Fair Housing Justice Center, Inc. v. Goldfarb Properties, Inc.

• Court: District Court, S.D. New York
• Decided: September 29, 2023
• Docket: 1:18-cv-01564
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT SOUTHERN DISTRICT OF NEW YORK

FAIR HOUSING JUSTICE CENTER, INC.,

Plaintiff,

– against –

PELICAN MANAGEMENT, INC., FORDHAM ONE COMPANY, LLC, and CEDAR TWO COMPANY, LLC, OPINION & ORDER

Defendants. 18-cv-1564 (ER) (OTW)

PELICAN MANAGEMENT, INC., FORDHAM ONE COMPANY, LLC, and CEDAR TWO COMPANY, LLC,

Counterclaimants,

− against −

Counterclaim-Defendant.

RAMOS, D.J.: �e Fair Housing Justice Center, Inc. (“FHJC”) brought this action alleging that Pelican Management, Inc., Fordham One Company, LLC, and Cedar Two Company, LLC (collectively “Defendants”), owners of rental buildings in New York City, adopted a minimum income requirement for prospective renters in the fall of 2015 that was unlawful under both the Fair Housing Act (the “FHA”) and New York City Human Rights Law (“NYCHRL”). Specifically, Defendants’ minimum income policy at that time required that all prospective renters earn an annual income of at least 43 times their total monthly rent. According to FHJC, that policy excluded virtually all renters who received rental subsidies, a large percentage of whom have disabilities. In January of 2019, Defendants adopted a new minimum income policy and thereafter asserted a counterclaim seeking a declaratory judgment that this new policy is lawful. A bench trial was conducted by this Court on February 7, 2022.1 �is Opinion constitutes the Court’s findings of fact and conclusions of law on whether Defendants’ 2015 Policy was unlawful and whether the 2019 Policy is lawful and an adequate remedy. For the reasons set forth below, the Court finds for FHJC on each of its claims under the FHA and NYCHRL, and finds that the 2019 Policy is unlawful. I. BACKGROUND FHJC filed this action on February 21, 2018 alleging that Defendants’ 2015 Policy (the “2015 Policy”) requiring that all prospective renters in their buildings earn an annual income of at least 43 times their total monthly rent was unlawful under the FHA and NYCHRL.2 Doc. 1. FHJC asserts that the 2015 Policy had a substantial adverse disparate impact on renters using one of three rental subsidies—the Olmstead Housing Subsidy (“OHS”), the New York City HIV/AIDS Services Administration (“HASA”) rental subsidy, and Section 8 or Housing Choice Vouchers.3 Doc. 86 ¶¶ 83–88. On April 23, 2018, the Defendants moved to dismiss the complaint. Doc. 25. On May 21, 2018, Judge Forrest, to whom the case was originally assigned, granted FHJC leave to file an amended complaint without ruling on Defendants’ motion. Doc. 34. In January 2019, nearly eleven months after the instant action was filed, Defendants adopted a new minimum income policy (the “2019 Policy”).

1 �e bench trial began on February 7, 2022 and was adjourned the next day. �e trial continued on October 27, 2022 and was concluded on October 28, 2022. 2 �e complaint was originally filed by both FHJC and an individual named Alfred Spooner. On May 2, 2019, the parties filed a stipulation of dismissal of all claims by Spooner with prejudice. See Doc. 75. �e original complaint was against Cedar Two Company, LLC., Deegan Two Company, and Goldfarb Properties, Inc. Deegan Two Company and Goldfarb Properties, Inc., were replaced by Pelican Management, Inc. and Fordham One Company in the second amended complaint. Doc. 86. 3 �e terms “subsidies” and “vouchers” are used interchangeably herein. On July 2, 2019, the parties stipulated to the filing of a second amended complaint (“SAC”). Doc. 85. On July 23, 2019, nine days after they were served with the SAC, Defendants filed an answer and asserted the counterclaim at issue here, in which they sought a declaration that the 2019 Policy is lawful. Doc. 94. On October 15, 2019, FHJC moved to dismiss that counterclaim. Doc. 107. �e motion was denied on July 24, 2020, with the Court finding that Defendants’ counterclaim was properly and timely asserted in response to the amended complaint, did not unduly expand the instant action or prejudice FHJC, and was ripe under the Declaratory Judgment Act. Fair Hous. Just. Ctr., Inc. v. Pelican Mgmt., Inc., No. 18 Civ. 1564 (ER), 2020 WL 4262291, at *5 (S.D.N.Y. July 24, 2020). �e Court commenced a bench trial on February 7, 2022. �e bench trial continued on February 8, 2022 and adjourned the next day due to the unavailability of two witnesses. �e trial continued on October 27, 2022 and concluded on October 28, 2022. �e following witnesses were called by FHJC in person or by affidavit:

• Shaye Belcon, FHJC’s investigative coordinator at the time of its investigation into Defendants’ policies;

• Kate Haggerty, an FHJC tester who posed as a potential renter to determine whether Defendants were engaging in housing discrimination;

• Lisa Darden, an FHJC tester who posed as a potential renter to determine whether Defendants were engaging in housing discrimination;

• Cathy Bowman, Director of the LGBTQ & HIV Advocacy Project at Brooklyn Legal Services with knowledge of the creation and administration of the HASA rental subsidy program;

• Trevor Schaper, Defendants’ Vice-President of Operations;

• Christopher Miller, Defendants’ Leasing Manager;

• Richard Dunn, the head of Defendants’ credit department; • Inga Ballard, an FHJC tester who conducted a test of Defendants’ rental policies;

• Philip Goldfarb, one of Defendants’ owners;

• Fred Freiberg, the Executive Director of FHJC at the time of FHJC’s investigation into Defendants’ rental policies;

• Dr. Justin Steil, Associate Professor of Law and Urban Planning at the Massachusetts Institute of Technology, was called by FHJC as its expert;

• Kiana Glanton, a vision impaired woman who submitted a rental application for one of Defendants’ properties using a Section 8 voucher;

• Alicia Suggs, a friend of Kiana Glanton who served as a guarantor in Glanton’s rental application.

The following witness was called by Defendants:

• Dr. Paul White, a Partner at Resolution Economics, LLC, was called as an expert.

At the conclusion of the bench trial, the Court directed the parties to submit proposed findings of fact and conclusions of law. II. FACTS A. Undisputed Facts FHJC is a non-profit organization in New York City whose mission is “to ensure that all people have equal access to housing opportunities in the New York City Region by eliminating housing discrimination; promoting policies and programs that foster open, accessible, and inclusive communities; and strengthening enforcement of fair housing laws.” Plaintiff’s Proposed Finding of Facts and Law ¶ 3. Among other things, FHJC conducts tests and other investigations of allegations of housing discrimination. Freiberg Aff., Pl. Ex. 68 ¶ 10. FHJC utilizes testers to pose as potential renters or homebuyers to determine whether illegal housing discrimination is taking place. Id. ¶ 12. a. OHS, HASA, and Section 8 Subsidies OHS helps to ensure safe and sustainable housing in the community for “high need” Medicaid beneficiaries who have become homeless or have been placed in nursing homes as a result of their disability. Defs. Ex. Q ¶ 21. OHS is a partial subsidy, and for the most part, OHS recipients are required to pay 30% of their income towards rent with

OHS providing a direct payment each month to the landlord to pay the balance. Id. ¶ 22; Pl. Ex. 29. In order to qualify for OHS, an applicant must have a disability. Accordingly, all OHS recipients are disabled under the FHA. HASA provides public benefits and services for people living with HIV/AIDS in New York City. All HASA clients are automatically eligible for a rent allowance above the standard public assistance levels. Pl. Ex. 66 ¶ 7. Because all HASA recipients are living with Symptomatic HIV, all recipients are disabled under the FHA. Steil Aff., Pl. Ex. 70 ¶ 53; see also Handicap definition, 24 C.F.R. § 100.201. A HASA subsidy can be for part or all of the rent. Pl. Ex. 66 ¶ 15. Section 8 is a partial subsidy and the most commonly used voucher program in

New York City. Steil Testimony, 10/28/2022 Tr. 190:12–17. Section 8 recipients are eligible to rent apartments using their vouchers up to the payment standard, the maximum amount of subsidy the program will pay to the owner on behalf of a voucher holder, with the tenant paying 30% of his or her income toward rent and Section 8 paying the balance. If a voucher recipient wants to rent an apartment with a rent greater than the payment standard, they may be permitted to do so as long as their share of the rent is no more than 40% of their adjusted gross income. Steil Aff., Pl. Ex. 70 ¶ 37. b. Defendants and the Test Calls Defendant Pelican Management manages multi-family residential buildings with approximately five to six thousand rental units in New York City and Westchester County, New York. 10/27/2022 Tr. at 64:15–19, 84:11–14, 101:17–20. Pelican Management’s public-facing name is Goldfarb Properties (“Goldfarb”), id. at 63:25–

64:04, and Philip Goldfarb is its owner. Id. at 59:9–18. Defendants Fordham One Company, LLC, and Cedar Two Company, LLC, each own one rental building in which Philip Goldfarb has an ownership interest. Id. at 97:19–98:1, 98:24–99:6. The buildings owned by these two defendants had apartments available for rent that Alfred Spooner, who was an original plaintiff in this case with FHJC, applied to rent in 2017. Pl. Ex. 40. At the time, Spooner held an OHS subsidy. The OHS subsidy guarantees that eligible individuals pay only 30% of their monthly income towards rent with the balance paid by OHS. Defs. Ex. Q ¶ 22. The apartment he applied to had a monthly rent of $1,384. Pl. Exs. 40, 41. His application was rejected under Goldfarb’s 2015 Policy for “insufficient income.” Pl. Ex. 44; 2/7/2022 Tr. at 143:06-09. Under the

