Opinion
Civil Action 22-CV-10800-AK
07-26-2023
MARY LOUIS and MONICA DOUGLAS, on behalf of themselves and similarly situated persons, and COMMUNITY ACTION AGENCY OF SOMERVILLE, INC., Plaintiffs, v. SAFERENT SOLUTIONS, LLC, and METROPOLITAN MANAGEMENT GROUP, LLC, Defendants.
MEMORANDUM AND ORDER ON DEFENDANTS' MOTIONS TO DISMISS
Hon. Angel Kelley, United States District Judge
Plaintiffs Mary Louis (“Louis”), Monica Douglas (“Douglas”), and Community Action Agency of Somerville, Inc. (“CAAS”), bring this putative class action against Defendants SafeRent Solutions, LLC (“SafeRent”), and Metropolitan Management Group, LLC (“Metropolitan”), on behalf of low-income and minority individuals who hold housing vouchers and were denied rental units. [Dkt. 15 at 1]. Plaintiffs allege that a tenant-screening service operated by SafeRent and Metropolitan's use thereof violate the Fair Housing Act, 42 U.S.C. § 3604, and Massachusetts antidiscrimination and consumer protection laws. [Id. at ¶ 1].
Metropolitan and SafeRent have filed motions to dismiss for lack of standing and for failure to state a claim pursuant to Federal Rules of Civil Procedure 12(b)(1) and 12(b)(6). [Dkt. 29; Dkt. 31]. Plaintiffs oppose these motions. [Dkt. 36]. For the following reasons, Metropolitan's motion to dismiss [Dkt. 29] is DENIED and SafeRent's motion to dismiss [Dkt. 31] is GRANTED IN PART and DENIED IN PART.
I. BACKGROUND
The Court recites here only those facts and law necessary to understand what has led to this action. Further details relevant to the Court's analysis will be discussed as needed. Unless otherwise noted, the facts are presented as alleged in the amended complaint. [See Dkt. 15 (“Am. Compl.”)]. Louis and Douglas are Black women who hold housing vouchers and whose rental applications were denied, at least in part, because of their credit histories and scores. [Id. at ¶¶ 7, 74-76, 84-85]. Metropolitan denied Louis' application for an apartment, and another housing provider not named as a defendant in this action initially rejected Douglas' application for an apartment, though it later approved her request through its appeal process. [Id. at ¶¶ 7475, 83-89]. Metropolitan and the housing provider that denied Douglas' application relied on SafeRent's tenant-screening services when denying these applications. [Id. at ¶¶ 20, 76, 85]. CAAS provides a variety of services to low-income, prospective renters, including those who have received housing vouchers. [Id. at ¶ 17]. Plaintiffs claim that the Defendants' use of SafeRent's tenant-screening services, in particular, the “SafeRent Score,” which relies on “credit histories and other information which bears little to no relationship to the risk that their rent will be paid,” disproportionally affects Black and Hispanic applicants and voucher holders in violation of federal and state antidiscrimination and consumer protection laws. [Id. at 1, ¶ 1].
The Court uses the same race and ethnicity descriptors used by the Plaintiffs in their amended complaint. [See generally Dkt. 15 (“Am. Compl.”)].
Federal housing vouchers, also known as “Section 8” vouchers, are long-term subsidies issued by local housing agencies. [Am. Compl. at ¶ 62]. The Section 8 housing program was enacted in 1974 to “aid[] low-income families in obtaining a decent place to live” and to “promot[e] economically mixed housing.” Burbank Apartments Tenant Ass'n v. Kargman, 48 N.E.3d 394, 399 (Mass. 2016) (citation omitted). The federal government funds the vouchers, and a state or local agency administers the program. Id. The agency pays the housing provider for all or part of a voucher holder's monthly rent, with the total amount paid by the voucher determined by evaluating the cost to lease a unit in the local housing marking, the voucher holder's income, and the cost to rent a particular unit. [Am. Compl. at ¶ 63].
A. The SafeRent Score
SafeRent designs, markets, and sells a variety of tenant-screening services to landlords, real estate agents, brokerages, and property managers nationwide and in Massachusetts. [Id. at ¶¶ 21-22]. These “self-service tenant screening solutions,” as advertised by SafeRent, are intended to help “identify top quality applicants.” [Id. at ¶ 22]. One of these services is the “SafeRent Score,” which uses an algorithm to calculate the risk of leasing a property to a particular tenant. [See id. at ¶¶ 24-25]. The SafeRent Score aggregates several factors, including credit history, bankruptcy records, past due accounts, payment performance, and eviction history, according “to their statistical significance in predicting lease performance,” and calculates a numerical rating between two hundred and eight hundred for rental applicants. [Id. at ¶¶ 25, 27, 30, 35]. This score is intended to measure the applicant's “lease performance risk,” and applicants with higher scores generally outperform applicants with lower scores. [Id. at ¶ 28]. The SafeRent Score does not consider the financial benefits of housing vouchers. [Id. at ¶¶ 31].
SafeRent was formerly known as CoreLogic Rental Property Solutions, LLC. [Am. Compl. at ¶ 1].
SafeRent then issues an “accept/decline/conditional decision” to housing providers based on the applicant's SafeRent Score and the specific housing provider's “predetermined decision points.” [Id. at ¶ 26]. Housing providers select a minimum SafeRent Score required for approval of a rental application, and they can also establish a range to “accept with conditions.” [Id. at ¶ 39]. SafeRent does not disclose the weight assigned to any of the factors considered in the SafeRent Score, nor does it provide the specific sources of its data. [Id. at ¶¶ 34-35]. Housing providers cannot change SafeRent's algorithm. [Id. at ¶ 33]. As such, housing providers select the minimum approval score without knowing how scores are calculated. [Id. at ¶ 39].
B. Reliance on Credit History and Score
According to Plaintiffs, “the SafeRent Score assigned to an applicant dictates their rental eligibility” and “is calculated based in large part on factors that produce disproportionately lower SafeRent Scores for Black and Hispanic applicants, and those using housing vouchers.” [Id. at ¶ 41]. They allege that the SafeRent Score is based “in significant part on the applicant's credit score and credit history, including non-tenancy debts,” which are not intended to gauge whether an individual will be a “good tenant.” [Id. at ¶¶ 45-46]. In particular, these credit reviews fail to account accurately for an applicant's ability to pay rent, because they do not consider income and assets in their calculations. [Id. at ¶¶ 47-48].
Individuals receive a housing voucher only if they qualify as “extremely low-income,” “very low-income,” or “low-income.” [Id. at ¶ 58]. Individuals with lower incomes often have lower credit scores than those with higher incomes. [Id. at ¶ 59]. Housing providers who rent to tenants with housing vouchers are guaranteed to receive at least some of their tenants' monthly rental payments, because the local housing authority disburses payment directly to the housing provider. [Id. at ¶ 65]. Moreover, voucher holders may also request exemptions from the minimum rent they are required to pay if they experience certain hardships, such as loss of assistance programs, the threat of eviction, decreased income, loss of employment, or a death. [Id. at ¶ 67]. This further insulates housing providers from the risk of non-payment for units rented to voucher holders. [Id.]. As such, credit scores and histories fail to predict whether a rental applicant would make a quality tenant, particularly when housing vouchers are involved. [See id. at ¶ 61].
