UNITED STATES DISTRICT COURT
DISTRICT OF CONNECTICUT

CONNECTICUT
CENTER et al.,

FAIR

HOUSING :
:
:
Plaintiffs,
:
:
v.
:
:
CORELOGIC RENTAL PROPERTY :
SOLUTIONS, LLC,
:
Defendant.
:
:

No. 3:18-CV-705 (VLB)

August 7, 2020

Memorandum of Decision on Motions for Summary Judgment [Dkts. 87, 112, 116]
Plaintiffs Connecticut Fair Housing Center (“CFHC”) and Carmen Arroyo
(“Ms. Arroyo”), individually and as next friend for Mikhail Arroyo (“Mr. Arroyo”)
(collectively, “Plaintiffs”) bring the instant litigation against Defendant CoreLogic
Rental Property Solutions, LLC (“Defendant” or “RPS”) alleging that RPS violated
the Fair Housing Act, 42 U.S.C. §§ 3601 et seq. (“FHA”), the Connecticut Unfair
Trade Practice Act, Conn. Gen. Stat. §§ 42-110a et seq. (“CUTPA”) and the Fair
Credit Reporting Act, 15 U.S.C. §§ 1681 et seq. (“FCRA”).
In April of 2016, Carmen Arroyo attempted to move her disabled son, Mikhail
Arroyo, for whom she was conservator, into her apartment complex ArtSpace
Windham, but his application was rejected. Two separate actions by defendant
CoreLogic Rental Property Solutions, LLC regarding that incident motivate the
instant lawsuit: first, CoreLogic RPS, through its CrimSAFE product, notified
apartment manager WinnResidential that “disqualifying records” were found for
Mr. Arroyo; second, RPS did not disclose Mr. Arroyo’s criminal records to Ms.

1

 Arroyo on behalf of Mr. Arroyo until the start of this litigation, despite her numerous
requests and provision of many documents.
RPS has filed a Motion for Summary Judgment as to the entirety of the
action. [Dkts. 112 (Redacted Version) and 114 (Unredacted Version)]. Plaintiffs have
filed two separate Motions for Partial Summary Judgment: one as to their file
disclosure claims [Dkt. 87], and one to their race and national origin discrimination
FHA and CUTPA claims. [Dkts. 116 (Redacted Version) and 118 (Unredacted
Version)]. The parties have filed oppositions and replies for each motion. For the
following reasons, the Court grants in part and denies in part RPS’s motion for
summary judgment and denies Plaintiffs’ motions for partial summary judgment.
I.

Material Facts1

A. Parties
Mikhail Arroyo is a Latino man. [Dkt. 118-1 (Pl.’s 56(a)1 Statement) ¶39]. Mr.
Arroyo is significantly disabled.

Id. ¶40.

His disabilities were caused by an

accident in July 2015. Id. ¶40. Mr. Arroyo was hospitalized until early 2016, when
he was transferred to a nursing home to continue to recover from his injuries. Ibid.

All facts are taken from the parties’ unredacted statements of undisputed facts
for the purposes of deciding these motions only. If a fact stated in one party’s
56(a)1 statement is admitted by the other party in its 56(a)2 statement and is
supported by the underlying exhibits, the Court cites to the statement in which the
fact first appeared. D. Conn. L. Civ. R. 56(a). For all other facts, the Court cites to
the underlying exhibit. The Court will issue an order on the parties’ motions to seal
in short order.
1

2

 The Connecticut Probate Court appointed Carmen Arroyo Mikhail Arroyo’s
conservator in August of 2015. 2,3.

The Connecticut Fair Housing Center (“CFHC”) is a housing advocacy nonprofit. RPS is a national tenant screening company that offers tenant screening
products under the rubric of “Rental Property Solutions,” which it has described

2

In Connecticut, a conservator may only be appointed in the following
circumstances:
(f) (1) If the court finds by clear and convincing evidence that the
respondent is incapable of managing the respondent's affairs, that the
respondent's affairs cannot be managed adequately without the
appointment of a conservator and that the appointment of a
conservator is the least restrictive means of intervention available to
assist the respondent in managing the respondent's affairs, the court
may appoint a conservator of his or her estate after considering the
factors set forth in subsection (g) of this section.
(2) If the court finds by clear and convincing evidence that the
respondent is incapable of caring for himself or herself, that the
respondent cannot be cared for adequately without the appointment
of a conservator and that the appointment of a conservator is the least
restrictive means of intervention available to assist the respondent in
caring for himself or herself, the court may appoint a conservator of
his or her person after considering the factors set forth in subsection
(g) of this section.
(3) No conservator may be appointed if the respondent's personal
needs and property management are being met adequately by an
agency or individual appointed pursuant to section 1-43, 19a-575a,
19a-577, 19a-580e or 19a-580g.
Conn. Gen. Stat. § 45a-650 (2015).
3 See [Dkt. 125-10 (Pls.’ Opp. to Def.’s Mot. Summ. J. Ex. 8: 6/14/16 Arroyo File
Disclosure) at 3]; [Dkt. 125-12 (Pls.’ Opp. to Def.’s Mot. Summ. J. Ex. 8: 11/15/16
Arroyo File Disclosure) at 5]. The Court finds that these documents themselves
would not be admissible evidence as they are not authenticated per Federal Rule
Evidence 901 and 902, but, per Federal Rule 56(c), they might point to the existence
of documents that would be admissible. Further, RPS has not disputed for the
purposes of this motion that Carmen Arroyo is Mikhail Arroyo’s conservator.
3

 as a “comprehensive leasing decision service to the single and multifamily
housing industry.” [Dkt. 114-1 (Def’s 56(a)1 Statement of Material Facts) ¶1]. One
of those products is “Registry CrimSAFE” (“CrimSAFE”). Id. ¶5. RPS provides this
service to managers of more than 120 properties in Connecticut. [Dkt. 118-1 ¶ 100].
Though not a party, WinnResidential also plays a central role in this
litigation. It is one of the largest property management companies in the country.
[Dkt. 118-1 ¶29]. WinnResidential has used RPS’s screening products since 2008.
Id. ¶30. In 2016, WinnResidential managed ArtSpace Windham, an apartment
complex in Connecticut where Carmen Arroyo resided and applied for housing on
behalf of Mikhail Arroyo. Id. ¶41.
B. CrimSAFE’s Role in Rental Housing Application Evaluations
For each housing applicant, CrimSAFE filters and then categorizes any
identified crimes according to their severity levels under varying state and federal
law, as well as the type of crime. Id. ¶5. CrimSAFE then applies the leasing criteria
chosen by the housing provider from the menu offered by CrimSAFE to any records
found and informs the housing provider whether “disqualifying” records are found.
Id. ¶5. Disqualifying records consist of both convictions and other charges,
including arrests which have not led to a conviction. See [Dkt. 118-4 (Ex. 22 to Pl.’s
Mot. Summ. J.].
To use CrimSAFE, a landlord fills out a short electronic form, generated by
RPS, that lists general categories of crimes for which CrimSAFE can screen. [Dkt.
118-1 ¶8]. A landlord establishes the leasing criteria by selecting from the list the
crimes which it wants CrimSAFE to screen. Id. These criminal categories track
4

 verbatim those created by the FBI’s National Incident Based Reporting System.
[Dkt. 129-1 (Def.’s 56(a)2 Statement of Facts) ¶8]. For each criminal category, the
landlord enters the maximum number of years back CrimSAFE should look “to
decline an applicant for the specified type of crime” (hereinafter, “lookback
period”). [Dkt. 118-1 ¶8]. The CrimSAFE configuration webpage explains that
“applicants whose criminal record[s] are older than the number of years for the
specified crime will result in an accept for your community.” Id. The maximum
lookback periods are 99 years for convictions and 7 years for non-convictions. Id.
¶ 11.
When a landlord receives a rental application, it provides RPS the applicant’s
first and last name, date of birth, and current address (and optionally the middle
name). [Dkt. 118-1 ¶14]. RPS then searches its database for criminal records that
match the applicant. Ibid. If a criminal record is matched to the applicant, RPS
determines the category, if the record is felony or not, and if conviction or not. Id.
¶16.4 After locating and categorizing a record that has been matched to the
applicant, RPS compares the age of the record with the lookback period for the
given category. Id. ¶17. If the applicant has a record within the landlord’s chosen

4

If the offense has been categorized in the past and appended to a master
offense table, this process is fully automated. Ibid. If the offense’s category has not
been appended to the master table, it is placed in a queue to be manually
categorized. Ibid. All offenses categorized in the master offense table have been
manually processed by employees and/or contractors of CoreLogic. Ibid. The RPS
database reflects records obtained from more than 800 different
jurisdictions/sources, for a total of more than 579 million records spanning more
than 10 million unique offense descriptions, which RPS has decades of experience
in standardizing and integrating into its database. [Dkt. 114-1 at ¶16].
5

 category and lookback period, CrimSAFE notifies the leasing agent of disqualifying
“record(s) found,” and directs the housing provider to “proceed with their
communities screening policies.” Id. ¶19.
In 2016, the period at issue here, the cover page of the RPS’s screening
report for leasing agents was titled “Lease Decision,” and listed a “Crim Decision,”
which tracked the CrimSAFE result, and stated “Record(s) Found” if disqualifying
records were found. Id. ¶ 20; see [Dkt. 114-2 (Kayani Decl.) at Ex. C (Leasing Agent
Version of Arroyo Background Screening Report)].
RPS has marketed and sold CrimSAFE as rendering a decision on an
applicant’s suitability for tenancy based on their criminal history. [Dkt. 118-1 at ¶5].
It has described the removal of “human bias or judgment” as a “benefit” of its
CrimSAFE product. Id. at ¶6. In configuration instructions it has provided, it states,
“A criminal record generally contains information on the type of crime, degree and
level of crime, and date of offense. With CrimSAFE, this public record information
is evaluated and used to provide a decision based on the client’s pre-determined
criminal decision policy.” Id. at ¶9.
RPS allows CrimSAFE customers to disclose or suppress information
underlying disqualification from its staff and housing applicants. Id. ¶ 26. If a
customer chooses to suppress disclosure of the underlying criminal record from
its onsite leasing staff, they see only whether disqualifying records are found or
not. Id. ¶ 26. Landlords have the option of having adverse action letters
automatically delivered to applicants via email when CrimSAFE has found

6

 disqualifying records in order to notify them that their applications have been
declined. Id. at ¶24.
CrimSAFE is only one of RPS’s screening products: RPS also offers
separate tenant screening products which simply identify and return criminal
public records of a housing applicant to the housing provider, but which do not
themselves filter or categorize the results in any way. [Dkt. 114-2 (Kayani Decl.)
¶¶4-5].
The parties provide conflicting evidence on whether CrimSAFE always
returns a copy of the underlying report that displays the full public data of an
applicant’s criminal record to someone at the client housing provider. RPS’s
executive Naeem Kayani declares that it does. [Dkt. 114-2 (Kayani Decl.) ¶7].
Plaintiffs point to a 2016 training which states that, if a housing provider unchecks
a box on the CrimSAFE configuration page, none of its users have access to the
reports containing the full public records. [Dkt. 118-3 (Ex. 13 to Pl.’s Mot. Summ.
J.: 2016 Training) at 37]; [Dkt. 118-4 [Ex. 22 to Pl.’s Mot. Summ. J.:

2016

Configuration Page Example)], [Dkt. 116-18 (Ex. 15 to Pl.’s Mot. Summ. J.: Thomas
Dep.) at 71-72].
In its proposal to WinnResidential, RPS wrote that, with CrimSAFE, “criminal
record search results are evaluated using our own advanced, proprietary
technology and an accept/decline leasing decision is delivered to your staff.” Id.
at ¶33]. RPS’s proposal to WinnResidential explains that “[u]sers who choose to
have their rental decisions automated using ScorePLUS® and CrimSAFE® may
suppress the full reports from the view of their on-site staff. WinnResidential
7

 currently uses this option and the site managers view a decision report.” Id.at ¶
36.5 WinnResidential also does not disclose the basis for an applicant’s denial to
the applicant, except to inform them that the denial is based on RPS’s screening
report. See [Dkt. 114-2]. RPS serves many customers with affordable/subsidized
properties. [Dkt. 114-4 (Dachtler Decl.) ¶5]. For instance, the majority of Winn
Residential’s rental units are federally-subsidized/affordable properties. Id.
Fewer than 7% of all rental housing applicants in Connecticut between 2016
and the present have had any “record found” through CrimSAFE. [Dkt. 118-1 ¶11].
C. CrimSAFE, Race and Ethnicity, and Criminal Records
CrimSAFE uses data from a national database of criminal records that RPS
aggregates from multiple sources, including incarceration records and court
records of criminal cases for both charges and convictions obtained from state
departments of corrections and administrative offices of the courts. [Dkt. 118-1
¶65]. RPS receives and records the race and ethnic background for close to 80%
of housing applicants who did match with a criminal record. [Dkt. 118-15 (Ex. 47 to
Pl.’s Mot. Summ. J.: RPS Documentation of Race)]). But RPS is not aware of the of
the race or ethnicity of housing applicants who did not match with such a criminal
record. [Dkt. 126-2 (Ex. 1 to Pl.’s Opp: Kayani Dep.) at 207:14-25].

5

As previously detailed, the parties dispute whether the full criminal records
are always delivered to someone at the client.

8

 Data reveal that disparities adverse to African Americans and Latinos6 and
in favor of whites7 exist at all stages of the criminal justice process: in arrest rates,
in jail detention rates, and in prison incarceration rates. Id. ¶73. African Americans
in the United States are more than four times as likely as whites, and Latinos twoand-a-half times as likely as whites, to have been either jailed or incarcerated at
some point in their lifetimes. [Dkt. 118-1 ¶72]. National data from 2015 demonstrate
that African Americans and Latinos are more likely to experience jail or prison
incarceration than Whites, regardless of income level. Id. ¶¶ 74-77. The disparity in
incarceration rates for African Americans in Connecticut is just over twice the
disparity at the national level, while for Latinos in Connecticut, the disparity is three
times the disparity at the national level. Id. ¶ 80. Overall, 10.61% of African
Americans nationally experience either jail or prison during their lifetime. Id. ¶ 81.
Among African Americans who were earning less than $30,000 in 2015, 14.34%
nationally had been in jail or prison in their lifetime. Id. ¶ 82.
The above data do not distinguish between innocent individuals who have
been charged but not convicted of a crime and guilty individuals who have been
convicted of committing a crime. The Court takes judicial notice of shorter-term
data that confirm that disparities exist both for individuals who are jailed and for
individuals who are imprisoned, though the disparity in imprisonment rates is

The Court follows the Plaintiffs in using the term “Latinos” as a noun to refer to a
group of people of Latin American family origin that includes at least one person
who does not identify as female.
7 The term “white” is here used to refer to non-Latino whites and the term AfricanAmerican is here used to refer to non-Latino African-Americans, except where
otherwise noted.
6

9

 greater. At year-end in 2015 in the United States, 169 in 100,000 white adults were
incarcerated in jails, 174 in 100,000 Latino adults were incarcerated in jails, and 607
in 100,000 African American adults were incarcerated in jails.8 Zhen Zeng, Jail
Inmates in 2016, Bureau of Justice Statistics (Feb. 2018), at Table 2, available at
https://www.bjs.gov/content/pub/pdf/ji16.pdf; see Fed. R. Evid. 201(b)-(c). Also at
year-end in 2015 in the United States, 312 in 100,000 whites were imprisoned for
sentences of more than 1 year, 820 in 100,000 Latino adults were imprisoned were
imprisoned for sentences of more than one year, and 1,745 in 100,000 African
American adults were imprisoned for sentences of more than one year. E. Ann
Carson and Elizabeth Anderson, Prisoners in 2015, Bureau of Justice Statistics at
2,

Figure

4

(Dec.

2016),

available

at

https://www.bjs.gov/index.cfm?ty=pbdetail&iid=5869.
As to arrests, Latinos comprised 42% of federal drug arrests made in 2014,
nearly three times their share of the population. [Dkt. 118-1 ¶71]. 9 In total, 64% of
The Court follows the Plaintiffs in using the term “Latino” as an adjective
when referring to a group of people of Latin American family origin that includes at
least one person who do not identify as female.
9 Citing Mark Motivans, Bureau of Justice Statistics, U.S. Department of
Justice, Federal Justice Statistics, 2013-14, at 10 (March 2017),
https://www.bjs.gov/content/pub/pdf/fjs1314.pdf; 2014 ACS 1-year Demographic
Estimates. Defendants object that this data is irrelevant to the proposition for
which it is used, given that it is only specific to federal drug arrests, but the Court
finds that it is sufficiently relevant, as drug arrests were the second most common
federal arrest in 2014 after immigration. Motivans at 7. Further, accurate data on
overall arrests of Latinos from criminal justice institutions is limited, as
demonstrated below for Connecticut.
Plaintiffs do not present national total arrest data, though Plaintiffs’ expert
Dr. Christopher Wildeman stated, that his certainty “would be extremely high” “that
the disparities [he] observed in the incarceration data also exist at the level of
arrest, charge and conviction” because “the transition probabilities from each
stage would have to be so much lower for Whites than for Native Americans and
8

10

 federal drug arrests were of Latinos and African Americans, who comprised 29%
of the total population. Id. Only 31% of federal drug arrestees were of whites, less
than half their share of the population. Id. African Americans and Latinos are more
likely than whites to be arrested, convicted, and sentenced for drug offenses even
though their rates of drug use are comparable to those of whites. Id.
In 2016, African Americans comprised 29.88% of all arrestees in the State of
Connecticut. State of Connecticut Department of Emergency Services and Public
Protection, Crime in Connecticut at 29 (2016) (hereineafter “Crime in Connecticut
2016”).10 But, as of 2016, African Americans comprised only 10.6% of Connecticut’s
population. U.S. Census Bureau, 2016 American Community Survey 1-year
Estimates, Table DP05: ACS Demographic and Housing Estimates (hereinafter
“2016 ACS 1-Year DP05”).11 Connecticut does not track arrests by ethnicity, so the
percentage of arrestees who are Latinos is unknown (and Connecticut’s reported
numbers of arrestees who are African American and white includes Latino
arrestees). Id.
The percentage of the population who is African American and the
percentage of the population who is Latino differs between Connecticut cities and

Hispanics and their starting rates of experiencing those events would have to be
so much higher that [non-disparities] would just be… a statistical aberration.” [Dkt.
116-46 (Wildeman Dep.) at 79:13-80:12].
10

Available at:
http://www.dpsdata.ct.gov/dps/ucr/data/2016/Crime%20in%20Connecticut%20201
6.pdf.
11 Available at data.census.gov. Reported figure has a 0.2% margin of error. An
additional 0.9% of the population reports being Black and White or Black and Native
American.
11