2015 Policy, all applicants who were rejected had to wait at least 60 days before they could apply again to rent an apartment. Defs. Ex. A; 2/7/2022 Tr. at 91:14–25. Upon receiving a complaint from Spooner, FHJC conducted four telephone tests to investigate his claim. Id. ¶¶ 18–19. The first test was conducted on August 17, 2017. During this test, a tester named Kate Haggerty called Goldfarb’s Far Rockaway, Queens Office and posed as the daughter of an OHS recipient. Pl. Ex. 59 ¶ 34; Pl. Ex. 60 ¶¶ 4– 26; Pl. Ex. 19. Haggerty conducted a second test on August 17 and 22, 2017. During this test, Haggerty called Goldfarb’s New Rochelle, New York office and again posed as the daughter of an OHS recipient. Pl. Ex. 59 ¶ 35; Pl. Ex. 60 ¶¶ 25–41; Pl. Ex. 20. The third test was conducted on September 13, 2017. During this test, a tester named Inga Ballard posed as a tenant with a Section 8 voucher. Pl. Ex. 59 ¶ 36; Pl. Ex. 21. The fourth test was conducted on October 13, 16, and 17, 2017. During this test, a tester named Lisa Darden called Goldfarb Properties’ Far Rockaway, Queens Office and posed

as the sister of a recipient of a HASA rental subsidy. Pl. Ex. 59 ¶ 37; Pl. Ex. 61 ¶¶ 4–50; Pl. Ex. 22. During all four tests, the Goldfarb staff confirmed their policy of requiring that all applicants earn 43 times the total rent in income, even if the applicant had a subsidy that would pay part of the rent or, in the case of HASA, the entire rent. The testers explained to Goldfarb employees how these subsidies worked, that the government would be paying large portions of the rent, and, most importantly, that it would be impossible to have an income low enough to receive a subsidy but high enough to satisfy the 43-times- the-rent requirement. In all instances, Goldfarb staff indicated that they understood these concerns, but that company policy was to apply these income tests to all applicants even

if they had a rental subsidy. See generally Pl. Exs. 60, 61, 67, 20–22. 2015 Policy In 2015, Goldfarb was experiencing an unusually high rates of tenant rent arrears. In response, Miller, Defendants’ Leasing Manager, and Schaper, Defendants’ Vice- President of Operations, created a company-wide policy establishing income and credit requirements for rental applicants.4 Goldfarb Testimony, 10/27/2022 Tr. at 82:13–15, 85:4–86:21. This new policy went into effect shortly after Goldfarb reached a settlement

4 Lawyers Michael Koenig and Hal Weiner also assisted in drafting the 2015 Policy. in another housing discrimination case that had been filed against it. 5 Goode v. Goldfarb Properties, Inc., No. 7913/2015 (N.Y. Sup. Ct.). In that settlement Goldfarb agreed to not discriminate against prospective renters on the basis of lawful source of income. See Schaper Testimony, 2/7/2022 Tr. at 58:22–59:4.

The 2015 Policy included the following requirements for applicants: (1) a credit rating of 650 or higher; (2) gross annual income equal to 43 times the monthly rent; and (3) a monthly net rent-to-income ratio of no more than 40%, that is to say, the tenant could not pay more than 40% of their net income in rent.6 See Defs. Ex. A; Defs. Ex. UU ¶¶ 17–18. The second and third requirements applied to rental applicants with subsidies even if the applicant would only be personally responsible for a small portion of the rent, or no rent at all. 2/27/2022 Tr. at 50:09–52:02, 53:15–18, 58:07–21, 110:04–21. When Defendants performed this calculation, it treated voucher payments toward rent as though it were cash income. For example, if an applicant were applying for a $1,000 apartment and had no income, but received a subsidy that would pay the full rent,

Defendants would treat that applicant as earning $12,000 in income (12 monthly payments of $1,000). This applicant would be rejected, however, because Defendants’ policy required that applicant to earn $43,000 in income, 43 times the monthly rent. 2/7/2022 Tr. at 58:07–21. Defendants used a software program called Yardi to process applications. Yardi had a feature that required applicants to answer pre-set questions, including whether the

5 �e plaintiff in that case, Goode, was a HASA recipient who was expressly told by a Goldfarb staff member that the company did not accept HASA subsidies. Goode Complaint, Pl. Ex. 26 ¶ 20. �is was later confirmed in a FHJC test call. Id. ¶ 27. 6 All applicants were also required to have good rental history, pass a criminal background check, and “be at current job for at least 3 months.” Defs. Ex. A. applicant has a housing subsidy. Defendants chose not to activate this feature when the 2015 Policy was in effect. Accordingly, Defendants had no consistent way to determine whether an applicant had a rental subsidy unless an applicant listed it as an “additional source of income.” 2/7/2022 Tr. at 144:16–146:02, 171:19–172:2.

The individuals in charge of reviewing the rental applications during this time were Schaper, Miller, and Dunn (the head of Defendants’ credit department). Id. at 65:04–20. Schaper attended a Fair Housing training in 2015 and 2017 conducted by Kevin Cremin, a trainer and former Department of Justice Civil Rights Division attorney.7 Schaper testified that he reviewed the materials distributed at the training sessions, which included as an example of unlawful source-of-income discrimination a policy requiring that “even if you have Section 8, you must make three times the rent.” Id. at 61:14–20, 64:20–65:03; see also Pl. Exs. 57, 58. Goldfarb also sent an online Fair Housing training course to all of its employees. The course included a test at the end of the training. 2/7/2022 Tr. at 60:9–13. Miller testified that he never consulted these

materials when developing the 2015 Policy. Id. at 109:13–16. Applications that did not include a subsidy were assessed by Miller until Dunn was hired in 2016, at which point Dunn reviewed all such applications. Schaper and Miller would review all applications that did include a subsidy and had the sole authority to decide whether to accept or reject the applicants. Id. at 65:04–16, 65:23–25, 110:22– 111:02.

7 Defendants’ employees were required to participate in this training as part of the settlement reached in the Goode matter. Dunn testified that it was his “regular practice” to apply a 5% variance to Defendants’ 40% net rent-to-income ratio requirement when reviewing applications that did not include a subsidy. That is, notwithstanding the Policy, Dunn would approve applicants with no subsidy who met Defendants’ other requirements so long as they had a

net rent-to-income ratio of 45% or lower. 10/27/2022 Tr. at 45:14–46:01. Moreover, Dunn also testified that he would often approve applicants who had a good rental payment history even if they did not meet Goldfarb’s income requirements. Id. at 48:14– 49:13, 49:20–50:24, 53:12–15. Schaper and Miller, on the other hand, testified that they were “fairly rigid” in applying the 40% net rent-to-income ratio requirement, and only provided exceptions under limited circumstances, such as when the applicant had exceptional credit, significant savings, or a guarantor. 2/7/2022 Tr. at 72:12–18, 95:13–18, 104:06–09, 117:24–119:07. Unlike Dunn, Schaper and Miller did not consider the rental payment history for applications that included a subsidy. Id. at 123:24–125:01, 126:03–07. In

other words, Defendants were more forgiving in reviewing the applications of those applicants who did not have a subsidy, versus those that did have a subsidy. 2019 Policy In May 2019, Goldfarb announced the new criteria for its 2019 Policy on its website and to its staff.8 For applicants with no subsidy, the criteria was unchanged.9

8 Defendants assert that they began applying the new policy to applications in January 2019, but acknowledge that there is no record of the 2019 Policy being applied between January and May 2019. 2/7/2022 Tr. at 150:12–20, 153:17–154:02; Pl. Ex. 55. 9 �e criteria includes a satisfactory credit history, proof of source of income, a gross annual income of at least 43 times the monthly rent, a rent-to-net income ratio of no more than 40%, good rental history, having been employed at least 3 months if income from employment is required to satisfy the rent to net income ratio, and an acceptable criminal history (date and offense are factors). Pl. Ex. 31. See 2019 Policy, Pl. Ex. 31. Additionally, the 60-day waiting period to reapply after an initial denial remained in place for all applicants under the 2019 Policy. Miller testified that Goldfarb made exceptions to the 60-day rule for both applicants with and without subsidies from time to time. 2/7/2022 Tr. 180:1–7.

a. Applicants with a Partial Subsidy Under the 2019 Policy, applicants with partial subsidies must meet all of the requirements as unsubsidized applicants, except they must earn 40 times the rent in income instead of 43 times as under the 2015 Policy. Partial subsidy applicants must still have a rent-to-income ratio of no greater than 40%. Instead of using the full rent amount for these calculations, however, Goldfarb uses the portion of the rent for which the applicant is responsible (rent minus voucher amount) when assessing an applicant’s financial eligibility. 2/7/2022 Tr. at 155:01–15.

b. Applicants with a Full Subsidy The only criteria for applicants with full subsidies are: having an acceptable criminal history,10 producing all sources of verifiable income, and confirming that the first month’s rent and security deposit will be tendered at lease execution. Under the 2019 Policy, Goldfarb no longer applies an income test for applicants with full subsidies. Pl. Ex. 31; 2/7/2022 Tr. at 154:23–25. For Section 8 recipients, Goldfarb did not use the adjusted gross income formula, and instead calculated the adjusted gross income of an applicant with a partial subsidy by

looking at their net income and deducting any taxes payable. Miller Testimony, 2/7/2022 Tr. at 161:23–25. The Section 8 program adjusts applicants’ gross income based on a