Reliance on conventional credit history disproportionately affects Black and Hispanic tenants, in addition to tenants who hold housing vouchers, because Black and Hispanic consumers have a lower median credit score than White consumers. [Id. at ¶¶ 50-51]. As of October 2021, Black consumers had a median credit score of 612 and Hispanic consumers had a median credit score of 661, while White consumers had a median credit score of 725. [Id. at ¶ 51]. Moreover, as of that date, 45.1% of Black consumers and 31.5% of Hispanic consumers had subprime credit scores, while only 18.3% of White consumers had subprime credit scores, which leads to less favorable credit terms. [Id. at ¶ 52]. These “[r]acial disparities in credit health reflect historical inequities that reduced wealth and limited economic choices for communities of color.” [Id. at ¶ 54]. For example, as of 2017, Black families owned less than seven cents for every dollar in wealth owned by White families, and Latino families owned less than eight cents of every dollar of wealth owned by White families. [Id.]. The past credit data factored into credit scoring models is “systemically and historically biased against non-white consumers.” [Id. at ¶ 53]. These racial disparities in credit health further “perpetuate wealth inequalities through reduced financial opportunities and fewer financial safety nets.” [Id. at ¶ 55 (emphasis omitted)].
C. Plaintiffs' Experiences
Louis is a 54-year-old Black woman who has a housing voucher that pays for approximately 69% of her rent. [Id. at ¶ 13]. Her rental application at Granada Highlands, which was managed by Metropolitan, was denied because of her SafeRent Score, which included non-tenancy related debt in its calculations. [Id. at ¶¶ 14, 20]. Metropolitan's rejection letter informed Louis that “the third-party service [Metropolitan] utilize[s] to screen all prospective tenants has denied [her] tenancy,” and “the service's SafeRent tenancy score was lower than is permissible under [Metropolitan's] tenancy standards.” [Id. at ¶ 76]. Although Louis attempted to appeal Metropolitan's decision, offering landlord and employment references, Metropolitan told Louis that it “do[es] not accept appeals and cannot override the outcome of the Tenant Screening.” [Id. at ¶ 78]. Louis then had to move into an apartment which cost more, had fewer amenities, and was located in a less desirable area with a high crime rate. [Id. at ¶¶ 79-80].
Douglas is a 65-year-old Black woman who has a housing voucher that pays for approximately 57% of her rent. [Id. at ¶ 15]. Her rental application at Millside at Heritage Park was similarly denied because of her SafeRent Score. The management company at Millside at Heritage Park, who is not named as a defendant in this action, informed Douglas that her “screening,” which was conducted by SafeRent, “came back with a deny comment based on unsatisfactory credit history as well as landlord tenant court record.” [Id. at ¶ 85]. Douglas appealed the decision with the help of a local housing advocacy group, and her rental application was ultimately accepted. [Id. at ¶¶ 86-89].
CAAS provides a variety of services to low-income, prospective renters, including those who have received housing vouchers. [Id. at ¶¶ 93-94]. CAAS helps voucher holders locate, apply for, and secure apartments where they can use their housing vouchers, and it frequently works with clients who have “excellent tenancy histories” but have non-tenancy related debt. [Id. at ¶¶ 94, 96]. Because leasing agents often tell CAAS advocates that only applicants with good credit histories or scores will pass screening, and CAAS has limited resources and therefore cannot support futile efforts, CAAS cannot regularly encourage its clients to apply to housing providers who use SafeRent's tenant-screening services. [Id. at ¶ 99]. This impedes CAAS's ability to accomplish its mission, and CAAS has had to expend “time and resources addressing the barriers that tenant screening companies impose” on CAAS's clients. [Id. at ¶¶ 99, 101].
D. The Fair Housing Act
The Fair Housing Act (“FHA”) prohibits discrimination in the sale or rental of housing on the basis of race, color, religion, sex, familial status, or national origin. See 42 U.S.C. § 3604. In the 1960s, segregation permeated society, affecting economic and social conditions nationwide, including housing. The FHA was enacted during “a period of considerable social unrest,” after an executive commission found that “both open and covert racial discrimination prevented [B]lack families from obtaining better housing and moving into integrated communities,” which was moving “our Nation . . . toward two societies, one [B]lack, one [W]hite-separate but unequal.” Tex. Dep't of Hous. & Cmt'y Affairs v. Inclusive Communities Project, Inc., 576 U.S. 519, 529 (2015). Various policies, practices, and prejudices, such as racially restrictive covenants preventing the conveyance of property to minorities, steering of potential buyers to consider homes in racially homogenous areas, and discriminatory lending practices, “created many predominantly [B]lack inner cities surrounded by mostly [W]hite suburbs.” Id. Congress enacted the FHA “to eradicate discriminatory practices within a sector of our Nation's economy,” including “housing restrictions that function unfairly to exclude minorities from certain neighborhoods without any sufficient justification.” Id. at 539.
E. The Amended Complaint and Procedural History
Plaintiffs filed an amended complaint on August 26, 2022. [See Dkt. 15]. Plaintiffs allege that SafeRent's tenant-screening services discriminate against voucher holders and, in particular, Black and Hispanic applicants in violation of the FHA, 42 U.S.C. § 3604, Massachusetts General Laws ch. 93A, § 9, and Massachusetts General Laws ch. 151B, §§ 4(6), 4(10). [Dkt. 15 at ¶ 1]. They bring a variety of claims against each Defendant individually and on behalf of race and source of income classes. Plaintiffs seek a declaratory judgment, an injunction, damages, fees, and costs. [Id. at ¶¶ 184-93].
The amended complaint states that the provision and use of the SafeRent Score has a disproportionate effect on Black and Hispanic applicants generally. However, Plaintiffs seek to bring race-based claims on behalf of “[a]ll Black and Hispanic rental applicants who use publicly funded housing vouchers and sought but were denied housing in Massachusetts because of their SafeRent Score at any property using Defendants' tenant screening services” after a particular date. [Am. Compl. at ¶¶ 104-05, 108 (emphasis added)]. As to the source of income class, Plaintiffs seek to bring claims on behalf of “[a]ll rental applicants who use publicly funded housing vouchers and sought but were denied housing in Massachusetts because of their SafeRent Score at any property using Defendants' tenant screening services” after a particular date. [Id. at ¶¶ 106-07].
As to SafeRent, Plaintiffs allege that the SafeRent Score violates the FHA by making housing unavailable on the basis of race and national origin in violation of 42 U.S.C. § 3604(a) and by providing different terms and conditions and discriminating in the provision of services in connection with housing on the basis of race and national origin in violation of 42 U.S.C. § 3604(b) (“Count I”). [Id. at ¶¶ 120-26]. They bring claims for race discrimination under Massachusetts General Law ch. 151B, § 4(6)(b) (“Count III”) and source of income discrimination under Massachusetts General Law ch. 151B, § 4(10) (“Count V”). [Id. at ¶¶ 13441, 150-57]. Plaintiffs also claim that SafeRent's provision of the SafeRent Score is an unfair business practice in violation of Massachusetts General Law ch. 93A, § 9 (“Counts VII and VIII”). [Id. at ¶¶ 165-83].
Only Louis brings claims against Metropolitan. Louis alleges that Metropolitan's practice of using the SafeRent Score to screen rental applicants violates the FHA by making housing unavailable on the basis of race and national origin in violation of 42 U.S.C. § 3604(a) and by providing different terms and conditions and discriminating in the provision of services in connection with housing on the basis of race and national origin in violation of 42 U.S.C. § 3604(b) (“Count II”). [Id. at ¶¶ 127-33]. She brings claims for race discrimination under Massachusetts General Law ch. 151B, § 4(6)(b) (“Count IV”) and source of income discrimination under Massachusetts General Law ch. 151B, § 4(10) (“Count VI”). [Id. at ¶¶ 14249, 158-64].
