (2002) (noting that most states have enacted resident preference laws). Residency requirements and preferences are also enacted at the municipal level for similar reasons as their state-level counterparts: the earmarking of local oppor­tunities for residents, the stimulation of the local economy, and parochialism. 24 See George T. Reynolds, Constitutional Law—Constitutional Assessment of State and Municipal Residential Hiring Preference Laws, 40 Vill. L. Rev. 803, 803–04 (1995) (noting the motivations behind state and municipal residency employment preferences).

Common residency requirements impose restrictions on who may vote, 25 See Michael J. Pitts, Against Residency Requirements, 2015 U. Chi. Legal F. 341, 344 (“[T]here is usually a durational residency requirement to qualify to vote.”). hold public office, 26 See id. (describing requirements that elected officials reside in certain jurisdictions or that they be eligible voters of the jurisdiction in which they seek office). receive construction contracts for public works, 27 See Doreen J. Piligian, Resident Preference Laws and the Award of Public Contracts, Construction Law., May 1990, at 10, 10 (discussing state and local policies that advantage residents in public works construction). and earn welfare benefits, 28 See Clark Allen Peterson, Comment, The Resurgence of Durational Residence Requirements for the Receipt of Welfare Funds, 27 Loy. L.A. L. Rev. 305, 306 (1993) (describing states’ durational residency requirements limiting welfare assistance to resi­dents of at least one year). among other activities. 29 See, e.g., Bryan H. Wildenthal, Note, State Parochialism, the Right to Travel, and the Privileges and Immunities Clause of Article IV, 41 Stan. L. Rev. 1557, 1561–62 (1989) (listing residency requirements limiting access to abortion, elk-hunting, and admission to the state bar—among others—some of which courts invalidated). Durational requirements discriminate between longstanding and recent residents, often by establishing waiting periods before new residents are eligible for public benefits such as welfare, voter eligibility, or in-state tuition. 30 See Farrell, supra note 18, at 549–50 (describing state classifications based upon duration of residence).

Residency preferences, while less restrictive than residency require­ments, nonetheless raise related legal and policy questions by conferring upon residents prioritized access to jobs, goods, or services. Residency preferences are utilized in affordable housing to provide residents of a “preference area” with prioritized access to local public (or publicly funded) housing. Local governments around the country have frequently proposed and implemented these preference policies, 31 See, e.g., Melinda Henneberger, Issue of Outsiders Kills Town’s Housing Plan, N.Y. Times (June 2, 1993), http://www.nytimes.com/1993/06/02/nyregion/issue-of-outsiders-kills-town-s-housing-plan.html?pagewanted=all (on file with the Columbia Law Review) (describing the insistence of elected officials in Harrison, New York that new af­fordable housing be accompanied by a residency preference); Srikrishnan, supra note 14 (describing the residency preference policy in National City, California); Andrew Theen, Gentrification: Can Portland Give Displaced Residents a Path Back?, Oregonian (Dec. 23, 2015), http://www.oregonlive.com/portland/index.ssf/2015/12/gentrification_can_
portland_gi.html [http://perma.cc/8Y2U-AY2M] (describing Portland’s residency pref­erence policy); Jeff Dillman, City of Dubuque to End Discriminatory Residency Preference in Housing Program, Fair Hous. Project (Apr. 16, 2014), http://www.fairhousingnc.org/
2014/city-of-dubuque-to-end-discriminatory-residency-preference-in-housing-program/ [http://perma.cc/3BZD-HYD8] (discussing the residency preference policy in Dubuque, Iowa); Finding Housing: Lottery Preferences, Bos. Planning & Dev. Agency, http://
www.bostonredevelopmentauthority.org/housing/lottery-preferences [http://perma.cc/
J8FL-B6SN] (last visited Oct. 20, 2017) (detailing Boston lottery preferences for affordable housing, including a residency preference).
which are the focus of this Note. Subsequent references to “residency preferences” in this Note refer specifically to residency preferences in the affordable housing context rather than to residency preferences generally.

2. Relevant Legal Boundaries. — Affordable housing residency prefer­ences, particularly those enacted in racially segregated areas, are most commonly challenged as violations of the FHA. Enacted as Title VIII of the Civil Rights Act of 1968, the Fair Housing Act 32 42 U.S.C. §§ 3601–3631 (2012). embraced a clear in­tegrationist purpose from the outset. The goals of the FHA, according to its cosponsor Senator Walter Mondale, were to cultivate “truly integrated and balanced living patterns,” 33 114 Cong. Rec. 3422 (1968) (statement of Sen. Mondale). address the problem of Americans “liv[ing] separately in white ghettos and Negro ghettos,” and promote “the principle of living together.” 34 Id. at 2276, 2279. Two major catalysts for the FHA’s passage were the assassination of Dr. Martin Luther King, Jr. and the release of the Kerner Report, commissioned by President Lyndon B. Johnson, which described the increasing segregation of U.S. society. 35 See Valerie Schneider, In Defense of Disparate Impact: Urban Redevelopment and the Supreme Court’s Recent Interest in the Fair Housing Act, 79 Mo. L. Rev. 539, 553 (2014) (“The [Kerner Report] recommended, among other things, the elimination of barriers to choice in housing and the passage of a national and enforceable ‘open housing law.’” (quoting Report of the National Advisory Commission on Civil Disorders 24 (1968), http://www.eisenhowerfoundation.org/docs/kerner.pdf [http://perma.cc/F4LH-5KA4])).

While the text of the FHA does not explicitly announce its integra­tionist aims, courts and scholars have understood its provisions to embrace that purpose in light of its legislative history. 36 See, e.g., Trafficante v. Metro Life Ins. Co., 409 U.S. 205, 211 (1972) (quoting Sen. Mondale’s statement in its analysis of the FHA’s intent); see also Robert G. Schwemm, Overcoming Structural Barriers to Integrated Housing: A Back-to-the-Future Reflection on the Fair Housing Act’s “Affirmatively Further” Mandate, 100 Ky. L.J. 125, 127–30 (2012) [hereinafter Schwemm, Overcoming Structural Barriers] (noting that the FHA’s legislative history is widely understood to endow the FHA with an anti-segregationist purpose). The FHA’s “affirmatively further” language, which instructs HUD to administer its programs “in a manner affirmatively to further the purposes” of the Act, 37 42 U.S.C. § 3608(d). is commonly understood as a “mandate to promote racial integra­tion.” 38 Philip Tegeler, Megan Haberle & Ebony Gayles, Affirmatively Furthering Fair Housing in HUD Housing Programs: A First Term Report Card, 22 J. Affordable Housing & Community Dev. L. 27, 28 (2013); see also Myron Orfield, Racial Integration and Community Revitalization: Applying the Fair Housing Act to the Low Income Housing Tax Credit, 58 Vand. L. Rev. 1747, 1750 (2005); Schwemm, Overcoming Structural Barriers, supra note 36, at 128–30. Regulations promulgated by HUD reassert the FHA’s integrationist mandate and indicate that disparate impact liability 39 Evidence of a law’s “disparate impact” on racial minorities—also referred to as its “discriminatory effect”—does not constitute a violation of the Fourteenth Amendment’s Equal Protection Clause (barring additional evidence of a discriminatory intent), but it may constitute a violation of certain statutes. Compare Griggs v. Duke Power Co., 401 U.S. 424, 430 (1971) (holding that facially neutral practices which exact a disparate impact may violate Title VII of the Civil Rights Act), with Washington v. Davis, 426 U.S. 229, 246 (1976) (holding that discriminatory intent is necessary to establish an equal protection violation). may constitute a violation. 40 See 24 C.F.R. § 91.225(a)(1) (2016) (directing HUD grant recipients to “take no action that is materially inconsistent with [the] obligation to affirmatively further fair housing”); id. § 100.500 (“Liability may be established under the [FHA] based on a practice’s discriminatory effect . . . even if the practice was not motivated by a discrim­inatory intent.”). In its 2015 decision in Texas Department of Housing and Community Affairs v. Inclusive Communities Project, Inc., 41 135 S. Ct. 2507 (2015). the Supreme Court affirmed that the FHA 42 Specifically, FHA § 3604, which makes it unlawful to “refuse to sell or rent” or “otherwise make unavailable or deny” a dwelling on the basis of protected class identity. 42 U.S.C. § 3604(a). prohibits both intentional discrimination and discriminatory con­sequences under a disparate impact standard. 43 Inclusive Cmtys., 135 S. Ct. at 2525; see also Robert G. Schwemm, Fair Housing Litigation After Inclusive Communities: What’s New and What’s Not, 115 Colum. L. Rev. Sidebar 106, 109–10 (2015) [hereinafter Schwemm, After Inclusive Communities]. Contra Inclusive Cmtys., 135 S. Ct. at 2533–35 (Alito, J., dissenting) (arguing that the plain meaning of the provision’s text is that it only reaches intentional discrimination).

In addition to claims brought pursuant to the Fair Housing Act, res­idency preference policies may also be subject to constitutional chal­lenges alleging racial discrimination in violation of the Fourteenth Amendment’s Equal Protection Clause or unconstitutional restriction of the fundamental right to travel and migration. 44 For further discussion of these doctrines, see infra sections II.B.3 and II.B.4, respectively.

B. Judicial Review of Residency Preferences

Until recently, 45 See generally First Amended Complaint, Winfield v. City of New York, No. 15-CV-05236-LTS (S.D.N.Y. Sept. 1, 2015), 2015 WL 10853937 [hereinafter Winfield, First Amended Complaint] (challenging the legality of New York City’s residency preference policy, implemented on the community district level); infra section I.C.2. legal challenges to residency preference policies have followed a predictable pattern: A suburban municipality with pre­dominantly white residents implements a residency preference for its affordable housing, and neighboring nonresidents claim that the policy discriminates against racial minorities. 46 See Schwemm, Overcoming Structural Barriers, supra note 36, at 136 & n.70 (identifying cases in which local governments “giv[e] preferences to local residents in predominantly white towns for subsidized housing” among the notable subcategories of FHA doctrine (citing Langlois v. Abington Hous. Auth., 234 F. Supp. 2d 33, 43, 69–70 (D. Mass. 2002); United States v. Hous. Auth. of Chickasaw, 504 F. Supp. 716, 717–18, 729–32 (S.D. Ala. 1980))); see also Florence Wagman Roisman, The Lessons of American Apartheid: The Necessity and Means of Promoting Residential Racial Integration, 81 Iowa L. Rev. 479, 518 (1995) (“Residency preferences often are used for the purpose of ex­cluding people of color . . . .”); Adam Gordon, Note, Making Exclusionary Zoning Remedies Work: How Courts Applying Title VII Standards to Fair Housing Cases Have Misunderstood the Housing Market, 24 Yale L. & Pol’y Rev. 437, 453 (2006) (noting that “the towns imposing . . . residency requirements generally are mainly white,” and there­fore “a residency requirement has a disparate impact on blacks”). This section explores the exist­ing case law and examines the situations in which courts have affirmed, invalidated, or called into question the legality of residency preference policies.

1. FHA Disparate Impact Claims. — Residency preferences are perhaps most susceptible to FHA disparate impact challenges that allege the poli­cies have a segregative effect. Such challenges emphasize the demo­graphic disparities between the “preference area” population and the nearby populations that are excluded or disadvantaged by the residency preference. In United States v. Housing Authority of Chickasaw, a virtually all-white city 47 504 F. Supp. 716 at 725 (“Chickasaw is commonly regarded as an all-Caucasian community desirous of maintaining itself as all-Caucasian . . . .”). The 1970 census re­ported that African Americans comprised 0.3% of Chickasaw’s population of 8,000 resi­dents, and at the time the decision was handed down the court found that “[n]o Negroes currently live in the City of Chickasaw.” Id. at 718. in Mobile County, Alabama, administered its low-rent housing program subject to a “citizenship requirement.” 48 Id. at 721. Under the “citizenship requirement,” applicants whose address fell outside Chickasaw were automatically rejected, and city employees conducted “physical check[s]” to confirm that applicants actually lived at the addresses provided. Id. (internal quotation marks omitted). The court found that, because of the racial disparities between Chickasaw and the remain­der of Mobile County, 49 The 1970 census reported that Mobile County’s overall population was 32.3% African American, and the population of Prichard, which bordered Chickasaw, was 50.5% African American. Id. at 718. the residency requirement effectively “exclude[d] non-Caucasians from ever establishing residency” in Chickasaw and therefore established a disparate impact. 50 Id. at 731. The court held that Chickasaw authorities had violated the FHA on that basis. 51 Id. at 732.

Similarly, in Langlois v. Abington Housing Authority, the District Court of Massachusetts—evaluating cross-motions for summary judg­ment—found a prima facie case of disparate impact based on the com­parative demographics of the suburbs with residency preferences 52 234 F. Supp. 2d 33, 69–70 (D. Mass. 2002). The defendants were eight PHAs in predominantly white suburbs that adopted identical residency preference policies, under which Section 8 waiting lists were determined by a lottery of applicants. While residents and nonresidents of the respective municipalities had an equal chance in the lotteries, a residency preference governed the waiting lists of lottery “winners,” such that applicants who lived or worked in the PHAs’ community placed ahead of applicants who did not. Id. at 42–43. and the sur­rounding urban areas. 53 Id. at 56 (“[T]he communities in this case have significantly lower percentages of minority residents than their urban neighbors . . . [and] fewer minority residents per capita than the state average. It follows logically, then, that any policy that facially favors the resi­dents of these communities will disproportionately favor whites over minorities . . . .”). When a “community has a smaller pro­portion of minority residents than does the larger geographical area from which it draws applicants,” the court indicated, a residency prefer­ence policy “cannot but work a disparate impact on minorities.” 54 Id. at 62.