10 As mentioned above, having an “acceptable criminal history” is a requirement for all applicants. number of deductions, including those for dependents, child care, disability assistance, unreimbursed medical expenses, and if there are elderly or disabled family members in the household. See Adjusted Income Mandatory Deductions, 24 C.F.R. § 5.611(a). The effect of these deductions is that the tenant’s rent share may be reduced and the

program’s share increased as the tenant’s circumstances change. Thus, for any Section 8 holder receiving these deductions, Goldfarb will assume the tenant’s share of the rent to be higher than it actually is. See 10/27/2022 Tr. 32:11–33:22 (Miller confirming that even if “Section 8 says from [their] perspective [that an applicant] make[s] $700 gross a month, [] from [Goldfarb’s] perspective… [an applicant] still make $1,000 gross a month”). c. Testimony of Kiana Glanton Kiana Glanton, a vision impaired woman, testified at trial that she applied for a

rental in a Goldfarb property in May of 2021, when the 2019 Policy was in effect. At the time she applied, Glanton had a Section 8 voucher. Because of her vision impairment, her friend, Alicia Suggs, assisted her in preparing and submitting the online application. The rent for the apartment she applied for was $2,110. Glanton Testimony, 10/28/2022 Tr. at 368:17. Goldfarb denied her application for “insufficient income” in June 2021 even though she had a monthly income of $1,900 and a Section 8 voucher. Id. at 370:17– 371:21; see also Section 8 May 2021 Voucher, Defs. Ex. BBB. The denial letter Glanton received in June 2021 informed her that she could reapply after 60 days (August 2021), which was after her May 2021 voucher was set to expire. 10/28/2022 Tr. at 371:09–21. She then requested an extension for her voucher, which was granted, with a new

expiration date of September 11, 2021. Id. at 372:15–24; see also Pl. Ex. 75. When she applied a second time in August 2021, 10/28/2022 Tr. at 376:24– 377:17, she did so with Suggs as her guarantor in hopes that Goldfarb would approve her application if she had a guarantor. Id. at 379:10–24. Suggs assisted in submitting the application again and Glanton’s recollection is that she included the extended voucher

with her second application. Id. at 377:03–08, 380:13–381:01, 390:24–391:04. However, the August 2021 application Glanton submitted to Goldfarb with the second application included the same voucher uploaded in May 2021, which had expired in July. Defs Ex. CCC. Glanton called the Goldfarb office to confirm she had uploaded the correct documents and informed the staff that because she was visually impaired she needed extra assistance to confirm all of the documents were correct. 10/28/2022 Tr. at 380:18–381:01. In early August, Glanton’s application was denied for a second time. When Glanton called Goldfarb, the staff told her that the apartment had already been rented out, and she later received a rejection letter. Id. at 399:6–9. The rejection letter stated “voucher insufficient” as the reason for the denial, with no further explanation as

to what that meant. Pl. Ex. 76. Glanton testified that Goldfarb told her that her application was denied because the voucher had expired. 10/28/2022 Tr. at 399:6–9. Glanton subsequently retained an attorney, Ali Frick. In November 2021, Frick emailed Goldfarb to schedule a viewing of an available apartment and attached a valid Section 8 voucher to the email. Defs. Ex. DDD. Pursuant to negotiations between Frick and Goldfarb, Glanton entered into an agreement with Goldfarb on December 9, 2021, whereby Goldfarb denied any discrimination and Glanton accepted a one year rent stabilized lease. Defs. Ex. EEE. B. Findings on Disputed Issues of Fact Both FHJC and Defendants rely on expert testimony to establish the adverse impacts, if any, of Goldfarb’s policies on applicants with subsidies. Dr. Steil, FHJC’s expert, submitted a report in August 2019 that analyzes the impact of both the 2015 Policy and the 2019 Policy.

Dr. White wrote three reports: one in October 2019, one in February 2021, and a third in April 2021 which amended the second report. The first report analyzed whether the selection rate of applications with a subsidy under the 2019 Policy was consistent with the selection rate of applications without a subsidy. Dr. White conducted his analysis by reviewing data from spreadsheets provided by Goldfarb for seven of its buildings. See October 2019 Report, Defs. Ex. Y at 2–3. His analysis covered two periods: November 2016–February 2018, and January–June 2019 (when Goldfarb alleges they instituted the 2019 Policy). The second report reviewed data collected for apartment applications in seven of Goldfarb’s properties for the period May 1, 2019 through August 31, 2020. Dr. White’s

analysis measured the effects of the 2019 income and credit requirements on the denial rate of applications which included a subsidy and those which did not include a subsidy. The third report was prepared in April 2021 in response to questions raised by Dr. Steil about data used by Dr. White. It amended the second report by conducting the same statistical analysis as the second report, but using updated data provided by Goldfarb. In an affidavit submitted on November 24, 2021, Dr. Steil conducted an analysis in response to Dr. White’s April 2021 report and critiques Dr. White’s findings as “present[ing] an incomplete and flawed picture of the effect of the 2019 [P]olicy.” Steil Aff., Pl. Ex. 70 ¶¶ 147, 152. The experts disagreed with respect to: whether the applicant pool each expert used was probative; whether applicants with partial and full subsidies should be treated as two distinct groups; and whether applications that included a subsidy denied for “insufficient voucher” or “no contact” should be included in the analysis. The Court first

summarizes the expert reports. Dr. Steil’s Expert Testimony Dr. Steil’s report concludes that the 2015 Policy had a significant and adverse impact on both disabled applicants and applicants with housing subsidies because it required tenants to (1) spend no more than 40 percent of their net income on rent, and (2) have a gross income equal to at least 43 percent the monthly rent, irrespective of any subsidy the tenant received. Steil Aff., Pl. Ex. 70 ¶ 131. In conducting his analysis, Dr. Steil looked to the pool of potential applicants for

housing units generally and not just applicants for Goldfarb apartments. Id. ¶ 102. He testified that this was the most relevant and appropriate available data because Goldfarb’s website statement—that interested applicants should not apply unless their gross income was at least 43 times the monthly rent— could have discouraged potential applicants from applying at all and made the actual Goldfarb applicant pool less probative. Id. Dr. Steil used this general pool of applicants to assess the effects of Goldfarb’s policies under two tests: the Comer test and the “four-fifths” test. In Comer, which take its name from a Second Circuit case, the court assessed disparate impact by comparing the percentage of minority families holding Section 8 vouchers with the percentage of minorities on the waiting list for housing. Comer v.

Cisneros, 37 F.3d 775, 793 (2d Cir. 1994). The Belmont Section 8 Program, which administered the Section 8 program for the town of Amherst and other suburban communities outside the city of Buffalo, adopted a local preference in favor of applicants who lived or worked in one of the suburban communities within Belmont’s administrative jurisdiction. Id. at 784. This local residency preference resulted in

minority families holding only 3% of the over 2,000 vouchers issued by Belmont, despite the fact that 21% of the over 3,800 families on the waiting list were minority families. Id. at 793. The Second Circuit held that this residency preference policy was racially discriminatory because it “impede[d] persons who live in the City of Buffalo, which has a higher percentage minority population than the suburbs, from obtaining rental assistance available under Belmont’s program, thereby giving subsidies to a disproportionate number of white suburban residents ahead of minority city dwellers.” Id. at 792. Dr. Steil drew the parallel between the findings in Comer and the data set in the instant case to conclude that the 2015 Policy had an adverse impact on persons with subsidies and people with disabilities.

First, he compared the share of protected class members (those with housing subsidies) in the applicant pool to the share of protected class members that could pass the Goldfarb income test. See Steil Aff., Ex. 70 at ¶ 123. He found that applicants with subsidies were 17 times less likely to pass the income test than applicants with no subsidies. Id. at ¶ 125. He also found that it was not possible for people eligible for OHS, people eligible for HASA, and 97% of the households eligible for a Section 8 voucher to satisfy Goldfarb’s rental requirements. Id. ¶¶ 74–100. He also compared all households that relied on subsidies with all households that did not. Taken together, 52% of participants in the OHS, HASA, and Section 8 voucher programs have a disabled household member and yet less than 3% of those applicants could meet the income test under the 2015 Policy.11 Id. By comparison, only 20% of households that did not rely on subsidies have a disabled household member. The parallel to Comer is that while 52% of applicants for

Goldfarb units would be expected to be disabled, less than 3% could meet the income test. Dr. Steil concluded that rejecting 97% of persons with OHS, HASA, and Section 8 vouchers has an adverse impact on persons with subsidies and people with disabilities under Comer. Id. ¶ 108. Additionally, his analysis found that households that rely on subsidies are more than 2.5 times as likely as households that did not rely on subsidies to have a disabled household member. Accordingly, Dr. Steil concluded that the 2015 Policy had an adverse impact on persons with disabilities under the Comer test. Id. The second test Dr. Steil conducted was the four-fifths test, adopted from the Equal Employment Opportunity Commission (“EEOC”) and used in an FHA case in the First Circuit to determine if a difference in selection rates constitutes a disparate impact.