In response to the amended complaint, Metropolitan filed a motion to dismiss for lack of standing and for failure to state a claim on October 27, 2022. [Dkt. 29]. That same day, SafeRent filed a motion to dismiss for failure to state a claim pursuant to Federal Rule of Civil Procedure 12(b)(6) and for lack of jurisdiction, specifically, lack of standing, pursuant to Federal Rule of Civil Procedure 12(b)(1). [Dkt. 31]. Plaintiffs filed a single opposition to those motions. [Dkt. 36]. On January 9, 2023, the United States submitted a statement of interest regarding the application and interpretation of the FHA to this action. [Dkt. 37]. Metropolitan and SafeRent's replies in support of their motions addressed Plaintiffs' opposition and the United States' amicus brief. [See Dkt. 46; Dkt. 47]. The Consumer Data Industry Association (“CDIA”) submitted a brief in support of SafeRent's motion to dismiss. [Dkt. 44]. With the Court's permission, Plaintiffs filed a sur-reply to the CDIA's memorandum on March 8, 2023. [Dkt. 51]. The Court held a motion hearing on July 7, 2023. [Dkt. 58].
II. LEGAL STANDARD
To survive a motion to dismiss under Federal Rule of Civil Procedure 12(b)(6), a complaint must allege sufficient facts to state a claim for relief that is “plausible on its face” and actionable as a matter of law. Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009) (quoting Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007)). Reading the complaint “as a whole,” the court must conduct a two-step, context-specific inquiry. Garda-Catalan v. United States, 734 F.3d 100, 103 (1st Cir. 2013). First, the court must perform a close reading of the complaint to distinguish factual allegations from conclusory legal statements. Id. Factual allegations must be accepted as true, while legal conclusions are not entitled to credit. Id. A court may not disregard properly pleaded factual allegations even if actual proof of those facts is improbable. Ocasio-Hernandez v. Fortuno-Burset, 640 F.3d 1, 12 (1st Cir. 2011). Second, the court must determine whether the factual allegations present a “reasonable inference that the defendant is liable for the misconduct alleged.” Haley v. City of Bos., 657 F.3d 39, 46 (1st Cir. 2011) (citation omitted). Dismissal is appropriate when the complaint fails to allege a “plausible entitlement to relief.” Rodriguez-Ortiz v. Margo Caribe, Inc., 490 F.3d 92, 95 (1st Cir. 2007) (quoting Bell Atl. Corp. v. Twombly, 550 U.S. 544, 559 (2007)).
“Although review of a Rule 12(b)(6) dismissal for failure to state a claim and review to ensure the existence of standing are conceptually distinct, the same basic principles apply in both situations.” Hochendoner v. Genzyme Corp., 823 F.3d 724, 730 (1st Cir. 2016); see Murphy v. United States, 45 F.3d 520, 522 (1st Cir. 1995) (explaining that review for dismissal for lack of subject matter jurisdiction under Federal Rule of Civil Procedure 12(b)(1) is “similar to that accorded a dismissal for failure to state a claim pursuant to” Federal Rule of Civil Procedure 12(b)(6)). That is, when “a district court considers a Rule 12(b)(1) motion, it must credit the plaintiff's well-pled factual allegations and draw all reasonable inferences in the plaintiff's favor.” Merlonghi v. United States, 620 F.3d 50, 54 (1st Cir. 2010). The party “invoking the jurisdiction of a federal court carries the burden of proving its existence,” and a “plaintiff cannot rest a jurisdictional basis merely on unsupported conclusions or interpretations of law.” Johansen v. United States, 506 F.3d 65, 68 (1st Cir. 2007) (internal quotation marks and citations omitted). A federal court's subject matter jurisdiction “must be apparent from the face of the plaintiffs' pleading,” id., and “the plausibility standard applicable under Rule 12(b)(6)” applies “to standing determinations at the pleading stage,” Hochendoner, 823 F.3d at 730. “Neither conclusory assertions nor unfounded speculation can supply the necessary heft” to establish standing at the pleading stage. Id. at 731.
III. DISCUSSION
Defendants raise several arguments in favor of dismissal. Metropolitan argues that Louis does not have standing to assert claims for housing discrimination under the FHA or Massachusetts law. [Dkt. 30 at 15-18]. SafeRent similarly contends that CAAS lacks organizational standing, and that Plaintiffs lack standing to bring claims pursuant to 42 U.S.C. § 3604(b) (“Section 3604(b)”) and Massachusetts General Laws ch. 151B, § 4(6)(b) (“Section 4(6)(b)”). [Dkt. 32 at 22-24]. SafeRent also claims that the FHA does not apply to it, which Plaintiffs and the United States amicus brief dispute. [Dkt. 32 at 15-18; Dkt. 36 at 17-24; Dkt. 37 at 12-14]. Defendants also argue that Plaintiffs have failed to state a claim for disparate impact housing discrimination under federal and state law. [Dkt. 30 at 7-15; Dkt. 32 at 6-14]. Finally, SafeRent argues that Plaintiffs have failed to state a claim under Massachusetts General Law ch. 93A, § 9 (“Chapter 93A”). [Dkt. 32 at 19-22].
A. Standing
Standing is a “prerequisite to a federal court's subject matter jurisdiction,” Hochendoner, 823 F.3d at 730, and “therefore must be resolved before a court may reach the merits,” Cowels v. Fed. Bureau of Investigation, 327 F.Supp.3d 242, 248 (D. Mass. 2018). The Constitution limits the jurisdiction of federal courts to actual cases or controversies. U.S. Const. art. III, § 2, cl. 1. For a “case or controversy” to exist, all plaintiffs must be able to demonstrate a “personal stake in the outcome of the controversy.” Katz v. Pershing, LLC, 672 F.3d 64, 71 (1st Cir. 2012) (quoting Baker v. Carr, 369 U.S. 186, 204 (1962)). To meet the “constitutional minimum” standing requirements, all plaintiffs must establish they have “(1) suffered an injury in fact, (2) that is fairly traceable to the challenged conduct of the defendant, and (3) that is likely to be redressed by a favorable judicial decision.” Spokeo, Inc. v. Robins, 578 U.S. 330, 338 (2016) (citing Lujan v. Defenders of Wildlife, 504 U.S. 555, 560-61 (1992)).
To allege an injury in fact, plaintiffs must show they have suffered an “invasion of a legally protected interest” that was “concrete and particularized” and “actual or imminent, not conjectural or hypothetical.” Id. at 339 (quoting Lujan, 504 U.S. at 560). Particularity requires plaintiffs to demonstrate that they have personally suffered some harm, and the “actual or imminent” requirement “ensures that the harm has either happened or is sufficiently threatening.” Katz, 672 F.3d at 71. “[I]t is not enough that the harm might occur at some future time,” and a “generalized grievance” is not enough. Id. at 71-72. A “mere interest in an event,” no matter how “passionate” or “charged with public import,” is not sufficient to provide an actual injury. United States v. AVX Corp., 962 F.2d 108, 114 (1st Cir. 1992). As to the second element, traceability, all plaintiffs must show “a sufficiently direct causal connection between the challenged action” of the defendant “and the identified harm.” Dantzler, Inc. v. Empresas Berrios Inventory & Operations, Inc., 958 F.3d 38, 47 (1st Cir. 2020) (citations omitted). This connection cannot be “overly attenuated” and cannot stem from the independent action of a third party. Id. “Although an indirect causal relationship is not prima facie fatal, an injury is less likely to satisfy this requirement where the causal chain between the defendant's action and the alleged harm depends on actions of a third party.” Delaney v. Baker, 511 F.Supp.3d 55, 68 (D. Mass. 2021). The final standing hurdle, redressability, requires the Court to “consider the relationship between ‘the judicial relief requested' and the ‘injury' suffered.” Efreom v. McKee, 46 F.4th 9, 21 (1st Cir. 2022) (citing California v. Texas, 141 S.Ct. 2104, 2115 (2021)).