 Connecticut suburbs or rural areas and is higher for renters looking for affordable
or subsidized housing than for those who are not. [Dkt. 114-1 ¶¶56-57].
D. CrimSAFE & the Purpose of Criminal Records
RPS has identified the purposes served by CrimSAFE and its policies as
tenant safety and landlord liability avoidance. [Dkt. 114-1 at ¶89]. RPS states that
the purpose of its criminal records screening products is to protect safety and
property in housing complexes because “[c]riminals can disrupt –and even
endanger –the entire neighborhood.” Id. at ¶90. It claims that its products allow
housing providers to more rapidly and accurately screen applicants according to
the standards of the housing providers. Id. ¶¶13-14.
The federal Bureau of Justice Statistics has found that 83% of released
prisoners are arrested again within 9 years, with an average of 5 arrests per
released prisoner. [Dkt. 114-1 ¶59]. The federal Bureau of Justice Statistics has
found that 18% of violent victimizations took place in the victim’s home, 16% took
place near the victim’s home, and 9% took place at a friend’s, neighbor’s, or
relative’s home. Id. ¶ 60.
RPS’s proffered expert, Jay Kacirk, stated in a September 2019 deposition
that “information about arrest records that did not lead to convictions… would not
be relevant to the decision whether a [rental housing] applicant should be accepted
or rejected because we are not able to use arrest records to base an approval or
denial on.” [Dkt. 126 at 28] (citing [Dkt. 126-1 ¶74] (citing [Dkt. 125-14 (Ex. 12 to Pl.’s
Opp.: Kacirk Dep.) at 43:8-21])). He later stated that pending charges “could affect
the tenancy of someone on the verge of incarceration,” but reiterated that “you

12

 have to be careful using arrests as far as decision-making.” [Dkt. 132 (Ex. J to Ex.
4 to Def.’s Opp: Kacirk Dep.) at 294]. Dr. Lila Kazemian, one of Plaintiff’s proferred
experts, said that “anybody who has their name already in the system becomes
more likely to have more contacts with the criminal justice system,” such that
individuals who were previously arrested have elevated statistical levels of rearrest. [Dkt. 129-1 ¶14].
E. CrimSAFE’s Role in Mikhail Arroyo’s April 2016 Rental Application Rejection
On April 4, 2016, HUD’s Office of General Counsel published a document
titled “Application of Fair Housing Act Standards to the Use of Criminal records by
Providers of Housing and Real Estate-Related Transactions.” The document
stated: “Nationally, racial and ethnic minorities face disproportionately high rates
of arrest and incarceration,” and that, “the fact of an arrest is not a reliable basis
upon which to assess the potential risk to resident safety or property posed by a
particular individual.” Id. at 3, 5 [hereinafter HUD OGC 4/4/2016 Guidance].
On April 15, 2016, over a week before Mikhail Arroyo’s application, RPS
shared with some of its clients an email which highlighted (a) per the HUD OGC
4/4/2016 Guidance, arrest records that don’t result in convictions are not reliable
bases to assess the potential risk resident safety or property; and (b) “according
to HUD, a blanket policy to deny any applicants with a criminal record may have a
disparate impact on African Americans and Hispanics.” [Dkt. 125-17 (Ex. 15: 4/15/16
RPS Email) at 2]; see also [Dkt. 125-18 (April 2016 RPS HUD Training PowerPoint)]].
In the email, RPS “recommends that our customers work with their legal counsel
to review their eligibility requirements and related policies around the use of

13

 criminal background data to… determine [whether] changes need to be made to
your CoreLogic product settings.” [Dkt. 125-17 at 2].
The email also states that “RPS is currently reviewing our products to
determine what changes, if any, to make in order to best support our clients in light
of this new guidance from HUD. Once this review is complete, any changes will be
communicated to clients with enough notice to allow for clients to adjust their
processes.” [Dkt. 125-17 at 3]. Despite the HUD OGC 4/4/2016 Guidance, as of
September 5, 2019, Stephanie Dachtler, a Relationship Manager for RPS, was not
aware of any changes made to CrimSAFE. [Dkt. 118-8 (Ex. 26 to Pl.’s Mot. Summ.
J.: Dachtler Dep.) at 75:18-23 There is no evidence on the record that RPS informed
its customers that they could change their CrimSAFE settings to reverse their
election to suppress from line staff and applicants the basis on which CrimSAFE
categorized applicants “disqualified,” so they could proceed with their
communities screening policies consistent with the HUD OGC 4/4/2016 Guidance.
WinnResidential has used RPS screening products since 2008 and used
CrimSAFE through at least July 31, 2019. [Dkt. 118-1 ¶30]. As of April 2016,
WinnResidential CrimSAFE settings included all charges for “theft” occurring
within the prior three years. [Dkt. 114-1 ¶23]. WinnResidential made changes to its
CrimSAFE settings in May of 2016 and July of 2016 in response to the HUD OGC
4/4/2016 Guidance. Id.
On April 26, 2016, Ms. Arroyo applied for housing at WinnResidential on
behalf of Mr. Arroyo at the Artspace Windham in Willimantic Connecticut. [Dkt. 1141 ¶25]. WinnResidential electronically requested from RPS a tenant screening

14

 report on Mr. Arroyo. Id. ¶ 26. That day, RPS provided WinnResidential a screening
report on Mr. Arroyo that included a “Score Decision” regarding Mr. Arroyo’s
credit-worthiness and a “Crim Decision” regarding his suitability as a tenant based
on his criminal background. [Dkt. 118-1 ¶44]. WinnResidential suppressed the
underlying criminal records from the view of its leasing agents. [Dkt. 126-1 ¶91].
The “Crim Decision” for Mikhail Arroyo stated “Record(s) Found,” which is
the text that appeared on reports RPS prepared for WinnResidential when it
determined that disqualifying records were found. Id. ¶45. The fourth page of the
screening report specifies, “Based upon your community CrimSAFE settings and
the results of this search, disqualifying records were found. Please verify the
applicability of these records to your applicant and proceed with your community’s
screening policies.” Id. ¶46; see [Dkt. 116-33 (Ex. 30 to Pl.’s Mot. Summ. J.: Adverse
Action Letter to Mikhail Arroyo)].
The parties dispute whether any other decisionmakers at WinnResidential
had Mr. Arroyo’s criminal record. Compare [Dkt. 116-41 (Aff. Ans. of
WinnResidential to Administrative Compl.) ¶23] and [Dkt. 116-35 (6/13/2017 Fact
Finding Hearing Tr.) at 50, 52, 68-71] with [Dkt. 114-2 (Kayani Decl.) ¶15] and [Dkt.
114-2 at Ex. D (Administrator Version of Mikhail Arroyo Screening Report
Generated May 3, 2018)].12 The Court finds there is a genuine issue of fact as to

The parties each argue that the other party’s evidence on this point is
inadmissible. RPS argues that Dkt. 116-41 (Aff. Ans. of WinnResidential to
Administrative Cmplt.) is inadmissible hearsay, see Fed. R. Evid. 802; Dkt. 114-2
(Kayani Decl.) and that Dkt. 116-35 (Carmen Arroyo Decl.) were not made on the
basis of personal knowledge, see Fed. R. Evid. 602, Roberts v. Ground Handling,
12

15

 whether WinnResidential knew the basis on which RPS categorized Mr. Arroyo as
disqualified because WinnResidential elected to suppress records and there is no
evidence it was given the option to or did change that election.
The sole criminal record upon which RPS relied in making the CrimSAFE
report for Mikhail Arroyo is a single charge in Pennsylvania for “grade S” retail
theft under 18 Pa. C.S.A. § 3929(a)(1) filed on July 18, 2014, when he was twenty
years old and prior to his accident. [Dkt. 118-1 at ¶51]. A “Grade S” in Pennsylvania
means “summary offense,” which is below the level of a misdemeanor and is often
called a non-traffic citation. Id. at ¶ 52. A charge for summary offense retail theft
indicates that this was his first offense and the value of the merchandise he
allegedly stole was under $150. 18 Pa. C.S.A. § 3929(b)(1). Ibid. The charge had not
led to a conviction as of the date it was reported. Id. ¶54.
Mr. Arroyo’s application was rejected. Id. ¶30. As a matter of law, Mr. Arroyo
was innocent of the charge. See Coffin v. United States, 156 U.S. 432 (1895) (“the
principle that there is a presumption of innocence in favor of the accused is the
undoubted law, axiomatic and elementary...”). Mr. Arroyo was never convicted
and, on April 20, 2017, the charge against Mr. Arroyo was withdrawn. Id. ¶¶53, 54.
Mr. Arroyo remained in a nursing home until June 2017. Id. ¶¶30, 63.

Inc., 499 F. Supp. 2d 340, 360 (S.D.N.Y. 2007) (“It is axiomatic that affidavits
submitted in support of or in opposition to a summary judgment motion must ‘be
made on personal knowledge”). Plaintiffs argue that [Dkt. 114-2 at Ex. D] cannot be
authenticated as a copy of a document that CoreLogic returned to WinnResidential
on 4/26/2016 as claimed, as it is a document generated on 5/3/2018, see Fed. R.
Evid. 901. However, neither party demonstrates that the other cannot produce
admissible evidence. see Fed. R. Civ. P. 56(c)(2). The Court finds that a dispute of
fact remains.
16

 F. Mr. Arroyo’s Consumer File Disclosure Request
RPS regularly processes requests from consumers for their files. [Dkt. 1141 (Def.’s Unredacted 56(a)1 Statement) ¶ 36]. It maintains written policies and
procedures for the ways in which consumers making file disclosures must be
authenticated. Id. ¶37. Those authentication procedures generally require
consumers to provide their personal identifying information, government
documentation, and/or answers to a series of personal security questions. Id.
In addition to regularly disclosing consumer files directly to the requesting
consumer, RPS processes the disclosure of consumer files to third-party legal
guardians acting on the consumer’s behalf. Id. ¶ 38. To protect consumer privacy
in the situation where a third party is seeking a copy of a consumer’s file, RPS’s
written policies generally require a notarized power of attorney, the consumer’s
name, proof of the address to where the disclosure should be mailed, and
confirmation of the last four digits of the consumer’s Social Security number. Id.
(citing [Dkt. 114-6 (Barnard Decl.) ¶ 8 & Ex. A at pp. 3]). Based on RPS’s written
authentication policy, in “any scenario” where those requirements cannot be
fulfilled, the RPS employee who is handling the file disclosure request must
escalate the request to a “supervisor.” [Dkt. 114-1 ¶39]. A situation in which a
consumer is disabled and cannot execute a power of attorney requires adjustment
of third-party authentication process and supervisory review. Id. (citing [Dkt. 114-6
at ¶ 13 & Ex. A at 3]).
Carmen Arroyo first contacted RPS on April 27, 2016 to request Mikhail
Arroyo’s consumer file. [Dkt. 114-1 ¶40]. RPS informed Ms. Arroyo of the process
17

 for obtaining the file in a third-party capacity, which required her to submit a
disclosure request form and certain documentation. Id. at ¶ 41. Three days later,
on April 29, 2016, RPS mailed Ms. Arroyo a consumer disclosure request form and
instructions. Id. ¶ 42. The form asked for Mr. Arroyo’s name, date of birth, social
security number or Tax Identification Number, phone number, current address, and
signature. [Dkt. 125-10 (Pls.’ Opp. to Def.’s Mot. Summ. J. Ex. 8: 6/14/16 Arroyo File
Disclosure) at 2].
Ms. Arroyo signed the form as Mr. Arroyo’s mother and mailed the first
consumer disclosure request form (the “First Disclosure Request”) to RPS on June
14, 2016. Ibid. RPS received it on June 27, 2016. Ibid. The First Disclosure Request
did not list Mikhail Arroyo’s Social Security number, it did not contain his complete
previous address information, and it was signed just below the line designated.
Ibid. With the First Disclosure Request, Ms. Arroyo submitted a copy of Mr.
Arroyo’s Pennsylvania driver’s license, a copy of her driver’s license, and a copy
of a certificate of conservatorship. Ibid. The certificate of conservatorship states
on its face that it is “NOT VALID WITHOUT COURT OR PROBATE SEAL
IMPRESSED.” [Dkt. 125-10 at 3]. It contains the name of the probate court, case
number, and case caption. Ibid. At the lower left hand corner, there is a circular
area that appears shaded or scratched out by a pen or pencil, through which one
can see the outlines of what appear to be letters, and in the center of which are the
typewritten words “Court Seal.” Ibid. The First Disclosure Request was escalated
to two supervisors. [Dkt. 114-1 ¶46]. The supervisors informed their employees that
they could not accept the “conservatorship court paper,” and that a power of

18

 attorney and Mr. Arroyo’s signature were needed. [Dkt. 112-8 (Ex. 9 to Def.’s Mot.
Summ. J., Barnard Decl. at Ex. B: 6/30/2016 Consumer Relation Remarks) at 11].
On June 30, 2016, RPS mailed a letter to Mr. Arroyo at the nursing home
address listed as his current address, asking him to contact RPS, which would
“accurately provide information [he] need[ed].” [Dkt. 114-1 ¶46]; see [Dkt. 114-6
(Barnard Decl.) at Ex. D]. The letter was returned as undeliverable. Id. The letter did
not mention any deficiencies in the form or conservatorship appointment and did
not inform Mr. Arroyo or Ms. Arroyo that she could submit a corrected form and
conservatorship appointment certificate with a legible raised seal. Id. Although Ms.
Arroyo’s address was listed on the copy of the certificate of conservatorship, RPS
did not send any mail to her or notify her that she needed to provide additional
information to complete Mr. Arroyo’s consumer disclosure request. [Dkt. 125-10];
[Dkt. 114-6 at Ex. D].
Three months later, on September 7, 2016, Ms. Arroyo called RPS to discuss
the status of the disclosure. [Dkt. 114-1 ¶47]. During that call, she was instructed
that she had to provide a notarized power of attorney. [Dkt. 114-6 (Barnard Decl.)
at Ex. B at 3]. RPS did not inform her of the other deficiencies in the application or
the critical absence of a legible raised seal on the Probate court conservatorship
appointment.
RPS’s next contact with Ms. Arroyo did not occur until November 1, 2016,
when she called to ask why RPS had not yet provided her with Mikhail Arroyo’s
consumer file. [Dkt. 118-1 ¶48]. RPS escalated the matter to its consumer relations
department, its compliance department, and then its internal legal department and

19

 outside attorneys, while staying in contact with Ms. Arroyo. Id. ¶¶ 49, 50. On
November 14, 2016, RPS asked Ms. Arroyo send RPS a new certificate of
conservatorship with the court seal visible, as well as proof of current address
documentation. Id. ¶50.
On November 15, 2016, Ms. Arroyo faxed additional documentation and a
new consumer disclosure form (the “Second Disclosure Request”) to RPS. Id. at
¶51. The Second Disclosure Request included Mr. Arroyo’s social security number,
and was signed by both Ms. Arroyo, who identified herself as Mr. Arroyo’s mother
and co-conservator, and by Tod Stimpson, who identified himself as Mr. Arroyo’s
co-conservator. [Dkt. 125-12 at 4]. Ms. Arroyo also attached an updated copy of the
conservatorship certificate which also stated it was invalid without a seal. Id. at 5.
At the lower left-hand corner, there is a faint circle of stray marks, more darkly
shadowed at the bottom, and in the center of which are the typewritten words
“Court Seal.” Ibid. No letters can be made out. Ibid.
The Second Disclosure Request and supporting documentation was
escalated to RPS’s compliance and legal departments, including consultation with
outside counsel. Id. ¶52. RPS found the documentation insufficient because the
conservatorship appointment did not have a visible court seal, Ms. Arroyo signed
her name instead of Mr. Arroyo’s, and there was no proof of address. [Dkt. 114-6 at
Ex. F at 1-2 (RPS Emails Regarding Request)]. On November 16 and November 18,
2016, RPS attempted to contact Ms. Arroyo by telephone to discuss the
documentation she submitted with the Second Disclosure Request. Id. at ¶53. Ms.

20

 Arroyo did not return these calls. Id. There is no evidence RPS sent Ms. Arroyo a
letter explaining the deficiencies.
In mid-December 2016, a paralegal at the CFHC contacted RPS and stated
the CFHC was assisting Ms. Arroyo in the file disclosure process. Id. ¶ 54. RPS
requested additional documentation from the CFHC in the form of a power of
attorney to establish that the CFHC was formally representing Ms. Arroyo. Id. RPS
then mailed and emailed the CFHC a consumer disclosure request form and
instructions on the documents the CFHC should submit. Id. RPS did not receive
any documentation from the CFHC in response. Id. RPS had no further contact with
CFHC or Ms. Arroyo until this suit was filed. Id. at ¶55.
Plaintiffs are not aware of any other conserved individual who has requested
a file disclosure from RPS, and RPS also has no record of any other conserved
individual requesting a file disclosure. [Dkt. 114 ¶45], [Dkt. 126-3 (Barnard Dep.)
120:15-124:24].
G. CFHC Involvement
Plaintiff CFHC is a nonprofit corporation incorporated in Connecticut. [Dkt.
1 (Compl.) ¶28]. Its mission is “eliminating housing discrimination and ensuring
that all people have equal access to housing of their choice.” Id. ¶¶ 19, 185-87, 192.
Ms. Arroyo retained the CFHC in 2017 to bring an administrative complaint
against WinnResidential for failing to reasonably accommodate Mr. Arroyo’s
disability by refusing to admit him. [Dkt. 114-1 ¶32]. WinnResidential appeared at

21

 an initial administrative fact-finding hearing on June 13, 2017, and a settlement was
reached following the hearing. Id. ¶ 34.
The

settlement

agreement

included

no

provision

requiring

that

WinnResidential allow Mr. Arroyo to move into the apartment. [Dkt. 125-9 (Ex. 7:
Conciliation Agreement)]. However, Mr. Arroyo moved into the apartment after the
hearing. [Dkt. 116-35 at Ex. A at 16-28, 68-71, 52].
CFHC received $13,00 in connection with that settlement as attorneys’ fees
for CFHC’s representation of the Arroyos, work that has not been claimed as
diversion damages. [Dkt. 125-13 (Ex. 11 to Pl.’s Opp, Kemple Decl.) ¶¶ 3-5]. CFHC
separately, from other sources, received grants totaling $380,000 to address
criminal record tenant screening in the housing application process. [Dkt. 114-1 ¶
63].
II.

Legal Standard
Summary judgment “shall be granted” if, construing the evidence in the light

most favorable to the non-movant, “there is no genuine dispute as to any material
fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a).
An issue is genuine if “the evidence is such that a reasonable jury could return a
verdict for the nonmoving party.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248
(1986). A fact is “material” if it “might affect the outcome of the suit under the
governing law.” Ibid.
“In ruling on a motion for summary judgment, ‘the evidence of the
nonmovant is to be believed, and all justifiable inferences are to be drawn in his
22

 favor.’” Tolan v. Cotton, 572 U.S. 650, 651 (2014) (quoting Anderson, 477 U.S. at
255)). “Credibility determinations, the weighing of evidence, and the drawing of
legitimate inferences form the facts are jury functions, not those of a judge.”
Anderson, 477 U.S. at 255. Put another way, “[i]f there is any evidence in the record
that could reasonably support a jury’s verdict for the nonmoving party, summary
judgment must be denied.” Am. Home Assurance Co. v. Hapag Lloyd Container
Line, GmbH, 446 F.3d 313, 315-16 (2d Cir. 2006) (internal citation and quotation
omitted).
A party asserting that a fact cannot be or is genuinely disputed must support
the assertion by:
(A) citing to particular parts of materials in the record, including
depositions, documents, electronically stored information,
affidavits or declarations, stipulations (including those made for
purposes of the motion only), admissions, interrogatory answers,
or other materials; or
(B) showing that the materials cited do not establish the absence
or presence of a genuine dispute, or that an adverse party cannot
produce admissible evidence to support the fact.
Fed. R. Civ. P. 56(c)(1). A party asserting that a fact is or is not true must present
admissible evidence to support their assertion. Fed. R. Civ. P. 56(c)(2).
Where the movant presents admissible evidence tending to show there is no
genuine issue of material fact for a jury to decide and she is entitled to judgment
as a matter of law, “the burden shifts to the nonmovant to point to record evidence
creating a genuine issue of material fact.” Salahuddin v. Goord, 467 F.3d 263, 273
(2d Cir. 2006). Fed. R. Civ. P. 56(c), (e). Rule 56(c) “mandates the entry of summary
judgment… against a party who fails to make a showing sufficient to establish the
existence of an element essential to a party’s case, and on which that party will
23

 bear the burden of proof at trial.” Bedor v. Friendly’s Ice Cream Corp., 392 F. Supp.
2d 367, 373 (2005) (quoting Celotex Corp v. Catrett, 477 U.S. 317, 322 (1986)).
III.