Langlois v. Abington Hous. Auth., 207 F.3d 43, 50 (1st Cir. 2000); Langlois v. Abington Hous. Auth., 234 F. Supp. 2d 33, 57–59 (D. Mass. 2002). Under this test there is a disparate impact if the selection rate for any protected group is less than four-fifths (or 80 percent) of the selection rate for the group with the highest selection rate.12

11 Dr. Steil took the average of 46% of all Section 8 households including a person with disabilities and 100% of OHS and HASA households including a person with disabilities to arrive at 52%. 12 This test consists of a four-step process to assess the adverse impact of a policy by: (1) calculating the rate of the selection for each group (by dividing the number of persons selected from a group by the number of applicants from that group); (2) observing which group has the highest selection rate; (3) calculating the impact ratios, by comparing the selection rate for each group with that of the highest group and; (4) observing whether the selection rate for any group is substantially less than the selection rate for the highest group. If it is, an adverse impact is indicated in most circumstances. Langlois v. Abington Hous. Auth., 207 F.3d at 50; Langlois v. Abington Hous. Auth., 234 F. Supp. 2d at 57–59. Dr. Steil compared the number of all households with and without disabled members and all households with or without disabled members that have OHS, HASA, or Section 8 subsidies. He concluded that the 2015 Policy disproportionately disqualified households with disabled members, because it effectively disqualified all OHS and

HASA applicants and 97% percent of Section 8 applicants. Steil Aff., Pl. Ex. 70 ¶ 116. For households without a disabled member, only 4% of applicants were denied based on the income test. On the other hand, 16% of households that rely on subsidies and have a disabled household member were disqualified. He goes on to state that households with a disabled … member are [] three times more likely than those without a disabled … member to be eligible for one of these three housing subsidies, and are therefore more than three times more likely than other renter households to be adversely affected by Goldfarb’s exclusion of Section 8, HASA, or OHS voucher applicants[] …. this disparity ratio of 3.79 is well above the 1.25 threshold used in the four-fifths test.

Id. ¶ 119. Thus, Dr. Steil concludes that because the 2015 Policy excludes more than 97 percent of applicants with housing subsidies, and households with a disabled household member are over three times more likely to be eligible for one of these three housing subsidies, there is a substantial disparate impact on subsidy and disabled renters. Dr. Steil also analyzed the effect of the 2019 Policy using all applicants with and without subsidies, treating the 2019 Policy as three different policies— one for applicants without a subsidy, one for applicants with a partial subsidy, and one for applicants with a full subsidy. He included all of the reasons for denial (credit denial, income denial, insufficient voucher, not qualified, no contact denial), and excluded those applications that were duplicates, pending, or canceled by the applicant. 13

13 “No contact” denials occur when Defendants are not able to reach an applicant for further communication. Dr. Steil concluded that comparing all applicants without a subsidy to all applicants with a full subsidy, 30% of applicants without a subsidy were approved and 53% of applicants with a full subsidy were approved. Id. ¶ 193. However, comparing all applicants with a partial subsidy to all applicants with a full subsidy, 30% of applicants

without a subsidy were approved, but only 13% of applicant with a partial subsidy were approved. In other words, applicants with a partial subsidy were approved at less than half the rate of applicants with no subsidy. Id. ¶ 195. Even excluding the “no contact” denials, the approval rate for applicants with a partial subsidy is only 20%, just two-thirds of the rate for applicants with no subsidy. Id. ¶ 197. Thus, Dr. Steil concluded, the 2019 Policy has a substantial adverse impact on applicants with partial subsidies. Id. ¶ 196. Dr. White’s Expert Testimony Defendants support their counterclaim that the 2019 Policy does not constitute

unlawful discrimination on the basis of source of income or disability with Dr. White’s expert testimony. Defs. Answer to SAC and Counterclaim, Doc. 94 ¶ 152. In his October 2019 report, Dr. White identified applicants with disabilities using three different definitions.14 Dr. White’s October 2019 Report, Defs. Ex. Y at 10. For the data set, he used a spreadsheet provided by Goldfarb that showed how many applicants applied that did not have subsidies, if they were accepted or rejected, and the reason for the decision. He only compared applications that did not have subsidies with applications that did include subsidies, without distinguishing between applicants with

14 First, he screened for any application that mentioned phrases such as “disability,” “HASA,” and “workers comp,” among others. Secondly, he included in the definition of disability those applicants that were on Social Security, reasoning that because people over 62 are more likely to be disabled than the population at large, Social Security can serve as a proxy for applicants with disabilities. �ird, he identified applications with HASA, OHS, or Section 8 subsidies. partial and full subsidies. He concluded that under the first definition of disability applications (applications that mentioned phrases such as “disability,” “HASA,” and “workers comp,” among others) the difference between the actual and expected number of qualified applications was neutral, in part because of the small number of disability

applications in 2019; under the second definition (including those applicants that were on Social Security), the difference between actual and expected number of qualified applications with disability was neutral; and under the third definition (applications with HASA, OHS, or Section 8 subsidies) the results were neutral, partly because of the low count of applications with HASA, OHS, or Section 8 subsidies. Id. at 29. Dr. White’s April 2021 report focused on the applications denied due to insufficient income or credit to test the effects of the new thresholds prescribed by the 2019 Policy. Dr. White excluded “insufficient voucher” and “no contact” as reasons for denials. Dr. White’s April 2021 Report, Defs. Ex. BB at 8. “Insufficient voucher” denials occurred when Goldfarb determined that the rent of the apartment was outside the

range that the applicant’s subsidy would allow and “no contact” denials occur when Defendants are not able to reach an applicant for further communication. He contends that these denials are for reasons other than Goldfarb’s 2019 Policy requirements. Id. at 10. Dr. White concluded that applications that included subsidies are not denied more than expected when compared to applications that did not include subsidies due to the new income and credit requirements, and that there were no statistically significant adverse effects on applicants that relied on subsidies regardless of subsidy type or decision being analyzed (insufficient credit, insufficient income, or combined). Id. at 22. Dr. Steil’s use of general publicly available data is appropriate Dr. White testified that Dr. Steil’s analysis is flawed because he used general population data instead of actual applicant data, which was available to FHJC and would be more probative of disparate impact. Dr. Steil argues that relying upon external data for the pool of potential applications is appropriate because of the possibility that

applicants may have been discouraged by the income requirement posted on Goldfarb’s website. The Court agrees that relying on external data is appropriate for this analysis because the applicant pool was limited in light of Goldfarb’s dissemination of its 2015 Policy through its website and rental agents. See Pl. Ex. 54 (Goldfarb’s webpage stating that applicants are required to have a income at least 43 times the amount of the monthly rent); see also Pl. Exs. 60, 61, 67, 20–22 (employees of Goldfarb Properties confirming Goldfarb’s policy of requiring that all applicants earn 43 times the total rent in income to FHJC’s test callers). Indeed, these types of [general population] data often form the initial basis of a disparate impact claim, especially in cases such as this one in which the actual applicant pool might not reflect the potential applicant pool, due to a self-recognized inability on the part of potential applicants to meet the very standards challenged as being discriminatory.

E.E.O.C. v. Joint Apprenticeship Comm. Of Joint Indus. Bd. of Elec. Indus., 164 F.3d 89, 97 (2d Cir. 1998). “Where … a defendant is alleged to have informed the potential applicant pool – here through advertising and external communications with … potential tenants – of the policy, an expert may properly account for potential applicants that may have been discouraged from applying due to an awareness of the policy….” Fortune Soc’y v. Sandcastle Towers Hous. Dev. Fund Corp., 388 F. Supp. 3d 145, 170 (E.D.N.Y. 2019). Applicants with partial and full subsidies should be treated as two distinct groups Dr. White treated applicants with partial and full subsidies as one group. Dr. White Testimony, 10/28/2022 Tr. at 290:2–7. In contrast, Dr. Steil took the position that the 2019 Policy is essentially three policies in which applicants with no subsidy, partial subsidies, and full subsidies are subject to different criteria. Whereas applicants with a full subsidy do not have an income criteria, applicants with a partial subsidy do. The Court agrees that these two groups should be treated differently in the analysis. Tellingly, looking only at the applicants with a partial subsidy reveals a statistically significant difference between the acceptance for applicants with a partial subsidy and

applicants with no subsidy, with 30% of applicants with no subsidy approved, compared with 13% of applicants with a partial subsidy. Steil Aff., Pl. Ex. 70 ¶¶ 171, 195–96. Even excluding “no contact” denials, the approval rate for applicants with a partial subsidy is two-thirds of the rate for applicants with no subsidy. Id. ¶ 197. It was appropriate to include applications that relied on subsidies denied for “insufficient voucher” in the analysis Dr. White argues that Dr. Steil was incorrect in including applicants with subsidies that that were denied for “insufficient voucher” in his analysis. These denials occurred when Goldfarb determined that the rent of the apartment was outside the range that the applicant’s subsidy would allow. Dr. White contends that this denial is for reasons other than Goldfarb’s 2019 Policy requirements. Defs. Ex. BB at 11. Dr. Steil

argues that excluding denials for “insufficient voucher” is problematic and makes it impossible to assess the full effect of the 2019 Policy. Steil Aff., Pl. Ex. 70 ¶¶ 173, 176. The determination of whether to include denials for “insufficient voucher” or not is pivotal. Indeed, Dr. White testified that “the insufficient voucher topic is the one that determines the difference between statistical significance and nonstatistical, and a disparate impact and [not a] disparate impact.” 10/28/2022 Tr. at 299:20–25. Dr. White’s position is that applications that are rejected for voucher insufficiency

occur because the applicant applied to an apartment with a higher rent than the voucher can cover. Dr. White April 2021 Report, Defs. Ex. BB at 9–10. But Dr. Steil maintains that this is not necessarily true and can change depending on a determination of three options: (1) what rent amount to use, which may be the rent for the unit the applicant applied for or the rent for another comparable available unit; (2) the amount the voucher would pay; and (3) as explicitly set out in Section 8 regulations, a determination of whether the applicant may be able to afford to contribute more to the rent than their expected share in order to afford a rent higher than the voucher payment standard. Steil Aff., Pl. Ex. 70 ¶ 175. Regarding the first option, the Court notes that Goldfarb did engage in this

practice at times. Miller testified that when an apartment an applicant applied for was no longer available, a leasing agent contacted the applicant to determine if the applicant would be interested in another available apartment in the building. 2/27/2022 Tr. at 168:20–169:4. On redirect, Dr. Steil insisted that there were other options Goldfarb could offer, such as offering apartments in another building or borough, and instituting a waitlist in which applicants would be contacted when units within their payment range became available. 10/28/2022 Tr. at 256:13–257:8. As to the second option, Steil testified that, for example, there is discretion in the HASA program to pay higher amounts if it is beneficial to the health of that applicant based on that applicant’s needs and the characteristics of the apartment (e.g., a building with an elevator). Id. at 190:7–11. Regarding the third option, Dr. Steil testified that under certain circumstances Section 8 voucher holders could make up the gap between the rent limits on their

vouchers and the rent by adding 10% of their adjusted gross incomes to the rent payments. Dr. Steil also testified that he found “more than 15” such cases in his analysis of the Goldfarb data.15 Steil Aff., Pl. Ex. 70 ¶¶ 187–88. Miller testified that Goldfarb takes account of the tenant option to add 10% of their rent limits of the Section 8 vouchers. See Miller Testimony ¶¶ 24–25. The Court finds that Dr. Steil’s methodology of including “insufficient voucher” as a denial category was proper because it accounts for “more than one out of every four” denials of applicants with subsidies. Steil Aff., Pl. Ex. 70 ¶ 176.