1. Louis (Counts II, IV, and VI)
Metropolitan argues that Louis has not pleaded facts sufficient to demonstrate an injury in fact, causation, or redressability. [Dkt. 30 at 15-18]. Louis alleges she was injured because she had to move into an apartment with fewer amenities and fewer bathrooms in a “less desirable area due to the area's high crime rate.” [Am. Compl. at ¶¶ 79-80]. She also states that the apartment she ultimately secured cost $200 more per month than the unit she sought to rent from Metropolitan. [Id. at ¶ 79]. Louis further alleges that this was “a direct result of [Metropolitan's] rejection” of her application. [Id.]. Such allegations, which are “personal and individual,” see Lujan, 504 U.S. at 560 n.1, are sufficient to establish an injury in fact and causation at this stage, see Czyzewsju v. Jevic Holding Corp., 580 U.S. 451, 464 (2017) (“For standing purposes, a loss of even a small amount of money is ordinarily an ‘injury.'”). As to redressability, Louis seeks damages and equitable relief that would prohibit Metropolitan from using the type of tenantscreening services that led to Louis' rental application denial. [See Am. Compl. at ¶¶ 184-93]. Such relief would redress Louis' injuries. See Gustavsen v. Alcon Labs., 903 F.3d 1, 9 (1st Cir. 2018) (“Nor can there be any doubt that plaintiffs' financial injury can be redressed by damages.”); see also Antilles Cement Corp. v. Fortuno, 670 F.3d 310, 318 (1st Cir. 2012) (explaining that the plaintiff “need only show that a favorable ruling could potentially lessen its injury” and “need not definitively demonstrate that a victory would completely remedy the harm”). Therefore, Louis has pleaded facts sufficient to establish standing.
2. CAAS (Counts I, III, V, VII, VIII)
CAAS, as an organization bringing claims here, is also subject to constitutional standing requirements. The Court therefore “conduct[s] the same inquiry as in the case of an individual: Has the plaintiff ‘alleged such a personal stake in the outcome of the controversy' as to warrant [its] invocation of federal-court jurisdiction?” Havens Realty Corp. v. Coleman, 455 U.S. 363, 378-79 (1982) (citation omitted). An organization has standing “if it can show that the defendant's actions cause a ‘concrete and demonstrable injury to the organization's activities' that is ‘more than simply a setback to the organization's abstract social interests,'” which is “not a demanding standard,” as “only a perceptible impairment of an organization's activities is necessary for there to be an injury in fact.'” Nat'l Ass'n of Consumer Advocates v. Uejio, 521 F.Supp.3d 130, 142 (D. Mass. 2021) (citations omitted). SafeRent claims that CAAS has provided only vague and conclusory allegations of “diverted” resources and a “frustrated” mission, which are insufficient to establish organizational standing. [Dkt. 32 at 23-24]. Plaintiffs respond that CAAS has alleged that it must spend more time and resources helping clients who are unlikely to pass SafeRent's screening, which is sufficient at the pleading stage to show an injury in fact. [Dkt. 36 at 32-35]. Plaintiffs prevail here.
CAAS alleges that it “provides a multitude of services to low-income housing-seekers,” working “with 29 tenants with vouchers in the last year or about 36% of their entire housing caseload,” and that 28% of the individuals who qualify for CAAS' services identify as Black and 20% identify as Hispanic. [Am. Compl. at ¶¶ 18, 93]. CAAS further claims that it “frequently encounter[s] clients who have excellent tenancy histories but have non-tenant related debt,” and that “[w]henever an applicant is denied an apartment because of credit screening or a low SafeRent Score, CAAS staff must expend additional time trying to help that applicant find another apartment.” [Id. at ¶¶ 96, 98]. CAAS elaborates:
Not being able to regularly encourage their clients to apply to housing providers who use SafeRent's tenant screening services limits the number of housing providers their clients can apply to. That, in turn, makes it harder for CAAS to accomplish its mission, particularly because the largest property management companies, with the most units, are most likely to use SafeRent or similar screening vendors....CAAS has expended time and resources addressing the barriers that tenant screening companies impose on their
clients. CAAS must therefore expend more money and resources in order to place their clients into other properties where they are not turned away because of their credit or other non-tenancy debt histories.[Id. at ¶¶ 99, 101]. As “broadly alleged,” Defendants' use of the SafeRent Score, which relies on credit history, has “perceptibly impaired [CAAS'] ability to provide counseling and referral services” for its clients, and “there can be no question that the organization has suffered injury in fact.” Havens Realty, 455 U.S. at 379.
Contrary to SafeRent's assertion, CAAS' allegations are neither conclusory nor insufficient, and their reliance on Equal Means Equal v. Ferriero, 3 F.4th 24, 30 (1st Cir. 2021) is misplaced. In Equal Means Equal, the organizations premised their standing on allegations that the defendant's unlawful conduct affected their lobbying activities and issue advocacy. Id. This type of harm is inapposite to the harm alleged here. Unlike the organizations in Equal Means Equal, CAAS states that it has already had to “expend more money and resources in order to place their clients into other properties where they are not turned away because of their credit or other non-tenancy debt histories,” which has “frustrated CAAS' mission by interfering with its ability to assist voucher-holders in Massachusetts obtain housing”; they do not allege mere advocacy. [Am. Compl. at ¶¶ 100-02]. “Such concrete and demonstrable injury to the organization's activities-with the consequent drain on the organization's resources-constitutes far more than simply a setback to the organization's abstract social interests.” Havens Realty, 455 U.S. at 379 (citing Sierra Club v. Morton, 405 U.S. 727, 739 (1972)). CAAS has pleaded facts sufficient to establish standing at this juncture.
3. Section 3604(b) and Section 4(6)(b) (Counts I-IV)
SafeRent states that Plaintiffs have not established an injury in fact sufficient to support claims for violations of Section 3604(b) and Section 4(6)(b), which prohibit discrimination in the “terms, conditions, or privileges” of the sale or rental of housing or the provision of services in connection therewith. [Dkt. 32 at 22-23]. According to SafeRent, neither Louis nor Douglas has alleged an injury in fact, because Louis was never approved for housing, and therefore could not have received less favorable terms and conditions, and Douglas never alleges that she was offered less favorable terms or received less favorable services as compared to non-Black and non-voucher renters. [Id.]. Plaintiffs reply that SafeRent's interpretation of those sections is too narrow. [Dkt. 36 at 29-31].
Plaintiffs must establish standing “for each claim that they press and for each form of relief that they seek.” In re Evenflo Co., Inc., Mktg., Sales Pracs. & Prods. Liab. Litig., 54 F.4th 28, 34 (1st Cir. 2022) (citing TransUnion, LLC v. Ramirez, 141 S.Ct. 2190, 2208 (2021)). Section 3604(b) and the corresponding state law make it unlawful to “discriminate against any person in the terms, conditions, or privileges of sale or rental of a dwelling, or in the provision of services or facilities in connection therewith, because of race, color, religion, sex, familial status, or national origin.” 42 U.S.C. § 3604(b); see Mass. Gen. Laws ch. 151B, § 4(6)(b) (prohibiting discrimination in the “terms, conditions or privileges of such accommodations or the acquisitions thereof, or in the furnishings of facilities and services in connection therewith”). Here, Louis and Douglas allege that the application process, particularly the use of the SafeRent Score, facilitated by the Defendants discriminated against them. [See Am. Compl. at ¶¶ 76, 84]. Such allegations are sufficient, at this stage, to confer standing for, at minimum, discrimination in the privilege of renting. See United States v. Hylton, 944 F.Supp.2d 176, 188-89 (D. Conn. 2013) (finding that the defendant discriminated against a potential sublessee “in the privileges of a rental by refusing to allow her to rent” the unit, as such refusal prevented the sublessee “from assuming residency and enjoying the privilege of renting”).