Analysis
RPS moves for summary judgment on all counts of the complaint: the FHA

disparate impact claim on the basis of race and national origin; the FHA
discriminatory treatment claim on the basis of race and national origin; the FCRA
claims; the FHA disparate impact claim on the basis of disability;

the FHA

discriminatory treatment claim on the basis of disability; the CUTPA claims on the
basis of CrimSAFE and on the basis of file disclosure; and CFHC’s claim for
compensatory damages. It further argues that Ms. Arroyo does not have individual
standing. Plaintiffs respond to RPS’s motion for summary judgment, and also
themselves move for partial summary judgment on their FHA disparate impact
claim on the basis of race and national origin and their CUTPA claim on the basis
of CrimSAFE. They also separately move for partial summary judgment on the
FCRA claims, the FHA disparate impact claim on the basis of disability, and the
CUTPA claim on the basis of file disclosure. The Court analyzes each claim in turn
but starts with standing.
A. Standing
To establish constitutional standing, “the plaintiff must show an ‘injury in
fact’ that is ‘fairly traceable’ to the defendant’s conduct and ‘that is likely to be
redressed by a favorable judicial decision.’” Bank of Am. Corp. v. City of Miami,
Fla., 137 S. Ct. 1296, 1302 (2017) (quoting Spokeo, Inc. v. Robins, 578 U.S. ––––, ––

24

 ––, 136 S.Ct. 1540, 1547 (2016)). A plaintiff must also satisfy “statutory standing,”
that is, demonstrate that her “interests fall within the zone of interests protected
by the law invoked.” Lexmark Int'l, Inc. v. Static Control Components, Inc., 572 U.S.
––––, ––––, 134 S.Ct. 1377, 1387-88 and n.4 (2017).
1. Carmen Arroyo’s Standing for Fair Housing Act (FHA) Claim
RPS argues that Carmen Arroyo does not meet either requirement for
standing for her FHA claim: she does not experience “injury in fact” required for
standing under any of FHA claims because she was not denied housing at any
point; and she does not have statutory standing because lack of companionship
and emotional distress are outside the zone of interests protected by the FHA. [Dkt.
114 at 44-45].
To allege constitutional injury-in-fact, a plaintiff must allege “an invasion of
a legally protected interest which is (a) concrete and particularized; and (b) actual
or imminent, not conjectural or hypothetical.” Citizens for Resp. & Ethics in Wash.
v. Trump, 953 F.3d 178, 189 (2d Cir. 2019), as amended (Mar. 20, 2020) (quoting
Lujan v. Def.’s of Wildlife, 504 U.S., 555, 560 (1992)).
Statutory standing under the FHA is “as broad[] as is permitted by Article III
of the Constitution.” Trafficante v. Metro. Life. Ins. Co., 409 U.S. 205, 209 (1972)
(citations omitted). The FHA allows any “aggrieved person” to bring a housingdiscrimination lawsuit. 42 U.S.C. § 3613(a). An “aggrieved person” includes “any
person who claims to have been injured by a discriminatory housing practice…”
42 U.S.C. § 3602(i)(1); see 24 C.F.R. 100.65 (Discriminatory housing practices

25

 include conduct that “limits the use of privileges, services, or facilities associated
with a dwelling because of race color…. of an owner, tenant or a person associated
with him or her”); 42 U.S.C. § 3604(f)(1)-(2) (prohibiting discrimination in the sale or
rental, or in terms, conditions, or privileges of sale or rental, “because of a
handicap of… (b) a person intending to reside in that dwelling after it is… rented
or made available; or (C) any person associated with a that buyer or renter.”). The
FHA provides a cause of action for individuals in a housing complex who are not
themselves denied housing on the basis of discrimination, but who allege
deprivation of relationships with individuals who are: it “allows suits by white
tenants claiming that they were deprived benefits from interracial associations
when discriminatory rental practices kept minorities out of their apartment
complex.” Bank of Am. Corp. v. City of Miami, Fla., 137 S. Ct. at 1303 (citing
Trafficante at 209-212).
Here, Ms. Arroyo alleges that RPS’s discriminatory housing practices
segregated her from her son on the basis of ethnicity and ability, and she claims
emotional injuries on that basis. Ms. Arroyo’s claims are similar to, and even more
compelling than, the plaintiffs’ claims in Trafficante, 409 U.S. 205. In Trafficante,
the Court found that two tenants, one who was white and one who was African
American, had constitutional and statutory standing to pursue FHA claims alleging
their landlord racially discriminated against rental applicants, depriving tenants of
the social and professional benefits of living in an integrated community. Id. at 20612. Like Ms. Arroyo, the aggrieved tenants were already living in the complex. Id.
at 206. Like Ms. Arroyo, the tenants’ grievance was based on their interest in an

26

 integrated community. Id. at 209. The Trafficante Court held protected the mutual
benefit of diverse associations and acquiring multi-cultural competency. Id.
Here, not only would Ms. Arroyo and her son benefit from mutual
association, Ms. Arroyo was also deprived of familial association, highly valued in
our society. She was also deprived of the ability to fully and effectively discharge
her maternal and legal duty as conservator of Mr. Arroyo to protect both his person
and interests as ordered by the Probate Court by having him live with her rather
than in a nursing home. These are concrete, particularized, and actual injuries
within the scope of interests protected by the FHA.
RPS cites two cases in opposition, but neither is apposite. First, in its
motion, RPS quotes Vaughn v. Consumer Home Mortgage Co., 297 F. App'x 23, 26
(2d Cir. 2008) for the proposition that FHA standing extends “only to those persons
who are ‘personally denied equal treatment by the challenged discriminatory
conduct.’” Id. at 26, cited by [Dkt. 114 at 45]. But this quotation misrepresents the
sentence: the sentence qualifies that it is only speaking to one “line of authority,”
among others, and that line of authority concerns “government-erected barriers,”
which are not issue here. Id. at 26. Second, in its reply, RPS quotes Wartluft v.
Milton Hershey School & School Trust, 400 F. Supp. 3d 91, 102-103 (M.D. Pa. 2019)
for the proposition that “a ‘[loss] of companionship’ and related emotional distress
damages as falling outside of the zone of interests protected by the FHA.” [Dkt. 140
at 18]. But in Wartluft the plaintiffs were parents bringing an FHA claim on the basis
that a school’s discriminatory expulsion of their daughter led to her suicide; they

27

 themselves did not live at the allegedly discriminatory residence or wish to live
there. Therefore, it also is not persuasive here.
The Court finds that Ms. Arroyo’s claim falls squarely within the zone of
interests protected by the FHA, and that she has standing.
2. Carmen Arroyo’s Standing for Connecticut Unfair Trade Practices Act
(CUTPA)
RPS also contends Ms. Arroyo does not have standing to pursue her CUTPA
claim. To sustain a claim under CUTPA a person must suffer an ascertainable loss
of money or property. Conn. Gen. Stat. § 42-110g. RPS argues that, as a matter of
undisputed fact, Ms. Arroyo did not suffer an adverse housing decision, and cannot
point to an ascertainable loss. In response, Plaintiffs argue that RPS’s actions
delayed Mr. Arroyo’s admission to ArtSpace Windham by about a year through
RPS’s CrimSAFE report regarding Mr. Arroyo and its misinforming Ms. Arroyo
about the documentation needed to obtain Mr. Arroyo’s CrimSAFE file. During this
year, Plaintiffs allege, Mr. Arroyo remained in a nursing home, Ms. Arroyo and Mr.
Arroyo had additional medical, travel, and housing expenses, and Ms. Arroyo had
higher housing expenses without Mr. Arroyo’s housing subsidy. See [Dkt. 1
(Compl.) at ¶102]. The Court finds that Ms. Arroyo’s claimed associational
deprivation and financial and emotional injuries are sufficient to establish her
statutory and constitutional standing to bring the CUTPA claim.
3. Standing for claims on behalf of African American rental applicants
The Arroyos are not African-American and, unlike the white plaintiff in
Trafficante, do not claim to be injured themselves by RPS’s alleged discrimination
28

 against African-American applicants, so they do not have statutory standing to
allege claims on the basis of discrimination against African Americans. Cf.
Trafficante v. Metro. Life Ins. Co., 409 U.S. 205, 208 (1972). But, in light of CFHC’s
inclusion as a Plaintiff, the Court holds that it is still appropriate to consider
Plaintiffs’ FHA and CUTPA claims on the basis of discrimination against African
Americans. See Rumsfeld v. Forum for Acad. & Institutional Rights, Inc., 547 U.S.
47, 53 (2006) (“The presence of one party with standing is sufficient to satisfy
Article III's case-or-controversy requirement.”).
Organizations who allege that a defendant’s actions have “frustrated the
organization [plaintiff]’s… services, with a consequent drain on resources” have
standing to bring FHA claims. Havens Realty Corp. v. Coleman, 455 U.S. 363, 369
(1982); Ragin v. Harry Macklowe Real Estate, 6 F.3d 898 (2d Cir. 1993) (reversing
finding that nonprofit agency lacked standing to bring discrimination claim against
real estate advertisers). On this basis, an organization may bring claims on behalf
of individuals not otherwise represented in the action. For example, in Saint-Jean
v. Emigrant Mortg. Co., where the plaintiffs were African American, the court held
that the fact that jury instructions mentioned alleged housing discrimination
against Hispanic communities was not a basis for retrial. 337 F. Supp. 3d 186, 198
n.4 (E.D.N.Y. 2018) (citing Ragin, 6 F.3 898); see also Veasey v. Perry, 29 F. Supp.
3d 896, 903-04 (S.D. Tex. 2014) (holding that organizations had standing to
challenge voter identification law as racially discriminatory against AfricanAmericans and Latinos based on pleaded mission statements including “to

29

 empower young people, particularly those of color,” and to “serv[e] as an advocacy
group for the working poor”).
In this case, Plaintiffs allege the disparate impact of CrimSAFE on African
American and Latino applicants frustrates CFHC’s mission is ensuring that all
people have equal access to the housing of the choice. [Dkt. 1 ¶19]. As to resource
drain, there is evidence that housing providers have reached out to CFHC for
guidance on the use of criminal records, [Dkt. 125-21 (Ex. 19: Kemple Dep.) at 98:1924], and that CFHC has changed its public trainings and presentations to account
for RPS’s policies regarding criminal records, id. at 97-111. Therefore, the Court
finds that CFHC has standing to bring FHA and CUTPA claims based on ethnicity
and race discrimination against both Latinos and African Americans.
B. FHA Disparate Impact Claim on the Basis of Race or Ethnicity
The FHA prohibits a person or entity from “mak[ing] unavailable or deny[ing]
a dwelling to any person because of race or national origin.” 42 U.S.C. § 3604(a).
Section 3604(b) prohibits discrimination “in the terms, conditions, or privileges” of
a rental. “Otherwise make[s] unavailable” is a “catchall phrase” that “look[s] to
consequences, not intent.” Tex. Dep't of Hous. & Cmty. Affairs v. Inclusive
Communities Project, Inc., 135 S. Ct. 2507, 2519 (2015).
Disparate impact claims are cognizable under the FHA. Id. at 2525. 24 C.F.R.
§ 100.500 sets forth the regulation of the Department of Housing and Urban
Development (“HUD”) on discriminatory effects, which the Second Circuit has
adopted to analyze FHA disparate impact claims:

30

 First, a plaintiff… must come forward with a prima facie case; and
second, the defendant… may rebut the prima facie case by proving
that the ‘challenged practice is necessary to achieve one or more
substantial, legitimate, nondiscriminatory interests of the respondent
or defendant.’ [Third], the burden of proof shifts back to the plaintiff to
show that the ‘substantial, legitimate, nondiscriminatory interests
supporting the challenged practice could be served by another
practice that has a less discriminatory effect.’
See Mhany Mgmt., Inc. v. Cty. of Nassau, 819 F.3d 581, 617 (2d Cir. 2016) (appeal
following a bench trial) (quoting 24 C.F.R. § 100.500(c)(3)); see Inclusive
Communities Project, 135 S. Ct. at 2514-15; 2522-23.
To establish a prima facie case, a plaintiff must show “that a challenged
practice caused or predictably will cause a discriminatory effect.” 24 C.F.R. §
100.500(c)(1). This standard has three elements: (i) “certain outwardly neutral
practices,” Mhany, 819 F.3d at 617; (ii) “a significantly adverse or disproportionate
impact on persons of a particular type, ibid.; and (iii) “a causal connection between
the facially neutral policy and the allegedly discriminatory effect,” Tsombanidis v.
W. Haven Fire Dep't, 352 F.3d 565, 575 (2d Cir. 2003) (bench trial). Both RPS and
Plaintiffs move for summary judgment on Plaintiffs’ FHA disparate impact
discrimination claim. The Court addresses first addresses their arguments as to
causal connection, and next addresses their arguments about disparate impact.
1. Prima Facie Case
i. Proximate Cause
In their motion for summary judgment, Plaintiffs argue that they have
established undisputed facts that show RPS was a proximate cause of housing
availability on the basis of race or ethnicity. [Dkt. 118 at 9-18]. RPS rejects this

31

 claim, arguing that it was not a proximate cause, is not an agent of WinnResidential,
and CrimSAFE is not a “decisioning” product. [Dkt. 129 at 24-33].
“As the agency charged with enforcement of the FHA, HUD's construction of
the statute ‘is entitled to great weight.’” Viens v. Am. Empire Surplus Lines Ins.
Co., 113 F. Supp. 3d 555, 567 n.11 (D. Conn. 2015) (quoting Trafficante, 409 U.S. at
210). HUD regulations provide:
It shall be unlawful, because of race,… handicap,… or national origin,
to engage in any conduct relating to the provision of housing or of
services and facilities in connection therewith that otherwise makes
unavailable or denies dwellings to persons.
24 C.F.R. § 100.70(b) (emphasis added).
A defendant makes housing unavailable “when [it] engages in a series of
actions that imposes burdens on… a protected class of residents or intended
residents, making it more difficult for the members of the protected class to obtain
housing or conveying a sense that the members of the protected class are
unwanted.” Gilead Cmty. Servs., Inc. v. Town of Cromwell, No. 3:17-CV-627 (VAB),
2019 WL 7037795, at *20 (D. Conn. Dec. 20, 2019). “[P]roximate cause under the
FHA requires ‘some direct relation between the injury asserted and the injurious
conduct alleged’;” “foreseeability” alone is not sufficient. Bank of Am. Corp. v. City
of Miami, Fla., 137 S. Ct. 1296, 1306 (2017) (quoting Holmes v. Securities Investor
Protection Corporation, 503 U.S. 258 (1992)). “A link that is too remote, purely
contingent, or indirec[t] is insufficient.” Empire Merchs., LLC v. Reliable Churchill
LLP, 902 F.3d 132, 141 (2d Cir. 2018) (quoting Hemi Grp., LLC v. City of New York,
559 U.S. 1, 9 (2010)) .

32

 Discrimination may have “multiple proximate causes,” and the possibility
that one decisionmaker may be overridden by a higher decisionmaker does not
“automatically render the link to the [subordinate’s] bias ‘remote’ or ‘purely
contingent’” for proximate cause purposes, especially where the ultimate
decisionmaker’s judgment is neither “independent” nor unforeseeable. Staub v.
Proctor Hosp., 562 U.S. 411, 419 (2011). In Staub, a Uniformed Services
Employment and Reemployment Rights Act case, the Supreme Court held that for
the purposes of showing illegal antimilitary bias, a biased supervisor’s unfavorable
report could be a proximate cause for the plaintiff’s ultimate discharge, even
though supervisor did not make ultimate decision. Id. “[T]he supervisor's biased
report may remain a causal factor if the [decisionmaker's] independent
investigation takes it into account without determining that the adverse action was,
apart from the [biased] supervisor's recommendation, entirely justified.” Id. at 421,
cited by Vasquez v. Empress Ambulance Serv., Inc., 835 F.3d 267, 274-75 (2d Cir.
2016) (recognizing that co-worker’s statements may proximately cause the plaintiff
to be fired).
An agency relationship is neither necessary nor sufficient to show proximate
cause. Both Staub and Vasquez address whether an employer may be held liable
for an employee’s discriminatory actions and animus, and conclude, on the basis
of agency principles, that an employer can be held liable for an employee’s
discriminatory motives. Staub, 562 U.S. at 419-21; Vasquez, 835 F.3d at 274-75. But
both treat the question of proximate causation independently from the agency

33

 analysis of motive attribution. Staub, 562 U.S. at 419-21; Vasquez, 835 F.3d at 27475.
Ultimately, “issues of proximate causation… involve application of law
to fact, which is left to the factfinder.” Exxon Co., U.S.A. v. Sofec, Inc., 517 U.S. 830,
840–41 (1996). Proximate cause becomes a question for a court only where
“…there are active and efficient intervening causes, or where reasonable
[factfinders]

could

reach

only

one

conclusion

regarding

the

issue

of proximate cause.” Margrave v. British Airways, 643 F. Supp. 510, 513 (S.D.N.Y.
1986). Whether there are multiple reasonable conclusions is a separate question
from whether there are disputed issues of fact. See Tolan v. Cotton, 572 U.S. 650,
651 (2014); Mosheuvel v. D.C., 191 U.S. 247, 252 (1903).
The Court finds that a reasonable factfinder could find RPS one proximate
cause of housing unavailability but need not necessarily do so; and leaves the
question of whether that unavailability is on a discriminatory basis for the following
sections. RPS markets CrimSAFE as “rendering a decision on an applicant’s
suitability for tenancy based on their criminal history.” [Dkt. 118-1 ¶¶5, 6, 9, 33].
CrimSAFE has many features which demonstrate the truth of the marketing
promise: It informs a housing provider whether there is a “record found” that would
be disqualifying under the criteria set by the client. [114-1 ¶8]. It allows clients to
disqualify innocent people who have been charged but not convicted, even where
the charges are non-criminal infractions. [Dkt. 118-1 ¶¶51-52]. It also allows clients
to suppress the remainder of the full reports from the view of their onsite leasing
staff, as WinnResidential did, so that onsite leasing staff see only the “records
34

 found” notification.” [Dkt. 118-1 ¶¶ 26, 36]. It chose to use the criminal categories
created by the FBI’s national incident-based reporting system as those available
as CrimSAFE configuration criteria. [Dkt. 129-1 ¶8]. It establishes the maximum
lookback periods of 7 years for non-convictions. Id. ¶ 11.
With its CrimSAFE product, RPS CoreLogic always returns information
about whether disqualifying records are or are not found but does not always the
full criminal record to the staff of the housing provider. If it always returned the
criminal record, the staff of the housing provider might consider only the fact of
the existence of a criminal record, but they would also be able to investigate it
themselves, and weigh the criminal record in light of everything else they knew
about a particular applicant – whether the charge was a conviction or a pending
arrest or a dismissal, whether it was a charge for first-degree assault or shoplifting
or a traffic accident involving damage, and whether the applicant provided any
other information in their application that would mitigate the particular records
found.
Instead, CrimSAFE always “categorizes” records: first into types of
offenses, and then into whether the record is disqualifying or not according to the
housing provider’s choices from CrimSAFE’s menu. See [Dkt. 114-1 ¶¶5, 6, 7, 8].
Importantly, categories do not just separate out offenses by distinguishing
characteristic, they also combine them, and, in the combination, CrimSAFE
reduces housing providers’ discretion. So, for instance, at the first level of
categorization: in the “Destruction Damage Vandalism of Property” category,
CoreLogic includes “traffic accidents involving damage,” which even CoreLogic’s

35

 expert concedes has no relationship to suitability for tenancy. [Dkt. 136-1 (Pl.’s
Reply 56(a)1 Statement) ¶103]. The same category also includes vandalism and
property damage, id., so a housing provider cannot exclude vandals without also
excluding people involved in traffic accidents. At the second level of
categorization, CoreLogic transforms the criminal records review process into a
yes/no switch, eliminating the possibility staff may be able to weigh a dismissed
arrest for theft differently than a pending charge for disorderly conduct differently
than a conviction for assault.
By allowing clients to elect to suppress the full criminal record information,
RPS allows clients to disable their staff from fully assessing the suitability of a
tenant applicant and enabled its clients to deny housing to individuals whose
records did not suggest they posed any risk to the property of its occupants. It
continued this practice even after HUD issued guidance that arrest records were
not a proper basis to disqualify a tenant although it could have readily informed
clients to alter their search parameters pending its review of its produce. And it is
no surprise that it did so: what distinguishes CrimSAFE from other RPS screening
products, its unique value-proposition, is the fact that it categorizes records for the
housing providers and simplifies decision-making. Were RPS clients to never use
CrimSAFE’s “record(s) found” message as a basis for a decision, CrimSAFE
logically could not provide RPS’s claimed benefits:

speeding up background

screenings and ensuring that housing providers’ employers were adhering to the
community standard. [Dkt. 114-1 ¶¶ 13-14].