It was appropriate to include applications with subsidies denied for “no contact” in the analysis Dr. Steil also included applicants with subsidies that were denied for “no contact” in his analysis. Dr. Steil argues that excluding “no contact” denials presents a misleading view of how voucher applicants fare under the 2019 Policy. Steil Aff., Pl. Ex. 70 at ¶ 180. These denials can be triggered either by the actions of the applicant or by the discretionary determinations of Goldfarb’s agents. Id. at ¶ 178. Defendants contend that this denial is for reasons other than Goldfarb’s 2019 Policy, reasoning that applicants who stop communicating with Goldfarb during the application process were not

15 Dr. Steil did not specifically identify any of those “more than 15” applicants. 10/28/2022 Tr. at 220:11– 221:5. “rejected” by Goldfarb, but rather voluntarily dropped out of the application process. Defendants’ Proposed Finding of Facts and Law ¶ 67. Goldfarb’s leasing agents have discretion as to when to contact the applicant, how many attempts to make to contact the applicant, what time of day to attempt contact, and

what means to make contact (email or phone call). See Miller Testimony, 2/27/2022 Tr. at 164:20–165:9. The leasing agent also has discretion to decide when an applicant can officially be categorized as “no contact.” Id. at 165:10–12. Dr. White admitted that it is possible that, absent any written policy, the ability to reach an applicant can depend partly on Goldfarb, and that its leasing agents can take actions that make certain applicants more or less likely to receive a “no contact” denial. Dr. White Testimony, 10/28/2022 Tr. at 295:20–296:14. Both Miller and Dr. White testified that they are not aware of any written policy regarding contact attempts or how long to wait before categorizing an applicant as “no contact,” and Goldfarb did not produce any records at trial showing the basis for these denials. Id.; see also 2/27/2022 Tr. at 165:13–15.

Perhaps most importantly, according to the data, applicants with subsidies are rejected for “no contact” 30% of the time, while applicants with no subsidy are rejected for “no contact” 14% of the time. As Dr. Steil testified, “it raises the question of why this rate is so different.” 10/28/2022 Tr. at 254:25–255:1. Accordingly, the Court finds that the inclusion of applicants with subsidies rejected for “no contact” is appropriate. Because the Court has found that it was appropriate to include applicants with subsidies who were denied due to “insufficient voucher” and “no contact,” and to distinguish between applications that included a full subsidy and applications that included a partial subsidy, the Court adopts Dr. Steil’s finding that there was a statistically significant difference between applicants with no subsidy and applicants with partial subsidies.

III. CONCLUSIONS OF LAW A. �e 2015 Policy Disability Discrimination Under the FHA A plaintiff asserting a claim for housing discrimination based on disability under a disparate impact theory of liability must initially prove (1) the occurrence of certain outwardly neutral practices; and (2) a significantly adverse or disproportionate impact on a group of persons produced by the defendant’s facially neutral acts or practices. Mhany Mgmt., Inc. v. Cnty. of Nassau, 819 F.3d 581, 617 (2d Cir. 2016). Once a plaintiff has established these two elements, the burden shifts to a defendant to “prove that its actions furthered, in theory and in practice, a legitimate, bona fide [] interest.” Mhany Mgmt., 819 F.3d at 617 (internal citation and quotation marks omitted). If defendants meet that burden the plaintiff must then show that the “substantial, legitimate, nondiscriminatory interests supporting the challenged practice could be served by another practice that has a less discriminatory effect.” Id. (citing 24 C.F.R. § 100.500(c)(3)). First, Defendants did have an outwardly neutral practice requiring rental applicants to meet a minimum income requirement based on the full rent. Secondly,

FHJC argues that because OHS and HASA voucher holders are used only by households with disabled persons, and 47% of Section 8 households include an individual with a disability, the 2015 Policy adversely affected applicants with subsidies that have a disabled person in the household. In response, Defendants contend that FHJC has offered no statistical evidence concerning the correlation between disability and source of income, and thus cannot prove that Defendants’ policies and practices have a disproportionate effect on persons with disabilities. The Court disagrees. FHJC did show a correlation between disability and source of income as it related to OHS and HASA subsidies because those programs are only used by households with a

disabled individual. Steil Aff., Pl. Ex. 70 ¶103. For Section 8 vouchers, FHJC demonstrated a disparate impact on applicants with disabled household members by making an inference from general population data showing that there is a higher than average number of households with disabled persons among Section 8 voucher holders. Id. ¶ 104. Defendants point to the ruling in another FHA disparate impact case in which FHJC was a plaintiff, Short v. Manhattan Apartments, Inc. 916 F. Supp. 2d 375 (S.D.N.Y. 2012). In that case, FHJC and a prospective tenant with AIDS who had a HASA subsidy, brought a housing discrimination action against real estate brokers, alleging that brokers had refused to rent to that tenant and others, based on disability and

source of income discrimination. The court found that plaintiffs failed to make their prima facie case of disability discrimination under the FHA because plaintiff’s disparate impact analysis conflated disability, which is a protected status under the FHA, with source of income, which is not. Id. at 393. The court also found that because “[p]laintiffs … offered no statistical evidence concerning the correlation between disability and source of income, the [c]ourt cannot conclude that Defendants’ policies and practices have a disproportionate effect on persons with disabilities.” Id. Unlike the plaintiffs in Short v. Manhattan Apartments, Inc. who offered no statistical evidence and showed no disproportionate impact of the policy on people with disabilities, FHJC has offered statistical evidence concerning the correlation between disability and the three specific subsidies it alleges Goldfarb’s policies have affected. Id. Nevertheless, Defendants argue that FHJC has not made its prima facie case because it “merely raises an inference of discriminatory impact.” Tsombanidis v. W.

Haven Fire Department, 352 F.3d 565, 575 (2d Cir. 2003). Defendants allege that FHJC has failed to show evidence of a robust causality because it used general population data rather than actual applicant data, which was available. Defendants argue that national or state general data are appropriate where actual applicant data is not available, citing to Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 308–09 n.13 (1977). However, this footnote does not support their argument. In Hazelwood, the Supreme Court noted that general statistics were highly probative, but that the actual data would be relevant if it was available. Id. In the instant case, the actual data on the number of applicants that had a disability is not available because Goldfarb does not inquire whether applicants had a disability. 10/28/2022 Tr. 313:3–314:2. In any event, as discussed infra, the Second

Circuit’s opinion in E.E.O.C., twenty years after Hazelwood, endorsed the use of general population data as appropriate in the exact circumstances presented in this case. E.E.O.C., 164 F.3d 89, 97 (2d Cir. 1998). See also Malave v. Potter, 320 F.3d 321, 327 n.4 (2d Cir. 2003) (holding that the District Court could not reject employee’s statistical evidence in support of his disparate impact claims simply because it failed to focus on applicant pool or eligible labor pool for at-issue promotions, where such data were not available; and observing that “[i]t may be that, even if the data were available, the applicant pool would not have been the most appropriate underlying labor pool…”). Having established the first two elements, the burden shifts to the Defendants to “prove that its actions furthered, in theory and in practice, a legitimate, bona fide [] interest.” Mhany Mgmt., 819 F.3d at 617. Goldfarb argues that it has a “crucial interest in renting only to tenants who can afford to pay the rent” and that reviewing an

applicant’s income is the only way to measure the ability of an applicant to do so. Defendants’ Post-trial Memorandum of Law at 6. Goldfarb argues that the 2015 Policy was created to reduce the high level of rent arrears among Goldfarb tenants. However, Goldfarb did not evaluate, let alone determine, that subsidy tenants contributed to those high arrears, and did not analyze the frequency with which subsidy holders did or did not pay their portion of the rent, despite having the data within the company’s business records to conduct such an analysis. And with respect to applicants with full subsidies specifically, which by definition cover 100% of the rent, Schaper acknowledged in his testimony that Goldfarb had no need to look at these applicants’ income. 2/7/2022 Tr. at 52:23–53:01, 81:10–81:17. Yet, Goldfarb imposed these income requirements on

applications with full subsidy for over four years. Id. at 50:10–52:02, 53:15–18, 58:07– 21. It cannot be the case that excluding applicants with full subsidies, whose voucher programs pay the full rent, was necessary to further Goldfarb’s interest of reducing the high level of rent arrears among its tenants. Because Defendants have not met their burden to prove that the 2015 Policy was necessary to achieve its interest, the Court finds for FHJC on the disability discrimination claim for Goldfarb’s 2015 Policy under the FHA. Disability Discrimination Under NYCHRL In evaluating claims based on a disparate impact theory of liability under both the FHA and NYCHRL, courts apply the burden shifting framework described above. Williams v. Regus Mgmt. Grp., LLC, 836 F. Supp. 2d 159, 171 (S.D.N.Y. 2011). In the context of a NYCHRL claim, however, the Court must “address the NYCHRL’s

‘uniquely broad and remedial purposes, which go beyond those of counterpoint State or federal civil rights laws.’” Id. (quoting Williams v. N.Y.C. Hous. Auth., 872 N.Y.S.2d 27, 31 (1st Dep’t. 2009)). Because the Court has found that FHJC has prevailed on its FHA disability claim, and the same analysis is applied, the Court finds for FHJC on the disability discrimination claim for Goldfarb’s 2015 Policy under the NYCHRL. Source of Income Discrimination Under the NYCHRL While source of income discrimination cannot be a basis for an FHA violation, it is prohibited by the NYCHRL. Section 8-107(5)(a) of the New York City Administrative

Code provides that it shall be an unlawful discriminatory practice for the owner, lessor, lessee, sublessee, assignees, or managing agent of, or other person having the right to sell, rent or lease or approve the sale, rental or lease of a housing accommodation . . . (1) To refuse to sell, rent, lease, approve the sale, rental or lease or otherwise deny to or withhold from any person or group of persons such a housing accommodation or interest therein because of . . . source of income of such person or persons.