2. Intentional Discrimination Claims. — In some cases, courts have also regarded a disparity in the racial demographics of the preference area and surrounding area as evidence of intentional discrimination or equal protection violations. In Comer v. Cisneros, the Second Circuit vacated the district court’s grant of summary judgment to defendants, including the dismissal of plaintiffs’ equal protection racial discrimination claims. 55 37 F.3d 775, 795 (2d Cir. 1994). Comer, a class action brought on behalf of minority residents of Buffalo, New York, challenged a residency preference awarded to applicants who lived within one of forty-one “consortium communities” in the Buffalo sub­urbs but not to residents of the city of Buffalo. Id. at 783–84. The court affirmed those claims’ potential validity based on the stark demographic distinction between the included and excluded popula­tions. 56 Demographic statistics from 1990 indicated that the proportion of minority resi­dents was much higher within Buffalo city limits than within neighboring suburbs, and minority families comprised twenty-one percent of the defendant PHA’s Section 8 waiting list but held only three percent of the vouchers and certificates it had issued. Id. at 793. It suggested that the preference policy operated as a “proxy for race” that obstructed minorities’ efforts to “integrate into suburban life.” 57 Id. The court also called the local preference policy a “government-erected bar­rier” that “effectively blocks African-American residents of [Buffalo] from moving to the suburbs.” Id.

More recently, in United States v. Town of Oyster Bay, the Eastern District of New York determined that intentional discrimination could be plausibly inferred from strong evidence of disparate impact. 58 66 F. Supp. 3d 285, 291 (E.D.N.Y. 2014). The case, brought by the Department of Justice against the Long Island town of Oyster Bay, challenged two residency preference policies as violations of the FHA based on their “pattern or practice of discrimination against African Americans.” Id. at 286–87. The two housing programs at issue were the “Next Generation” program, which encouraged the development of housing for first-time homebuyers whose incomes approximated the town median, and the “Golden Age” pro­gram, which incentivized developers to build below-market housing for senior citizens. Id. at 287–88. Both programs offered priority to Oyster Bay residents and their children (in the case of “Next Generation”) or parents (in the case of “Golden Age”). Id.
Fewer than one percent of Oyster Bay residents who were eligible for the “Next Generation” program were African American and roughly ninety percent were white, whereas African Americans comprised approximately ten percent of the eligible popu­lation in Suffolk and Nassau Counties and more than twenty percent of the eligible popu­lation in the New York City metropolitan area. Id. at 288. Similarly, African Americans comprised no more than 0.4% of the population of Oyster Bay residents eligible for the “Golden Age” program, at least three percent of the eligible pool in Nassau and Suffolk Counties, and at least ten percent of the eligible pool in the New York City metropolitan area. Id. at 289.
The deci­sion identified stark racial disparities as an “important starting point” 59 Id. at 291 (internal quotation marks omitted) (quoting United States v. Yonkers Bd. of Educ., 837 F.2d 1181, 1221 (2d Cir. 1987)). for intentional discrimination and found that Oyster Bay’s stated goal of prioritizing its own residents could be plausibly interpreted “to suggest a discriminatory motive” in light of those demographic disparities. 60 Id. at 292. The defendant claimed that the “Next Generation” program’s purpose was “keep[ing] our children here” and “providing our young people” with opportunities to become homeowners. Id. The court denied petitioner’s motion to stay the proceedings pending the Supreme Court’s decision in Inclusive Communities regarding disparate impact liability under the FHA, 61 In Texas Department of Housing and Community Affairs v. Inclusive Communities Project, Inc., the Supreme Court affirmed the FHA’s prohibition of conduct that effectuates a racially discriminatory impact, as well as intentional discrimination. 135 S. Ct. 2507 (2015); see also supra notes 41–43 and accompanying text. because Oyster Bay’s preference policy rendered intentional discrimination a separate, cognizable cause of action. 62 Town of Oyster Bay, 66 F. Supp. 3d at 292–93.

3. Validity of Governmental Interests. — A major theme that runs throughout residency preference case law is the interrogation of governmental justifications for administering preference policies. In Chickasaw, authorities justified the “citizenship requirement” on two grounds: It allowed the city to better provide for the needs of its own low-income residents, and it prevented Chickasaw’s affordable housing from becom­ing a “dumping ground for social undesirables.” 63 United States v. Hous. Auth. of Chickasaw, 504 F. Supp. 716, 723 (S.D. Ala. 1980). The former Housing Authority Chairman insisted that the aversion to “undesirables” was not race-based but rather a desire to insulate the city from drug addicts and prostitutes. Id. at 718, 729 & n.2. The court deemed both justifications to be “legitimate” concerns that precluded any infer­ence of discriminatory intent, thereby defeating the federal government’s intentional discrimination claim. 64 Id. at 728–29. The court found county commissioners’ testimony that no intent to discriminate existed to be genuine and the proffered justifications to be “sincerely ex­pressing their actual reasons for adopting the residency requirement.” Id. at 728. Similarly, in Fayerweather v. Town of Narragansett Housing Authority, the District Court of Rhode Island found a residency preference to be rationally related to the town’s valid interest in prioritizing its own residents’ housing needs. 65 848 F. Supp. 19, 22 (D.R.I. 1994). Fayerweather is unique among the cases discussed in this section in that it did not involve claims brought under the FHA or alleg­ing racial discrimination. The case involved a constitutional right-to-travel claim. The plaintiffs, who were relegated from a “local” to “non-local” housing waitlist upon their moves from Narragansett to neighboring towns, challenged Narragansett’s residency pref­erence pol­icy, in part, as a violation of their constitutional right. Id. at 20–21.
The court relied on the distinction between bona fide and durational residency pref­erences in its decision to apply rational basis review. Id. at 21–22. A durational require­ment, the court suggested, would act as “a condition on a public benefit” and therefore implicate the fundamental right to travel, meriting strict scrutiny, whereas a bona fide re­quirement raised no issue of fundamental rights and warranted merely rational basis review. Id.

More recent case law, however, casts doubt on the rationales ap­proved by the Chickasaw and Fayerweather courts, which suggest that resi­dency preferences can be justified by a desire to prioritize local residents’ interests. The Langlois court denounced this kind of circular justification, which treats the desire to prioritize local residents as a legitimate basis for prioritizing local residents, as invalid. 66 Langlois v. Abington Hous. Auth., 234 F. Supp. 2d 33, 69 (D. Mass. 2002) (suggesting that if the court accepted reasons prioritizing local residents “as legitimate justifications, residency preferences in and of themselves would forever justify the disparate impacts that they cause”). It held that the defendants’ proffered rationales were extensions of that logic. 67 Id. The defendants’ justifications included the facilitation of current residents’ abilities to remain in their communities, the protection of administrative fees, and the “community morale” engendered by the knowledge that PHAs prioritized current resi­dents. Id. Defendant PHAs could only rebut the plaintiffs’ prima facie case of disparate impact, the court determined, by offering a “record of local conditions and needs” that justified the residency preferences and by showing that no less discriminatory alternative was available to address those needs. 68 Id. at 70. The Oyster Bay court demonstrated a similar skepticism of circular justifications by treating the defendant’s stated desire to prioritize and benefit its own residents as evidence of discriminatory purpose, given the demographic discrepancies between the preference area and surrounding vicinity. 69 See United States v. Town of Oyster Bay, 66 F. Supp. 3d 285, 292 (E.D.N.Y. 2014).

4. Case Law Patterns and Trends. — Even when courts have invalidated the specific residency preference at issue, they have taken care to affirm the general validity of such preferences. The Chickasaw court noted the “valid purpose” of prioritizing the housing needs of “established community members vis-a-vis newcomers.” 70 United States v. Hous. Auth. of Chickasaw, 504 F. Supp. 716, 731 (S.D. Ala. 1980). The eventual settlement for the Comer parties did not provide for the wholesale abolishment of resi­dency preferences in the Buffalo area but rather for an expansion of the preference area to the entire county, which effectively extended the preference to minority residents of Buffalo whom it had previously ex­cluded. 71 See Corinne Anne Carey, The Need for Community-Based Housing Development in Integration Efforts, 7 J. Affordable Housing & Community Dev. L. 85, 89 (1997) (discussing the settlement’s extension of the preference area). For an analysis of the Comer settlement, including stipulations that distinguish between minority and nonminority applicants, see id. at 89–90. A 2011 New York state court decision granting a preliminary injunction against proposed rezoning suggested that the operation of New York City’s residency preference system perpetuated segregation, but the court posited that an extension of the preference to residents of the neighboring community district “might act to correct the imbalance in the applicant pool.” 72 Broadway Triangle Cmty. Coal. v. Bloomberg, 941 N.Y.S.2d 831, 833–34, 837 (Sup. Ct. 2011). The lawsuit contended, in part, that New York City’s plan to build affordable units in Community District 1 (encompassing the predominantly white neigh­borhoods of Williamsburg and Greenpoint) as opposed to nearby Community District 3 (encompassing the predominantly African American neighborhood of Bedford-Stuyvesant) would exact a racially discriminatory impact. Id. at 833–34. These cases suggest that courts’ objections to residency preferences are usually confined to the preferences’ specific applications and that courts may cure identified problems through mod­ification rather than abolishment.

On the whole, the trajectory of the case law on residency preferences exhibits two particularly notable trends. First, the decisions indicate an increasing judicial willingness to view disparate impact not only as a harm itself but also as evidence of intent or unconstitutionality. While the influ­ence of the Oyster Bay decision should not be overstated, it does indi­cate a possible shift in the jurisprudence toward treating disparate impact as a “starting point” and probing facially neutral residency preferences for discriminatory intent. This is a far cry from the Chickasaw court’s unwillingness, thirty-four years prior, to infer discriminatory intent even from the stated purpose of keeping “undesirables” out of an all-white suburb. 73 See Hous. Auth. of Chickasaw, 504 F. Supp. at 728–29.

Second, the case law evinces increasing skepticism of circular justifi­cations and a reluctance to treat residency preferences as presumptively valid. Decisions such as Comer, Langlois, and Oyster Bay suggest that when significant demographic disparities exist, the prioritization of local resi­dents’ access to affordable housing may not be legitimate simply for rea­sons of parochialism.

C. New York’s Community Preference Policy and the Winfield Lawsuit

1. The Community Preference Policy. — New York City’s residency pref­erence system, which the city calls the “community preference policy,” gives applicants from each community district 74 New York City is made up of fifty-nine community board districts. See Winfield v. City of New York, No. 15CV5236-LTS-DCF, 2016 WL 6208564, at *2 (S.D.N.Y. Oct. 24, 2016). a preference in securing new affordable housing units within that same district, relative to appli­cants who reside in other districts. 75 Id. at *3–4. The policy was established in the late 1980s with the original stated purpose of enabling residents of low-income neighborhoods to take advantage of the city’s redevelopment efforts. 76 See Rafael Cestero, An Inclusionary Tool Created by Low-Income Communities for Low-Income Communities, N.Y.U. Furman Ctr.: Dream Revisited (Nov. 2015), http://furmancenter.org/research/iri/essay/an-inclusionary-tool-created-by-low-income-communities-for-low-income-commu [http://perma.cc/UXW3-RWPU]; see also Corinne Ramey & Laura Kusisto, Group Challenges New York City on Housing Allocations, Wall St. J. (July 7, 2015), http://www.wsj.com/articles/group-challenges-new-york-city-on-housing-allocations-1436310942 (on file with the Columbia Law Review).
In New York’s telling, the policy emerged as a response to the “demands” of low-income residents, who had “suffered from disinvestment” and spent “years . . . watching the neighborhood’s infrastructure and services deteriorate around them.” See Cestero, supra; see also FAQ, NYC Housing Connect, http://a806-housingconnect.nyc.gov/
nyclottery/lottery.html#faq [http://perma.cc/9VPV-HAU6] (last visited Oct. 20, 2017) (“[T]he community preference was established to provide greater housing opportunities for long-time residents of New York City neighborhoods where [city agencies] have made a significant investment in housing.”).