B. Application of the FHA and Massachusetts Law to SafeRent (Counts I, III and V)
SafeRent submits that the FHA does not apply to SafeRent, a tenant-screening service, because SafeRent does not make housing decisions. [Dkt. 32 at 15-17]. SafeRent also states that the Massachusetts housing discrimination laws, which are even narrower in reach, do not apply to it. [Id. at 17-18]. Plaintiffs counter that SafeRent's tenant-screening services result in a discriminatory housing practice, and it is this consequence that renders SafeRent liable under the FHA. [Dkt. 36 at 17-21]. Plaintiffs also assert that SafeRent is liable under Massachusetts law because it had a role in the “tenant selection process” and, alternatively, is a “credit service” subject to the state housing discrimination laws. [Id. at 21-24].
SafeRent relies on its assertion that it “is not a landlord and does not otherwise have authority to make housing decisions.” [Dkt. 32 at 17]. This, however, does not accurately capture SafeRent's role in the rental application process, at least as alleged by Plaintiffs here, and it conflicts with the intent and clear statutory text of the FHA. Plaintiffs allege that SafeRent has violated two provisions of the FHA: Section 3604(a), which prohibits conduct that “otherwise make[s] unavailable . . . a dwelling to any person because of race . . . or national origin,” and Section 3604(b), which proscribes discrimination in the “terms, conditions, or privileges of a sale or rental of a dwelling, or in the provision of services” because of race or national origin. 42 U.S.C. § 3604. Contrary to SafeRent's assertions, neither provision limits liability to people or entities that “make housing decisions.” [See, e.g., Dkt. 32 at 15-17]. Section 3604 instead “focuses on prohibited acts,” Meyer v. Holley, 537 U.S. 280, 285 (2003), and the Supreme Court has explained that the “language of the Act is broad and inclusive,” Trafficante v. Metropolitan Life Ins. Co., 409 U.S. 205, 209 (1972). That is, “Congress' use of the phrase ‘otherwise make unavailable' refers to the consequences of an action rather than the actor's intent,” and it is “results-oriented.” Inclusive Communities, 576 U.S. at 534. Moreover, regulations issued by the Department of Housing and Urban Development (“HUD”) provide that an actor may be held liable under the FHA for its “own conduct that results in a discriminatory housing practice.” 24 C.F.R. § 100.7(a)(1)(i) (emphasis added). While “housing providers are often the target of FHA claims . . . other entities are frequently held liable under the FHA,” and “[n]othing in the language of the statute precludes” SafeRent's liability here. Conn. Fair Housing Ctr. v. CoreLogic Rental Prop. Solutions, LLC, 369 F.Supp.3d 362, 374-75 (D. Conn. 2019).
Plaintiffs allege that SafeRent markets and sells its SafeRent Score to housing providers to “automate human judgment” and to “identify top quality applicants” and “better renters.” [Am. Compl. at ¶¶ 22, 25, 29, 41-42]. While SafeRent delivers “an accept/decline/conditional decision” based on the housing provider's “predetermined decision points,” those housing providers “cannot change the screening algorithm” and do not know how the SafeRent Score is calculated before selecting its minimum score for applicant approval. [Id. at ¶¶ 24, 26, 33-35, 39]. In fact, Plaintiffs allege that SafeRent does not share any details about its algorithm's design with housing providers, rental applicants, or the public generally. [Id. at ¶¶ 34, 36-37]. Moreover, Plaintiffs state that housing providers rely on SafeRent's evaluations-the SafeRent Score-as the basis for any decisions to accept or deny a rental application. [Id. at ¶ 38]. Metropolitan explained to Louis that “the third-party service” it utilized “to screen all prospective tenants has denied [her] tenancy,” and it “cannot override the outcome of the Tenant Screening.” [Id. at ¶¶ 44, 47]. Contrary to SafeRent's characterization of its actions, the amended complaint plausibly claims that SafeRent did have the “authority to make housing decisions” [see Dkt. 32 at 16], because SafeRent “effectively controls the decision to approve or reject a rental application,” as it “has sole control over how scores are calculated” [Am. Compl. at ¶ 43]. Taken together, the allegations in the amended complaint suggest that SafeRent's provision of the SafeRent Score is “directly related” to the rental transaction “because it determined who was qualified to occupy a housing unit.” CoreLogic, 369 F.Supp.3d at 373. Because that determination may disqualify otherwise qualified rental applicants and, as alleged, results in a disparate impact on protected groups, SafeRent is subject to the FHA. See id. at 374-75 (noting that the application of the FHA “beyond direct housing providers” is a “logical extension[] which effectuate[s] the purpose of the FHA,” as “a housing provider could simply use an intermediary to take discriminatory and prohibited actions on its behalf and defeat the purpose of the FHA”).
SafeRent attempts to differentiate Conn. Fair Housing Ctr. v. CoreLogic Rental Prop. Solutions, LLC, 369 F.Supp.3d 362, 375 (D. Conn. 2019), by arguing that the court there found that the tenant-screening company-SafeRent, albeit going by a different name-could be held liable under the FHA because it “knew of and had the power to correct and end a discriminatory practice” and because it was the housing provider's agent. [Dkt. 32 at 17]. However, SafeRent's description does not accurately capture that court's decision. The portions of the opinion that are cited by SafeRent largely appear in the court's discussion of how a Second Circuit decision supported its finding that screening companies can be held liable under the FHA. CoreLogic, 369 F.Supp.3d at 374. By focusing on the court's evaluation of narrow theories of liability, SafeRent ignores CoreLogic's central holding. Elsewhere in the decision, the court explains that the screening company could be held directly liable under the FHA for several reasons, observing that the defendant “held itself out as a company with the knowledge and ingenuity to screen housing applicants” and “specifically advertised its ability to improve ‘Fair Housing compliance.'” Id. at 372. The court also notes that “[a]llowing a screening company to facilitate discrimination by disqualifying qualified applicants on an impermissible basis or by allowing a customer to set impermissible qualification standards with impunity would subvert the purpose of the FHA.” Id. Additionally, in distinguishing the case from another decision relied upon by the defendant, the court observes that the policy challenged before it “is directly related to the real estate transaction because it determined who was qualified to occupy a housing unit,” and the “clear relationship between the challenged practice and the attempted property rental” weighed in favor of finding that the tenantscreening company could be held directly liable. Id. at 373.