36

 RPS argues in response that it is not a direct cause of any housing
discrimination because its housing manager customers set the criteria for deciding
which criminal records should result in rejection and determine whether to
suppress the full reports from onsite housing staff. [Dkt. 129 at 24-32]. RPS further
argues that its housing manager clients can override RPS’s recommendation of
denial based on individualized review, and that, when doing so, they can take
account of the underlying criminal record because the reports are always available
to someone at the client housing provider. [Dkt. 129 at 24-32]; [Dkt. 114 at 27].
The fact that RPS’s statement of “record(s) found” may be overridden by its
client does not eliminate its responsibility—discrimination may have multiple
causes and parties other than final decisionmakers may be liable. See Staub v.
Proctor Hosp., 562 U.S. 411, 419 (2011); Vasquez, 835 F.3d at 274-75. Therefore,
the Court finds that a reasonable factfinder could find RPS a proximate cause of
housing unavailability, but would not necessarily do so.
The parties’ arguments do not compel another decision. The only decision
Plaintiffs cite that is specific to proximate cause holds only that a fact-finder could
reasonably find proximate cause where there is another decisionmaker, not that
such a finding is compelled or that the opposite is unreasonable. See [Dkt. 136 at
5] (citing Staub v. Proctor Hosp., 562 U.S. 411, 419 (2011)) (reversing and remanding
Seventh Circuit court holding that defendant was entitled to judgment as a matter
of law, itself a reversal of the jury verdict).
The three cases RPS cites also do not compel a grant of summary judgment.
RPS first cites Zabriskie v. Federal National Mortgage Association, 912 F.3d 1192

37

 (9th Cir. 2019), in which the Ninth Circuit considered whether an entity that
provided software that “automatically applies [underwriting] guidelines and
requirements” to consumer credit information input by a potential lender to assess
“a loan’s eligibility for purchase” was a consumer reporting agency such that it
could be held liable under the FCRA. Id. at 1195. In finding that the defendant could
not be held liable because it did not “evaluate” applications, the Ninth Circuit stated
the “commonsense principle” that “when a person uses a tool to perform an act,
the person is engaging in the act; the tool’s maker is not.” Id. But, the opinion was
amended, and the amended opinion omitted this reasoning, and instead based its
conclusion on the fact that the information gathered was used for a different
purpose than making consumer reports, a conclusion irrelevant to the instant
issue. Zabriskie v. Fannie Mae, 940 F.3d 1022, 1025 (9th Cir. 2019). Further, unlike
the instant case, the question was not of proximate cause but instead of the scope
of the FCRA. 912 F. 3d. at 1192.
Next, RPS cites National Fair Housing Alliance v. Deutsche Bank, No.
18CV0839, 2018 U.S. Dist. LEXIS 196636 (N.D. Il. Nov. 19, 2018), in which the court
dismissed an

FHA claim that “Defendants’ [maintenance] delegation practice

resulted in poorly-executed property maintenance, which led to racially-disparate
effects,” finding that the claim alleged “chain-link causation” with “intermediate
steps” that the FHA does not permit. Id. at *38-39. But when considering a second
motion to dismiss the amended complaint, which alleged the same FHA claims as
those dismissed in the original complaint, the court found the plaintiffs had
sufficiently alleged proximate cause, noting that intervening Circuit decisions had

38

 rejected its previous “method of counting ‘steps’ between an action and an injury”
as too prone to manipulation by the counter. Nat'l Fair Hous. All. v. Deutsche Bank
Nat'l Tr., No. 18 CV 839, 2019 WL 5963633, at *5 (N.D. Ill. Nov. 13, 2019) (citing
Kemper v. Deutsche Bank AG, 911 F.3d 383 (7th Cir. 2018) and City of Miami v.
Wells Fargo & Co., 923 F.3d 1260 (11th Cir. 2019), cert. granted, judgment vacated
as moot sub nom. Wells Fargo & Co. v. City of Miami, Fla., 140 S. Ct. 1259 (2020))).
For the same reason, the Court finds the reasoning cited by RPS
unpersuasive. RPS and WinnResidential acted hand-in-glove to deny Mr. Arroyo
housing. RPS allowed screening on the basis of charges that did not lead to a
conviction and allowed its customer to conceal from its line staff the basis for an
“unqualified” classification. In so doing RPS was an integral participant in the
denial of housing by WinnResidential to persons charged with an offense even
though the charges were dismissed. Parties cannot escape liability by sharing
decision making and shielding one another because no single entity is wholly
responsible.
Third, RPS cites Republic of Iraq v. ABB AG, 920 F. Supp. 2d 517 (S.D.N.Y.
2013), a case in which the court considered a RICO claim against a bank that had
set up and serviced an escrow account, facts distant from the instant ones. The
plaintiff alleged that money was transferred out of the escrow account by a
fraudulent actor, the U.N. Id. In rejecting the claim for lack of proximate cause, the
court held that “because [the bank] BNP released and accepted funds into the U.N.
escrow account at the UN’s direction, however, BNP’s servicing of that account
cannot have been the proximate cause of Iraq’s injury. . . . Contingent relationships
39

 of this sort are too indirect to support [any] recovery. ” Id. at 550. But this case is
not analogous. RPS did not perform a ministerial or administrative function like
BNP; WinnResidential relied on RPS’s expertise, and WinnResidential’s options
were determined by RPS. RPS’s CrimSAFE product is unique, so its relationship
with its customer is not “contingent” in the way the bank’s was, and RPS
determined the framework of the criteria which its customers could use, so it was
not simply following a customer’s instructions.
ii. Statistical Showing of Causation of Disparate Impact
Plaintiffs and RPS both assert that there is no genuine issue of material fact
as to whether Plaintiffs have shown disparate impact. Plaintiffs assert that they
have. [Dkt. 118 at 18-24]. RPS asserts that Plaintiffs have not made the required
statistical showing and cannot do so. [Dkt.114 at 27-34.]. After considering the law
and the facts presented, the Court finds that Plaintiffs have presented sufficient
statistical evidence to put into dispute whether RPS’s practice of reporting housing
applicants’ criminal records to housing providers as potentially disqualifying
records has a disparate impact on African American and Latino people.
Disparate impact liability exists where a discriminatory policy “actually or
predictably results in a disparate impact on a [protected] group…” 24 C.F.R. §
100.500(a) (emphasis added). “A robust causality requirement ensures that ‘[r]acial
imbalance… does not, without more, establish prima facie case of disparate
impact’ and thus protects defendants from being held liable for racial disparities
they did not create.” Texas Dep't of Hous. & Cmty. Affairs v. Inclusive Communities
Project, Inc., 135 S. Ct. 2507, 2523 (2015). “A plaintiff who fails to allege facts… or

40

 produce statistical evidence demonstrating a causal connection cannot make out
a prima facie case of disparate impact.” Id. “A plaintiff has not met its burden if it
merely raises an inference of discriminatory impact.” Tsombanidis, v. W. Haven
Fire Department, 352 F.3d 565, 575 (2d Cir. 2003) (appeal following bench trial). To
make out a prima facie case of disparate impact, “plaintiffs must… utilize the
appropriate comparison groups. They must first identify members of a protected
group that are affected by the neutral policy and then identify similarly situated
persons who are unaffected by the policy.” Tsombanidis, 352 F.3d at 576–77.
National or state general population statistics may be used as the
appropriate comparison groups in at least three situations: First, national or state
statistics are appropriate where “there is no reason to suppose” that the local
characteristics would differ from the national statistics. Dothard v. Rawlinson 433
U.S. 321, 330 (June 27, 1977). 13 In Dothard v. Rawlinson, the Court held that use of
national height and weight statistics was appropriate to find a discriminatory effect
on Alabama women where “there was no reason to suppose that physical height
and weight characteristics of Alabama men and women differ markedly from those

13

Defendants cite Townsend v. Nassau Cty. Med. Ctr., 558 F.2d 117, 119 (2d Cir.
June 30 1977) for the proposition that “a statistic relating only to the general
population, and not to the employment practices of the particular defendant” is not
sufficient to demonstrate that a job prerequisite “operates to exclude” minorities.
Id. at 119-20. But that holding relies on an interpretation of Griggs v. Duke Power
Co., 401 U.S. 424, 431 (1971) that the Supreme Court had rejected just three days
before in Dothard. Compare Dothard, 433 U.S. at 330, with Townsend, 558 F.2d at
120. The case is also distinguishable, see infra.

41

 of the national population.” Id; see U.S. Dept. of Housing and Urban Development,
April 4, 2016 Office of General Counsel Guidance.
Second, “studies based on general population data and potential applicant
pool data” may be the “initial basis of a disparate impact claim, especially in cases
[where] 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 discriminatory.” E.E.O.C. v. Joint Apprenticeship Comm.
of Joint Indus. Bd. of Elec. Indus., 186 F.3d 110, 119 (2d Cir. 1999) (employment
discrimination context). In this situation, the potential applicant pool data may
provide a more accurate depiction of the true discriminatory impact than the actual
applicant pool data, though there is a question of who is reasonably a part of the
potential applicant pool. For instance, in Wards Cove Packing Co. v. Atonio, an
employment discrimination case, the Court held that a “proper comparison” in a
disparate impact case is “between the racial composition of the [at issue jobs] and
the racial composition of the qualified population in the relevant labor market,” so
the general population “cannot be used as a surrogate for the class of qualified job
applicant” for specialized job. 490 U.S. 642, 650-51 (1989). T
Third, national or state general statistics are appropriate where actual
applicant data is not available. Hazelwood Sch. Dist. v. United States. 433 U.S. 299,
308-09 n.13 (1977). In Hazelwood, an employment discrimination case, the Court
held that data on pool of eligible candidates is appropriate to consider where
reliable actual applicant data was not available. Id. Defendants cannot insulate

42

 themselves from disparate impact claims by failing to keep records of the race of
applicants.
The Second Circuit has signaled that it is open to statistics based on the
pool of potential applicants:
[P]laintiffs might have been able to meet their burden by
providing statistical evidence (1) that x% of all of the
[members of protected class] in West Haven need (or
have good reason) to live in the “group settings”
prohibited by the facially neutral fire regulations at issue,
(2) that y% of all of the [similarly situated persons outside
the protected class] in West Haven need (or have good
reason) to live in such group settings prohibited by the
fire regulations, and, crucially, (3) that x is significantly
greater than y.

Tsombanidis, 352 F.3d at 577. In that case, the defendant city and city fire
department set policies for landlords but were not themselves landlords. A postInclusive-Communities Project case confirms the validity of the use of the potential
applicant pool: in Reyes v. Waples Mobile Home Park Ltd. P'ship, 903 F.3d 415, 428
(4th Cir. 2018), cert. denied sub nom. Waples Mobile Home Park Ltd. P'ship v. de
Reyes, 139 S. Ct. 2026, (2019), the Fourth Circuit found that plaintiffs sufficiently
alleged a prima facie case of disparate impact where they pled undocumented
immigrants constitute 36.4% of the Latino population in Virginia compared with
only 3.6% of the non-Latino population, demonstrating that Latinos are ten times
more likely than non-Latinos to be adversely affected by the policy requiring
documentation evidencing legal status.
Here, it is undisputed that actual applicant data is unavailable. RPS does not
receive the race of housing applicants from its clients. And, even if a housing
43

 applicant is matched with a criminal record, the record only sometimes includes
the race of the applicant: RPS has no record of the race or ethnicity of
approximately 19% of the individuals who it reported had criminal records, where
the applicant had a Connecticut address and had applied to rent an apartment. [Dkt.
118-1 ¶84 n. 19].14
In the place of statistics on actual applicants, Plaintiffs present
uncontroverted evidence that, nationally and in Connecticut, and at every income
level, African Americans and Latinos are more likely to be arrested than whites and
are more likely to be incarcerated than whites. [Dkt. 118-1 ¶¶ 72-82].15 Plaintiffs
present uncontroverted evidence that, nationally, African Americans and Latinos
are more likely to be arrested for federal drug crimes than whites, [Dkt. 118-1 ¶71],
and, in Connecticut, African Americans are more likely to be arrested than whites.
Compare Crime in Connecticut 201 at 29, with 2016 ACS 1-Year DP05. The question,
then, is whether the statistics offered by the Plaintiff reflect the experiences and
profiles of the eligible rental applicant pool for RPS’s Connecticut clients.

14

RPS argues that Plaintiffs should have taken discovery of WinnResidential to
gather this information, even if RPS itself did not have it. [Dkt. 129 at 18]. However,
Plaintiffs’ claims are not limited to ArtSpace Windham, to Plaintiff’s knowledge
WinnResidential no longer manages that property, and RPS has not produced the
list of properties using CrimSAFE in Connecticut. See [Dkt. 124 (Order on Mot. to
Compel)]. Therefore, the Court finds that Plaintiffs have not failed in their discovery
obligations.
15 Other courts have questioned whether incarceration data is a sufficient proxy
for conviction data but have not granted summary judgment on the basis of the
argument. See Fortune Soc'y v. Sandcastle Towers Hous. Dev. Fund Corp., 388 F.
Supp. 3d 145, 175-76 (E.D.N.Y. 2019).
44

 The Court finds that this is a disputed question of fact because “although
the decisions of Plaintiff's experts to rely on a broader potential applicant pool
outside of the actual pool of… applicants is reasoned, adequately based in law and
sufficient for the Court to find the testimony admissible, the more tangential nature
of the analysis may diminish the weight a fact-finder would afford the conclusions.”
Fortune Soc'y v. Sandcastle Towers Hous. Dev. Fund Corp., 388 F. Supp. 3d 145,
176 (E.D.N.Y. 2019) (collecting cases and denying summary judgment on FHA
disparate impact claim on the basis of race discrimination where housing provider
allegedly did not accept tenants with criminal convictions). Specifically, the parties
dispute whether the statistics offered by Plaintiffs reflect the reasonable eligible
applicant pool on two bases: that Plaintiffs’ statistics are, at most granular, statewide, and that Plaintiffs’ statistics are not specific to city renters. After considering
the presented evidence, the Court finds that a reasonable fact-finder might find for
either party.
First, the Court finds that there is a disputed question of material fact as to
whether Plaintiffs have presented sufficient evidence that Connecticut is the
market area of RPS’s Connecticut clients so that there is no gap between the
people reflected in the statistics offered by Plaintiffs and the eligible rental
applicant pool for RPS’s Connecticut clients. Connecticut is a small state
consisting of 4,842 square miles of land, with a water surface area of 701 square
miles. See U.S. Census Bureau, Geography Division, TIGERweb Decennial:
Connecticut (Census 2010), available at https://tigerweb.geo.census.gov/tigerweb/.
It is a commutable state serviced by three interstate highways, I-84, I91, and I-95,

45

 and several state highways which traverse the entirety of the state. See
Connecticut Department of Transportation, State Highway System Map (2019),
available at https://portal.ct.gov/DOT/PP_Bureau/Documents/Maps. Since RPS
serves more than 120 properties in the small state of Connecticut, [Dkt. 118-1 ¶
100], the entire state may be within the market area for one or more of the clients
served by RPS. See R.I. Comm'n for Human Rights v. Graul, 120 F. Supp. 3d 110,
125 (D.R.I. 2015) (granting plaintiff summary judgment on FHA disparate liability,
finding statewide rental market for housing complex in middle of small state served
by major highways); Fortune Soc'y, 388 F. Supp. 3d a 169 (denying both parties
summary judgment for 8 county market). Although RPS states “many of [its]
customers operate subsidized/affordable housing communities,” it does not
provide evidence that there are parts of the state which are not in the market area
for at least one of its clients. [Dkt. 114-4 ¶5]. But since “it is not the role of the
district court to make ultimate conclusions as to the persuasiveness of the
proffered evidence,” the Court does not grant the Plaintiffs summary judgment on
this issue. See Deutsch v. Novartis Pharm. Corp., 768 F. Supp. 2d 420, 434 (E.D.N.Y.
2011) (quoting Quiet Tech. v. Hurel–Dubois UK Ltd., 326 F.3d 1333, 1340–41 (11th
Cir.2003)).16

16

There are two ways to collect data on arrests and convictions by race:
bottom-up, by asking individuals whether they have been arrested or convicted or
incarcerated in a certain time period, or top-down, by asking law enforcement
agencies, courts and prisons about the characteristics of who they have arrested,
convicted, and imprisoned in a certain period. Both have limits: it is difficult to
construct and administer a national- or state-representative survey of individuals
and their answers may not be accurate, and it is difficult to collect and harmonize
46

 Next, the Court considers RPS’s argument that a reasonable fact finder could
not find Plaintiffs’ evidence specific enough to reflect the arrest profile of the pool
of applicants to affordable/subsidized rental housing in Connecticut. Were the
Court to find that RPS had demonstrated that the eligible rental applicant pool
should be narrowed to individuals eligible for and/or interested in subsidized and
affordable housing in cities, similar to the applicant pool for the ArtSpace
Windham, the relevant question would be whether theexperiences of African
Americans and Latinos statewide in terms of arrests and incarcerations differ from
the experiences of African Americans and Latinos eligible for and/or interested in
subsidized and affordable housing in cities in terms of arrests and incarcerations.
Plaintiffs have demonstrated that African Americans and Latinos face higher rates
of arrest and incarceration regardless of their income and regardless of their state
geography. This evidence gives the Court reason to believe that, since African
Americans and Latinos face higher rates of arrest regardless of their socioeconomic status, state-wide arrest and incarceration statistics reflect the arrest
and incarceration profile of the pool of applicants for affordable/subsidized
housing.