NYC Admin. Code § 8-107(5)(a). NYCHRL discrimination claims follow the same burden shifting framework as federal discrimination claims. Williams, 836 F. Supp. 2d at 171. Within that framework, however, “courts must address the NYCHRL’s ‘uniquely broad and remedial purposes, which go beyond those of counterpoint State or federal civil rights laws.”’ Id. (quoting Sealy v. Hertz Corp., 688 F. Supp. 2d 247, 258 (S.D.N.Y. 2009)). Thus, FHJC must prove two elements: (1) the occurrence of certain outwardly neutral policies or practices; and (2) a significantly adverse or disproportionate impact on persons of a particular type produced by the defendant’s facially neutral acts or practices. Mhany Mgmt., Inc., 819 F.3d at 617. Once these two elements are established, the

burden shifts to the defendant to “prove that its actions furthered, in theory and in practice, a legitimate, bona fide [] interest.” Id. (citation and internal quotation marks omitted). If Defendants can meet their burden then FHJC must show that the “interests supporting the challenged practice could be served by another practice that has a less discriminatory effect.” Id. a. Disparate Impact First, FHJC has identified a neutral policy in Defendants’ 2015 Policy of requiring all applicants, including subsidy holders, to meet the 43-times-the-rent

requirement. Schaper testimony, Defs. Ex. UU ¶¶ 24–25. Secondly, FHJC has established disparate impact with Dr. Steil’s finding that 100% of OHS applicants, 100% of HASA applicants, and 97% of Section 8 applicants would not be able to satisfy Goldfarb’s 2015 Policy.16 Steil Aff., Pl. Ex. 70 ¶¶ 92–93, 97. The burden now shifts to Defendants to show that there was a legitimate business justification for the 2015 Policy. Defendants argue that the 2015 Policy was facially neutral with respect to the applicants’ sources of income and that the criteria treated all sources of income the same, including rent subsidies, which Goldfarb included as income in evaluating applications. See Schaper Testimony, Defs. Ex. UU ¶¶ 11, 25. However, under a theory of disparate

16 Dr. White did not perform any analysis under the 2015 Policy as to whether there is an adverse impact on applicants with subsidies, irrespective of disability status. See Defs. Ex. Y; see also White Testimony, 10/28/2022 Tr. at 305:6–22. impact, the policy is unlawful if it has an unjustified significant adverse effect on a protected group, in this case subsidy holders, regardless of its neutral application. See Levin v. Yeshiva Univ., 96 N.Y.2d 484, 491 (2001) (“[A] claim is established where a plaintiff demonstrates that a defendant’s policy or practice results in a disparate impact to

the detriment of any group protected under the City Human Rights Law”) (internal quotations omitted). Accordingly, the Court will analyze whether Defendants meet their burden to show that there was a legitimate business justification for this requirement. As discussed above, Defendants’ purported justification of reducing the high level of rent arrears among Goldfarb tenants was not sufficient to support their 43-times-the- rent policy for applicants with subsidies. Defendants insist that insofar as the 2015 application criteria had an adverse disparate effect on the applicants with subsidies, “the effect was the result of applying neutral criteria in literal compliance with the statutes forbidding discrimination based on source of income.” Defendants’ Post-trial Memorandum of Law at 45. This reasoning is unpersuasive.

While Defendants are correct that subsidies are “sources of income,” that does not mean that vouchers can be construed like cash income. That is to say, rental subsidies are not disposable income available to be used for just any expense. Rather, rental subsidies are guaranteed rent payments made directly to the landlord. Indeed, at trial Schaper acknowledged that treating subsidies like cash income and applying the 2015 Policy criteria was in effect, crediting only 40% of the voucher which can only be used to pay rent directly to a landlord. See Schaper Testimony, 2/7/2022 Tr. 77:21–80:01. In addressing the NYCHRL’s “uniquely broad and remedial purposes,” Sealy, 688 F. Supp. 2d 247, at 258, it is evident that the intention behind the source of income discrimination provision of § 8-107(5)(a) was to avoid precisely what Goldfarb’s 2015 Policy accomplished—discriminatory outcomes for applicants with subsidies. Subsidies are intended to be used solely for rent payments in order to guarantee recipients the ability to afford housing. Goldfarb’s 2015 Policy was discriminatory and prevented

applicants with full or partial subsidies from accessing housing. Because Defendants have not provided evidence that there was a legitimate business justification for this requirement, FHJC need not show that the “interests supporting the challenged practice could be served by another practice that has a less discriminatory effect.” Mhany Mgmt., Inc., 819 F.3d at 617 (citation and internal quotation marks omitted). Accordingly, the Court finds for FHJC on the discrimination claim based on source of income pursuant to the NYCHRL.

b. Intentional Discrimination/Disparate Treatment FHJC also claims intentional discrimination, or disparate treatment, based on source of income under the NYCHRL. A plaintiff need only show that the protected class was treated “less well . . . because of a discriminatory intent.” Mihalik v. Credit Agricole Cheuvreux N. Am., Inc., 715 F.3d 102, 110 (2d Cir. 2013). In evaluating such claims, courts consider the “totality of the circumstances” as “the overall context in which the [challenged conduct occurs] cannot be ignored.” Id. at 111 (quoting Hernandez v. Kaisman, 957 N.Y.S.2d 53, 59 (1st Dep’t 2012)). Defendants argue that they have acted in accordance with their good faith understanding of the law and therefore did not commit intentional discrimination against

applicants with subsidies. Defendants’ Memorandum of Law at 31, 45. For instance, Schaper and other Goldfarb staff attended a seminar given by an expert on housing discrimination and Goldfarb retained two fair housing lawyers to help develop the 2015 Policy. 10/27/2022 Tr. at 105:21–25, 106:18–107:7. The overall context of the treatment of applicants with subsidies pursuant to the

2015 Policy is such that Defendants, at best, should have known about the disparate treatment of such applicants, and at worst, intentionally treated applicants with subsidies less well because of discriminatory intent. For example, prior to the implementation of the 2015 Policy, Goldfarb entered a settlement in Goode after being sued for source of income discrimination. After Defendants implemented the 2015 Policy, Goldfarb was aware that subsidy holders could not meet the 43-times-the-rent requirement from their interactions with the FHJC test callers in relation to the instant case. See FHJC Test Caller Affidavits, Pl. Exs. 60, 61, 67 (detailing how they informed multiple Goldfarb employees of how subsidies worked and that applicants with subsidies would not be able to meet the income requirement). Defendants were aware of how these subsidies worked,

as well as the difference between subsidies and income as it relates to making rent payments from the test calls and from the housing discrimination training mandated by the Goode settlement. Further, written materials provided at these training sessions stated explicitly that a policy requiring applicants with subsidies to earn “x times the rent” is discrimination based on lawful source of income. Pl. Ex. 57 at 18. Nonetheless, no Goldfarb staff member took any action to change Goldfarb’s policy. Defendants’ 2015 Policy also treated applicants with subsidies less favorably on its face. While Goldfarb assessed the income of applicants with no subsidies only against their portion of the rent (i.e., the full rent), Goldfarb assessed the income of applicants with subsidies against both their rent share and the voucher share of the rent for which they were not personally responsible. This disparate treatment alone reflects intentional discrimination, as applicants with subsidies were forced to satisfy an unreachable and illogical standard (i.e., to prove they could pay rent for which they had no personal

responsibility). Further, Goldfarb evaluated applications with subsidies differently than applications with no subsidy. Goldfarb employee Dunn was responsible for evaluating applications with no subsidy while Miller and Schaper were reviewing applications with subsidies. Dunn regularly applied a 5% variance to Goldfarb’s net income policy, thus approving applicants with no subsidy with a rent-to-income ratio as high as 45%. Meanwhile, Schaper and Miller applied the 40% rent-to-income ratio requirement rigidly, and only granted exceptions in limited circumstances. Goldfarb “engaged in discrimination with willful or wanton negligence, or recklessness, or a conscious disregard of the rights of others or conduct so reckless as to

amount to such disregard.” Chauca v. Abraham, 30 N.Y.3d 325, 334 (2017) (citation and internal quotation marks omitted). After being sued by Goode and becoming aware of its discriminatory practices against applicants with subsidies, and receiving fair housing training, Goldfarb adopted its 2015 Policy without conducting any analysis of how this policy would affect applicants with subsidies. See Schaper Testimony, 2/7/2023 Tr. 59:1–60:2. The totality of the circumstances of Goldfarb’s reckless indifference amount to intentional discrimination.17 Accordingly, the Court finds for FHJC on its intentional discrimination on the basis of source of income claim under the NYCHRL.