New York City facilitates the development of new affordable housing units through incentives for developers, including direct subsidies, site acquisition, and tax credits and exemptions. 77 See Winfield, First Amended Complaint, supra note 45, at 13. Developers are required to select residents for affordable units by soliciting applicants and conducting a lottery 78 Winfield, 2016 WL 6208564, at *2. Through the lottery, each applicant is assigned a randomized number in the “log” for that particular development. See Winfield, Motion to Dismiss, supra note 5, at 8. and must consider certain mandated “set-asides” or prefer­ences when assigning units. 79 Declaration of Commissioner Vicki Been in Support of Defendant’s Motion to Dismiss at 3–4, 8, Winfield, No. 15-CV-05236-LTS [hereinafter Winfield, Been Declaration].
In addition to the community preference, which applies to fifty percent of units, seven percent of units are designated for applicants with disabilities (five percent for those with mobility impairments and two percent for those with hearing or visual impairments). An additional five percent of units must be assigned, when possible, to city employees. As a result, only thirty-eight percent of units are not subject to set-asides and are available on equal terms to all applicants who reside in New York City and meet the income and family-size criteria. N.Y.C. Dep’t of Hous. Pres. & Dev., Marketing Handbook: Policies and Procedures for Resident Selection and Occupancy 34–35 (2016), http://www1.nyc.gov/
assets/hpd/downloads/pdf/developers/marketing-handbook.pdf [http://perma.cc/
F6VK-6CF3].
New York City expanded the application of the community preference from thirty percent to fifty percent of units in 2002, and it has remained at that level since. 80 See Winfield, First Amended Complaint, supra note 45, at 14. An intermunicipal res­idency preference policy also applies: All applicants residing in New York City must be processed and assigned before any nonresidents. 81 N.Y.C. Dep’t of Hous. Pres. & Dev., supra note 79, at 35–36. The citywide preference works slightly differently for the disability and municipal employee set-asides: Residents who fall into those categories are processed before eligible nonresidents, but eligible nonresidents are processed next (rather than residents who do not meet the desig­nated criteria). Id. at 34.

2. The Winfield Lawsuit. — In July 2015, three plaintiffs represented by the Anti-Discrimination Center filed a lawsuit against New York City alleging that the community preference policy perpetuates segregation and violates the FHA and New York City’s Human Rights Code. 82 See Winfield, First Amended Complaint, supra note 45, at 2, 4. The plaintiffs are African American women, residents of New York City, and income eligible for New York City’s affordable housing. Each entered lot­teries for new affordable housing developments located in Manhattan community districts in which she did not reside and was not selected. 83 Id. at 4. The complaint argues that the city’s “outsider-restriction policy” impairs low-income residents’ mobility, making it more difficult for them to ob­tain housing in “neighborhoods of opportunity.” 84 Id. at 2; Andrea McArdle, Winfield v. City of New York: Testing the Limits of Disparate-Impact Liability After Texas Department of Housing and Community Affairs v. The Inclusive Communities Project Inc., 24 J. Affordable Housing & Community Dev. L. 287, 295 (2015) [hereinafter McArdle, Winfield].

The complaint illustrates New York City’s residential segregation at the community district level. 85 See Winfield, First Amended Complaint, supra note 45, at 8–10. The city’s overall population, according to 2010 census data, is 22.8% African American and 28.6% Latino, but in fourteen community districts (including two of the three to which plaintiffs applied) the combined African American and Latino population is under twenty percent, and in seventeen other districts those two demographic groups account for at least eighty percent of the population. Id. at 9–10.
The “relative difference” between the citywide African American population and the African American population within a given community district—measured as the difference between the two percentages divided by the citywide percentage—exceeds fifty percent in forty-two of New York City’s fifty-nine community districts. Id. The relative dif­ference between the citywide and community district Latino populations exceeds fifty per­cent in thirty-three community districts. Id. The complaint argues that measuring demo­graphic disparity by relative difference “puts the scope of a variation in context.” Id. at 9 n.3.
Given these patterns, the lawsuit alleges, the city’s preference policy renders it liable for violating the FHA under both a disparate impact theory and an intentional discrimination theo­ry. 86 Id. at 29–30. It claims that New York City’s segregated and discriminatory history, its rejection of alternative policies that would promote integration, and its continued implementation of the “outsider-restriction policy” despite awareness of (or deliberate indifference to) its segregative impact “de­monstrate that the . . . policy constitute[s] intentional discrimination.” 87 Id. at 27–28.

New York maintains that longstanding residents of gentrifying neigh­borhoods have earned preferential opportunities to remain and enjoy the benefits of revitalization, because they endured disinvestment and “persevered through years of unfavorable living conditions.” 88 Winfield, Motion to Dismiss, supra note 5, at 5; see also Winfield, Been Declaration, supra note 79, at 4 (decrying the unfairness of displacement “for people who have en­dured years of unfavorable conditions, and who deserve a chance to participate in the re­naissance of their neighborhoods”).
Media coverage of the community preference policy highlights a related justification by focusing on the particular interests of elderly (and often long-term) residents in avoiding displacement from their neighborhoods. See, e.g., Ramey & Kusisto, supra note 76 (“Younger residents, regardless of race, were inclined to move, whereas older ones said they had put down roots that were more important than a new apartment.”); Bobbie Sackman, CityViews: How Community Preference Supports the Right to Age in Place, City Limits (Sept. 14, 2016), http://citylimits.org/2016/09/14/cityviews-how-community-preference-supports-the-right-to-age-in-place/ [http://perma.cc/UEX6-T6D2] (describing the value of the community preference policy for older residents and arguing that in­creased “mobility” is less likely to serve their interests).
New York’s reliance on longstanding residency as a justification for community preferences is undermined, potentially, by the nondurational nature of the policy. The Winfield plaintiffs criticize the policy on these grounds, noting that it assigns preferences “regardless of length of residency in the community district” and “even if [the applicant] established residen­cy in the community district on the final day of the application period.” 89 Winfield, First Amended Complaint, supra note 45, at 15. Durational residency preferences, however, are disfavored by HUD 90 See 24 C.F.R. § 5.655(c)(1)(C)(v) (2016) (“A residency preference must not be based on how long an applicant has resided or worked in a residency preference area.”); id. § 982.207(b)(iv) (same). and may violate the constitutional right to travel and migration. 91 See infra note 188 and accompanying text (discussing judicial decisions indi­cating that durational residency preferences may contravene the constitutional right to travel). As a result, the city may be unable to legally tailor community preferences to the “longstanding resident” argument through which it attempts to justify them.

New York City’s additional justification for its policy is a pragmatic one: It serves as an effective means of “overcoming local resistance” to development and construction and therefore facilitates the creation of affordable housing. 92 Winfield, Motion to Dismiss, supra note 5, at 4. Community preferences are often popular with lo­cal residents and community organizations, who might otherwise oppose affordable housing development in their own neighborhoods. 93 See, e.g., Brad Lander, N.Y.C. Councilmember, My Statement on the Lawsuit Against NYC’s 50% Community Preference Policy for Affordable Housing (July 8, 2015), http://bradlander.nyc/news/updates/my-statement-on-the-lawsuit-against-nyc-s-50-community-
preference-policy-for-affordable [http://perma.cc/2GPA-KQ7K] (citing the policy as criti­cal to cultivating “community support” for new developments and as “one meaningful reason why” New York City sees relatively little NIMBY-like opposition to affordable housing).
As the city pursues a significant initiative to increase affordable housing, 94 See Andrea McArdle, Challenges to Achieving New York City’s Affordable Housing Goals: Reconciling Mandatory Inclusionary Housing, Community Preference Requirements, and Fair Housing Laws, City Square by Fordham Urb. L.J. (May 26, 2016), http://urbanlawjournal.com/challenges-to-achieving-new-york-citys-affordable-housing-goals-reconciling-mandatory-inclusionary-housing-community-preference-requirements-and-fair-housing-laws/ [http://perma.cc/8VGN-TC4P] (describing New York City’s “am­bitious affordable housing initiative” and “continuing community opposition” to it); see also Mireya Navarro, Segregation Issue Complicates de Blasio’s Housing Push, N.Y. Times (Apr. 14, 2016), http://www.nytimes.com/2016/04/15/nyregion/segregation-new-york-city-and-de-blasio-affordable-housing.html (on file with the Columbia Law Review) (discuss­ing New York City’s push to build 80,000 affordable housing units over a decade). it may rely particularly heavily on the neighborhood-level goodwill that community preferences generate 95 In fact, community activists and local politicians opposing a recent development effort in Crown Heights, Brooklyn, have called for an increase in the community pref­erence policy for that development, from fifty percent to eighty percent, as a condition of their support. See Megan Carpentier, Brooklyn Lawmakers Enter Gentrification Feud over Crown Heights Neighborhood, Guardian (Oct. 19, 2016), http://www.theguardian.com/
us-news/2016/oct/19/brooklyn-gentrification-crown-heights-bedford-union-armory [http://
perma.cc/7QSS-337H].
—a factor that the Winfield complaint dismisses as mere political expedience. 96 See Winfield, First Amended Complaint, supra note 45, at 26.

In October 2016, the Winfield court denied New York City’s motion to dismiss. 97 Winfield v. City of New York, No. 15CV5236-LTS-DCF, 2016 WL 6208564, at *1 (S.D.N.Y. Oct. 24, 2016). The court held that the plaintiffs pleaded sufficient facts to allege both their disparate impact and intentional discrimination theories. 98 Id. at *7. The decision ascribed to the community preference policy the “very pur­pose” of preserving “the existing racial and ethnic makeup of local com­munities.” 99 Id. at *6. As of March 2018, the litigation is ongoing with the parties engaged in discovery. 100 See generally Winfield v. City of New York, No. 15-cv-05236, 2018 WL 840085 (S.D.N.Y. Feb. 12, 2018) (setting forth the most recent order from the magistrate judge, as of this Note’s publication, and acknowledging the long, contentious, and continuing discovery process).

D. San Francisco’s Neighborhood Resident Housing Preference

1. Ordinance Enacted. — Unlike New York’s community preference policy, which New York City predominantly justifies in race-neutral terms, San Francisco enacted its “resident housing preference” ordinance in November 2015 with the express purpose of addressing race-specific gen­trification and displacement issues. 101 Hannah Albarazi, Supes Shift Housing Preference Toward Neighborhood Residents, SFBay (Nov. 17, 2015), http://sfbay.ca/2015/11/17/supes-shift-housing-preference-toward-neighborhood-residents/ [http://perma.cc/33BH-CZZU]. Interest­ingly, San Francisco officials have cited New York City’s preference policy as a model and lamented that San Francisco is “decades behind” New York in addressing displacement concerns. Id. Under the ordinance, the lotteries for forty percent of new affordable housing units prioritize applicants who reside either within the project’s supervisorial district or within a one-half-mile “buffer zone.” 102 Neighborhood Resident Housing Preference, S.F. Mayor’s Office of Hous. & Cmty. Dev., http://sfmohcd.org/neighborhood-resident-housing-preference [http://perma.cc/6ART-UYMN] (last visited Oct. 20, 2017).

Local politicians and civil rights advocates lauded the ordinance as a possible antidote to the “alarming rate of displacement” among San Francisco’s African American population, which declined from 13.4% in 1970 to 5.5% in 2014. 103 See Albarazi, supra note 101 (quoting a local NAACP board member who called the ordinance a “step toward doing the right thing”). Even those city officials who opposed the ordi­nance seemed to be more concerned with its particulars than its principle. 104 One detractor reportedly based a “no” vote on a concern that the designation of forty percent of units for neighborhood residents was too high, and other supervisors expressed concern about the handling of neighborhoods that straddled multiple supervisorial districts. Id.

2. San Francisco and HUD Clash. — In August 2016, HUD, then un­der the leadership of Obama appointee Julián Castro, denied San Francisco’s proposal to implement the supervisorial district preference plan for the Willie B. Kennedy Apartments, 105 Letter from Gustavo Velasquez, Assistant Sec’y for Fair Hous. & Equal Opportunity, U.S. Dep’t of Hous. & Urban Dev., to Olson Lee, Dir., S.F. Mayor’s Office of Hous. & Cmty. Dev. 1 (Aug. 3, 2016) [hereinafter, Letter from Velasquez, Aug. 3, 2016] (on file with the Columbia Law Review). a new affordable housing development for senior citizens located in the historically African American neighborhood Western Addition. 106 See Editorial, HUD Agrees to Offer Residents in Areas with High Displacement Priority in Federal Housing, S.F. Chron. (Sept. 24, 2016), http://www.sfchronicle.com/opinion/editorials/article/HUD-is-right-to-allow-SF-try-a-neighborhood-9242954.php (on file with the Columbia Law Review).
The Tenderloin Neighborhood Development Corporation (TNDC) developed the ninety-eight-unit building, named for a deceased San Francisco community leader and city official, into affordable senior-citizen housing. See New Senior Housing Named for Longtime San Francisco Supervisor, Tenderloin Neighborhood Dev. Corp. (Sept. 21, 2016), http://www.tndc.org/news/new-senior-housing-named-for-longtime-san-francisco-supervisor/ [http://perma.cc/46KN-5CLT]. Because San Francisco and TNDC planned to rehabilitate the building under HUD’s Rental Assistance Demonstration program, its resident-selection policies were subject to HUD’s approval. For discussion of these res­ident-selection policies, see generally Tenderloin Neighborhood Dev. Corp., Affirmative Fair Housing Marketing Plan & Resident Selection Criteria: Willie B. Kennedy (2016), http://www.tndc.org/wp-content/uploads/AFHMP-RSC-WBK-ALL-All-Verbiage-FINAL-08-18-2016.pdf [http://perma.cc/SA3A-667U].
HUD indicated that the proposed policy “could limit equal access to housing and perpetuate seg­regation,” and that it “may also violate the Fair Housing Act.” 107 Letter from Velasquez, Aug. 3, 2016, supra note 105, at 1. HUD did, however, conditionally approve a citywide preference policy that extended priority for one hundred percent of the units to applicants who lived or worked within San Francisco. Id.