CoreLogic proceeded to a bench trial. See Conn. Fair Housing Ctr. v. CoreLogic Rental Prop. Solutions, LLC, No. 18-CV-705-VLB, 2023 WL 4669482, at *1 (D. Conn. July 20, 2023). In the court's memorandum of decision and order, the court noted: “[B]efore the Court can evaluate whether the Plaintiffs have met their burden on the elements of their disparate impact and treatment claims, the Plaintiffs must prove that CoreLogic denies or otherwise makes housing unavailable.” Id. at *17. In other words, the court had to determine whether, on the evidence provided, CoreLogic was subject to the FHA in that case. The court explained that “[b]ased on the facts presented during trial, the Court concludes that neither of the Plaintiffs' theories of liability [of how CoreLogic ‘denies or makes housing unavailable'] are supported by a preponderance of the evidence.” Id. CoreLogic was decided after the benefit of discovery, and the case before this Court is at a different procedural posture. Moreover, CoreLogic involved a different aspect of the company's tenant-screening services, and the facts and theories presented to the CoreLogic court are not before this Court.
SafeRent's attempt to escape liability under Massachusetts law also fails. While Sections 4(6) and 4(10) do list a number of actors to which they apply, see Crossing Over, Inc. v. City of Fitchburg, 161 N.E.3d 432, 443 n.17 (Mass. App. Ct. 2020) (noting that Section 4(6) “list[s] a series of private actors . . . to whom [it] appl[ies,], while the FHA[] is drafted more broadly”), another subsection provides that it is unlawful “to aid, abet, incite, compel or coerce the doing of any of the acts forbidden under this chapter,” which includes Sections 4(6) and 4(10), Mass. Gen. Laws. ch. 151B, § 4(5). Plaintiffs allege that SafeRent provides Metropolitan and other housing management companies decisions on rental applications. [See Am. Compl. at ¶¶ 22-24, 27-28, 39-44]. The Massachusetts Supreme Court has found that defendants who play a role in the tenant selection process may be held liable under, at minimum, Section 4(10), even if they only “aid[ed] or abet[ted]” a violation of Section 4(10). See DiLiddo v. Oxford St. Realty, Inc., 876 N.E.2d 421, 430 (Mass. 2007). Such logic similarly applies here to both the Section 4(6) and 4(10) claims, and they survive the motion to dismiss on this basis. See Mass. Gen. Laws. ch. 151B, § 9 (proclaiming that “[t]his chapter shall be construed liberally for the accomplishment of its purposes”).
Because the Court finds a sufficient basis for the Section 4(6) and 4(10) claims against SafeRent to survive the motion to dismiss stage, it does not evaluate the parties' arguments about whether SafeRent furnishes credit services such that it can be held liable under Section 4(10). [See Dkt. 32 at 18; Dkt. 36 at 23-24; Dkt. 47 at 12].
C. Disparate Impact under the FHA and Massachusetts Law (Counts I-VI)
The heart of Defendants' arguments for dismissal is that Plaintiffs have not pleaded a prima facie claim of disparate impact discrimination for race or source of income under federal or state housing laws. [Dkt. 30 at 8-15; Dkt. 32 at 6-15]. In particular, they submit that Plaintiffs have failed to allege that the use of the SafeRent Score is an “artificial, arbitrary, and unnecessary” policy that creates a barrier for Black, Hispanic, or voucher-holding applicants; that the amended complaint does not plead a disparate impact because it does not allege any significant statistical disparity in housing outcomes for Black, Hispanic, or voucher-holding applicants; and that Plaintiffs have failed to establish that the Defendants' actions caused any such disparity in housing outcomes. [Dkt. 30 at 8-15; Dkt. 32 at 6-15]. Plaintiffs counter that their allegations regarding the disparities in credit histories and scores, when viewed alongside their allegations that the SafeRent Score incorporates such information into its tenant-screening, is sufficient to plead a claim of disparate impact here. [Dkt. 36 at 5-17].
Disparate impact claims are actionable under the FHA. Inclusive Communities, 576 U.S. at 539 (noting that “[r]ecognition of disparate-impact claims is consistent with the FHA's central purpose”). They are also cognizable under Massachusetts housing discrimination laws, which follow the same burden-shifting framework as federal housing discrimination claims. Burbank Apartments Tenant Ass'n v. Kargman, 48 N.E.3d 394, 407, 411 (Mass. 2016). As such, Plaintiffs' federal and state law discrimination claims rise or fall together. A disparate impact claim “challenges practices that have a ‘disproportionately adverse effect on minorities' and are otherwise unjustified by a legitimate rationale.” Inclusive Communities, 576 U.S. at 524 (citing Ricci v. DeStefano, 557 U.S. 557, 577 (2009)). “A plaintiff can make out a claim for disparate impact by showing that a defendant's actions actually or predictably result in racial discrimination.” Barrow v. Barrow, No. CV 16-11493-FDS, 2017 WL 2872820, at *3 (D. Mass. Jul. 5, 2017) (citing Macone v. Town of Wakefield, 277 F.3d 1, 7 (1st Cir. 2002)) (modifications omitted). A plaintiff must plead (1) a specific and actionable policy, (2) a disparate impact, and (3) facts raising a sufficient inference of causation. Miller v. Countrywide Bank, N.A., 571 F.Supp.2d 251, 255 (D. Mass. 2008). Courts must engage in “a rigorous examination on the merits at the pleading stage” to “balance the interests of both property owners and protected classes.” Burbank Apartments, 48 N.E.3d at 411 (citation omitted).
1. Specific and Actionable Policy
Plaintiffs must point to “specific policies” that they allege result in racial discrimination to satisfy the first element of a disparate impact claim. See Barrow v. Barrow, No. CV 16-11493-FDS, 2016 WL 6996996, at *5 (D. Mass. Nov. 29, 2016). Plaintiffs have alleged that Defendants' provision and use of the SafeRent Score when evaluating a rental application is such a policy, because the SafeRent Score relies on credit history, and Black, Hispanic, and low-income voucher holders are disproportionately likely to have poor credit history, resulting in a lower SafeRent Score and causing a disproportionate denial of housing. [See Am. Compl. at ¶¶ 122-23, 129-30, 136-37, 144-45, 152-53, 160-61]. Plaintiffs have therefore pleaded a “specific and actionable” policy. See Barrow, 2017 WL 2872820, at *3.
Defendants rely on Inclusive Communities, 576 U.S. at 543, to insist that Plaintiff's have failed to plead that the provision and use of the SafeRent Score to evaluate rental applications create an “artificial, arbitrary, and unnecessary barrier[]” for Black, Hispanic, and voucherholding applicants. [See Dkt. 30 at 9-11; Dkt. 32 at 7-9]. Inclusive Communities, however, did not add such a requirement at the pleading stage. Rather, Inclusive Communities invoked the “artificial, arbitrary, and unnecessary” language to illustrate the types of policies that the FHA is intended to reach, that is, the “ultimate purpose of disparate-impact liability.” Nat'l Fair Housing Alliance v. Travelers Indemnity Co., 261 F.Supp.3d 20, 29 (D.D.C. 2017) (quoting Inclusive Communities, 576 U.S. at 543). As such, Defendants' attempt to require such allegations at the pleading stage fails. And even if it did not, Plaintiffs have alleged facts suggesting that the SafeRent Score's reliance on credit history is arbitrary and unnecessary.[See Am. Compl. at ¶¶ 30, 35, 45-50, 72, 131-32, 138-39, 146-47, 154-55, 162-63].
Defendants rely on several decisions from other circuits to support their argument. See, e.g., Ellis v. City of Minneapolis, 860 F.3d 1106, 1114 (8th Cir. 2017). Those decisions are not binding on this Court, and the Court disagrees with their reading of the “artificial, arbitrary, and unnecessary” language in Tex. Dep't of Hous. & Cmt'y Affairs v. Inclusive Communities Project, Inc., 576 U.S. 519, 543 (2015), which addressed a district court's partial grant of summary judgment and the following bench trial.