B. Defendants’ Counterclaim Defendants bring a counterclaim for a declaration that the 2019 Policy is lawful. FHJC asserts that the 2019 Policy is not lawful and asks that the Court dismiss the counterclaim. FHJC makes four arguments in support of its position: (1) the 2019 Policy directly contradicts NYCHRL agency guidance; (2) the 2019 Policy fails to account for subsidy programs’ determination of what share of rent is affordable for a particular household; (3) Kianna Glanton’s application experience demonstrates that the 2019 Policy still creates a significant unjustified exclusion of voucher holders; and (4) Dr. Steil’s analysis reflects that the 2019 Policy continues to have an adverse impact on

applicants with partial subsidies. Finally, FHJC asserts that even if the Court were to find that the 2019 Policy is lawful, it is inadequate as a remedy for the years of discrimination Defendants engaged in and demand injunctive relief, compensatory damages, and punitive damages. First, FHJC argues that Defendants are in direct contravention of the agency responsible for implementing the NYCHRL, the New York City Human Rights Commission, and that the Commission is entitled to deference. Plaintiff’s Proposed

17 FHJC argues that Goldfarb’s 60-day rule (barring applicant from reapplying after being denied an apartment) “has a clear disproportionate impact on applicants with subsidies of which Goldfarb was likely aware.” Plaintiff’s Post-trial Memorandum of Law at 23 n.12. However, this rule applies to applicants with and without subsidies alike, and FHJC has presented no evidence that the policy disproportionately impacted applicants with subsidies. Moreover, Goldfarb has made exceptions for both applicants with and without subsidies in the past. Finding of Fact and Conclusion of Law ¶¶ 243–44. During the bench trial, FHJC offered the Commission’s “Best Practices for Housing Providers to Avoid Source of Income Discrimination” (“Best Practices”) into evidence in support of their position that Goldfarb cannot require applicants with subsidies to meet a minimum income

requirement. Pl. Ex. 69. The Best Practices materials provide frequently asked questions to help property managers meet their obligations as an owner, and in relevant part provides: Q4. May I require all applications to make a minimum income such as $68,000 annually or 43* the monthly rent? May I require co-signor or guarantor?

A4. When a voucher program calculates the tenant(s)’ rent based on their income, the government has already determined that they can afford to pay their required portion. In some instances, tenants will pay no portion of the rent out of pocket. Where the tenants’ rental portion is calculated based on the tenants’ income, it is a violation of the Law to impose any additional income requirements on applicants for housing. You may not require guarantors or co-signers.

Id. (emphasis added). “[A]n administrative agency’s construction and interpretation of its own regulations and of the statute under which it functions are entitled to great deference.” Arif v. N.Y.C. Taxi & Limousine Comm’n, 3 A.D.3d 345, 346 (1st Dep’t 2004) (citations omitted). However, agency guidance like the Best Practices materials do not receive the deference accorded formal regulations adopted after notice and hearing procedures. See Schneider v. Feinberg, 345 F.3d 135, 142 (2d Cir. 2003) (“Interpretations such as those in opinion letters—like interpretations contained in policy statements, agency manuals, and enforcement guidelines, all of which lack the force of law—do not warrant Chevron- style deference.”) (quoting Christensen v. Harris County, 529 U.S. 576, 587 (2000)). Consequently, far from FHJC’s suggestion that the Commission’s interpretation of the NYCHRL should be upheld, “interpretations contained in formats such as opinion letters are entitled to respect ... but only to the extent that those interpretations have the power to persuade.” Catskill Mountains Chapter of Trout Unlimited, Inc. v. City of New

York, 273 F.3d 481, 490 (2d Cir. 2001), adhered to on reconsideration, 451 F.3d 77 (2d Cir. 2006) (internal quotes omitted). Further, agency positions, including rules and regulations, must be “in harmony with” and must not contravene the statute construed. Weiss v. City of New York, 95 N.Y.2d 1, 4–5 (2000) (invalidating an elevator safety regulation that expanded statutory liability on non-operating owners because the regulation was inconsistent with a statute that imposed liabilities on operators). Here, similarly, there is nothing in the statute at issue, the source of income discrimination statute (§ 8-107(5)(a) of the New York City Administrative Code) stating that landlords cannot make an assessment of an applicant’s ability to pay rent. The source of income statute only forbids discrimination; it does not

require landlords to accept applicants with subsidies regardless of the amount of their incomes. Additionally, because Goldfarb owns properties in Westchester County as well, it is appropriate to consider a local statute that specifically authorizes landlords to make their own assessment of prospective tenants’ ability to pay, considering sources of income including government subsidies: The Charter and Administrative Code of Westchester County §700.20(V)(1)-(2). This statute provides that it shall not be considered discriminatory to make decisions based on an individual’s income and defines source of income to include “any form of federal, state or local public assistance or housing assistance, grant or loan program, including the federal housing subsidy known as ‘Section 8.’” Secondly, FHJC argues that the 2019 Policy leaves the door open to justified exclusions by failing to account for the various deductions that can be applied to Section

8 tenants. According to FHJC, Goldfarb should take into account the deductions that the Section 8 program applies for dependents, childcare, disability assistance, unreimbursed medical expenses, and for elderly or disabled family members in the household. However, Goldfarb does not know whether a Section 8 applicant qualifies for any of the deductions while they are processing the application. See Miller Testimony, 2/7/2023 Tr. 162:11–12. Moreover, Section 8 applicants can lose these deductions as circumstances change, causing an increase to their portion of the rent payments. Miller Testimony, 10/27/2022 Tr. 34:18–35:6. Nevertheless, FHJC maintains that Goldfarb should rely on each agency’s determination and that Defendants have not offered any explanation for why Goldfarb must undertake its own calculation. See Plaintiff’s Post-trial Memorandum

of Law at 28 n.15. However, Goldfarb is correct that it is not unlawful for a landlord to ask about income or assess an applicant’s ability to pay for housing, so long as they do not discriminate based on the subsidy, a lawful source of income. Third, FHJC points to Kianna Glanton’s application experience with Goldfarb as evidence that the 2019 Policy continues to create a significant unjustified exclusion of voucher holders. However, the record reflects that in her second application Glanton did in fact include an expired voucher, and not the renewed voucher she was granted. Defs Ex. CCC. Thus, Goldfarb properly rejected her application for “insufficient voucher.” This is far from a “baseless exclusion” of applicants with subsidies as FHJC alleges. Fourth, FHJC argues that Dr. Steil’s analysis reflects that the 2019 Policy continues to have an adverse disparate impact based on source of income, specifically on applicants with partial subsidies. Accordingly, the Court conducts the burden shifting analysis described above. FHJC has presented a prima facie case of disparate impact as

the Court has adopted Dr. Steil’s report which indicates, among other things (1) including all denial reasons, 30% of applicants with no subsidy are approved under the 2019 Policy compared with only 13% of applicants with partial subsidies, a statistically significant adverse difference and; (2) even excluding “no contact” denials, applicants with partial subsidies are only approved 20% of the time compared to applicants with no subsidy, who are approved 30% of the time, also a statistically significant adverse impact. The burden thus shifts to Defendants to prove that the “challenged practice is necessary to achieve one or more substantial, legitimate, nondiscriminatory interests.” Mhany Mgmt., 819 F.3d at 617 (quoting 24 C.F.R. § 100.500(c)(1)-(2) (2013)). The Court finds that Goldfarb’s 2019 Policy, as it relates to applicants with partial subsidies, is not necessary

to achieve its interest—assessing the ability of applicants for to pay the rent. Defendants have not offered any evidence reflecting that partial subsidy holders are not able to pay their portion of the rent absent the 2019 Policy, or provided any evidence reflecting that partial subsidy holders are unlikely to pay the portion of their rent that their programs have deemed them qualified to pay. Because Defendants have not met their burden, FHJC need not show that the “interests supporting the challenged practice could be served by another practice that has a less discriminatory effect.” Id. Accordingly, the Court finds that the 2019 Policy is not lawful, but only as to the partial subsidy requirements. C. FHJC’s Remedy FHJC argues that Defendants’ 2019 Policy does not serve as an adequate remedy for the four years in which Goldfarb excluded OHS, HASA, and Section 8 applicants and as a result, it is entitled to injunctive relief, compensatory damages, and punitive damages for Defendants’ four years of discrimination.

Injunctive relief The FHA and NYCHRL authorize injunctive relief both to end future discriminatory practices and to provide appropriate action to eliminate the discriminatory effects of past illegal policies. 42 U.S.C. § 3613(c)(1); N.Y.C. Admin. Code § 8-502(a); see also U.S. v. Yonkers Bd. of Educ., 837 F. 2d 1181, 1236 (2d Cir. 1987) (“A district court has ‘not merely the power but the duty to render a decree which will so far as

possible eliminate the discriminatory effects of the past as well as bar like discrimination in the future.’”) (internal citation omitted). The Court must craft injunctive relief with a view toward the statute’s goals of preventing future violations and removing lingering effects of past discrimination. See Rogers v. 66–36 Yellowstone Blvd. Coop. Owners, Inc., 599 F. Supp. 79, 83 (E.D.N.Y. 1984) (internal citation omitted). The scope of the injunction is to be determined by the nature and extent of the legal violation. Id. The characteristics of the injunction must serve the “twin goals of eradicating the past discrimination and securing future adherence to our nation’s equal housing commitment.” Id.

In the instant case, Defendants have engaged in a consistent pattern of source of income discrimination. In 2015 Defendants were sued for their policy that openly refused to accept HASA subsidies. As part of the settlement, they were required to abide by a general injunction not to make unavailable or withhold apartments because of a lawful source of income. Pl. Ex. 68 ¶ 17; Pl. Ex. 27. Shortly thereafter, Defendants adopted the 2015 Policy which, as the Court has found, continued to make it virtually impossible for applicants with subsidies to qualify under their income policy.