data from diverse criminal justice institutions. (This point is, in fact, the basis of
RPS’s business model. See Dkt. 114-1 ¶13.).
Here, there is an additional question of who the appropriate pool is: while
this case is focused on Connecticut, a Connecticut housing applicant may have,
like Mr. Arroyo, a criminal record from another state or from the federal
government, so top-down statistics from Connecticut criminal justice institutions
only may not completely capture the criminal record of Connecticut housing
applicants. However, Plaintiffs provide national data in addition to state-level data,
and RPS does not argue that there are state-by-state disparities not captured by
national level data.
47

 RPS argues at length that there is a large variation in the population
demographics within Connecticut. It provides evidence that the percentage of the
population who is African American and the percentage of the population who is
Latino differs between Connecticut cities and Connecticut suburbs or rural areas,
and between renters looking for affordable or subsidized housing and those who
are not. [Dkt. 114-1 ¶¶56-57]. But RPS’s population statistics do not definitively
speak to the key question: whether differences in rates of arrest and incarceration
by race and ethnicity differ between geographic localities and income levels and
propensity to rent. For example, the fact that 38% of the population of Hartford is
African-American while 7% of the population of Willimantic is African-American,
[Dkt. 114-5 at Ex. I], does not necessarily say anything about whether the arrest
rate of African Americans in Willimantic is more or less disproportionately high
than the arrest rate of African Americans in Hartford. See Inclusive Communities
Project, 135 S. Ct. at 2522 (cautioning against drawing conclusions about policies
from bare statistical disparity). RPS simply has not given the Court any definitive
“reason to suppose that” relative rates of criminal justice experience of urban
Connecticut renters “differ markedly from those of the [Connecticut] population.”
See Dothard, 433 U.S. at 330.
RPS’s cases are not to the contrary, as they are all from the employment
discrimination context. See [Dkt. 114 at 20-24]; Townsend v. Nassau County
Medical Center, 558 F.2d 117 (2d Cir. 1997) (finding that national data on those with
and without college degrees insufficient to answer question about populations with
and without a B.S. degree); Mandala v. NTT Data, Inc., No. 18-CV-6591 CJS, 2019

48

 WL 3237361, at *4 (W.D.N.Y. July 18, 2019) (finding that national statistics regarding
conviction and arrests could not be assumed to reflect statistics for those qualified
to be “viable candidates” for positions of Salesforce and web developers ); Wards
Cove Packing Co. v. Atonio, 490 U.S. at 650 (finding that statistics on cannery
workers insufficient given question of who would be qualified to be managers);
Chin v. Port Auth. of New York & New Jersey, 685 F.3d 135, 152 (2d Cir. 2012)
(finding data on all employees insufficient given question of who was eligible for a
promotion among employees who had passed required test). But the employment
discrimination context is materially different from the housing context because
jobs often require unusual additional qualifications—consider the requirements
described in the cases above—and so there is a reason to suppose that the
characteristics of the “eligible labor pool” differ in systematic and relevant ways
from the characteristics of the general state population. Chin, 685 F.3d at 152. 17 In
contrast, “eligibility” for renting an apartment largely depends on income and
geographic preference, characteristics Plaintiffs have accounted for.
In its reply brief, RPS argues that, without evidence of the number of African
Americans and Latinos who apply to rent housing from RPS’s Connecticut clients,
Plaintiffs cannot prove a prima facie case. In support, it cites Ungar v. New York
City Hous. Auth., 363 F. App'x 53, 56 (2d Cir. 2010) (Summary Opinion):

In the employment discrimination context, “in the typical disparate impact case
the proper population for analysis is the applicant pool or the eligible labor pool.”
Chin v. Port Auth. of New York & New Jersey, 685 F.3d 135, 152 (2d Cir.
2012).17RPS’s citation of this authority omits the phrase, “the eligible labor pool,”
a mis-citation which incorrectly heightens the standard statistics must meet to be
acceptable. [Dkt. 114 at 19-20].
17

49

 To know whether TSAP has disparate impact on Hasidic Jews, we
would need to know, at a minimum, the percentage of the
approximately 80,000 people who apply for public housing each year
that is Hasidic. Because plaintiffs have not provided this information,
they failed to establish a prima facie case of discrimination.
But in light of the support of Tsombiandis, 352 F.3d at 577, and Reyes, 903
F.3d at 428m for the use of a potential applicant pool, and in light of the distinction
that the instant Plaintiffs have offered a causal story for the disparate impact while
the Ungar plaintiffs did not, the Court does not find Ungar persuasive.
Finally, RPS argues, without legal citation, that even if there is a disparate
impact, it is not “substantial,” as fewer than 7% of all rental housing applicants in
Connecticut between 2016 and the present have had any “record found” through
CrimSAFE. [Dkt. 114 at 16-17]; [Dkt. 114-1 ¶11]. In the Second Circuit, “courts
should take a ‘case-by-case approach’ in judging the significance or substantiality
of disparities, one that considers not only statistics but also all the surrounding
facts and circumstances.” Chin v. Port Auth. of New York & New Jersey, 685 F.3d
135, 153 (2d Cir. 2012) (quoting Waisome v. Port Auth. of N.Y. & N.J., 948 F.2d 1370,
1376 (2d Cir.1991)). Disparate impact claims have been found valid even where a
relatively small percentage of individuals are affected. See, e.g., Jones v. City of
Bos., 752 F.3d 38, 44-45, 52-53 (1st Cir. 2014) (disparate impact was established
even though less than 1% were affected by neutral rule). Here, in light of the facts
discussed above and below, the Court finds that Plaintiffs have provided sufficient
evidence to put into question whether there is a disparate impact.
For these reasons, the Court does not grant either party summary judgment
as to this issue.

50

 2. Business Purpose
If a statistical disparate impact is shown, the “defendant has the burden of
proving that the challenged practice is necessary to achieve one or more
substantial, legitimate, nondiscriminatory interests,” 24 C.F.R. § 100.500(c)(2).
“HUD has clarified that this step of the analysis ‘is analogous to the Title VII
requirement that an employer's interest in an employment practice with a disparate
impact be job related.’” Inclusive Communities Project, Inc., 135 S. Ct. at 2514-15.
(quoting Implementation of the Fair Housing Act's Discriminatory Effects Standard,
78 Fed.Reg. 11460, 11470 (2013)). Here, any business justification must address
(a) why excluding applicants with arrests and convictions is justified and (b) why
categorizing criminal records is justified. As to (a), RPS argues first that screening
for arrests and convictions is legally required for federally subsidized properties
and, second, screening for arrests and convictions is permitted to protect health
and safety; as to (b), RPS argues CrimSAFE’s categorization has unique
advantages in accurately categorizing the risk level of arrests and convictions, as
well as in minimizing bias in the screening process. [Dkt. 129 at 33-38; Dkt. 114 at
24-26].
Landlords of federally-assisted housing must reject applicants recently
evicted for drug activity or who are registered sex offenders and must perform
“necessary criminal history background checks in the State where the housing is
located and in other states where the household members are known to have
resided.” 24 C.F.R. §§ 5.856; see 24 C.F.R. § 5.854. With respect to an applicant
“evicted… for drug-related criminal activity,” landlords must consider individual
51

 factors, such as whether the individual has completed a rehabilitation program or
no longer resides with household members seeking housing, though a tenant may
have been evicted without satisfying a criminal conviction standard of proof for the
activity. 24 C.F.R. §§ 5.854(a), 5.861. But these statutes do not provide a justification
in this case because they apply to only a very narrow set of applicants—sex
offenders and applicants who have previously been evicted for drug-related
criminal activity—of which Mr. Arroyo is not one.
42 U.S.C. § 13661(c) permits criminal background screening to detect any
other criminal activity that would harm the “health, safety, or right to peaceful
enjoyment of the premises by other residents the owner, or public housing-agency
employees.” Except in limited circumstances not applicable here, a consumer
reporting agency may not make a consumer report containing “records of arrest
that, from date of entry, antedate the report by more than seven years or until the
governing statute of limitations has expired, whichever is the longer period.” 15
U.S.C. § 1681c(a)(2). Under the statute, a record of a conviction of a crime which
antedates the report by more than seven years may be included. Id. at 1681c(a)(5).
RPS suggests that because 15 U.S.C. § 1681c allows screening reports to reflect
the criminal records for up to seven years for non-convictions and for no time limit
for convictions, Plaintiffs cannot challenge the time periods under the FHA. [Dkt.
129 at 33] (quoting Eldred v. Ashcroft, 537 U.S. 186, 208 (2003) (“[C]ourts are not at
liberty to second-guess congressional determinations and policy judgments…,
however debatable or arguably unwise they may be”)).

52

 RPS provides neither legal nor empirical support for the proposition that the
lone fact that an applicant has a pending arrest record is sufficient for a housing
provider to determine that an individual poses a threat to the health and safety of
a residential community, although it does provide sufficient evidence to put into
dispute whether an older conviction is sufficient evidence of a threat to health and
safety.
15 U.S.C. § 1681c and 42 U.S.C. § 13661(c) do not establish that Congress
determined that a pending arrest record alone may be the basis for a housing
denial. “It is at best treacherous to find in congressional silence alone the adoption
of a controlling rule of law.” United States v. Wells, 519 U.S. 482, 496 (1997) (quoting
NLRB v. Plasterers' Local Union No. 79, 404 U.S. 116, 129–130 (1971)). 15 U.S.C. §
1681c does not explicitly endorse the use of screening reports reflecting criminal
records of non-convictions and older convictions; it only does not prohibit them.
The statute’s silence is especially treacherous in light of the answering
silence in the FHA. The FHA states that its protections do not apply to any decision
denying housing because an applicant “has been convicted by any court of
competent jurisdiction of the illegal manufacture or distribution of a controlled
substance.” 42 U.S.C. § 3607(b)(4). If silence speaks, then “the specific carveout
for drug convictions provides a strong inference that Congress presumed the Act
could sometimes require housing providers to overlook other types of criminal
records to avoid having discriminatory effects on members of protected
classes….” Simmons v. T.M. Assocs. Mgmt., Inc., 287 F. Supp. 3d 600, 603 (W.D.
Va. 2018).

53

 And this interpretation of congressional silence has been validated by the
administering agency: HUD has released guidance that clarified that an arrest
record by itself—in the absence of consideration of a police report or other
additional facts—may not be the basis for denying admission, terminating
assistance, or evicting tenants from public and other federally-assisted housing,
see Guidance for Public Housing Agencies (PHAs) and Owners of FederallyAssisted Housing on Excluding the Use of Arrest Records in Housing Decisions,
HUD

PIH

Notice

2015-19,

(November

2,

2015),

available

at:

http://portal.hud.gov/hudportal/documents/huddoc?id=PIH2015-19.pdf
(hereinafter “HUD Nov. 2, 2015 Guidance”).
HUD states “that the fact that an individual was arrested is not evidence that
he or she has engaged in criminal activity” sufficient to warrant denial of
admission.” Id. at 3. A housing owner may only “make an adverse housing decision
based on the conduct underlying an arrest if the conduct indicates that the
individual is not suitable for tenancy and the PHA or owner has sufficient evidence
other than the fact of arrest that the individual engaged in the conduct,” “such as
police reports detailing the circumstances of the arrest, witness statements, and
other relevant documentation to assist them in making a determination that
disqualifying conduct occurred.” Id. at 3-4. This requirement that housing owners
look at conduct is specific to arrests: “reliable evidence of a conviction for criminal
conduct that would disqualify an individual for tenancy may also be the basis for
determining that the disqualifying conduct in fact occurred.” Id. at 3.

54

 This guidance is not a product of notice-and-comment and is neither
binding, Ctr. for Auto Safety v. Nat'l Highway Traffic Safety Admin., 452 F.3d 798,
807 (D.C. Cir. 2006), nor entitled to Chevron deference. Christensen v. Harris Cty.,
529 U.S. 576, 587 (2000) (“ Interpretations contained in policy statements… do not
warrant Chevron-style

deference.”)

(citing Chevron

U.S.A.

Inc.

v.

Natural

Resources Defense Council, Inc., 467 U.S. 837 (1984)). Rather, this guidance is
accorded weight according to Skidmore v. Swift & Co., 323 U.S. 134 (1944). Id.
“Under Skidmore, the weight [courts] accord an agency interpretation depends
upon “the thoroughness evident in its consideration, the validity of its reasoning,
its consistency with earlier and later pronouncements, and all those factors which
give it power to persuade.” Boykin v. KeyCorp, 521 F.3d 202, 208 (2d Cir. 2008)
(quoting Skidmore, 323 U.S. 134); see also United States v. Mead Corp., 533 U.S.
218, 228 (2001).
Here, the Court finds that the November 2, 2015 HUD Guidance is persuasive:
it is thorough, synthesizing a review of the relevant case law, a review of the
specifics of the governing regulations, and relevant governmental statistics
regarding arrests. Id. at 3-4. Its reasoning is careful, drawing a distinction between
relying on an arrest record itself and other evidence of the conduct underlying the
arrest. And it is consistent with later pronouncements, particularly the April 4, 2016
HUD Guidance. Therefore, the Court finds that the November 2, 2015 Guidance
undermines RPS’s claim that 15 U.S.C. § 1681c and 42 U.S.C. § 13661(c) establish
the time limits during which a pending arrest charge or an old conviction alone may
be the basis for a housing application denial.

55

 With regard to empirical evidence, RPS provides that the federal Bureau of
Justice Statistics has found that 83% of released prisoners are arrested again
within 9 years, with an average of 5 arrests per released prisoner. [Dkt. 114-1 ¶59].
18% of violent victimizations took place in the victim’s home, 16% took place near
the victim’s home, and 9% took place at a friend’s, neighbor’s, or relative’s home.
Id. ¶ 60. In response, Dr. Lila Kazemian, one of Plaintiff’s offered experts, stated
that “anybody who has their name already in the system becomes more likely to
have more contacts with the criminal justice system,” such that individuals who
were previously arrested have elevated statistical levels of re-arrest. [Dkt. 129-1
¶14]. She also states that, based on her review of the relevant academic literature,
“[t]here is no compelling empirical evidence to suggest that old criminal records
are predictive of future offending” because “the more time that passes since the
last crime, the less likely it is that the individual will engage in the crime in the
future.” [Dkt. 118-1 ¶95].
This evidence puts into dispute whether individuals with old convictions
may be excluded to protect health and safety, especially since the term “old
conviction” remains undefined and is used to mean convictions from five to ninetynine years old. But none of this evidence supports the proposition that individuals
with pending arrests are threats to health and safety. The evidence from the Bureau
of Justice Statistics does not speak to the dangerousness of individuals who have
been arrested, but not charged. Dr. Kazemian’s comments only beg the question
of the dangerousness of arrestees: the fact that someone who has been arrested
once is more likely than others to be arrested again only demonstrates that

56

 whatever characteristics are associated with being arrested likely persist over
time—and many characteristics, including implicit bias, cultural incompetence,
race and place of residence, persist over time.18

This evidence demonstrates that there is a material dispute of fact as to
whether there is a business justification for screening for old convictions, that is,
whether the justifications offered by RPS, with the support offered by RPS,
demonstrate that screening for old convictions is “necessary” to achieve one or
more substantial, legitimate, nondiscriminatory interests. But, in light of the HUD
Guidance and empirical evidence cited, no reasonable fact finder could find that
there is a business justification for screening solely on the basis that someone has
a pending arrest, in the absence of the details of the arrest.

18

Neither party presents any evidence of the percentage of arrests that lead to
convictions or other measurements of risk. But, as an aside, the Court notes that
state and local government data demonstrate that conviction rates have changed
over time and that they vary considerably depending on the charge. Compare Brian
A. Reaves, Bureau of Justice Statistics, U.S. Dep’t of Justice, Felony Defendants in
Large
Urban
Counties,
2009,
at
22,
Table
21
(2013),
http://www.bjs.gov/content/pub/pdf/fdluc09.pdf (in the 75 largest counties in the
United States in 2009, approximately one-third of felony arrests did not result in
conviction, with about one-quarter of all cases ending in dismissal.), cited by U.S.
H.U.D., Office of Public and Indian Housing, Guidance for Public Housing Agencies
(PHA) and Owners of Federally-Assisted Housing on Excluding the Use of Arrest
Records in Housing Decisions, (PIH 2015-19; Nov. 2. 2015) at 3 n. 8, with, Issa
Kohler-Hausmann, Misdemeanorland: Criminal Courts and Social Control in an
Age of Broken Windows Policing 68-69 (2018) (author’s analysis of data from New
York State Department Division of Criminal Justice Services showing that, in New
York City between 2010 and 2015, more than 50 percent of misdemeanor arrests
were dismissed, an increase from the 1985 dismissal rate), with, Malcolm Feeley,
The Process is the Punishment: Handling Cases in a Lower Criminal Court at xxviii,
127 (1979) (author’s analysis of 1970s New Haven Court of Common Pleas
dispositions).
57

 The offered evidence is especially insufficient in the face of the bedrock
principle of our legal system that a person who is arrested is presumed innocent –
and that innocence alone does not undermine probable cause to arrest. See
Schware v. Bd. of Bar Exam’rs, 353 U.S. 232, 241 (1957) (“The mere fact that a man
has been arrested has very little, if any, probative value in showing that he has
engaged in any misconduct. An arrest shows nothing more than that someone
probably suspected the person apprehended of an offense.”); United States v. Di
Re, 332 U.S. 581, 595 (1948) (“the presumption of innocence is not lost or impaired
by neglect to argue with a policeman”); Coffin v. United States, 156 U.S. 432 (1895)
(“the principle that there is a presumption of innocence in favor of the accused is
the undoubted law, axiomatic and elementary, and its enforcement lies at the
foundation of the administration of our criminal law.”); Illinois v. Wardlow, 528 U.S.
119, 126 (2000) (“Indeed, the Fourth Amendment accepts that risk in connection
with more drastic police action; persons arrested and detained on probable cause
to believe they have committed a crime may turn out to be innocent.”); Panetta v.
Crowley, 460 F.3d 388, 395–96 (2d Cir. 2006) (“an officer's failure to investigate an
arrestee's protestations of innocence generally does not vitiate probable cause” to
arrest).
And even if RPS had demonstrated that screening for old convictions
indisputably contributes to safety, a dispute remains whether CrimSAFE, which
characterizes and categorizes criminal records, allows property managers to more
quickly and accurately screen for these safety risks. Compare [Dkt. 120 (Jay Kacirk
Decl.) at 3, 6, 9, 11] & [Dkt. 128-3 ¶26] (applicants dispute RPS’s tenant screening

58

 reports in less than 1% of instances), with, e.g., [Dkt. 116-31 (Ex. 28: Kazemian Rep.)
at 2] (estimating that industry-wide, approximately 30% of criminal history reports
contain inaccuracies). The parties also dispute whether CrimSAFE helps to remove
potential explicit or implicit bias at the individual property manager level, or
whether it gives individual property managers more information. Compare [Dkt.
114-2 (Kayani Decl.) ¶¶18, 19], with id. at Ex. C at p. 4 (CrimSAFE result listing a
“race” category, but listing Mr. Arroyo’s race as “unknown”).
Finally, RPS makes two additional arguments: first that the Court should not
rely solely on statistics, as some crimes, though low probability, may have
dramatic consequences on a community; and second, that the Court should
consider the liability for failing to review criminal records that housing providers
might face. [Dkt. 129 at 38-39]. But, as the Court will go on to note, the alternative
to CrimSAFE is returning more detailed reports of criminal records to line staff
decisionmakers, rather than not returning criminal records reports at all. This
alternative does not ignore the possibility of a low probability event, but allows for
it, and the Court sees no reason why the alternative would necessarily lead to
worse community consequences or greater housing provider liability.
For these reasons, Court finds that RPS and Plaintiffs provide sufficient facts
to demonstrate the existence of a material dispute as to whether there is a business
purpose for screening for convictions and denies the parties’ motions for summary
judgment on that basis. But the Court finds that the Plaintiffs have shown that there
is no business justification for screening applicants on the basis of the fact of a
pending arrest alone.
59