The outcry against HUD’s rejection of the resident preference plan was widespread and vehement. Civil rights advocates decried the deci­sion, and the president of the local NAACP chapter called on the city to fight back in court. 108 See J.K. Dineen & Emily Green, HUD Gets Pushback from Supervisors, NAACP on Housing Ruling, S.F. Chron. (Aug. 18, 2016), http://www.sfchronicle.com/bayarea/
article/HUD-gets-pushback-from-supervisors-NAACP-on-9171902.php (on file with the Columbia Law Review).
Both local and national politicians lambasted HUD’s decision and lobbied HUD on the policy’s behalf. 109 See J.K. Dineen, Feds Reject Housing Plan Meant to Help Minorities Stay in SF, S.F. Chron. (Aug. 17, 2016), http://www.sfchronicle.com/politics/article/Feds-reject-housing-plan-meant-to-help-minorities-9146987.php (on file with the Columbia Law Review) (de­scribing San Francisco city officials’ criticisms of HUD’s failure to recognize the plight of San Francisco’s African American community); Letter from Dianne Feinstein, U.S. Senator, to Julián Castro, Sec’y, U.S. Dep’t of Hous. & Urban Dev. 2 (Aug. 19, 2016) (on file with the Columbia Law Review) (calling on HUD to work with San Francisco to develop “anti-displacement strategies that pass legal and regulatory muster”); Letter from Herrera, supra note 4, at 1 (calling HUD’s decision “wrong as a matter of law and public policy for several reasons”); Letter from Nancy Pelosi, Democratic Leader, U.S. House of Representatives, to Julián Castro, Sec’y, U.S. Dep’t of Hous. & Urban Dev. 1 (Aug. 31, 2016) (on file with the Columbia Law Review) (expressing “strong[] disagree[ment]” with HUD’s decision and urging its reversal). Implicit in the reaction was a suggestion that the FHA’s traditional integrationist aims might be inapt in the face of rapid gentrification and that concerns about segregation within San Francisco should be superseded by the con­cern that minority populations were being displaced from the city entire­ly. As San Francisco’s City Attorney articulated in a letter to HUD, “San Francisco’s Plan addresses gentrification forces that were unknown when the Fair Housing Act was passed in 1968, and is not what Congress in­tended the Fair Housing Act to address.” 110 Letter from Herrera, supra note 4, at 2.

3. Displacement Preference. — On September 21, 2016, HUD re­affirmed its disapproval of the neighborhood-based preference but approved an alternative plan that the city had proposed: Forty percent of units in the Willie B. Kennedy Apartments would be subject to a prefer­ence for San Francisco residents at an “elevated risk of displacement.” 111 Letter from Gustavo Velasquez, Assistant Sec’y for Fair Hous. & Equal Opportunity, U.S. Dep’t of Hous. & Urban Dev., to Edwin Lee, Mayor of S.F. 1 (Sept. 21, 2016) (on file with the Columbia Law Review). This pref­erence was extended to all income-eligible lottery applicants who resided in “neighborhoods undergoing extreme displacement pres­sure,” as determined by census tract. 112 News Release, Office of the Mayor, City of S.F., Mayor Lee Applauds Federal Approval of Anti-Displacement Housing Policy (Sept. 22, 2016) (on file with the Columbia Law Review). Qualifying census tracts were identified based upon research by University of California at Berkeley. Id. Residents from at least five neighbor­hoods, including Western Addition, were eligible. 113 J.K. Dineen, Federal Agency OKs Preferences at New SF Senior Housing Complex, SFGate (Sept. 22, 2016), http://m.sfgate.com/news/article/Federal-agency-OKs-preferences-at-new-SF-senior-9238463.php [http://perma.cc/2XP4-T3CM] [hereinafter Dineen, Federal Agency OKs Preferences].
Preliminary data from the Willie B. Kennedy housing lottery indicated that twenty-five of the ninety-eight preference-based units were assigned to African American families, whereas only fourteen would have gone to African American households without the preference. See J.K. Dineen, Preferences Help Keep Black Seniors in SF, S.F. Chron. (Oct. 17, 2016), http://www.sfchronicle.com/bayarea/article/Antidisplacement-preference-boosts-blacks-at-9979037.php (on file with the Columbia Law Review).

City officials celebrated the decision as a “monumental victory” and downplayed the distinction between the policy they had initially pro­posed and the one that HUD approved. 114 News Release, Office of the Mayor, supra note 112; see also Richard Gonzales, Feds to Allow Preferences for Low-Income Applicants in S.F. Housing Complex, NPR (Sept. 23, 2016), http://www.npr.org/2016/09/23/495237494/feds-to-allow-preferences-for-low-income-applicants-in-s-f-housing-complex (on file with the Columbia Law Review) (“San Francisco officials are claiming victory in a dispute with federal housing officials regarding a city effort to combat gentrification.”); Ericka Cruz Guevarra & Matt Beagle, Federal Officials Approve Preferences at New Senior Housing Complex in S.F., KQED (Sept. 22, 2016), http://ww2.kqed.org/news/2016/22/federal-officials-approve-preferences-at-new-senior-housing-complex-in-s-f/ [http://perma.cc/W9VN-S4CX] (quoting a city supervisor’s claim that the new policy is “just as good and just as effective” as the original proposal). National politicians and jour­nalists joined in hailing the new preference plan as a triumph and a model for other cities. 115 See, e.g., Guevarra & Beagle, supra note 114 (reporting Sen. Feinstein’s state­ment that the policy would serve as a model for other California cities); Editorial, HUD Agrees to Offer Residents in Areas with High Displacement Priority in Federal Housing, S.F. Chron. (Sept. 23, 2016), http://www.sfchronicle.com/opinion/editorials/article/
HUD-is-right-to-allow-SF-try-a-neighborhood-9242954.php (on file with the Columbia Law Review) (last updated Sept. 24, 2016) (indicating, incorrectly, that HUD decided to “support the neighborhood preference provision” and trumpeting the decision as a “his­toric shift” and a “flashing light for cities across the country”).
The celebration over HUD’s acquiescence to an anti-displacement policy threatens to obscure the significant distinction between the policy that HUD rejected and the one that it approved. 116 Among public officials, only Representative Pelosi emphasized the distinction: She applauded the anti-displacement preference as an “achievement” but maintained that “preserving the principle of neighborhood preference will be essential” in protecting mi­nority populations against displacement. Guevarra & Beagle, supra note 114. HUD’s response to the initial proposal demonstrates a seeming indiffer­ence toward the specific population that a residency preference is de­signed to exclude or to benefit. The anti-displacement strategy may in­deed be a model for future affordable housing preferences; if so, it is likely because the law does not recognize a distinction between a policy like San Francisco’s and those enacted by white suburban enclaves.

II. Residency Preferences as Anti-Displacement Efforts: Inverting or Extending a Discriminatory Practice?

In the face of rapid urban gentrification and rising housing costs, affordable housing is in high demand and low-income communities face increasing displacement pressures. 117 See infra notes 126–131 and accompanying text. Against this backdrop, local gov­ernments may turn to residency preferences as an anti-displacement tool. In one sense, these residency preferences share an inherent parochialism with their exclusionary counterparts. To treat displacement as a problem is to presume that those who are currently in a place possess a superior claim to it. Yet there is also something distinct about residency prefer­ences deployed to preserve the very communities that such preferences have often been exercised to exclude: low-income, urban-dwelling racial minorities. Residency preferences in New York City and San Francisco purport to reorient a discriminatory tool toward an inclusionary end.

Although preferences designed to preserve minority communities arguably serve a different objective than those that are designed to ex­clude such communities, the distinction may not be legally meaningful. The “anti-displacement” policies, like their exclusionary counterparts discussed in section I.B, strive to keep existing residents in place. 118 See supra notes 76, 88–93, 101–102, and accompanying text (describing the goals of the New York and San Francisco policies). In doing so, they reinforce existing housing patterns and demographics. Therein lies the problem: Intracity residency preferences may be a valua­ble tool for local governments to combat displacement pressures on low-income minority residents, but they are likely not a valid one. If courts treat affirmative-action-minded preferences in a manner consistent with existing doctrine, such policies may be subject to equal or even greater legal vulnerability than their more classically exclusionary counterparts.

This Part examines both the value and vulnerabilities of neighbor­hood-level residency preferences enacted to preserve minority communi­ties. Section II.A describes the underlying displacement pressures that motivate and inform these policies and discusses residency preferences’ potential to mitigate those effects. Section II.B examines the potential legal validity of such policies under each of the prominent applicable federal doctrines: FHA disparate impact liability, FHA intentional dis­crimination liability, equal protection law, and the right to travel.

A. Residency Preference Policies as an Anti-Displacement Measure

1. Urban Gentrification and Displacement. — The relationship between gentrification and displacement is at once intuitive and elusive. An influx of higher-income residents into a community and the ensuing elevation of the local cost of living can compel preexisting, lower-income residents to relocate. 119 See, e.g., Chase M. Billingham, The Broadening Conception of Gentrification: Recent Developments and Avenues for Future Inquiry in the Sociological Study of Urban Change, 29 Mich. Soc. Rev. 75, 92–93 (2015) (noting this effect and suggesting that it is in­disputably “gentrification-induced displacement”). Even so, the displacement narrative of gentrification exists alongside an opposing (though perhaps not incompatible) narrative of “social mixing,” which suggests that middle-income residents’ migration into lower-income neighborhoods yields increased integration and en­hances community resources to the benefit of the preexisting residents who remain. 120 See Loretta Lees, Gentrification and Social Mixing: Towards an Inclusive Urban Renaissance?, 45 Urb. Stud. 2449, 2450–51 (2008) (describing and criticizing the “social mixing” theory of gentrification); see also Kathe Newman & Elvin K. Wyly, The Right to Stay Put, Revisited: Gentrification and Resistance to Displacement in New York City, 43 Urb. Stud. 23, 26 (2006) (acknowledging challenges in resolving the distinction between “equitable reinvestment” and displacement).

Although concerns and research about gentrification date back to the 1960s, by most accounts gentrification in the United States (and the attention paid to it) became increasingly pervasive in the late 1990s and ensuing years. 121 See Billingham, supra note 119, at 76 (noting “substantial gains in economic status” of neighborhoods since the 1990s); Edward Goetz, Gentrification in Black and White: The Racial Impact of Public Housing Demolition in American Cities, 48 Urb. Stud. 1581, 1581 (2011) [hereinafter Goetz, Gentrification] (noting that the recent wave of gentrification was largely driven by public investment efforts); Newman & Wyly, supra note 120, at 27 (“Interest in displacement re-emerged towards the end of the 1990s as a new gentrification wave once again pushed these questions to the forefront.”). The effects of gentrification are varied, context depend­ent, and difficult to quantify—in part because studies of displacement pursue the difficult task of measuring absence and because displaced in­dividuals are difficult to identify, locate, and survey. 122 See Elvin Wyly et al., Displacing New York, 42 Env’t & Plan. 2602, 2603 (2010) (identifying the challenges of measuring displacement); see also Jackelyn Hwang, Gentrification in Changing Cities: Immigration, New Diversity, and Racial Inequality in Neighborhood Renewal, 660 Annals Am. Acad. Pol. & Soc. Sci. 319, 323 (2015) (describing how most studies measuring gentrification “lack direct indicators of neigh­borhood upgrading and are unable to distinguish gentrification from other forms of neighborhood ascent”). Additionally, the distinction between forced relocation and voluntary relocation is not al­ways clear-cut. The decision to relocate in response to rising costs may fall along a continuum of voluntariness 123 See Wyly et al., supra note 122, at 2603 (suggesting that many housing decisions are neither wholly voluntary nor involuntary). and may be attributable to a range of interrelated and indirect factors. 124 See Mark Davidson, Spoiled Mixture: Where Does State-Led ‘Positive’ Gentrification End?, 45 Urb. Stud. 2385, 2388 (2008) (describing the indirect forms that displacement may take). Though displacement can often be directly attributed to a surging housing market, it may also result from gentrification-induced actions such as housing demolition, evictions, and redevelopment. 125 See Newman & Wyly, supra note 120, at 27. Moreover, low-income residents who remain in gentrifying neighborhoods may suffer from the loss of neighborhood culture, community social networks, and local resources. See id.; see also Davidson, supra note 124, at 2389–92 (describing indirect economic, community, and neighborhood resource displacements).

The prevailing understanding is that gentrification causes displace­ment, 126 See, e.g., Lees, supra note 120, at 2457 (describing gentrification as “part of an aggressive, revanchist ideology designed to retake the inner city for the middle classes” and noting several studies that have concluded it leads to displacement); Newman & Wyly, supra note 120, at 45–46, 51 (describing the “tremendous pressure” that New York City’s rapid gentrification in the 1990s placed on low-income residents and theorizing that empirical studies underestimate displacement figures); Wyly et al., supra note 122, at 2620 (describing large-scale displacement in New York City following neighborhood re­vitalization). But see Lance Freeman & Frank Braconi, Gentrification and Displacement: New York City in the 1990s, 70 J. Am. Plan. Ass’n 39, 48 (2004) (finding lower mobility rates in gentrifying neighborhoods). with the burden often falling disproportionately on the lowest-income residents of gentrifying neighborhoods. 127 See Billingham, supra note 119, at 92 (“[T]he decisions of households and small businesses to relocate from gentrifying areas occur at the margin, with those least able to afford the escalating costs likely to be the first to depart.”). Displacement is most often studied at the neighborhood level: Certain neighborhoods become sites of displacement, others become destinations for displaced popula­tions, and still others fill both roles. 128 See Wyly et al., supra note 122, at 2612–14 (discussing “origin” and “destination” neighborhoods for forced displacement in New York City and identifying Bedford-Stuyvesant in Brooklyn as one example of a neighborhood that fills both roles). However, displacement also occurs at the municipal and regional levels. 129 See Billingham, supra note 119, at 80–81 (disputing the characterization of gen­trification as an exclusively neighborhood-level phenomenon). Low-income minority populations are more likely than low-income white populations to live in concentrated poverty, 130 Edward G. Goetz, Desegregation in 3D: Displacement, Dispersal, and Development in American Public Housing, 25 Housing Stud. 137, 138 (2010). and urban displacement disproportionately affects African Americans. 131 See Goetz, Gentrification, supra note 121, at 1583 (noting that gentrification pro­duces both race-based and class-based demographic changes).