The parties make much of the Department of Housing and Urban Development's guidance on the use of credit scores and history when evaluating rental applications. [See Dkt. 30 at 10-11; Dkt. 32 at 7-9; Dkt. 36 at 6-10]. While such guidance may be relevant to whether Defendants' actions ultimately created an “artificial, arbitrary, and unnecessary” barrier to housing, the Court need not evaluate that issue at the motion to dismiss stage. See Inclusive Communities, 576 U.S. at 543.
2. Disparate Impact
Defendants argue that the Plaintiffs do not allege facts demonstrating a disparate impact on Black and Hispanic applicants and voucher holders. [Dkt. 30 at 12-13; Dkt. 32 at 9-11]. In particular, SafeRent contends that the amended complaint provides no data regarding the housing outcomes of the two protected groups. [Dkt. 32 at 10]. Additionally, SafeRent claims that the statistical evidence provided in the amended complaint is inadequate because Plaintiffs ask the Court to infer a statistical disparity in housing outcomes based on racial disparities in credit history, but credit history is just one element of the formula used to calculate the SafeRent Score. [Id. at 10-11]. Metropolitan similarly argues that allegations in the amended complaint “say nothing about the actual people who applied to rent” at Metropolitan's properties, nor do they state anything about whether Black and Hispanic applicants and voucher holders are more frequently denied apartments at Metropolitan's properties. [Dkt. 30 at 12]. In other words, according to Metropolitan, the amended complaint “alleges disparate credit scores amongst the general population, but there are no allegations of a disparate impact at [Metropolitan's property] corresponding with these disparate credit scores.” [Id.]. Plaintiffs counter that the amended complaint is “replete” with allegations regarding the credit histories of Black, Hispanic, and low-income individuals, and it alleges “in detail” how SafeRent Scores incorporate such information into its calculations. [Dkt. 36 at 10]. According to Plaintiffs, this “predictable disparity is sufficient to satisfy pleading standards.” [Id. at 11].
To satisfy the second element of a disparate impact claim, a plaintiff must allege “specific facts showing a disparate impact.” Barrow, 2016 WL 6996996, at *5. These facts must “tie to the acts or practices of [the] defendants].” Id. “A showing of disparate impact is usually made using statistical evidence.” Id. (citing Hallmark Developers, Inc. v. Fulton Cnty., 466 F.3d 1276, 1286 (11th Cir. 2006)); see Mandala v. NTT Data, Inc., 975 F.3d 202, 209 (2d Cir. 2020) (“To nudge a disparate impact claim across the line from conceivable to plausible- and, indeed, to ultimately prove such a claim-plaintiffs typically rely on statistical evidence to show a disparity in outcome between groups.”).
Plaintiffs have alleged significant statistical disparities in the credit histories of Black and Hispanic consumers as compared to White consumers and of voucher holders as compared to individuals who do not use vouchers. [Am. Compl. at ¶¶ 51-61]. The question is whether these allegations are sufficient at the pleading stage to allow the Court to infer that such disparities result in “an adverse effect on the protected group.” Greater New Orleans Fair Hous. Action Ctr. v. U.S. Dep't of Hous. & Urban Dev., 639 F.3d 1078, 1085-86 (D.C. Cir. 2011) (quoting Garcia v. Johanns, 444 F.3d 625, 633 (D.C. Cir. 2006)). They are.
Plaintiffs have alleged that Defendants have a policy of providing and using the SafeRent Score to evaluate rental applications, and that the SafeRent Score relies heavily on credit history. Plaintiffs further allege that this policy has a disproportionate impact on Black and Hispanic applicants and voucher holders because they are significantly more likely to have poor credit histories, which results in a disproportionate rate of housing denials for these protected groups. In other words, if a lower credit score makes it more likely that a rental applicant will be denied, and Black and Hispanic individuals and voucher holders are disproportionately likely to have lower credit scores, it follows that any policy that relies on credit scores to evaluate rental applications will have a disproportionate impact on Black and Hispanic applicants and voucher holders who apply. This risk is heightened if the policy does not account for the impact a voucher has on a tenant's ability to pay rent each month. Louis and Douglas also specifically allege that they were denied housing because of their SafeRent Scores. See Miller, 571 F.Supp.2d at 259 (noting that the alleged facts gave rise to an inference of disparate impact “especially [] in light of the individual named plaintiffs' allegations”). Combined, these allegations are sufficient at the pleading stage to raise an inference of disparate impact, that is, that the “challenged practice caused or predictably will cause a discriminatory effect.” Burbank Apartments, 48 N.E. 3d at 411 (emphasis added); see CoreLogic, 369 F.Supp.3d at 379 (“Defendant's claim that dismissal is required unless Plaintiffs allege statistical evidence of how the policy affects the protected and unprotected group again sets Plaintiffs' bar too high.”); see also Nat'l Fair Housing Alliance, 261 F.Supp.3d at 30 (explaining that courts do not need “to abandon common sense or necessary logical inferences that follow from the facts alleged” (citing Inclusive Communities, 576 U.S. at 543)).
The fact that credit history is only one piece of SafeRent's formula does not doom Plaintiffs' disparate impact claims at this stage. Plaintiffs allege that the SafeRent Score relies “in significant part on the applicant's credit score and credit history,” in addition to other factors, such as “bankruptcy records, past due accounts, payment performance, and eviction history.” [Am. Compl. at ¶¶ 35, 45]. With the exception of eviction history, all of these factors may contribute to credit history and the wealth inequalities that Plaintiffs allege exist and result in a disparate impact here. [See id. at ¶¶ 50-60]. Whether it is credit history or other SafeRent Score factors that drive any disparate impact on the protected groups is a question for a later stage in this proceeding.
3. Causality
Defendants argue that the amended complaint does not plausibly allege that either Defendant's actions caused any disparity in housing decisions. SafeRent and Metropolitan maintain that Plaintiffs fail to allege facts supporting an inference that the SafeRent Score's use of credit history, and not some other factor, causes any disparity in housing outcomes. [Dkt. 30 at 13-15; Dkt. 32 at 12]. SafeRent also contends that it is the housing providers, not SafeRent, who makes the housing decisions. [Dkt. 32 at 12]. Plaintiffs respond that the factors considered by the SafeRent score, as listed in the amended complaint and including credit history, are related to and reinforce each other. [Dkt. 36 at 13-15]. Because Plaintiffs have “alleged disparities across the range of inputs to the SafeRent Score,” the amended complaint states facts sufficient to support causality. [Id.]. Plaintiffs also reply that the amended complaint states facts sufficient to show how SafeRent's policies will cause predictable disparities in housing and that SafeRent is not absolved of liability simply because housing providers can choose to use SafeRent Scores. [Id. at 16].
The Supreme Court has made it clear that disparate impact claims under the FHA have a “robust causality requirement,” which ensures that “racial imbalance, does not, without more, establish a prima facie case of disparate impact and thus protects defendants from being held liable for racial disparities they did not create.” Inclusive Communities, 576 U.S. at 542 (internal modifications and quotation marks omitted). There must be a “causal connection between the identified policy and the alleged disparate impact.” Miller, 571 F.Supp.2d at 259. At the pleading stage, a plaintiff must “allege facts . . . or produce statistical evidence demonstrating a causal connection [to] make out a prima facie case of disparate impact.” Inclusive Communities, 576 U.S. at 543. “[A] disparate-impact claim that relies on a statistical disparity must fail if the plaintiff cannot point to a defendant's policy or policies causing that disparity.” Id. at 542; see Burbank Apartments, 48 N.E.3d at 411. The question ultimately comes down to this: Have the Plaintiffs alleged facts demonstrating that the policy at issue-the provision and use of the SafeRent Score, which relies at least in part on credit history, to evaluate rental applications-causes the alleged disparate impact, that is, a disproportionate denial of rental applications for Black and Hispanic individuals and voucher holders generally?