 3. Less Discriminatory Alternatives
After a defendant has shown a legitimate business interest for its facially
neutral practice, the burden shifts to the Plaintiffs to show that there is a “less
discriminatory alternative,” which must be 24 C.F.R.§ 100.500(c)(3). “[A] a less
discriminatory alternative must serve the respondent's or defendant's substantial,
legitimate nondiscriminatory interests, must be supported by evidence, and may
not be hypothetical or speculative.” Implementation of the Fair Housing Act's
Discriminatory Effects Standard, 78 Fed. Reg. at 11,473 (Feb. 15, 2013); see also
Inclusive Communities Project, 135 S. Ct. at 2518 (stating that before rejecting a
business or public interest, "a court must determine that a plaintiff has shown that
there is 'an available alternative . . . practice that has less disparate impact and
serves the [entity's] legitimate needs'" (emphases added)).
In their motion for summary judgment, Plaintiffs advance four less
discriminatory alternatives to achieve the goals of protecting safety and property:
(1) RPS could exclude arrests that have not or did not result in a conviction from
being considered as a basis for a CrimSAFE decision; (2) RPS could set a
“reasonable, evidence-based” cap on the lookback period for convictions, and
exclude older arrests from being considered as a basis for a CrimSAFE decision;
(3) RPS could evaluate each criminal record on an individualized basis by
considering the record and relevant mitigating circumstances outside the criminal
record itself to determine the actual risk to safety before reporting a housing
provider that the applicant is disqualified; and (4) RPS could provide the underlying
information about the criminal history to the housing provider without providing a
60

 leasing decision the landlord can do an individualized assessment. [Dkt. 118 at 29
-31].
RPS first argues that the first two alternatives are precluded under Rule 37(c)
because plaintiffs did not disclose these “alternatives” during discovery. [Dkt. 129
at 41-42]. Rule 37(c)(1) states that “[i]f a party fails to provide information or identify
a witness as required by Rule 26(a) or (e), the party is not allowed to use that
information or witness to supply evidence on a motion, at a hearing, or at a trial,
unless the failure was substantially justified or is harmless.” Fed. R. Civ. P. 37. Rule
26(e)(1) states: “A party who has made a disclosure under Rule 26(a)—or who has
responded to an interrogatory, request for production, or request for admission—
must supplement or correct its disclosure or response in a timely manner. . . .” Fed.
R. Civ. P. 26(e)(i).
This argument is unavailing. Far from sandbagging RPS, Plaintiff produced
all relevant information during discovery. When RPS asked Plaintiffs to “describe
… the factual basis of your [Complaint’s] allegation that there exist ‘less
discriminatory alternatives’,” Plaintiffs responded that RPS “has available to it the
less discriminatory alternative of . . . consideration of such factors as the … the
outcome or disposition of the case” and “the amount of time since the criminal
activity occurred” (emphasis added). [Dkt. 136-1 (Pl.’s Reply Facts) ¶¶ 47, 109]. In
addition to the interrogatory response, Plaintiffs disclosed expert reports to RPS,
which specifically included that “arrest[s] … should not even be included” in
CrimSAFE.”, id. ¶ 14, and that an appropriate “guideline for policy” based on the

61

 empirical studies is “five to nine years” provided this is considered “in conjunction
with some of the other factors” the expert had discussed. Id.
Although RPS argues that the Rule 26(e) burden applies to “theories,” in
addition to information, it cites only Agence France Presse v. Morel, 293 F.R.D. 682,
686 (S.D.N.Y. 2013), in support. See [Dkt. 129 at 41-42]. But, there the “theory” at
issue was a damages calculation, Morel, 293 F.R.D. at 686, and Rule 26 specifically
requires parties to disclose “a computation of each category of damages claimed
by the disclosing party.” There is no corresponding Rule or regulation requiring
discovery disclosure of less discriminatory alternative theories. So, the Court
considers the alternatives on their merits.
First, the Court finds that it is undisputed that RPS could cease considering
arrests that do not result in conviction. However, since disputes of fact remain as
to whether this would diminish the disparate impact caused by excluding such
people, this is not a basis in and of itself for a grant of summary judgment.
As to the second alternative, a “reasonable, evidence-based” cap on the
lookbook period for conviction is not sufficient to win Plaintiffs’ summary
judgment as too many disputed facts remain. First, Plaintiffs do not specify what
an “evidenced-based” number of years, and Plaintiffs’ expert could not specify
what she thought was an “evidence-based” lookback period for any particular
CrimSAFE category of crime. [Dkt. 132 at 228-29]. The Court cannot find that RPS
could undertake an alternative the outline of which even Plaintiffs do not know.
Second, the number of years is in dispute in light of the BJS study discussed
above.

62

 Plaintiffs’ third alternative fails because, as RPS points out, RPS does not
have the majority of the information that would lead to a leasing decision. It
receives an applicants’ identifying information. [Dkt. 118-1 ¶14]. It then returns
information about the applicant’s offense record, if any. Id. ¶19. But it does not
have other information, and so it could not itself undertake a holistic individualized
review of the applicant. Since there is no evidence that RPS’s clients would give
RPS such information, it is speculative and not truly available. See ICP at earlier
cite; Allen v. City of Chicago, 351 F.3d 306, 313-14 (7th Cir. 2003) (On appeal of
summary judgment, “Without any evidence that the officers’ alternative of
increasing merit promotions would lead to a workforce substantially equally
qualified, we cannot accept the officers’ alternative as substantially equally valid.”).
Plaintiffs’ fourth alternative is not a satisfactory basis for summary judgment
in their favor. At a minimum, RPS provides its clients with the option to view the
underlying criminal report information, though the client may choose suppress that
information from on-site managers and other levels of administrators. [Dkt. 118-3
at ARROYO0001750]. The parties submit dueling evidence on whether RPS always
provides at least one person in the client company with access to the full records,
or whether RPS may prevent everyone at the company from accessing the full
records, so it is disputed. Compare id. (From 2016 training, “Consider the Backup
Reports setting on this screen as the “MAIN SWITCH” for making backup reports
viewable. When unchecked, NO USERS will have access to criminal backup
reports, not even administrators.” ) with [Dkts. 129-2 (Kayani Decl.) at ¶¶ 11 (“The
CrimSAFE section of a report is always accompanied by another portion of the

63

 report that displays the full public data of any record(s) identified via CrimSAFE,
including the date the offense was committed, the severity level, and its current
status…. The administrative version of the report, with the full details of any
records found by the CrimSAFE product, is made available to the identified
supervisor(s) at the customer simultaneous with the leasing-agent version of the
report via hyperlink., 24, Ex. D (Arroyo BackUp Report)] and [Dkt. 129-3 (Dacthler
Decl.) at 18]. While RPS responds that CrimSAFE should include the full criminal
history with every report, and eliminate the features that enable suppression of
details, there is, as explained above, a dispute of facts as to whether there is a
legitimate business justification for RPS’s suppression.
Because of these remaining factual disputes, the Court denies the parties’
motions for summary judgment as to Plaintiffs’ FHA claim for disparate impact on
the basis of race. These claims will proceed to trial.
C. Fair Housing Act Claim for Disparate Treatment on the basis of Race or
National Origin
RPS argues that Plaintiffs’ FHA claim for discriminatory treatment on the
basis of race or national origin in Count I must fail because there is no evidence of
racial animus because RPS was not aware of Mr. Arroyo’s race or ethnicity. [Dkt.
114 at 15-17]. In response, Plaintiffs argue that the Court can and should consider
the “totality of the circumstances” to determine the disparate treatment claim on
the basis of race [or national origin], and that such evidence suffices for the claims
to proceed to trial. [Dkt. 126 at 25-29]. The Court agrees with Plaintiffs.
“Discriminatory

intent

may

be

inferred

from

the

totality

of

the

circumstances.” L.C. v. LeFrak Org., Inc., 987 F. Supp. 2d 391, 400 (S.D.N.Y. 2013)
64

 (quoting Reg'l Econ. Cmty. Action Program, Inc. v. City of Middletown, 294 F.3d 35,
48 (2d Cir.2002)). “A plaintiff can establish a prima facie case of disparate treatment
‘by showing that animus against the protected group was a significant factor in the
position taken by the municipal decision-makers themselves or by those to whom
the decision-makers were knowingly responsive.’” Mhany Mgmt., Inc. v. Cty. of
Nassau, 819 F.3d 581, 606 (2d Cir. 2016)(government actor context) (quoting
LeBlanc-Sternberg v. Fletcher, 67 F.3d 412, 425 (2d Cir. 1995)) “Because
discriminatory intent is rarely susceptible to direct proof, a district court facing a
question of discriminatory intent must make ‘a sensitive inquiry into such
circumstantial and direct evidence of intent as may be available. The impact of the
official action whether it bears more heavily on one race than another may provide
an important starting point.’” Ibid. (2d Cir. 2016) (quoting Village of Arlington
Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 266, (1977)). “But unless a ‘clear
pattern, unexplainable on grounds other than race, emerges,’ ‘impact alone is not
determinative, and the Court must look to other evidence.’” Id. (quoting Arlington
Heights, 429 U.S. at 266); see Columbus Bd. of Ed. v. Penick, 443 U.S. 449, 464
(1979) (“disparate impact and foreseeable consequences, without more, do not
establish a constitutional violation.”).19 Other relevant considerations for

See also Soberal-Perez v. Heckler, 717 F.2d 36, 42 (2d Cir. 1983) (“while it is true
that…. the fact that a particular action has a foreseeable adverse impact may be
relevant evidence in proving an equal protection claim… standing alone that fact
is insufficient to establish discriminatory intent”); Washington v. Davis, 426 U.S.
229, 241-42 (1976 (discriminatory impact alone does not show discriminatory intent
but it is a relevant factor to consider along with the totality of relevant facts). See
also United States v. City of New York, 717 F.3d 72, 82 (2d Cir. 2013); United States
19

65

 discerning a racially discriminatory intent include” ‘[t]he historical background of
the decision ... particularly if it reveals a series of official actions taken for invidious
purposes,’ ‘[d]epartures from the normal procedural sequence,’ ‘[s]ubstantive
departures,’ and ‘[t]he legislative or administrative history ... especially where there
are contemporary statements by members of the decisionmaking body, minutes of
its meetings, or reports.’” Id. (quoting Arlington Heights, 429 U.S. at 266-68).
“Questions of subjective intent can rarely be decided by summary judgment.”
United States v. City of New York, 717 F.3d 72, 82 (2d Cir. 2013) (citing Harlow v.
Fitzgerald, 457 U.S. 800, 816 (1982)) (claim for disparate treatment in hiring under
Title VII).
Here, Plaintiffs point to evidence that, as of mid-April of 2016, RPS was aware
that its CrimSAFE screening product’s use of arrest records may have a
disproportionate and arbitrary effect on African Americans and Latinos, and that it
has not taken affirmative steps to end its screening product’s use of those records.
On April 4, 2016, HUD’s Office of General Counsel published a document titled
Application of Fair Housing Act Standards to the Use of Criminal records by
Providers of Housing and Real Estate-Related Transactions, in which it stated
“Nationally, racial and ethnic minorities face disproportionately high rates of arrest
and incarceration,” and that, “the fact of an arrest is not a reliable basis upon which
to assess the potential risk to resident safety or property posed by a particular

v.

City

of

Yonkers,

96

F.3d

600,

66

603

(2d

Cir.

1996)

(post-trial).

 individual.” HUD OGC 4/4/2016 Guidance at 3-5. On April 15, 2016—over a week
before Mikhail Arroyo’s application—RPS shared with some of its clients an email
which highlighted (a) per HUD, arrest records that don’t result in convictions are
not reliable bases to assess the potential risk resident safety or property; and (b)
“according to HUD, a blanket policy to deny any applicants with a criminal record
may have a disparate impact on African Americans and Hispanics.” [Dkt. 125-17 at
2]; see also [Dkt. 125-18]. The email also stated that RPS was considering changes
to its products:
RPS is currently reviewing our products to determine what changes,
if any, to make in order to best support our clients in light of this new
guidance from HUD. Once this review is complete, any changes will
be communicated to clients with enough notice to allow for clients to
adjust their processes.

[Dkt. 125-17 at 3]. But, as of September 5, 2019, Stephanie Dacthler, a Relationship
Manager for RPS, was not aware of any changes made to CrimSAFE based on the
HUD Guidance. [Dkt. 118-8 at 75:18-23].
This scenario is far from one in which “the issue of race was first introduced
[to the defendants] upon filing this action.” Cf. Favourite v. 55 Halley St., Inc., 381
F. Supp. 3d 266, 280 (S.D.N.Y. 2019) (granting summary judgment on disparate
treatment claim). Instead, there is at least a disputed question of fact as to whether
RPS’s decision to continue to allow its housing clients to use CrimSAFE to screen
for arrest records in the face of the legal interpretation of HUD’s Office of General
Counsel that African Americans and Latinos are disproportionately more likely to
be arrested is motivated by discriminatory intent. For this reason, the Court denies

67

 RPS’s motion for summary judgment as to the FHA claim for disparate treatment
on the basis of race or ethnicity. The claim will proceed to trial.
D. FCRA Claims
In Count IV and V of the Complaint, Plaintiffs allege that RPS violated the Fair
Credit Reporting Act (“FCRA”), specifically, 15 U.S.C. §§ 1681g and 1681h, by
failing to disclose Mikhail Arroyo’s file to Carmen Arroyo and by failing to establish
reasonable requirements for proper identification, and that such violations were
both negligent and “willful.”
The FCRA balances protecting individuals from identity theft by requiring
the submission of personally-identifying information before disclosure with
promoting access to consumer files by minimizing the personally-identifying
information required. 15 U.S.C. § 1681g requires that a consumer reporting agency
“shall, upon request, and subject to section 1681h(a)(1) of this title, clearly and
accurately disclose to the consumer… all information in the consumer’s file at the
time of the request.” Section 1681h(a)(1) in turn provides that “a consumer
reporting agency shall require, as a condition of making the disclosures required
under section 1681g of this title, that the consumer furnish proper identification.”
The FCRA does not define “proper identification,” however, and the case law
illustrates how the tension between its two goals has frustrated courts’ filling in of
the gap. In Howley v. Experian Info. Sols., Inc., a court denied the defendant’s
motion for summary judgment on the plaintiff’s § 1681h(a) claim against a
consumer reporting agency for disclosing his information to a third party who
shared the same first name as the plaintiff and shared all digits but the last of his
68

 social security number. 813 F. Supp. 2d 629 (D.N.J. 2011). There, the court found
that there was a question of fact as to whether the agency had disclosed the
information without proper identification. 813 F. Supp. 2d 629. Id. On the other
hand, in Menton v. Experian Corp., a court denied a related defendant’s motion to
dismiss a plaintiff’s § 1681g(a) claim for failing to disclose his file where he had
provided a copy of his driver’s license, a bank statement with his name and
address, his law firm website, and a notarized copy of his signature. No. 02 CIV.
4687 (NRB), 2003 WL 941388, at *1 (S.D.N.Y. Mar. 6, 2003). There, the court held that
there was “no reason that Experian could not have verified Mr. Menton’s identity
and provided him with his credit report soon after receiving the various alternative
forms of identification which he did furnish.” Id. at 3. By Menton’s lights, proper
identification may include documents whose validity is determined by local state
law: for example, state driver’s licenses, or notarized documents, such as a power
of attorney or written authorization. See, e.g., id.
A consumer reporting agency must implement consumer identification
requirements that “ensure that the information is sufficient… to match consumers
with their files” and “commensurate with an identifiable risk of harm arising from
misidentifying the consumer.” 12 C.F.R. § 1022.123(a). RPS, a nationwide specialty
consumer reporting agency, must implement a file disclosure which “[c]ollect[s]
only as much personal information as is reasonably necessary to properly identify
the consumer….” 12 C.F.R. § 1022.137(a)(2)(ii). Further, “[i]n the event that a
consumer requesting a file disclosure cannot be properly identified in accordance
with the FCRA, [a consumer report agency must] provid[e] a statement that the

69

 consumer’s identity cannot be verified; and directions on how to complete the
request, including what additional information or documentation will be required
to complete the request, and how to submit such information,”… or “accept the
request.” 12 C.F.R. § 1022.137(a)(2)(iii)(C), (e)(1)-(2) (promulgated as an
implementation of 15 U.S.C. § 1681g, among others).
Here, the only FCRA violations that Plaintiffs allege arise out of violations of
Mikhail Arroyo’s FCRA rights, and so RPS can only be liable for damages to him.
The FCRA authorizes a third party to “accompany” a consumer to receive
disclosures provided that person provides “reasonable identification,” which may
include the consumer’s “written statement granting permission… to discuss the
consumer’s file in such person’s presence.” 15 U.S.C. § 1681h(d); see 15 U.S.C. §
1681b(a)(2) (“any consumer reporting agency may furnish a consumer report… in
accordance with the written instructions of the consumer to whom it relates.”). But
15 U.S.C. 1681g(a) does not give a person “a right to receive information from a
third party’s file.” Neclerio v. Trans Union LLC, 983 F. Supp. 2d 199, 219 (D. Conn.
2013) (VLB); see Oses v. Corelogic Saferent, LLC, 171 F. Supp. 3d 775, 782 (N.D. Ill.
2016) (same). And third parties do not have remedies under the FCRA—a person
who negligently or willfully fails to comply with the FCRA “with respect to any
consumer is liable to that consumer” for damages including “actual damages
sustained by the consumer.” 15 U.S.C. §§ 1681n(a),1681o(a) (emphases added).
Plaintiffs argue that RPS violated its duty to make the required consumer
disclosures because it did not disclose Mr. Arroyo’s information to Ms. Arroyo on
behalf of Mr. Arroyo until the start of this litigation. [Dkt. 87 at 9-10]; [Dkt. 105 at 170

 5]. RPS argues that Plaintiffs cannot show damages and because Ms. Arroyo failed
to present proper identification. The Court preliminarily addresses RPS’s argument
regarding damages, and then addresses the parties’ arguments regarding
disclosure and proper identification.
RPS argues that Plaintiffs cannot show damages based on RPS’s conduct
throughout the disclosure process because (1) Plaintiffs have not developed
testimony from WinnResidential that Ms. Arroyo could have used the file
disclosure information to persuade WinnResidential to overlook Mr. Arroyo’s
criminal history; and (2) such testimony would not be helpful, because
WinnResidential already had the criminal record but refused to grant access to
housing until it was sued. [Dkt. 114 at 34].
The Court denies RPS summary judgment on this basis because the relevant
facts are disputed. Plaintiffs have presented some evidence that WinnResidential
might have allowed Mr. Arroyo to move in before June 2017 after Plaintiffs provided
WinnResidential

with

details

about

Mr.