Gentrifying neighborhoods often undergo stark demographic transi­tions in both socioeconomic and racial composition. A 2016 report on the effects of gentrification in New York City between 1990 and 2010 identified fifteen of fifty-five city neighborhoods as “gentrifying,” meaning that they were low-income areas in 1990 that experienced rent growth above the city median over the ensuing two decades. 132 N.Y.U. Furman Ctr., Focus on Gentrification, in State of New York City’s Housing and Neighborhoods in 2015 4, 5 (2016), http://furmancenter.org/files/sotc/Part_1_
Gentrification_SOCin2015_9JUNE2016.pdf [http://perma.cc/72DQ-PHYN].
Between 1990 and 2010–2014, mean household rents in gentrifying neighbor­hoods increased by 34.3%—over twelve percent more than the citywide increase of 22.1%. 133 Id. Average household income among New York City residents, adjusted for inflation, remained relatively steady over the same period 134 Id. at 9. Inflation-adjusted average household income in New York City was $78,500 in 1990 and $79,950 in 2010–2014. Id. but rose by nearly fourteen percent in gentrifying neighbor­hoods. 135 See id. In gentrifying neighborhoods, the average household income was $51,400 in 1990 and $58,550 in 2010–2014. Id. Moreover, during the same period, the white population in gentrifying areas increased, despite the fact that it declined significantly within the city as a whole. 136 Id. at 12. The citywide white population decreased from 43.4% in 1990 to 33.4% in 2010, but in gentrifying neighborhoods it increased from 18.8% in 1990 to 20.6% in 2010. Id. Meanwhile, the black population declined very slightly citywide while declining steeply in gentrifying areas. 137 Id. The city’s overall black population was 25.6% in 1990 and 23.6% in 2010 but decreased from 37.9% to 30.9% in gentrifying neighborhoods over the same period. Id. Simi­larly, in Western Addition, the San Francisco neighborhood of the Willie B. Kennedy Apartments, the African American percentage of the popula­tion declined from roughly eighty percent in 1970, to thirty percent in 2000, 138 See Ann-Marie Alcántara, New Map Shows the Decline of SF’s Black Population, Bold Italic (Dec. 17, 2014), http://thebolditalic.com/new-map-shows-the-decline-of-sf-s-black-population-the-bold-italic-san-francisco-651aba4e199a [http://perma.cc/UJT2-P7TY] (summarizing demographic changes in census tract 158 according to maps avail­able from the Anti-Eviction Mapping Project); see also Loss of Black Population, Anti-Eviction Mapping Project, http://www.antievictionmappingproject.net/black.html [http://
perma.cc/AXE6-H9RE] (last visited Feb. 9, 2018).
to fifteen percent in 2010. 139 S.F. Planning Dep’t, San Francisco Neighborhoods Socioeconomic Planning Profiles 78 (2011), http://sf-planning.org/sites/default/files/FileCenter/Documents/
8501-SFProfilesByNeighborhoodForWeb.pdf [http://perma.cc/6PKP-NNEY].

2. Potential Value of Residency Preferences. — Research indicates that public interventions such as rent regulation and subsidized housing are the most effective way to counterbalance displacement pressures. 140 See Newman & Wyly, supra note 120, at 47–48 (discussing the public efforts that help enable at-risk residents to combat displacement pressures). With affordable housing in high demand as gentrification pressures mount, the application of residency preferences is a topic of paramount con­cern—both for residents who wish to take advantage of the preferences to avoid displacement and for those seeking to relocate to neighbor­hoods in which the preferences limit their ability to obtain affordable housing.

Residency preferences operate to the benefit of existing low-income residents in the neighborhoods in which such preferences are implemented. Because racial minorities in the United States experience dispropor­tionately high levels of poverty, 141 According to a 2017 report from the U.S. Census Bureau, the 2016 poverty rate was just 8.8% for non-Hispanic white U.S. residents, whereas Hispanic U.S. residents and black U.S. residents had poverty rates of 19.4% and 22.0%, respectively. U.S. Census Bureau, Income and Poverty in the United States: 2016 12 (2017), http://www.census.gov/
content/dam/Census/library/publications/2017/demo/P60-259.pdf [http://perma.cc/
K6SW-4JPA].
and because the racial wealth gap is particularly severe in urban areas, 142 For the period spanning 2011–2015, the U.S. Census Bureau reports that non-Hispanic white residents of urban areas had a poverty rate 13.7% below the poverty rate of Hispanic residents of urban areas and 16.4% below the poverty rate of black residents of urban areas. In rural areas, by contrast, the non-Hispanic white poverty rate was 13.1% below the Hispanic poverty rate and 15.9% below the black poverty rate. See Alemayehu Bishaw & Kirby G. Posey, A Comparison of Rural and Urban America: Household Income and Poverty, U.S. Census Bureau: Blog (Dec. 8, 2016), http://www.census.gov/newsroom/
blogs/random-samplings/2016/12/a_comparison_of_rura.html [http://perma.cc/6QR3-AQYR].
some interested parties frame the operation of intracity preferences as a civil rights issue. In San Francisco, local politicians expressed particular frustration with HUD’s treatment of the preference policy as a discriminatory device rather than an inclu­sionary tool. 143 See supra section I.D.2. The policy, they insisted, would work to the ad­vantage of minority communities. 144 See Letter from Herrera, supra note 4, at 2 (“The Plan takes a tool that commu­nities used in the past to keep protected minorities out and flips it on its head, to help residents remain in their neighborhood instead.”). The national media seized on this theme with articles that painted the policy as a lifeline for minority communities and cast its potential contravention of the FHA as an unfor­tunate paradox. 145 See, e.g., Emily Badger, Why Affordable Housing in a Black Neighborhood May Not Help Black Residents, Wash. Post (Aug. 19, 2016), http://www.washingtonpost.com/
news/wonk/wp/2016/08/19/why-affordable-housing-in-a-black-neighborhood-may-not-help-black-residents/?utm_term=.402ab9cb8b40 [http://perma.cc/82XM-Z5LT] (de­scribing the policy as an inversion of an exclusionary practice designed to help minorities remain in their communities); Richard Gonzales, How ‘Equal Access’ Is Helping Drive Black Renters Out of Their Neighborhood, NPR (Sept. 16, 2016), http://www.npr.org/
2016/09/16/494266208/how-equal-access-is-helping-drive-black-renters-out-of-their-neighborhood [http://perma.cc/6XUK-S6J4] (implying that the FHA—a “law designed to give African-Americans a fair shake in getting housing”—now causes their potential displacement).

In New York, where the preference applies to all community districts regardless of demographics, the city and its allies also emphasize the pol­icy’s particular value to minority communities. 146 See, e.g., Winfield, Motion to Dismiss, supra note 5, at 4 (describing the community preference policy as a “critical tool to ensuring that low income households, who are often disproportionately racial and ethnic minorities, are able to stay in [their] neighborhoods” when housing prices rise); Cestero, supra note 76 (deeming the policy essential to New York’s racial diversity, as “large numbers of black and Latino households [are] displaced from neighborhoods with rising economic fortunes”). One housing developer, for example, advertised that the preference would “help the area retain its traditional Latino identity,” 147 El Barrio’s Artspace PS109, Facebook, http://www.facebook.com/pg/
artspacePS109/about/?ref=page_internal [http://perma.cc/XPU5-LMXH] (last visited Oct. 19, 2017).
and the city contends that abolishing the preference policy “would turn the FHA on its head.” 148 Winfield, Motion to Dismiss, supra note 5, at 3. Certain com­mentators endorse a targeted application of the policy but balk at its extension to more affluent and majority-white neighborhoods. 149 See, e.g., Sam Tepperman-Gelfant, Local Preferences Require Local Analysis, N.Y.U. Furman Ctr.: Dream Revisited (Nov. 2015), http://furmancenter.org/research/
iri/essay/local-preferences-require-local-analysis [http://perma.cc/KTY2-EZTD] (arguing the preference policy should apply only in neighborhoods “meet[ing] some minimum threshold of racial diversity”).
However, both sides of the Winfield litigation reject the notion that the prefer­ence should be applied only to neighborhoods with high concentrations of low-income minority residents. The policy’s proponents suggest that it serves low-income minorities even in whiter and more affluent neigh­borhoods, 150 See Cestero, supra note 76 (suggesting community preferences are particularly essential in New York’s more affluent neighborhoods, which possess lower-income popu­lations crucial to those neighborhoods’ character and cultural fabric). while its detractors insist that its segregative effect harms racial minorities in any setting. 151 See Craig Gurian, Building Justice: How Community Preferences Enforce Racial Segregation in NYC, City Limits (Sept. 6, 2016), http://citylimits.org/2016/09/06/
building-justice-how-community-preferences-enforce-racial-segregation-in-nyc/ [http://
perma.cc/C238-FG6J] (“There is no good kind of segregation. It’s not the good kind to be stuck with under-performing schools or to have less access to medical care.”).

Clearly, residency preference policies can operate to preserve minor­ity communities in the face of gentrification pressures—whether one un­derstands that to be a desirable result is another matter. For cities with this goal, these policies may prove to be an appealing tool for mitigating displacement and preserving racial and socioeconomic diversity. The effectiveness of such policies, however, will depend upon their legal validity.

B. Legal Vulnerabilities Faced by Intracity, Anti-Displacement Residency Preferences

This section extrapolates from existing case law to assess how neigh­borhood-level preferences will fare under each of the major grounds for legal challenge. Because the existing law deals almost exclusively with challenges to residency preferences in predominantly white communities, 152 See supra section I.B. it is uncertain whether courts will interpret intracity, anti-displacement preferences as subject to the same legal standards. The most likely sce­nario, however, is that courts will take a traditionally antagonistic approach to such policies’ furtherance of existing racial demographics in segregated neighborhoods. As a result, neighborhood-level residency preferences enacted to mitigate gentrification-induced displacement will likely bear the same risk of disparate-impact-based in­validation as their exclusionary precursors and possibly an elevated risk of intentional-discrimination-based invalidation.

This section examines how neighborhood-level residency prefer­ences can be understood in the context of each of their four major fed­eral legal obstacles: FHA disparate impact liability, FHA intentional dis­crimination liability, equal protection racial discrimination claims, 153 Policies limiting access to affordable housing would not pose other problems under the Fourteenth Amendment because courts have not recognized housing as a fundamental right. See Lindsey v. Normet, 405 U.S. 56, 74 (1972) (“We do not denigrate the importance of decent, safe, and sanitary housing. But the Constitution does not pro­vide judicial remedies for every social and economic ill. We are unable to perceive in that document any constitutional guarantee of access to dwellings . . . .”). and the constitutional right to travel.