The Court has already noted that Plaintiffs allege that the SafeRent Score “considers applicants' credit history” and “other credit-related information, including non-tenancy debts.” [Am. Compl. at ¶ 30]. As explained, this “other credit-related information,” such as “bankruptcy records, past due accounts, [and] payment performance,” contributes to credit history and scores, thereby affecting the SafeRent Score as alleged by the Plaintiffs. [See id. at ¶ 35]. Indeed, as alleged by Plaintiffs, the only factor not related to financial data is eviction history, and even that could be due to non-payment of rent. [See id.]. Plaintiffs' allegations of disparities in credit scores, credit health, debt collection rates, the use of payday and other alternative loan sources, and wealth inequality plausibly encompass many of these factors. [See id. at ¶¶ 51-60]. Because these factors relate to and reinforce one another, and given that Plaintiffs have alleged disparities across the range of factors considered by the SafeRent Score, it is not necessary to know the precise weight assigned to each factor at the pleading stage. Indeed, it is impossible for Plaintiffs to plead such information, as SafeRent does not disclose the data sources considered or the weights assigned to such data in calculating the SafeRent Score. [Id. at ¶ 34]. The Court will not penalize Plaintiffs for the “informational disadvantage” they suffer at this early stage in light of their other well-pleaded allegations. See Mandala, 975 F.3d at 212. While the existence of “other factors” could explain the disparate impact alleged, Plaintiffs have pleaded facts sufficient at this phase to support a plausible claim that the Defendants' use of the SafeRent Score, which relies in part on credit history and related factors, disproportionately affects Black and Hispanic applicants and voucher holders generally. It does not matter that Defendants did not create the longstanding inequities in access to wealth and wealth-building mechanisms that have contributed to racial and income disparities in credit scores and histories; Defendants decided to rely on those inequities when providing and using the SafeRent Score, which includes credit scores and histories in the algorithm, and such reliance, as alleged by Plaintiffs, has a disparate impact on housing opportunities for Black and Hispanic applicants and voucher holders.
SafeRent's argument that housing providers, not SafeRent, make the final housing decision fails for the same reasons articulated in the Court's discussion of whether the FHA applies to SafeRent. SafeRent created and provided the tenant-screening process to Metropolitan and other housing providers, marketing its service as a means to “automate human judgement” and refusing to disclose the details of its algorithm or to allow providers to alter any of the factors considered. [See Am. Compl. at ¶¶ 25, 33-37]. Thus, at this stage, the relevant policy and the resulting disparate impact can be traced to SafeRent, even if housing providers also play a role.
D. Chapter 93A (Counts VII-VIII)
To state a claim for a Chapter 93A violation, a plaintiff must show that (1) the defendant engaged in trade or business and committed an unfair or deceptive act; (2) this act resulted in economic injury to the plaintiff; and (3) there is a casual connection between the practice and the plaintiff's economic injury. See Mass. Gen. Laws. ch. 93A, §§ 2, 9, 11. SafeRent maintains that Plaintiffs' Chapter 93A claims are derivative of their disparate impact claims and must be dismissed for the same reasons. [Dkt. 32 at 19]. In the alternative, SafeRent argues that SafeRent's challenged conduct falls within Chapter 93A's “safe harbor” provision and that Plaintiffs allege “nothing more than good-faith or (at most) negligent conduct” that does not rise to an “unfair” practice. [Id. at 19-20]. Plaintiffs counter that the facts alleged do state a plausible claim for unfair acts or practices that are not covered by the safe harbor provision.[Dkt. 36 at 24-27].
Plaintiffs assert their Chapter 93A claims under a theory of “unfairness,” not deceptiveness. [Dkt. 36 at 24 n.16].
The Court has already explained why Plaintiffs' disparate impact claims survive dismissal, and therefore SafeRent's first argument is moot. Conduct is “unfair” under Chapter 93A if it (1) falls “within at least the penumbra of some common-law, statutory, or other established concept of unfairness”; (2) is “immoral, unethical, oppressive, or unscrupulous”; and (3) causes “substantial injury to consumers.” Walsh v. TelTech Sys., Inc., 821 F.3d 155, 160 (1st Cir. 2016) (citations omitted). “To rise to the level of an ‘unfair' act or practice, the defendant's conduct must generally be of an egregious, non-negligent nature.” Id.; see also Baker v. Goldman, Sachs & Co., 771 F.3d 37, 51 (1st Cir. 2014). Chapter 93A does not apply to “transactions or actions otherwise permitted under laws as administered by any regulatory board or officer acting under statutory authority of the commonwealth or of the United States.” Mass. Gen. Laws. ch. 93A, § 3 (“Section 3”). The “burden is on the defendant[]” to show that Section 3 “applies to the conduct in question,” which requires the defendant to “show more than the mere existence of a related or even overlapping regulatory scheme that covers the transaction.” Cablevision of Bos., Inc. v. Public Imp. Comm'n, 38 F.Supp.2d 46, 61 (D. Mass. 1999) (citations omitted). This “burden is a difficult one to meet.” Bierig v. Everett Square Plaza Assocs., 611 N.E.2d 720, 728 n.14 (Mass. 1993). A defendant must “show that such scheme affirmatively permits the practice which is alleged to be unfair.” Cablevision, 38 F.Supp.2d at 61 (citation omitted).
At the motion hearing, Plaintiffs suggested that the requirement that conduct be “egregious” was limited to situations in which a defendant's actions sound in negligence, which they claim is not the case here. A close reading of the caselaw suggests otherwise. See Walsh v. TelTech Sys., Inc., 821 F.3d 155, 160 (1st Cir. 2016).
Plaintiffs do not dispute that HUD allows, in certain circumstances, housing providers to consider credit history when screening applicants and to reject an applicant for poor credit history, which could support a finding that SafeRent's actions are protected by Section 3. See Dep't of Housing & Urban Dev., HUD Handbook 4350.3: Occupancy Requirements of Subsidized Multifamily Housing Programs, at 4-23, § 4-7(F) (Nov. 2013). Instead, Plaintiffs argue that “[j]ust because an act or practice may be authorized by statute or regulation does not mean that it ‘can never be an unfair or deceptive act or practice' as applied.” [Dkt. 36 at 26 (citing Schubach v. Household Fin. Corp., 376 N.E.2d 140, 142 (Mass. 1978))]. This is true. See Tomasella v. Nestle USA, Inc., 962 F.3d 60, 79 (1st Cir. 2020) (“Under this rubric, the legality of the challenged act or practice is not dispositive of its unfairness.”). However, Plaintiffs have not alleged facts to suggest that SafeRent, in creating and providing the SafeRent Score to housing providers, acted in an immoral, unethical, oppressive, or unscrupulous manner such that its conduct was of an egregious nature sufficient to raise a plausible claim of unfairness under Chapter 93A.
Whether actions are protected by Section 3 has no bearing on whether they have a disparate impact on protected groups under antidiscrimination laws.
IV. CONCLUSION
For the foregoing reasons, Metropolitan's motion to dismiss [Dkt. 29] is DENIED and SafeRent's motion to dismiss [Dkt. 31] is GRANTED IN PART and DENIED IN PART. Plaintiffs' Chapter 93A claims against SafeRent, Counts VII and VIII, are DISMISSED.
SO ORDERED.