Arroyo’s

criminal

history

that

WinnResidential had located on their own. E.g. [Dkt. 116-35 at Ex. A (6/13/2017 Fact
Finding Hearing Transcript)]:20 Plaintiffs present evidence that they provided
WinnResidential with new information about Mr. Arroyo’s Pennsylvania arrest after
the CHRO Complaint and at the mediation. Id. at 68:5-71:10. Plaintiffs present
evidence that relevant decision-making WinnResidential employees did not have

20

Defendants object that this affidavit is hearsay but, since the Exhibit includes
statements by WinnResidential employees with personal knowledge of the events
discussed, Court finds that the affidavit is admissible for the purposes of summary
judgment. [Dkt. 140-1 (Def’dt’s Rule 56(a)(1 Reply Statement of Facts) at ¶87].
71

 this information, and that WinnResidential employees considered “the crux of
issue” as to why “Mikhail Arroyo ha[d] not been allowed to move in to ArtSpace at
Windham” to be that they had not had that information. Id. at 52:14-24, 50:16-25.
But, Plaintiffs also present evidence that WinnResidential did not immediately
allow Mr. Arroyo to move in once it received his criminal records—WinnResidential
went forward with the fact-finding hearing. This is sufficient evidence for a
reasonable fact-finder to possibly but not necessarily find that (a) WinnResidential
did not have Mr. Arroyo’s criminal history;21 (b) Mr. Arroyo’s actual criminal record
would have changed Winn Residential’s decision; and (c) therefore, if RPS had
provided the Arroyos with the file disclosure earlier, Mr. Arroyo would have had a
better opportunity to enjoy the housing of his choice. Although RPS objects that
WinnResidential only permitted Mr. Arroyo to move into the complex after being
sued and settling that lawsuit, [Dkt. 140-1 at ¶87], the objection does not definitively
undermine the importance of the Mr. Arroyo’s criminal history: settlement terms

21

The parties agree that WinnResidential suppressed the underlying
criminal records from the view of its leasing agents since 2015. [Dkt. 126-1 ¶91].
However, the parties dispute whether any other decisionmakers at WinnResidential
had Mr. Arroyo’s criminal record, and neither party offers admissible evidence on
the question, see Fed. R. Civ. P. 56(c)(2): Dkt. 116-41 (Aff. Ans. of WinnResidential
to Administrative Cmplt.) is inadmissible hearsay, see Fed. R. Evid. 802; Dkt. 114-2
(Kayani Decl.) & Dkt. 116-35 (Carmen Arroyo Decl.) were not made on the basis of
personal knowledge, see Fed. R. Evid. 602, Roberts v.Ground Handling, Inc., 499 F.
Supp. 2d 340, 360 (S.D.N.Y. 2007) (“It is axiomatic that affidavits submitted in
support of or in opposition to a summary judgment motion must ‘be made on
personal knowledge”) and [Dkt. 114-2 at Ex. D] cannot be authenticated as a copy
of a document that CoreLogic returned to WinnResidential on 4/26/2016 as claimed,
as it is a document generated on 5/3/2018, see Fed. R. Evid. 901.

72

 depend on the strength of the parties’ positions, for which Mr. Arroyo’s criminal
history was a factor.22
The Court next considers the parties arguments as to “proper identification:”
whether, under the FCRA, Ms. Arroyo submitted “proper identification” for herself
as conservator for Mr. Arroyo. The Court holds that, on the undisputed facts, she
did not. See [Dkt. 101 at 10-14].23 A conservatorship certificate with an impressed
seal is necessary for “proper identification” of a Connecticut conserved person
under the FCRA. Where state law defines the validity of an identification document,
state law defines “proper identification” under the FCRA. See Menton, 2003 WL
941388. Ms. Arroyo was appointed to be Mr. Arroyo’s conservator under
Connecticut state law. Under Connecticut law, as stated plainly on the face of a
conservatorship seal itself, a conservatorship certificate “is not valid without a
probate seal impressed.” See [Dkt. 125-10 at 3]; Johnson v. Raffy's Cafe I, LLC, No.
CV106002069S, 2015 WL 2166123, at *3 (Conn. Super. Ct. Apr. 6, 2015) (finding
probate certificate valid for purposes of establishing jurisdiction “after reviewing
the probate certificate” and finding “that it contained the raised seal”). The
heightened state standard for conservatorship documents is consistent with the
FCRA’s requirement that proper identification be

“commensurate with an

Plaintiffs also allege damages on the basis of Ms. Arroyo’s time and mental
annoyance spent following up, [Dkt. 126 at 30-31], but, since the FCRA claim is for
the disclosure of Mr. Arroyo’s files, and FCRA liability is only to the consumer, 15
U.S.C. §§ 1681n &1681o, the Court finds that damages to Ms. Arroyo do not support
Mr. Arroyo’s FCRA claim.
22

23

The Court recognizes that, in its motion for summary judgment, RPS does
not make this argument on the FCRA claim, [Dkt. 114 at 35-36], but notes that RPS
does earlier make this argument as a reason for granting it summary judgment on
the FHA disability disparate impact claim. Id. at 31.
73

 identifiable risk of harm arising from misidentifying the consumer” in light of the
heightened risks of identity theft for a person who would qualify to be conserved
under Connecticut law. See Conn. Gen. Stat. §§ 45a-644(c); 45a-650.
Plaintiffs argue that courts have held that the absence of a visible embossed
seal on a copy of a document does not make it invalid, as long as the original
document carries the seal. See, e.g. [Dkt. 105 at 4-6] (citing In re Robinson v. Chase
Home Fin. LLC, 403 B.R. 497, 503 (Bankr. S.D. Oh. 2008) (applying Ohio law to
mortgage copy held in public records); Schwab v. GMAC Mortg. Corp., 333 F.3d
135, 138 (3d Cir. 2003) (applying Pennsylvania statute regarding mortgage validity,
which did not require embossment to be legible on mortgage copy); Warfield v.
Byron, 137 Fed. App’x 651, 655 (5th Cir. 2005) (finding that absence of embossed
seal on photocopy of summons did not invalidate district court’s jurisdiction);
Oliver v. NY State Police, 2019 WL 453363, at 6 (W.D.N.Y. 2019) (same); Smith v.
Nat’l Credit Sys., Inc., 2015 WL 12780446, at *2 (N.D. Ga. 2015) (lack of visible
embossed notary seal on copy of affidavit filed with court did not affect validity of
affidavit). But Plaintiffs’ cited cases are simply inapposite: none were decided in
Connecticut courts or speak to Connecticut law, none address cases in which the
document states on its face that it is not valid without an embossed seal, and none
concern proper identification for the purposes of the FCRA. If anything, the cited
cases demonstrate that it is commonly understood that impressed seals are not
visible on photocopies. See Schwab, 33 F.3d at 138 and Warfield, 137 Fed. App’x at
655.

74

 The parties agree that Ms. Arroyo submitted copies of her conservatorship
certificates in June 2016 and November 2016. [Dkt. 114-1 ¶¶44, 51]. The parties
submitted copies of conservatorship certificates with their motions for summary
judgment. See [Dkt. 125-10 at 3]; [Dkt. 125-12 at 5]. Upon review, the Court holds
that no reasonable factfinder could find that that the copies of the certificates
demonstrate that the probate seal was impressed. As to Ms. Arroyo’s June 2016,
submission, in the lower left-hand side of the page, there is a shaded-in circle,
through which some white outlines are visible. [Dkt. 125-10 at 3]. White marks
which may be the remnants of a shaded-out impressed seal are not themselves an
impressed seal. As to Ms. Arroyo’s November 2016 submission, there are only
some stray marks in the same area, and nothing that establishes the presence of
an impressed seal. [Dkt 125-12 at 5]. “Although the Court can understand
[Plaintiffs]’ frustration with [RPS]’s extreme attention to detail, it is mindful that
[RPS] has a duty to protect the confidentiality and security of [Mr. Arroyo’s]
information.” Ogbon v. Beneficial Credit Servs., Inc., No. 10 CIV. 3760 PAE, 2013
WL 1430467, at *9 (S.D.N.Y. Apr. 8, 2013).
But the Court goes on to consider whether, having found that Mr. Arroyo
could “not be properly identified in accordance with the FCRA,” RPS nevertheless
violated its duty under 15 U.S.C. § 1681g by failing to “provid[e] a statement that
the consumer’s identity cannot be verified; and directions on how to complete the
request, including what additional information or documentation will be required
to complete the request, and how to submit such information.” See 12 C.F.R. §
1022.137(a)(2)(iii)(C). The Court finds that Plaintiffs have submitted sufficient

75

 evidence to put into question whether RPS violated this duty: while RPS mailed Mr.
Arroyo a letter in June of 2016 asking him to contact RPS to discuss the First
Disclosure Request, [Dkt. 114-1 ¶46], it is not clear that they would have instructed
her that they could not accept the “conservatorship court paper” because of the
seal defect or because of the missing power of attorney. See [Dkt. 114-6 at 11].
Further, when Ms. Arroyo called RPS in September 2016 to discuss the status of
the disclosure, [Dkt. 114-1 ¶47], she was instructed that she had to provide a
notarized power of attorney. [Dkt. 114-6 (Barnard Decl.) at Ex. B at 3]. The Court
finds that these facts are sufficient to put into question whether RPS violated its
duty to disclose by providing directions on how to complete a disclosure request
for the period starting with RPS’s response to Ms. Arroyo’s first set of
documentation and ending with its second call trying to reach Ms. Arroyo.
Therefore, the Court grants RPS summary judgment as to Plaintiffs’ FCRA claims
before June 30, 2016 and after November 18, 2016, and denies the parties’ motions
for summary judgment as to Plaintiffs’ FCRA claims for the period from June 30,
2016 and November 18, 2016.
i. Willfulness
“[A] company subject to FCRA does not act in reckless disregard of it unless
the action is not only a violation under a reasonable reading of the statute's terms,
but shows that the company ran a risk of violating the law substantially greater
than the risk associated with a reading that was merely careless.” Safeco Ins. Co.
of Am. v. Burr, 551 U.S. 47, 69, 127 S. Ct. 2201, 2215 (2007). A company’s
interpretation of the statute is objectively unreasonable when courts or the

76

 overseeing regulatory agencies have offered guidance that “might have warned it
away from the view it took.” Id. at 70. Where there is a “dearth of guidance” and a
“less-than-pellucid statutory text,” a misreading of the statute is not objectively
unreasonable. Safeco, 551 U.S. at 70.
Mr. Arroyo seeks punitive damages for a “willful” violation of the FCRA. [Dkt.
1 at Counts IV and V]. Plaintiffs argue that RPS’s failure to disclose was willful
because its alleged policy of requiring a power of attorney was reckless, it failed to
correct its erroneous policy despite multiple opportunities, and the necessary
correction would have been simple. [Dkt. 87 at 10-13]. RPS argues that not
accepting certificates of conservatorship with non-visible impressed seals that
state on their face that they are not to be accepted without a seal cannot be
considered “willful” because of the lack of specification in the statute and the lack
of binding or even apposite case precedent. Plaintiffs reply that there is a disputed
question of fact as willfulness: first, RPS’s failure to accept such certificates is
unsupported by law; second, RPS failed to follow reasonable procedures for
evaluating disclosure requests by failing to escalate Ms. Arroyo’s request and by
informing her that a power of attorney was needed.
The Court finds that, in light of the regulations and facts outlined above,
there is a disputed question of fact as to whether RPS acted “objectively
unreasonably” in failing to provide accurate directions on how to complete Ms.
Arroyo’s request on behalf of Mr. Arroyo, including what additional information or
documentation would be required to complete the request, and how to submit such
information. Therefore, the Court denies the parties’ motions for summary
77

 judgment as to the Plaintiff’s willfulness claims for the period from June 30, 2016
to November 18, 2016, but otherwise grants RPS summary judgment as to
Plaintiffs’ willfulness claim.
E. Fair Housing Act Disparate Impact on the Basis of Disability
In Count II, brought on behalf of all Plaintiffs, Plaintiffs allege that RPS’s file
disclosure “policy” of refusing to provide disclosures to conservators has a
disparate impact on disabled individuals in violation of the Fair Housing Act, 42
U.S.C. §§ 3601 et seq.
Liability may be established under the Fair Housing Act where a “practice”
“actually or predictably results in an disparate impact on a group of persons or
creates, increases, reinforces, or perpetuates segregated patterns because of race,
color, religion, sex, handicap, familial status, or national origin.” 24 C.F.R. §
100.500(a). The Second Circuit “evaluate[s] claims that a defendant discriminated
‘because of’ a disability under the burden shifting framework established in
McDonnell Douglas Corp. v. Green, 411 U.S. 792…’” the same three-part burden
shifting framework laid out for the Plaintiffs’ previous FHA claims. Rodriguez v.
Vill. Green Realty, Inc., 788 F.3d 31, 40 n.11 (2d Cir. 2015) (quoting Mitchell v. Shane,
350 F.3d 39, 47 (2d Cir.2003) (FHA case)).
Although “statistical proof almost always occupies center stage in a prima
facie showing of a disparate impact claim,” Robinson v. Metro-North Commuter
R.R., 267 F.3d 147, 160 (2d Cir. 2001), to inevitably require statistical proof when a
policy categorically applies to a protected class would be to equate what is real
with what is measured (not even with what is measurable!). Thus, in Cripe v. City
78

 of San Jose, 261 F.3d 877, 889-90 (9th Cir. 2001), the Ninth Circuit found that the
plaintiffs had shown a prima facie case without statistical analysis for purposes of
the Americans with Disabilities Act where the City’s policy required all police
officers to serve in a beat-patrol assignment before obtaining a specialized
assignment, a policy which rendered the class of disabled plaintiffs categorically
ineligible for specialized assignments. Statistics are not necessary if a challenged
policy categorically applies to a protected class. See 24 C.F.R. § 100.500(a)
(disparate impact liability may apply where a policy “predictably results in a
disparate impact”).
But a plaintiff must “identify the targeted practice with sufficient
particularity… that defendants have adequate notice of precisely what actions” are
at issue. Rodriguez v. Bear Stearns Companies, Inc., No. 07-CV-1816 (JCH), 2009
WL 5184702, at *6 (D. Conn. Dec. 22, 2009). In Inclusive Communities Project, Inc.,
the Supreme Court held, “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.” 135 S.Ct. at 2523. It reasoned that without such
identification, defendants may be “held liable for… disparities they did not create.”
Id. What is true for claims based on a statistical disparity is even more true for
claims based on a policy’s categorical application to a protected class: if plaintiffs
cannot point to a specific policy, then there is no basis for deducing that the policy
categorically applies to a protected class, and so there is no basis for the claim at
all.

79

 Plaintiffs characterize RPS’s process as: “policies and practices of (1)
refusing to allow court-appointed conservators or guardians to receive the
consumer file of the individual subject to the conservatorship or guardianship; (2)
requiring that third-parties, including court-appointed conservators or guardians,
submit a “power of attorney” executed by the consumer in order to receive the
consumer file; and/or (3) requiring that court-appointed conservators or guardians
provide more onerous documentation of their authority than an individual holding
a power of attorney designated by a consumer in order to request and receive a
consumer file.” [Dkt. 1 at ¶165].
The Court finds that the undisputed facts have demonstrated that RPS does
not implement the first two policies but does implement the third. RPS processes
the disclosure of consumer files to third-party legal guardians acting on the
consumer’s behalf. [Dkt. 118 at ¶38]. To protect consumer privacy in the situation
where a third party is seeking a copy of a consumer’s file, RPS’s written policies
generally require a notarized power of attorney, the consumer’s name, proof of the
address to where the disclosure should be mailed, and confirmation of the last four
digits of the consumer’s Social Security number. Id. (citing [Dkt. 114-6 (Barnard
Decl.)] at ¶ 8 & Ex. A at 3). But, based on RPS’s written authentication policy, in
“any scenario” where those requirements cannot be fulfilled, the RPS employee
who is handling the file disclosure request is required to escalate the request to a
“supervisor.” [Dkt. 114-1 ¶39]. A situation in which a consumer is disabled and
cannot execute a power of attorney would require adjustment of third-party

80

 authentication process and supervisory review. Id. (citing [Dkt. 114-6] at ¶ 13 & Ex.
A at 3).
The Court finds that Plaintiffs have shown sufficient evidence of a disparate
impact to prove a prima facie case. The relevant comparison is between those
disabled individuals under a conservatorship who requested disclosure from RPS
and all persons who requested disclosure through a third-party. Conserved
persons are categorically unable to execute powers of attorney.24 As RPS
acknowledges in its statement of facts, all cases in which a consumer is disabled
such that she cannot execute a power of attorney “require… supervisory review,”
adding an additional step to the process that conserved persons face. [Dkt. 114-1
¶39].
Next, the Court considers whether RPS’s third party authentication policy
serves a legitimate and statutorily required interest of safeguarding consumer
policy. RPS argues that its requirement that a Connecticut conservator submit a
valid certificate of conservatorship is mandated by the FCRA, 15 U.S.C. §
1681h(a)(1), and, as explained above, the Court agrees. [Dkt. 114 at 31].

24

A probate court may only order an involuntary conservatorship if it is the least
restrictive means of intervention to assist the individual in managing his or her
affairs and caring for him or herself. Conn Gen. Stat. § 45a-650. Because a power
of attorney is significantly less restrictive than a conservatorship, a person with
the mental capacity to execute a power of attorney should not be involuntarily
conserved. A valid durable power of attorney that a person signed before being
conserved may, after October 1, 2016, remain valid as well. See Conn. Gen. Stat. §
45a-650.

81

 The Court therefore considers whether there exists a less-discriminatory
alternative to requiring more onerous documentation from conservators. Plaintiffs
argue that there is: making the fully visible seal requirement part of a clear
documentation

policy

and

readily

communicating

that

requirement

to

conservators. [Dkt. 126 at 40-42]. But this proposed alternative is not actually an
alternative to the RPS’s conservator documentation requirement—instead, it is a
proposed addition to RPS’s conservator documentation requirement.
Or, to put it another way: this proposed alternative demonstrates that
Plaintiffs’ targeted practice is not really RPS’s documentation requirement, but
instead RPS’s communication about its documentation requirement. And Plaintiffs
have not provided sufficient evidence that RPS’s communication about its
conservatorship documentation requirement “actually or predictably results in a
disparate impact:” Plaintiffs have not provided statistical evidence that RPS
invariably fails to communicate its conservatorship requirement, and Plaintiffs
have not provided evidence that RPS has a policy that would mean that it would
invariably fail to communicate its policy on conservatorship documentation. On
the other hand, RPS has provided evidence that, at least once, in November 2016,
it did attempt to communicate its conservatorship documentation requirement.
[Dkt. 114-1 ¶¶51-53].

25

Therefore, Plaintiffs have not demonstrated an “actual or

predictable” disparate impact.