1. FHA Disparate Impact Claims. — The most common basis for legal challenges to residency preference policies is an assertion that such policies contravene the Fair Housing Act. 154 See, e.g., Comer v. Cisneros, 37 F.3d 775, 795 (2d Cir. 1994); United States v. Town of Oyster Bay, 66 F. Supp. 3d 285, 291 (E.D.N.Y. 2014); Langlois v. Abington Hous. Auth., 234 F. Supp. 2d 33, 37 (D. Mass. 2002); United States v. Hous. Auth. of Chickasaw, 504 F. Supp. 716, 732 (S.D. Ala. 1980). HUD regulations lay out the standard for disparate impact liability under the FHA, indicating that “[a] practice has a discriminatory effect where it actually or predictably results in a disparate impact” or “creates, increases, reinforces, or per­petuates segregated housing patterns.” 155 24 C.F.R. § 100.500(a) (2016). Such a practice may none­theless be lawful, however, if supported by a “legally sufficient justi­fication”—one necessary to achieve a legitimate interest that could not be served by an alternative practice with a less discriminatory impact. 156 Id. § 100.500(b). In 2015, the Supreme Court’s decision in Inclusive Communities affirmed disparate impact liability under the FHA. 157 Tex. Dep’t of Hous. & Cmty. Affairs v. Inclusive Cmtys. Project, Inc., 135 S. Ct. 2507, 2525 (2015). The decision clarified that such claims require both a showing of a harmful impact on a protected class and either of two requirements: (1) proof that the defendant lacked a legitimate interest in implementing its practice or (2) proof that the defendant could have achieved its interest with a less discriminatory alternative. 158 Id. at 2514–15; see also Schwemm, After Inclusive Communities, supra note 43, at 119. The Court also insisted on a “robust causality requirement” linking the challenged policy with the disparate impact. Inclusive Cmtys., 135 S. Ct. at 2512. And it indicated that the “mere awareness of race” in remedial measures “does not doom that endeavor at the outset,” though it seemed to endorse only “race-neutral tools.” Id.; see also McArdle, Winfield, supra note 84, at 294. The Court also emphasized the importance of giving defendant PHAs and developers a sufficient opportunity to demonstrate valid interests underlying their policies, suggesting that it would be “onerous” to find liability merely because “some other priority might seem preferable” to a valid interest such as revital­ization. Inclusive Cmtys., 135 S. Ct. at 2512. Some scholars have interpreted this language as limiting the Court’s potential embrace of residential mobility as a remedial strategy. See McArdle, Winfield, supra note 84, at 293. In the wake of Inclusive Communities, scholars have suggested that residency preferences might be particularly ripe for challenge under the disparate impact theory. 159 See, e.g., McArdle, Winfield, supra note 84, at 295; Schwemm, After Inclusive Communities, supra note 43, at 125.

a. Harmful Impact. — The first and most crucial question in evaluating a disparate impact challenge levied against an anti-displacement residency preference is whether the preservation of existing neighbor­hood populations yields a discriminatory impact. Under the prevailing view, neighborhood-level residency preferences have a clearly discrim­inatory effect because they seek to perpetuate existing housing patterns in segregated cities. The Winfield court appeared to subscribe to this perspective; in denying New York City’s motion to dismiss, it attributed to New York City the goal of preserving existing residential demo­graphics. 160 Winfield v. City of New York, No. 15CV5236-LTS-DCF, 2016 WL 6208564, at *6 (S.D.N.Y. Oct. 24, 2016). HUD, in rejecting San Francisco’s proposed residency preference, articulated a similar position by suggesting that the policy might perpetuate segregation. 161 See supra note 107 and accompanying text. From this perspective, any housing policy that reinforces segregated patterns necessarily effectuates a disparate impact.

b. Legitimate Interest. — If a discriminatory impact is found, the suc­cess of challenges brought against neighborhood preferences will hinge on how receptive courts are to cities’ justifications for the preference. An anti-displacement rationale for neighborhood preferences could be ar­ticulated in at least three different ways: as an interest in protecting current residents against displacement, as an interest in sustaining low-income and minority communities, and as an interest in promoting neighborhood stability.

The first of these interests, if framed as the retention of current resi­dents, arguably suffers from the very circularity that the Langlois court rejected. 162 See supra notes 66–68 and accompanying text; see also Robert G. Schwemm, The Community Preference Policy: An Unnecessary Barrier to Minorities’ Housing Rights, N.Y.U. Furman Ctr.: Dream Revisited (Nov. 2015), http://furmancenter.org/research/
iri/essay/the-community-preference-policy-an-unnecessary-barrier-to-minorities-housin [http://perma.cc/ZLP5-9SAZ] [hereinafter Schwemm, Unnecessary Barrier] (dismissing circular justifications in light of Langlois).
Given the increasing skepticism courts have shown toward such rationales, 163 See supra notes 66–69 and accompanying text. a court might refuse to regard this as a legitimate jus­tification. If the implicated interest is framed, however, as protecting individuals against a looming threat of displacement rather than as retaining the neighborhood’s specific residents, it may be more viable. Local governments might argue that they are prioritizing existing resi­dents not because they are residents but because they are the population most vulnerable to the consequences of gentrification in their own neighborhood. 164 New York City puts a spin on this justification by claiming that longstanding resi­dents have earned a right to enjoy the fruits of gentrification. See supra notes 76, 88–93 and accompanying text. But by framing the individual interest more in terms of an earned benefit than a mitigated risk, New York makes the interest duration-based. Case law from outside the housing context indicates that a proffered government interest in favoring older residents over newer ones is not legitimate. See Zobel v. Williams, 457 U.S. 55, 57, 65 (1982) (holding that, in the context of a state dividend program that compensated locals based on number of years of residency, favoring established residents over new ones is “constitutionally unacceptable”).

The second category of interest focuses on sustaining minority communities within the city, as San Francisco advertised that its residency preference plan was designed to do. 165 See supra notes 103–104 and accompanying text. While this interest is not plagued with the circularity problem, it faces another obstacle: The preservation of racially and culturally specific communities is arguably contrary to the clear integrationist mandate of the FHA. 166 See supra notes 32–43 and accompanying text. Race-conscious affirmative action measures have been deemed permissible under Title VII of the Civil Rights Act, 167 See United Steelworkers v. Weber, 443 U.S. 193, 208–09 (1979) (holding that an employment affirmative action plan was permissible under Title VII). and an extension of that case law to Title VIII would pave the way for affirmative action housing measures.

Case law indicates that the FHA’s integrationist mandate may be set aside when it conflicts with the Act’s antidiscrimination mandate. In United States v. Starrett City Associates, the Second Circuit struck down ra­cial “ceiling quotas” that were integrative in that they promoted a racially heterogeneous population within a housing development but discriminatory in that they disproportionately deprived minority appli­cants of access to the development. 168 840 F.2d 1096, 1101 (2d Cir. 1988) (describing the tension between racial quotas that “promote Title VIII’s integration policy” but “contravene its antidiscrimination policy” before invalidating those quotas).

That decision might provide precedential support for upholding an inverse policy: one that is segregative but antidiscriminatory. While this presents a plausible pathway for cities to defend a neighborhood prefer­ence’s disparate impact through an affirmative action rationale, it re­quires that courts willingly conflate anti-displacement objectives with antidiscrimination objectives. It is not clear that courts would embrace a governmental interest in preserving a neighborhood’s racial and ethnic composition under a statute enacted to disrupt those very patterns. 169 While courts might differ from HUD in their willingness to endorse affirmative action measures under the FHA, HUD’s rejection of San Francisco’s residency preference plan demonstrates a knee-jerk resistance to any plan that seeks to preserve a neighbor­hood’s existing demographics. See supra notes 107–110 and accompanying text.

The third possible category of governmental interest—one directed at neighborhood stability—certainly sounds legitimate, but this justifica­tion might crumble upon interrogation of the specific instability at issue. If the instability is the displacement of current residents and the influx of new residents, this argument merely reframes the circular rationale. If the instability is the shifting of neighborhood demographics and the ero­sion of “culture,” then the justification is a different spin on the “pre­serve minority communities” justification.

Other forms of instability might indeed be valid concerns, but resi­dency preferences are unlikely to be the least discriminatory means of achieving them. 170 See Schwemm, Unnecessary Barrier, supra note 162 (suggesting that even if a governmental defendant could articulate a legitimate justification for its residency pref­erence policy’s disparate impact, less discriminatory means of achieving those goals might exist). For example, while mitigating elevated housing costs and a lack of socioeconomic diversity is presumably a legitimate govern­mental interest, it can be addressed simply through the development of affordable housing units in the neighborhood; assigning those units ac­cording to a residency preference policy is not a necessary measure.

Similarly, while New York may have struck upon a valid interest in claiming that its community preference policy helps to mitigate NIMBY-like opposition to new development, 171 See supra notes 92–96 and accompanying text. An acronym for “not in my backyard,” “NIMBY” is used as shorthand for localized opposition to the siting of de­velopment or programs in one’s own immediate neighborhood, even if one supports the development or programs generally. See, e.g., Peter W. Salsich, Jr., Affordable Housing: Can NIMBYism Be Transformed into OKIMBYism?, 19 St. Louis U. Pub. L. Rev. 453, 455 (2000). it is hard to imagine that a disparate impact-inducing residency preference plan would be deemed the best possible means to achieve that interest. While neigh­borhood-level preferences might be a useful tool to combat community oppo­sition, they are far from the only strategy. 172 Community activists in Crown Heights, Brooklyn, for example, conditioned their support for a proposed development on a number of factors besides the expansion of the residency preference, including demands that the development’s architecture be con­sistent with that of the neighborhood and that the development contain more family units and fewer one-bedroom units than initially proposed. Simone Wilson, Imagine if the Bedford Union Armory Were Transformed into 100% Affordable Housing, Patch (Oct. 17, 2016), http://patch.com/new-york/prospectheights/imagine-if-bedford-union-armory-were-transformed-100-affordable-housing [http://perma.cc8LZT-7YGX]. A city defendant would be hard-pressed to prove that no alternative exists that would create a less severe disparate impact.

Finally, city defendants will not be able to justify neighborhood-level residency preferences aimed at mitigating the displacement effects of gentrification through the well-established rationale of municipal protec­tionism. In cases involving intercity residency preferences, courts have often recognized a government’s desire to ensure that its services are available to its own residents as a legitimate interest. 173 See supra notes 63–65 and accompanying text. Whether this valid interest is persuasive enough to overcome a finding of disparate impact is another question. In Chickasaw, for example, it was not. See United States v. Hous. Auth. of Chickasaw, 504 F. Supp. 716, 731–32 (S.D. Ala. 1980). In the wake of Langlois, it is less clear whether this admittedly circular justification is val­id; 174 See supra notes 66–68 and accompanying text. what is clear, however, is that it is unavailable to cities with neigh­borhood-level—rather than intermunicipal—preferences. The de­sire to ensure that city services are available for city residents does not explain a policy that prioritizes certain city residents over others.

Considering the available justifications and their likelihoods of suc­cess, anti-displacement residency preferences enacted at the neighbor­hood level may be particularly vulnerable to FHA disparate impact liability. Absent a judicial embrace of affirmative-action-oriented ratio­nales that justify neighborhood-level preferences through their po­tential to preserve communities of racial minorities, such preferences lack a reliably “legitimate” justification. These policies may be no more likely than their exclusionary, intercity counterparts to survive disparate impact challenges, and they may be even more vulnerable without the ability to lean on the once-reliable protectionist justification.

2. FHA Intentional Discrimination Claims. — It is possible that anti-displacement neighborhood preferences might also suffer an elevated susceptibility to FHA intentional discrimination claims due to the race-conscious nature of neighborhood preferences. If the Oyster Bay decision and Winfield memorandum are any indication, courts may be increasingly willing to entertain claims of intentional discrimination. 175 See supra notes 59–62, 73, 99 and accompanying text. And given the FHA’s strictly integrationist ambitions, 176 See supra notes 32–43 and accompanying text (discussing the widespread understanding of the FHA as most centrally promoting integrationist objectives). courts may condemn the delib­erate perpetuation of segregated housing patterns regardless of whether its purported purpose is to preserve or exclude minority communities.

To be sure, courts are unlikely to reach for intentional discrimination liability under the FHA when disparate impact liability is cognizable, barring an egregious display of animus or deliberate discrimination—and preferences aimed at protecting low-income communities against dis­placement are particularly unlikely to be deemed egregious. The osten­sibly inclusionary aim of such policies, however, means that the pol­icies are more likely to be overtly race conscious, which may, in turn, make segregative intent easier to prove. San Francisco city officials, for exam­ple, were much more candid about their intention that the residency preference policy operate to preserve existing racial demographics 177 See supra notes 101–104 and accompanying text. than the defendant in Chickasaw—a case in which the court deemed credible the defendant’s claims that it did not intend the policy’s segregative effects. 178 See supra notes 63–64 and accompanying text.

Because the alleged disparate impacts created by anti-displacement, neighborhood-level residency preferences may be more plainly deliber­ate (or even the preferences’ very purpose), such policies are more likely to invite FHA intentional discrimination claims. If courts decline to en­dorse affirmative action efforts to preserve minority communities under the FHA, administrators of anti-displacement preferences might be uniquely vulnerable to intentional discrimination liability.

3. Equal Protection Racial Discrimination Claims. — Equal protection racial discrimination claims are rarely brought to challenge residency preferences, in part because the existence of disparate impact liability under the FHA makes a statutory violation much easier to establish than a constitutional violation, 179 Equal protection jurisprudence makes it clear that disparate impact alone does not amount to a Fourteenth Amendment violation. See Washington v. Davis, 426 U.S. 229, 246 (1976). Equal protection violations require a finding of discriminatory intent. See Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 266–68 (1977). Intent may be proved through stark patterns “unexplainable on grounds other than race,” the events preceding the challenged action, or the relevant legislative or administrative history. Id. and in part because even policies that effec­tuate a stark disparate impact are likely to be facially race neutral. 180 This was true even in United States v. Housing Authority of Chickasaw, in which the preference policy effectively preserved Chickasaw as an all-white suburb. See supra notes 47–50, 63–64 and accompanying text. Nonetheless, the Second Circuit in Comer v. Cisneros looked favorably up­on the plaintiffs’ claim that they had suffered a constitutional harm under the Equal Protection Clause, at least so far as to hold that plaintiffs could survive summary judgment. 181 See supra notes 55–57 and accompanying text.