However, in 2019, Defendants adopted a new policy which this Court has found does not have a disparate impact on full subsidy holders, but does have a disparate impact on partial subsidy holders. In general, injunctive relief is necessary where a defendant has not brought its policies and practices into compliance with the law and there is a likelihood that the illegal conduct will continue. USA v. Space Hunters, Inc., No. 00 Civ. 1781 (RCC), 2004 WL 2674608, at *8–9, aff’d 429 F.3d 416 (2d Cir. 2005). In the instant case, Defendants have improved their rental criteria to lessen the discriminatory impact but did not entirely remove those criteria that limited access to Defendants’ apartments for renters with partial subsidies. Accordingly, the Court grants FHJC’s proposed injunctive relief enjoining Defendants from:

• Denying or withholding rental apartments, or otherwise making rental apartments unavailable on the basis of disability or lawful source of income, including by applying minimum income requirements to applicants with rental subsidies or vouchers;

• Discriminating against applicants with rental subsidies or vouchers in the terms or conditions of the rental process because of disability or source of income, including by offering variances and waivers to rental criteria to non-subsidy applicants and not to subsidy applicants, and requiring rejected applicants to wait a minimum period of time before reapplying for apartments to rent; and

• Coercing, intimidating, threatening, or interfering with any person, including FHJC, on account of having aided or encouraged any other person in the exercise or enjoyment of any rights granted or protected by the FHA and the NYCHRL. Doc 169-1 (Plaintiff’s Proposed Injunction). Additionally, Defendants shall (1) adopt written non-discriminatory rental criteria, (2) contract with a third party to provide a fair housing training program as outlined, (3) maintain sufficient records to verify compliance with the terms of the injunction and make such records available to FHJC. Id.

The non-discriminatory rental criteria shall not include: a minimum gross or net income requirement, including net rent-to-income ratios for applicants with partial pay or 100% pay rental subsidies or vouchers or minimum time periods for rejected applicants to reapply. Id. Compensatory Damages

FHJC seeks compensatory damages in the form of (1) the value of the diversion of its resources to investigate and eliminate Defendants’ discriminatory policies and practices; and (2) the value of the injury to FHJC’s mission or purpose in the form of the future costs to monitor compliance with the injunctive order and repair for some of the community harm caused by Defendants’ policies and practices. In support of its diversion of resources claims, FHJC has produced time records, invoices from Dr. Steil, and testimony describing the number of hours spent on its testing investigation of Defendants; its post-testing activities to evaluate the information it obtained and decide how to respond to end the challenged policy and modify Defendants’ future practices; and its work during the course of this multi-year litigation including the trial. Also, FHJC has established its ordinary hourly billable rates at the time of its testing investigation and seeks those same hourly rates even though the organization’s

hourly rates have increased in the four years since that time. Plaintiff’s Proposed Finding of Facts and Law ¶ 172. The Court finds FHJC’s damage award request to be reasonable, well-documented, and consistent with the evidence presented at trial. Accordingly, the Court awards FHJC $52,300.00 in diversion of resources compensatory damages. FHJC also seeks future compensatory damages for frustration of mission to counteract the injury caused by Defendants to FHJC’s mission and compensation for

three future activities: (1) monitoring Defendants’ rental advertisements, website, and rental application records; (2) developing and distributing consumer rights information, including about housing discrimination laws, to organizations and government agencies to assist renters with rental subsidies searching for housing; and (3) conducting compliance testing of Defendants. Id. ¶¶ 174–77. The Court has reviewed the testimony and exhibits by FHJC describing each of these activities and agrees with Plaintiff that they are necessary to counteract Defendants’ injury to the organization and the community it serves in New York City and Westchester County where Defendants manage thousands of rental units. The proposed hourly rates for future compliance and outreach work are based on FHJC’s 2018 hourly billable rates.

Additionally, FHJC seeks frustration of mission damages for a period of five years, since Defendants instituted their discriminatory rental requirements for at least five years. Defendants argue testing for violations of fair housing laws is part of FHJC’s mission, therefore there is no interference with their mission, and damages are unwarranted. Defendants’ Proposed Finding of Facts and Law ¶ 42. FHJC’s mission is “to ensure that all people have equal access to housing opportunities in the New York City Region by eliminating housing discrimination; promoting policies and programs that foster open, accessible, and inclusive communities; and strengthening enforcement of fair housing laws.” Plaintiff’s Proposed Finding of Facts and Law ¶ 3. Conducting test calls is one tool FHJC utilizes in accomplishing its mission. Courts have held that “a defendant’s act that reduces the efficiency or success of the organization’s activities can be adequate” to frustrate an organization’s mission. Hous. Rts. Initiative v. Compass, Inc., No. 21 Civ. 2221 (SHS), 2023 WL 1993696, at *7 (S.D.N.Y. Feb. 14, 2023),

reconsideration denied, No. 21 Civ. 2221 (SHS), 2023 WL 2989048 (S.D.N.Y. Apr. 18, 2023). The Court finds that the 2018 rate is reasonable and that five years of monitoring and education is appropriate, since Defendants have employed their discriminatory rental criteria for more than five years. Accordingly, the Court orders Defendants to pay $188,240 in frustration of mission compensatory damages, for a total of $240,540 in compensatory damages.

Punitive Damages Punitive damages are available under the NYCHRL where the defendant “engaged in a discriminatory practice or discriminatory practices with malice or with reckless indifference to the … protected rights of an aggrieved individual.” Robinson v. Instructional Sys., Inc., 80 F. Supp. 2d 203, 209 (S.D.N.Y. 2000). Likewise, the FHA expressly provides for the recovery of punitive damages by plaintiffs who have suffered discriminatory housing practices. 42 U.S.C. § 3613(c)(1). Defendants argue punitive damages are not justified because Goldfarb “did not deliberately exclude tenants with housing subsidies in the development of its 2015 Policy” and “implemented facially neutral application criteria in an attempt to make sure all applicants were treated the same.” Defendants’ Proposed Finding of Facts and Law ¶

49. However, the Court has found that Defendants created both discriminatory policies and practices at issue. As described above, Defendants took no efforts to evaluate whether the 2015 Policy complied with fair housing laws and it was only after two lawsuits that they changed the minimum income policy for applicants with subsidies. And although the 2019 Policy is an improvement, there are still aspects of the policy that

the Court finds questionable, such as its income calculation for applicants with partial subsidies and its “no contact” and “insufficient voucher” designation procedures. When determining an appropriate amount of punitive damages, the Supreme Court has set forth three “guideposts” to consider: (1) the reprehensible nature of the conduct; (2) the ratio of punitive damages to compensatory damages; and (3) civil sanctions for comparable misconduct. BMW of N. Am., Inc. v. Gore, 517 U.S. 559, 575– 582 (1996). FHJC requests $1.5 million in punitive damages. Defendants have had ample opportunity and time to correct their policies, have failed to do so, and thereby have acted discriminatorily toward two of the most vulnerable populations in our society— people with disabilities and low-income renters.

Furthermore, Defendants’ large size (over 6,000 apartments) and annual gross revenue of over $72 million indicate that their discriminatory conduct affected many people. While the Court has found that Defendants’ policies, and its 2015 Policy in particular, violated the FHA and the NYCHRL, it finds that its efforts to remediate the violations, while falling short, suggest its conduct was not “reprehensible.” Therefore, the Court awards FHJC $750,000 in punitive damages. With respect to the ratio of punitive damages, while the Supreme Court has “rejected the notion that the constitutional line is marked by a simple mathematical formula” the Court has indicated that a ratio of punitive damages to compensatory damages of ten to one or less is generally appropriate. Id. at 582. FHJC has requested $240,540 in compensatory damage, thus making $750,000 in punitive damages reasonable. Lastly, with respect to comparable civil penalties, $750,000 is also reasonable.

As of January 2018, the United States could seek a civil penalty of up to $102,606 for a first violation of the FHA (as of 2022 the number has increased to $115,054). See 28 CFR 85.3(b)(3)(i). Under the 2018 notice, the penalty then doubles to $205,211 for each subsequent violation. 28 C.F.R. 85.3(b)(3)(ii). Thus, a $750,000 punitive damages award would reflect the maximum penalty for roughly three violations.18 Likewise, the City Commission can impose a penalty of up to $125,000 for any violation. N.Y.C. Admin. Code § 8-126(a). Again, an award of $750,000 would reflect six violations, even though it is likely that far more applicants were harmed by Defendants over the course of the 2015 Policy. “Under well established precedent in this Circuit, it is the defendant’s burden to

show that his financial circumstances warrant a limitation of the award.” Mathie v. Fries, 121 F.3d 808, 816 (2d Cir. 1997) (citation and internal quotation marks omitted). Defendants have not proffered any evidence at trial that warrants a limitation of the award. Accordingly, the Court orders Defendants to pay $750,000 in punitive damages to FHJC.

18 �e Court notes that Defendants are likely to have committed more than three violations in light of their 6,000-unit portfolio over four years. IV. CONCLUSION For the foregoing reasons, the Court finds for the FHJC on each of its claims under the FHA and NYCHRL. The Court also finds that the 2019 Policy is not lawful, but only as to the partial subsidy requirements. The Court awards FHJC $240,540 in compensatory damages and $750,000 in punitive damages. Additionally, the Court will enter an injunction against Defendants concurrently with this Opinion & Order. The Clerk of Court is respectfully directed to enter judgment and close the case.

It is SO ORDERED.

Dated: Setember 29, 2023 : } New York, New York bt \ 2 Edgardo Ramos, U.S.D.J.