As discussed above, race-conscious policymaking is more likely to be provable in the affirmative action context. 182 See supra notes 166–167, 176–178 and accompanying text (describing the FHA’s integrationist mandate and noting that segregative intent may be easier to prove in the case of affirmative action policies, which are more likely to be overtly race conscious, than in the case of policies that discriminate against minorities). Policies that target African American and Latino communities for preservation and protection against displacement may more transparently consider race, thereby rendering equal protection liability somewhat more plausible. While ex­clusionary intercity residency preferences have largely been insulated from equal protection liability, 183 See supra section I.B.2 (discussing the limited case law invalidating residency preferences under an intentional discrimination standard). ostensibly inclusionary intracity prefer­ences raise obvious equal protection concerns. By treating neighborhood residency as, in the words of the Comer court, a “proxy for race,” 184 Comer v. Cisneros, 37 F.3d 775, 793 (2d Cir. 1994). these policies would merit strict scrutiny. The paradox here is a familiar one from the affirmative action context: Practices that prioritize racial minor­ities are more constitutionally vulnerable than facially race-neutral poli­cies that impose an adverse disparate impact upon minorities. 185 See, e.g., City of Richmond v. J.A. Croson Co., 488 U.S. 469, 477, 511 (1989) (holding that a city ordinance requiring that thirty percent of construction contracts be issued to minority-owned businesses violated the Equal Protection Clause); Washington v. Davis, 426 U.S. 229, 238–39 (1976) (holding that strict scrutiny does not apply to a facially race-neutral police application test, despite its disparate impact on minority applicants). The more care­fully residency preferences are targeted at protecting nonwhite commu­nities against displacement, the more constitutionally problematic they become.

4. Right-to-Travel Claims. — Residency requirements are most often deemed to be in violation of the constitutional right to travel 186 While the Constitution does not explicitly acknowledge a fundamental right to travel, the Supreme Court has consistently interpreted Article IV’s Privileges and Immunities Clause as implying that right. See, e.g., United States v. Guest, 383 U.S. 745, 758 (1966) (noting that “freedom to travel . . . has long been recognized as a basic right under the Constitution”). The Court has also suggested that this right may be based in other Constitutional provisions, including the Commerce Clause, the Equal Protection Clause, and the Due Process Clause. For one discussion of the constitutional basis for this right, see Shapiro v. Thompson, 394 U.S. 618, 666–71 (1969) (Harlan, J., dissenting). Under its right-to-travel jurisprudence, the Court has recognized three fundamental com­ponents: the right to interstate ingress and egress, the right to be welcomed into states as a visitor, and the right of new permanent residents of a state to be treated equally to the state’s other citizens. Saenz v. Roe, 526 U.S. 489, 500 (1999). when they discriminate not merely on the basis of state residency 187 Bona fide residency requirements, which give special treatment to in-state residents but place all permanent residents on equal footing, are generally deemed constitutional. See Martinez v. Bynum, 461 U.S. 321, 328 (1983) (“A bona fide residence requirement, appropriately defined and uniformly applied, furthers the substantial state interest in assuring that services provided for its residents are enjoyed only by residents.”). but on the basis of duration of state residency. 188 Durational residency requirements are more likely to “penalize” interstate migration than bona fide requirements and are therefore unconstitutional “unless shown to be necessary to promote a compelling government interest.” Shapiro, 394 U.S. at 634 (emphasis omitted); see also Att’y Gen. of N.Y. v. Soto-Lopez, 476 U.S. 898, 903 (1986) (Brennan, J., plurality opinion) (“[O]ur recent cases have dealt with state laws that, by classifying residents according to the time they established residence, resulted in the unequal distribution of rights and benefits among otherwise qualified bona fide residents.”). As a result, residency preferences for affordable housing are consistently devoid of durational components. There is no indication that cities enacting neighborhood-level residency preferences are likely to break with this precedent, particularly in light of administrative regulations that prohibit durational preferences for PHA-administered waiting lists. 189 See 24 C.F.R. § 982.207(b)(1)(iv) (2016).

Nonetheless, anti-displacement rationales for residency preferences are particularly intertwined with a duration-based logic. New York City’s defense of its preference policy leans heavily on the idea that longstand­ing residents have built equity in their neighborhoods, and the Winfield complaint criticizes New York’s policy for failing to distinguish between longstanding residents and recent arrivals. 190 See supra notes 88–91 and accompanying text. Cities could turn to dura­tional preferences to more closely target the goal of retaining longtime residents. In that case, as with equal protection liability, an odd irony arises: The more a city tailors a residency preference to protecting the desired population—in this case, longstanding residents of a neighbor­hood—the more likely its policy is to violate the Constitution.

It is also possible that the intracity–intercity distinction may matter in the case of right-to-travel liability. The Supreme Court has made it clear that only state-level durational residency requirements, which contravene the fundamental right to interstate travel, are unconsti­tutional. 191 See, e.g., Johnson v. City of Cincinnati, 310 F.3d 484, 495 (6th Cir. 2002) (noting that the Supreme Court has never “formally recognized a limited right to intrastate travel”). Appellate courts have generally extended that same consti­tutional protection to intrastate travel between municipalities. 192 See id. at 498 (“In view of the historical endorsement of a right to intrastate travel and the practical necessity of such a right, we hold that the Constitution protects a right to travel locally through public spaces and roadways.”); see also Wellford v. Battaglia, 485 F.2d 1151, 1152 (3d Cir. 1973) (holding that a Wilmington charter provision imposing a five-year residency requirement upon mayoral candidates implicated the constitutional right to travel and thus merited strict scrutiny); King v. New Rochelle Mun. Hous. Auth., 442 F.2d 646, 648 (2d Cir. 1971) (holding that the logic underlying Shapiro extends to intrastate as well as interstate travel).
In King, the Second Circuit held that a five-year durational residency requirement for admission to public housing violated the Equal Protection Clause even though it principally infringed upon claimants’ right to intrastate—rather than interstate—travel. Id. Noting that the Supreme Court in Shapiro attributed the right to travel to “‘our constitutional concepts of personal liberty’” rather than to a “particular constitutional provision,” the Second Circuit held that such personal liberty concerns must logically extend to intrastate activity. Id. It suggested that “[i]t would be meaningless to describe the right to travel between states as a fundamental precept of personal liberty and not to acknowledge a correlative constitutional right to travel within a state.” Id.
It is conceivable that the law might not recognize such a right on the hyper­local, intramunicipal level; at some point, perhaps, the alleged right is too geographically limited to be understood as “travel” or “migration.” If so, cities might be able to implement durational neighborhood-level res­idency preferences without running afoul of the constitutional right to travel.

Overall, the efficacy of neighborhood-level residency preferences is severely undermined by their legal vulnerabilities. Residency preferences implemented at the neighborhood level to combat gentrification-in­duced displacement may, as their proponents contend, turn exclusion­ary residency preferences on their head. 193 See Letter from Herrera, supra note 4, at 2 (arguing that San Francisco’s pro­posed residency preference “takes a tool that communities used in the past to keep pro­tected minorities out and flips it on its head, to help residents remain in their neighbor­hoods instead”). And yet, the legality of these policies appears to be at best uncertain and—if courts adhere to a tradi­tional reading of the FHA that strictly condemns all segregated neighborhoods—perhaps even unlikely. An inherent mismatch exists between the existing law and emerging policies. These preferences aim to effectuate the FHA’s goal of promoting fair housing for low-income and minority residents, but they do so in response to displacement pres­sures that the FHA does not contemplate 194 As San Francisco City Attorney Dennis Herrera wrote to HUD, neighborhood-level residency preferences are a response to housing challenges that were scarcely imagin­able upon the FHA’s enactment in 1968. Id. and in a manner that clashes with the FHA’s anti-segregationist objective.

III. “Legalizing” Anti-Displacement Residency Preferences: Potential Solutions to the Policy-Law Mismatch

If neighborhood-level residency preferences are to be effectively and legally utilized to address issues of urban displacement, either courts’ approaches to such policies or the policies themselves must evolve. This Part advocates for a combination of these strategies, with primary reli­ance on an alternative understanding of how neighborhood-level hous­ing patterns relate to integrationist goals. Section III.A argues that courts should interpret such policies as consistent with the FHA’s integrationist aims when the impending displacement would result in a less integrated and diverse municipality. Section III.B identifies five possible adjustments that would render neighborhood-level residency preferences more le­gally viable.

A. Reframing the Conversation: Residency Preferences as Integration-Preservation Measures

While San Francisco’s attempt to utilize residency preferences to preserve the African American population in Western Addition was blocked by HUD because it ostensibly perpetuated segregation, 195 See infra section I.D (discussing HUD’s response to San Francisco’s residency preference proposal). it might well be reinterpreted as an effort to preserve integration. After all, San Francisco’s dwindling African American population, which declined from 13.4% in 1970 to 5.5% in 2014, 196 Albarazi, supra note 101. suggests that many residents displaced from neighborhoods with concentrated African American pop­u­lations leave the city entirely. In the face of encroaching homogenization, policies that perpetuate segregated neighborhoods may none­theless serve a larger-scale integrationist purpose.

The crucial factor here is the breadth of the applicable geographical and conceptual scope. Gentrification scholars have called for a broader inquiry that examines gentrification as a municipal and regional phe­nomenon rather than a strictly neighborhood-level occurrence. 197 See, e.g., Billingham, supra note 119, at 80–81 (arguing that sociologists should broaden the geographical scope of their gentrification analyses). Critics of this “[g]eographic myopia” 198 Id. at 82. argue that it overlooks critical factors in the gentrification analysis and cite both academic and practical benefits to widening the geographic lens. 199 Id. at 79–82 (“[M]unicipal and regional policies affecting economic activity and class-specific migration patterns have profound influence on the trajectory of gentrification.”). By similarly broadening the scope through which one views housing patterns, the perpetuation of certain neighborhood-level segregation might be viewed as a means of promot­ing comparatively macroscopic integration. Legal scholars have drawn attention to the tractable nature of interpretive lenses, which are ex­panded or constricted to facilitate a particular perspective and, often, outcome. 200 For the most well-known of these scholarly accounts, see Mark Kelman, Interpretive Construction in the Substantive Criminal Law, 33 Stan. L. Rev. 591, 593 (1981). Narrower lenses—sometimes applied unconsciously—may simplify the narrative at the expense of context or nuance. 201 See id. at 664 (describing the pitfalls of unconscious “[n]arrow time-framing”).

In the case of housing, an inquiry into integration and segregation as strictly neighborhood-level patterns may miss the forest for the trees. If minority residents are displaced from their city at an elevated rate, the preservation of certain segregated neighborhoods may in fact be a cor­rective to segregation at the municipal or regional level. This argument should not be misunderstood as advocating the abandonment of efforts to achieve neighborhood integration or as conflating integration and diver­sity; rather, it promotes the pragmatic recognition that integration requires diversity. When residency preference policies seek to preserve the diversity of a population against the alternative of homogenizing dis­placement, they may act in support of integrationist goals—even if their localized effect is to perpetuate segregated patterns.

Furthermore, residency preference policies in gentrifying neighbor­hoods may be reinterpreted as necessary to effectuate the integrationist or “social mixing” potential of gentrification. 202 See Freeman & Braconi, supra note 126, at 39 (“If it proceeds without widespread displacement, gentrification also offers the opportunity to increase socioeconomic, racial, and ethnic integration.”). Arguably, a preference policy that seeks to preserve a neighborhood’s preexisting demographics perpetuates segregation only if the neighborhood demographics are stat­ic and homogenous; in gentrifying neighborhoods, such a policy can help to realize and stabilize integrated housing patterns. Without residency preferences, a gentrifying neighborhood may move from low income and predominantly minority to higher income and predominantly white, with only a fleeting transitional window of integrated living.

Racial and socioeconomic demographics in gentrifying neighbor­hoods are attributable not only to the identities of those who are displaced but also to the identities of new arrivals. 203 See Lance Freeman, Displacement or Succession? Residential Mobility in Gentrifying Neighborhoods, 40 Urb. Aff. Rev. 463, 485–87 (2005) (arguing that “in-movers rather than out-movers are the driving force behind neighborhood change in gen­trifying neighborhoods” and that “[h]igher incomes and being White are associated with an increased likelihood of moving into such a neighborhood”). As a result, demo­graphics in gen­trifying neighborhoods may change swiftly. Data from New York City indicate that the average household income in gentrifying neighbor­hoods (adjusted for inflation) rose 6.1% between 2005 and 2010–2014. 204 N.Y.U. Furman Ctr., supra note 132, at 9. The average household income rose from $55,400 in 2005–2009 to $58,550 in 2010–2014. Id. During the same period, the citywide average income rose just 0.06%. 205 The average household income was $79,900 in 2005–2009 and $79,950 in 2010–2014. Id. Between 2000 and 2010, the share of white residents in New York City overall decreased by over seven percent, 206 White residents made up 35.8% of New York’s population in 2000 and 33.4% in 2010. Id. at 12. but the share of white resi­dents increased by over twenty percent in gentrifying neighborhoods. 207 The white share of gentrifying neighborhoods was 17.1% in 2000 and 20.6% in 2010. Id. Given the pace of demographic shifts in gentrifying neighborhoods, the idea that residency preferences in such neighborhoods perpetuate preexisting patterns seems misguided. Rather, these preferences are better understood as promoting a more persistent and less transitory kind of integration in the midst of rapidly changing demographics.

This reframed approach calls into question the assumption that neighborhood-level residency preferences perpetuate segregation. When viewed in the context of gentrification’s homogenizing potential, residency preference policies that seek to preserve minority communities are a weapon against segregation, not its facilitator. When properly imple­mented, they should not be understood to create discriminatory or segregative effects subject to disparate impact liability. In this light, resi­dency preferences are a tool of integration preservation consistent with the FHA’s provisions and purpose.

B. Rethinking Residency Preferences

Neighborhood-level residency preferences are not inherently inva­lid, and their potential legal vulnerabilities can be mitigated by strategic adjustments. This section addresses five possible strategies for rethinking anti-displacement residency preferences so that they are more likely to both avoid and survive legal challenge.

1. Extend Fewer Preferences. — First, neighborhood-level residency preferences may be both less objectionable and more legal when they apply to a smaller proportion of available housing units. Opponents of residency preferences in New York and San Francisco have identified the extent of those preferences 208 Under New York City’s community preference policy, residents of the community district have preferred access to fifty percent of available units. See Winfield, First Amended Complaint, supra note 45, at 2. The preference was expanded from thirty percent in 2002. Id. at 14. San Francisco’s neighborhood resident housing preference applies to forty percent of available units. See S.F. Mayor’s Office of Hous. & Cmty. Dev., supra note 102. as one basis for their criticism. 209 See, e.g., Winfield, First Amended Complaint, supra note 45, at 15–17 (criticizing New York’s fifty percent preference policy as overly extensive because it leaves only thirty-eight percent of units open to equal competition); Albarazi, supra note 101 (reporting that one San Francisco City Supervisor voted against the preference policy because he felt the forty percent total was excessive). By applying the preference to a smaller portion of units in a given develop­ment, cities and housing authorities might provoke less controversy.

Narrowing the extent of residency preferences could also help such policies survive legal challenge. While a narrower preference may not be more closely tailored to any of the likely governmental justifications, it might render such justifications less necessary by reducing the prefer­ence’s disparate impact. A reduced preference might therefore be re­garded as less discriminatory than a more expansive one, because it does not so much perpetuate existing housing patterns as prevent them from utter disruption. 210 See Schwemm, Unnecessary Barrier, supra note 162 (suggesting that New York’s policy might be less discriminatory if the city reduced the preference from fifty percent to twenty-five percent).

For proponents of neighborhood-level residency preferences, the curtailed approach described here bears an obvious downside. A reduced preference will serve fewer residents and protect a more limited subset of the existing population against displacement. The proposed adjustment, therefore, is not one that would strengthen residency preferences so much as strike a compromise.

2. Expand the Geographic Scope of Preference Areas. — A second adjust­ment to neighborhood-level residency preferences would strategically expand the geographic preference area to encompass more racially diverse populations. Concerns that residency preferences exacerbate segre­gated housing are most prominent and forceful where the preferences apply to geographical areas whose populations are made up of either predominantly white or predominantly minority residents. 211 For example, the Winfield complaint points to the uneven racial composition of New York City community districts as the basis for its claim that the preference policy perpetuates segregation. See Winfield, First Amended Complaint, supra note 45, at 9–10 (pointing out that while African American and Latino residents comprise a combined 51.4% of the New York City population, thirty-one of the city’s fifty-nine community districts have a combined African American and Latino population that is either below twenty percent or above eighty percent). By expanding the geographic scope to more diverse areas or pairing demo­graphically distinct neighborhoods together into a single preference area, residency preferences could help protect against displacement without directly preserving the specific racial composition of individual neighborhoods.

Versions of this approach have been among the most popular solu­tions to the problem of segregation-perpetuating residency preferences. The Comer settlement expanded the challenged residency preference to the entirety of Erie County, so that residents of Buffalo were included in—rather than excluded by—the preference’s scope. 212 See Carey, supra note 71, at 89 (describing the terms of the Comer settlement, including the establishment of countywide residency preferences). The Langlois decision spoke approvingly of a “tempered approach” to residency pref­erences, in which urban and suburban PHAs would partner and ex­tend preferences reciprocally to one another’s residents. 213 Langlois v. Abington Hous. Auth., 234 F. Supp. 2d 33, 70 (D. Mass. 2002). A New York state court decision suggested that the “imbalance” in the community pref­erence policy’s applicant pool might be mitigated by the merging of two community districts into a single preference area. 214 Broadway Triangle Cmty. Coal. v. Bloomberg, 941 N.Y.S.2d 831, 837 (Sup. Ct. 2011).

One potential pitfall of this approach is that its most effective itera­tion would require a race-conscious design of expanded preference areas, which could invite controversy and legal challenges. Additionally, larger preference areas diminish the preference’s ability to protect against displacement at the hyperlocal level and to promote community preservation in individual neighborhoods. Residents of low-income, pre­dominantly minority communities would compete for affordable housing in their neighborhoods on equal footing with residents of certain nearby—and potentially majority-white—neighborhoods, though they would also have equal access to affordable units in those other neighborhoods.

3. Limit Residency Preferences to Particular Neighborhoods. — A third approach to neighborhood-level residency preferences is to apply the preference only to neighborhoods that meet certain criteria—ideally criteria tied to the city’s justification for administering the preference. Under this approach, a city would identify specific eligible neighbor­hoods and extend residency preferences to all income-eligible residents of those neighborhoods. A preference policy intended to insulate resi­dents from rising housing costs, for example, could be applied only in neighborhoods that display some threshold increase in rental prices; an expressly anti-displacement policy would be administered only in neigh­borhoods with sufficient patterns of displacement. 215 See Schwemm, Unnecessary Barrier, supra note 162 (suggesting that residency preferences be applied only in neighborhoods without “stability-enhancing factors”).

FHA case law has blocked racial quotas that impose ceilings on minority populations but has suggested that “‘access’ quotas” designed to increase housing opportunities for racial minorities may be permissible under certain conditions. 216 United States v. Starrett City Assoc., 840 F.2d 1096, 1102 (2d Cir. 1988) (quoting Burney v. Hous. Auth. of Beaver, 551 F. Supp. 746, 763 (W.D. Pa. 1982)). Therefore, it is conceivable that a city might be able to selectively implement residency preferences in neighborhoods with threshold levels of diversity 217 Some commentators have advocated this precise approach in New York City. See, e.g., Tepperman-Gelfant, supra note 149 (proposing that New York City implement neighborhood preference policies “only when a neighborhood currently meets some minimum threshold of racial diversity”). —so long as it could convincingly frame its goal as the advancement of an integrated community rather than the preservation of a segregated one. 218 See Starrett City, 840 F.2d at 1102 (describing permissible racial quotas as those that promote adequate minority representation where it might not otherwise exist).

Preference policies administered with these criteria would arguably operate on steadier legal ground because they would be narrowly tai­lored to the city’s primary proffered justification. Even if the policies were found to create a disparate impact, the city might be better posi­tioned to argue that no less discriminatory alternative existed. Moreover, the preference policy’s close relationship to a valid governmental interest could help to rebuff any intentional discrimination or equal protection challenges.

A traditional residency preference policy administered under this approach would allow eligible residents in eligible neighborhoods to compete for affordable housing exclusively in their own neighborhoods. (Affordable housing in ineligible neighborhoods would, presumably, be equally available to all applicants regardless of their geographic resi­dence.) Another permutation of this approach, by contrast, might give residents of eligible neighborhoods preferred access to affordable hous­ing citywide regardless of its location.

San Francisco’s anti-displacement preference, which HUD approved after rejecting its neighborhood-level residency preference, targets neigh­borhoods in this latter manner. It extends the preference to resi­dents of specific census tracts that have experienced acute displacement and is therefore contingent on applicants’ geographic residence but not on the location of the affordable housing development to which they apply. 219 See Dineen, Federal Agency OKs Preferences, supra note 113 (noting that the preference will apply to “residents who live in low-income neighborhoods undergoing dis­placement and experiencing advanced gentrification” and identifying five eligible neigh­borhoods); News Release, Office of the Mayor, supra note 112 (describing San Francisco’s anti-displacement preferences as prioritizing low-income applicants living in “census tracts that have been identified as having the greatest risk of displacement” citywide). Commentators and public officials widely hailed this revised ap­proach as effective and comparatively uncontroversial, and HUD’s acquiescence signals that it may also be a more legally viable solution. 220 See supra notes 111–116 and accompanying text (detailing HUD’s approval of the anti-displacement policy and the ensuing reaction).

4. Duration-Based Preferences. — A fourth suggested approach also imposes strategic criteria on the operation of neighborhood-level resi­dency preferences but does so by limiting applicant eligibility rather than neighborhood eligibility. Under this approach, residents of any neighbor­hood within the implementing city might be preference eligible but only if they have lived in their neighborhood for a sufficient duration. Du­ration-based residency preferences would function as a sort of earned benefit, treating longevity of residence as a proxy for virtues such as commitment to the local community.

As discussed above, New York City’s defense of its community pref­erence policy relies heavily on the claim that longstanding residents of gentrifying neighborhoods have earned a right to remain by enduring years of poor living conditions. 221 See supra note 88 and accompanying text. This is a potentially compelling justifi­cation, and its specific focus on longtime residents of previously impover­ished neighborhoods minimizes the circularity problem. It is poorly suited, however, to justifying a policy that extends preferences without regard to duration. 222 The Winfield plaintiffs emphasize this mismatch in their complaint. See supra note 89 and accompanying text; see also Gurian, supra note 151 (“[The] preference exists even if the applicant has only lived in the community district for 10 minutes and even if the applicant has been living in comfortable circumstances.”).

The obvious solution is a duration-based residency preference, but such an approach raises immediate pragmatic difficul­ties. Is the pref­erence extended only to applicants with a threshold duration of resi­dency (and administered equally to all who meet the threshold), or is it available to all residents and scaled based on duration? How does the preference apply to households with members of varying duration or who inherited their current housing from a family member? Does the preference apply to every neighborhood, or does it vary according to each neighborhood’s trajectory of disinvestment and gentrification? (In theory, it could be combined with the preceding ap­proach so that only longstanding residents of specific neighborhoods would be eligible.)

Additionally, durational preferences often violate the fundamental right to travel. 223 See supra section II.B.4 (analyzing “right-to-travel” challenges to neighborhood-level residency preferences). It is unclear, however, whether neighborhood-level du­rational preferences violate that right. 224 See supra note 191 and accompanying text. While scaled durational pref­erences—those available to all residents but tiered according to du­ration of residency—might be less likely to invite right-to-travel liability, case law from outside the housing context indicates that such provisions may violate the Equal Protection Clause. 225 See Zobel v. Williams, 457 U.S. 55, 56–57, 65 (1982) (holding that a state divi­dend-distribution plan that distributed one “dividend unit” for each year of residence after 1959 impermissibly favored established residents over new residents and violated the Equal Protection Clause). If constitutional challenges can be avoided or defeated, a durational residency preference seems like a well-tailored policy for cities concerned with rewarding longstanding residents’ endurance, though determining the contours of eligibility would pose an administrative headache.

5. Residency as a “Plus Factor.” — Finally, city governments could rep­licate a strategy from affirmative action doctrine by treating residency as a “plus factor” that enhances an applicant’s candidacy rather than as a criterion considered in isolation. 226 See, e.g., Grutter v. Bollinger, 539 U.S. 306, 334 (2003) (stating that affirmative action programs may treat race as a “‘plus’ factor” but may not impose racial quotas). This approach would grant neighbor­hood residents preferred access to local affordable housing while also allowing outsiders an opportunity to compete for the same units. 227 See Schwemm, Unnecessary Barrier, supra note 162 (imagining an approach in which other plus factors allow outsiders to “demonstrate their commitment to the target neighborhood”). “Res­idence” would be accorded numeric value within a larger quantitative system.

A “plus factor” policy would mitigate equal protection concerns even if residency were viewed as a proxy for race, given that the Supreme Court has endorsed an analogous tactic in the educational setting. 228 See Grutter, 539 U.S. at 334 (“Universities can, however, consider race or ethnicity more flexibly as a ‘plus’ factor in the context of individualized consideration of each and every applicant.”). Al­lowing nonresidents of the neighborhood to compete for every available unit might also alleviate objections to the policy, because residency itself would not be solely determinative. 229 The Winfield complaint emphasized the alleged unfairness of a system in which “outsiders” could only compete for thirty-eight percent of units. See Winfield, First Amended Complaint, supra note 45, at 15 (“The outsider-restriction policy operates in fa­vor of a current resident . . . even if that person established residency in the community district on the final day of the application period.”). The specifics of the system (in particular, the factors considered as “pluses” and the weight accorded to them) would determine the policy’s effectiveness in mitigating displace­ment and the extent of any disparate impact it created.

A legally defensible version of this approach would require that the preference system be tailored to a compelling, noncircular justification. If residency is merely one factor for preferred access to affordable hous­ing rather than the determinative factor, the policy’s objective cannot be the prioritization of existing residents. A successful preference system in this mold would require an array of “plus factors” directed at the specific effects of gentrification that the city wishes to address. Other “plus fac­tors” could include involvement in community organizations, employ­ment in the neighborhood, or other characteristics that evince a participatory approach toward community preservation.

Conclusion

Proponents of neighborhood-level residency preference policies recognize their potential as anti-displacement measures and regard such policies as an inclusionary reappropriation of discriminatory intercity preferences. Under existing law, however, anti-displacement residency preferences might be deemed less an inversion of discriminatory policies and more an extension of them. The demographics of the target com­munities may be different, but the goal of insulating an existing popula­tion is arguably unchanged.

For neighborhood-level residency preferences to operate as a useful and legally viable tool for cities seeking to mitigate gentrification-induced displacement, the demographic consequences of local housing patterns must be considered at a broader geographic level and strategic adjust­ments must be made to the way in which preferences are administered. Cities can utilize residency preferences to turn an exclusionary tool on its head, but a true inversion will require reframing the legal conversation and revising governmental approaches.