RPS also argues that a disparate impact claim may not be based on a “one
isolated decision.” [Dkt. 140 at 12], [Dkt. 114 at 30] (citing Reidt v. Cty of
Trempealeau, 975 F.2d 1336, 1341 (7th Cir. 1992)). Plaintiffs are not aware of any
25

82

 The Court understands Mr. Arroyo’s frustration with RPS’s 2016 process.
However, in light of the undisputed facts, Plaintiffs do not state an FHA claim for
disparate impact based on disability. The Court therefore grants RPS’s motion for
summary judgment as to this claim.
F. Fair Housing Act Disparate Treatment on the Basis of Disability.
In Count II of the Complaint, Plaintiffs claim that RPS intentionally
discriminated against Mr. Arroyo on the basis of his disability.
The Second Circuit “evaluate[s] claims that a defendant discriminated
‘because of’ a disability under the burden shifting framework established in
McDonnell Douglas Corp. v. Green, 411 U.S. 792…’” the same three-part burden
shifting framework laid out for the Plaintiffs’ previous FHA claims.” Rodriguez v.
Vill. Green Realty, Inc., 788 F.3d 31, 40 n.11 (2d Cir. 2015) (quoting Mitchell v. Shane,
350 F.3d 39, 47 (2d Cir.2003) (FHA case)). Where plaintiffs have not “submitted any
evidence demonstrating that the non-discriminatory reasons articulated by [the
defendant] for [the challenged decisions] were mere pretexts,” summary judgment
is warranted. Lee v. ITT Standard, 268 F. Supp. 2d 315, 346 (W.D.N.Y. 2002), report

other conserved individual who has requested a file disclosure from RPS, and RPS
also has no record of any other conserved individual requesting a file disclosure.
[Dkt. 114 ¶45] [Dkt. 126-3 (Barnard Dep.) 120:15-124:24]. But the facts of Reidt
demonstrate its un-persuasiveness in this circuit: Plaintiff Debra Reidt alleged that
her employer, a sheriff’s department, violated Title VII when it denied her
application to a certain position because she was a woman. The court found that
the decision was an “isolated” one because she was the only woman who applied
to the position. Id. at 1339. This kind of reasoning—if a discriminatory policy
discourages enough individuals in a protected class, then it is not discriminatory—
has been rejected by the Second Circuit. See E.E.O.C., 186 F.3d at 119.
83

 and recommendation adopted in part sub nom. Estate of Lee v. ITT Standard, 268
F. Supp. 2d 356 (W.D.N.Y. 2003) (granting motion for summary judgment)
Here, it is undisputed that RPS had in place written policies and procedures
for processing consumer file disclosure made by third parties. [Dkt. 114-1 ¶38]. It
is undisputed that RPS processed Mr. Arroyo’s claim in the way that it did because
the claim was brought by a third party, Ms. Arroyo. Id. ¶¶41, 46, 48-50. It is also
undisputed that RPS’s decision to deny Ms. Arroyo access to Mr. Arroyo’s
consumer file was based on the documentation submitted, and that the decision
was made after multiple levels of review by the legal and compliance teams. Id.
¶¶44, 46, 48-50, 52. Plaintiffs have not presented any evidence that RPS’s
application of its third-party disclosure guidelines was a mere pretext, and in fact,
has not opposed their motion for summary judgment on this claim. Therefore, the
Court grants RPS summary judgment as to this claim.
G. Fair Housing Act Failure to Accommodate Claim
In Count III, brought on behalf of the Arroyos, Plaintiffs allege that RPS
refused to make a “reasonable accommodation” to allow Ms. Arroyo access to Mr.
Arroyo’s consumer file in violation of 42 U.S.C.§ 3604(f)(2) and (f)(3).
“To prove a failure-to-accommodate claim, a plaintiff must show (1) that the
plaintiff or a person who would live with the plaintiff had a handicap within the
meaning of § 3602(h); (2) that the defendant knew or reasonably should have been
expected to know of the handicap; (3) that the accommodation was likely
necessary to afford the handicapped person an equal opportunity to use and enjoy

84

 the dwelling; (4) that the accommodation requested was reasonable; and (5) that
the defendant refused to make the requested accommodation.” Olsen v. Stark
Homes, Inc., 759 F.3d 140, 156 (2d Cir. 2014).
Plaintiffs argue that Mr. Arroyo had a handicap, that RPS knew of the
handicap, that the accommodation was likely necessary to afford him equal access
to housing, that his requested accommodation, that Ms. Arroyo as conservator
request and receive Mr. Arroyo’s consumer file on his behalf, was reasonable, and
that RPS refused the accommodation. [Dkt. 87 at 14-18]. Defendants argue
Plaintiffs’ reasonable accommodation claim fails for two reasons. First, Defendants
argue that Plaintiffs fail to show that, but for the accommodation, they likely were
denied an equal opportunity to enjoy the housing of their choice, as (a) RPS has
no authority to override any housing decision by WinnResidential and (b) Plaintiffs
have not supported their claim with any testimony from WinnResidential. [Dkt. 114
at 32]. Second, Defendants argue that Plaintiffs did not request a reasonable
accommodation, because the request that RPS ignore the requirement of a fully
visible seal on the conservatorship form is not reasonable. [Dkt. 114 at 33]. After
evaluating the evidence, the Court grants summary judgment on the basis of RPS’s
second argument.
“Requested accommodations are reasonable where the cost is modest and
they do not pose an undue hardship or a substantial burden on the housing
provider.” Olsen v. Stark Homes, Inc., 759 F.3d 140, 156 (2d Cir. 2014). RPS argues
that Mr. Arroyo’s request that RPS ignore its requirement of a fully visible seal on
the conservatorship form is not reasonable, as the authenticating features are

85

 there for the protection of the conserved person and are legally required. [Dkt. 114
at 33] and [Dkt. 140 at 12-14]. Plaintiffs respond that a fully visible seal on the
conservatorship form is not necessary to establish “proper identification,” which
is all that is required under the FCRA, and that the seal on the forms Ms. Arroyo
submitted was visible enough. As discussed above, the Court finds that an
impressed seal on the conservatorship form is necessary to establish “proper
identification” under the FCRA, so the request to waive that requirement is not
reasonable. The Court therefore grants RPS summary judgment on this claim.
H. CUTPA Claim
Plaintiffs plead a claim under CUTPA, Conn. Gen. Stat. § 42-110b(a) in Count
VI of the Complaint.
CUTPA prohibits “unfair methods of competition and unfair or deceptive
acts or practices in the conduct of any trade or commerce.” See Conn. Gen. Stat. §
42-110b. A CUTPA claim may be brought by “[a]ny person who suffers any
ascertainable loss of money or property.” See Conn. Gen. Stat. § 42-110g(a). When
analyzing the first element, whether plaintiffs alleged an unfair act, the Court must
apply the “cigarette rule” which considers whether the act: (1) “offends public
policy as it has been established by statutes, the common law, or otherwise”; (2)
is “immoral, unethical, oppressive, or unscrupulous”; or (3) “causes substantial
injury to consumers.” See Harris v. Bradley Memorial Hosp. and Health Ctr., Inc.,
296 Conn. 315, 351 (2010).
All three criteria do not need to be satisfied to support a finding of
unfairness. A practice may be unfair because of the degree to which it
meets one of the criteria or because to a lesser extent it meets all three
86

 . . . Thus a violation of CUTPA may be established by showing either
an actual deceptive practice or a practice amounting to a violation of
public policy.
Caires v. JP Morgan Chase Bank, N.A., 880 F. Supp. 2d 288, 299 (D. Conn.
2012) (internal citation and quotation marks omitted). Consumers may prove a
CUTPA violation using either the unfairness standard or the deception standard.
See, e.g., Caldor, Inc. v. Heslin, 215 Conn. 590, 577 A.2d 1009 (1990) (deception);
Conaway v. Prestia, 191 Conn. 484, 493, 464 A.2d 847, 852 (1983) (unfairness).
Plaintiffs move for summary judgment on the basis that undisputed facts
establish that RPS’s CrimSAFE product constitutes an unfair practice because it is
against public policy, causes substantial injury, and facilitates discrimination, [Dkt.
118 at 32-42], and that RPS’s file disclosure policies violate the FCRA. [Dkt. 87 at
18 - 19]. RPS moves for summary judgment on the basis that (1) Plaintiffs’ claim
based on CrimSAFE must fail because there is no evidence that it has engaged in
“unfair” or “deceptive acts; and (2) Plaintiffs’ claim on the basis of file disclosure
must fail because file disclosure is not part of “trade” or “commerce,” and there is
no evidence of any damages alleged caused by RPS. [Dkt. 118 at 36-38]. RPS
further opposes Plaintiffs’

CrimSAFE claim on the grounds that (a) Plaintiffs

cannot prove a violation of the FHA, (b) RPS does not “facilitate” discriminatory
conduct, and (c) the Connecticut state legislature recently failed to pass a 2019
legislative proposal to “limi[t] criminal records lookback period that a landlord may
use when evaluating the housing application of a prospective tenant,” was
defeated in the Connecticut state legislature. CT. S.B. No. 54 (2019).[Dkt. 129 at 4446].
1. File Disclosure
87

 The Court denies the parties’ motions for summary judgment as to Plaintiff’s
CUTPA claim on the basis of file disclosure, as there is a disputed question of
material fact regarding whether RPS’s file disclosure policies did violate 15 U.S.C.
§ 1681g. And, as the Court has previously stated, RPS’s conduct is sufficiently
related to its business to established relationship to trade or commerce. See Dkt.
41 at 32-33 (Order on Mot. to Dismiss); Nastro v. D’Onofrio, 263 F. Supp. 2d 446,
457-58 (D. Conn. 2003) (denying motion to dismiss CUTPA claim where defendants’
property transfers were sufficiently related to their underlying business to
establish a relationship to trade or commerce); see also Macomber v. Travelers
Prop. & Cas. Corp., 261 Conn. 620, 643 (2002) (explaining that CUTPA does not
require a consumer relationship). Finally, the Court has established that there is a
question of material fact as to whether Plaintiffs suffered any damages from RPS’s
actions in not disclosing Mr. Arroyo’s file. See supra Section III.F.
2. CrimSAFE
i. Public Policy & Facilitation
The Court finds that there remains a material question of fact as to whether
RPS violated CUTPA “through its CrimSAFE product, violating the Fair Housing
Act to the detriment of housing applicants with criminal records, who are
disproportionately likely to be African American or Hispanic.” See [Dkt. 1 (Compl)
¶ 226.g.].
Conduct that violates the FHA offends Connecticut public policy and is
actionable under CUTPA. See Green v. Konover Residential Corp., No. 95CV1984,
1997 WL 736528, at *7 (D. Conn. Nov. 24, 1997) (“The Connecticut courts have read

88

 CUTPA broadly enough to encompass the claims of plaintiffs, which include . . .
violations of the Fair Housing Act by virtue of defendants' discriminatory repair
practices.”). Connecticut has adopted the public policy goals of the FHA. In its
Constitution, Connecticut provides that “no person shall be… subjected to
segregation or discrimination in the exercise or enjoyment of his civil … rights
because of … race, color, ancestry or national origin.” Connecticut Constitution,
Sec. 20. And, in 1990, Connecticut passed comprehensive fair housing legislation
modelled after the FHA. See Conn. Gen. Stat. § 46a-64b, et seq; see also Statement
of Senator Blumenthal, 33 S. Proc., Pt. 11, 1990 Sess. 3494 (“[t]his is landmark
legislation ... that sets out a separate fair housing act with all the standards and
assurances that exist under Federal law. Indeed, it incorporates the federal
standards into our state statute ....”). Connecticut courts construct Connecticut’s
FHA consistent with the federal courts’ interpretation of the analogous provisions
of the FHA. Comm’n on Human Rights & Opportunities v. Savin Rock Condo. Ass’n,
Inc., 273 Conn. 373, 384-85, 870 A.2d 457, 463 (2005).
The Court cannot determine as a matter of law that RPS did or did not violate
the FHA. See Section III.B., supra. For the same reasons that there is a question of
whether RPS violated the FHA, there is a question of whether RPS violated CUTPA
as a matter of public policy or facilitation of housing providers’ discrimination:
CrimSAFE may be, but is not necessarily as a matter of law, a proximate cause of
housing discrimination against African Americans and Latinos, including Mr.
Arroyo.

89

 Finally, RPS argues that the Connecticut state legislature recently failed to
pass a 2019 legislative proposal to “limi[t] criminal records lookback period that a
landlord may use when evaluating the housing application of a prospective
tenant.” [Dkt. 129 at 44-46] (citing CT. S.B. No. 54 (2019)). But, of course, “it is at
best treacherous” to rely on “congressional silence,” and there are any number of
reasons such a measure might have failed to pass. United States v. Wells, 519 U.S.
482, 496 (1997).
ii. Substantial Injury
For the same reason the Court finds that there is a question as to RPS’s
proximate cause of housing denials generally, see supra Section II.B.1.i, the Court
finds that there is a question of whether RPS caused Mr. Arroyo’s housing denial.
Therefore, the Court denies Plaintiffs’ motion for summary judgment as to this
claim.
iii. Ascertainable Loss
Because the Court denies Plaintiffs’ motion for summary judgment as to its
CUTPA claim based on CrimSAFE on other grounds, the Court does not address
the question of ascertainable loss.

I. CFHC’s Damages Claim
Alleging

that

a

particulardefendant’s

actions

have

“frustrated

the

organization [plaintiff]’s [services], with a consequent drain on resources” suffices
to allege organizational injury-in-fact.” Havens Realty Corp. v. Coleman, 455 U.S.
363, 369 (1982).

The CFHC seeks compensatory damages consisting of
90

 “frustration of mission” and “diversion of resources.” RPS moves for the Court to
grant it summary judgment dismissing these claims. RPS argues that the claims
are not compensable because (1) CFHC’s contemplated educational marketing
campaign is prospective and undeveloped; (2) CFHC has not demonstrated the
link between the diverted resources and RPS; and (3) any damages have been
offset by third-party awards. [Dkt. 114 at 38-44]. CFHC responds that prospective
damages may be compensated, that they have provided sufficient evidence to put
into question RPS’s responsibility for the diversion, and that third-party grants are
“collateral sources” that should not be setoff against RPS’s damages. [Dkt. 126 at
47-53]. The Court agrees with the Plaintiffs.
In the Second Circuit, an organization’s expenses for investigation of a
particular defendant’s conduct and advocacy against that particular defendant,
including litigation expenses, demonstrate injury-in-fact through diversion of
resources. See Mental Disability Law Clinic, Touro Law Ctr. v. Hogan, 519 F. App'x
714, 717 (2d Cir. 2013) (citing Ragin v. Harry Macklowe Real Estate Co., 6 F.3d 898,
904 (2d Cir.1993)) (affirming litigation expenses demonstrate diversion-ofresources injury in fact); Olsen v. Stark Homes, Inc., 759 F.3d 140, 158 (2d Cir.
2014) (affirming that investigation and advocacy on behalf of specific clients
demonstrates diversion-of-resources injury in fact). But, “there must be evidence
directly tying these damages to the defendant’s alleged wrongdoing.” Miami Valley
Fair Hous. Ctr., Inc. v. Connor Grp., No. 3:10-cv-83, 2015 WL 853193, at *9 (S.D. Ohio
Feb. 26, 2015) (citing Ragin, 6 F.3d at 909).

91

While CFHC may have received

 compensation, the sufficiency of that compensation is not established, particularly
given the fact that litigation continues long thereafter.
Further, the Fair Housing Act authorizes the Court to award “such affirmative
action as may be appropriate.” 42 U.S.C. § 3613(c)(1); see United States v. Hylton,
944 F. Supp. 2d 176, 194 (D. Conn. 2013), aff'd, 590 F. App'x 13 (2d Cir. 2014).26 The
Court must tailor such relief to vindicate “the statute’s goals of preventing future
violations and removing lingering effects of past discrimination.” United States v.
Space Hunters, Inc., 2004 U.S. Dist. LEXIS 23699, at *22 (S.D.N.Y. Nov. 22, 2004),
aff’d in part, vacated and remanded in part on other grounds, 429 F.3d 416, 421 (2d
Cir. 2005). “To recover, a fair housing organization must establish that
expenditures in education, counseling and/or outreach are necessary to
counterbalance the effects of a defendant's discriminatory practices.” Fair Hous.
of Marin v. Combs, No. C 97-1247 MJJ, 2000 WL 365029, at *4 (N.D. Cal. Mar. 29,
2000). The Fair Housing Act is a remedial statutory scheme designed to rid the
nation of costly and destructive discriminatory housing practices and the work of
entities like CFHC is instrumental to the fulfilment of its objectives.

26

RPS cites two out-of-circuit cases on injury-in-fact for the purposes of
standing for the proposition that CFHC cannot recover damages for its future work.
[Dkt. 114 at 38-39]. (citing Fair Hous. Council of Suburban Philadelphia v.
Montgomery Newspapers, 141 F.3d 71, 77 (3d Cir. 1998), and Chesapeake Climate
Action Network v. Exp.-Imp. Bank of the United States, 78 F. Supp. 3d 208, 232
(D.D.C. 2015)). Neither are persuasive in light of the apposite statutory and incircuit authority.

92

 CFHC claims $82,639.93 in diversion of resources, up to $350,000 for an
educational marketing campaign to inform the public that blanket bans are illegal,
and its attorneys’ fees. [Dkt.114-5 at Ex. H]. Of its $82,639 in claimed diversion
damages, $9,447 is for CFHC’s work advocating for the Arroyos against RPS; the
remainder is for “education and outreach,” “testing,” “testing costs,” “client
work,” and “grant writing.” [Dkt. 114-5 at Ex. G ].
The Court finds that CFHC has introduced sufficient evidence for its
damages claims to go forward. CFHC may recover for prospective damages. See
Hylton, 944 F. Supp. 2d at 194. CFHC has demonstrated clear ties between the
litigation expenses for this case and the $9,447 “CoreLogic” tab of its diversion log
to RPS to support its claim. As to the proposed marketing campaign and the
remainder of the $82,639 in diverted expenses, CFHC has introduced sufficient
evidence to demonstrate a dispute of fact as to whether those expenses “are
necessary to counterbalance the effects of a defendant's discriminatory practices,”
in light of its evidence of RPS’s marketing efforts, its evidence that housing
providers have reached out to it for guidance, [Dkt. 125-21 at 98:19-24], its evidence
that it has changed its public trainings and presentations to account for RPS’s
policies, id. at 97-111, and its evidence that it has consulted with an advertising
company on how such a campaign would be conducted, id. at 140:7-141:21.
Finally, RPS argues that any damages owed to CFHC should be offset by the
$380,000 in grant funding CFHC has received to address criminal record tenant
screening in the housing application process. [Dkt. 114-1 ¶65]. The parties disagree
about whether the grant funds CFHC has received are better characterized as
93

 “recovery” for the same injury or as “collateral sources.” On the one hand, “a
plaintiff may not recover twice for the same injury…. When a plaintiff receives a
payment from one source for an injury, defendants are entitled to a credit of that
amount against any judgment obtained by the plaintiff as long as both payments
represent common damages.” Phelan v. Local 305 of United Ass'n of Journeymen,
& Apprentices of Plumbing & Pipefitting Indus. of U.S. & Canada, 973 F.2d 1050,
1063 (2d Cir. 1992). On the other hand, “[t]he weight of common law authority is
that collateral sources are not deductible from a tort damage award.” Equal
Employment Opportunity Comm'n v. Enterprise Ass'n Steamfitters Local No.
638, 542 F.2d 579, 591 (2d Cir.1976). On this basis, a Connecticut district court
found in a Fair Housing Act Case that the funds a plaintiff received from a state
agency and tax credit “were collateral sources and that the defendants are not
entitled to a set off of those amounts in the event of an award of damages by the
Court in plaintiffs' favor,” although they “ in part, defrayed expenses that are
claimed as damages suffered by plaintiff.” Valley Hous. LP v. City of Derby, No. 06CV-1319 TLM, 2011 WL 2144633, at *2-3 (D. Conn. May 31, 2011).
Here, CFHC did not receive the grant funding from another “source of the
injury,” so Plaintiffs argue that the grant funding is from a “collateral source,” and
does not offset any damages that RPS may owe. This Court finds that reasoning
compelling. Funds allocated for systemic or programmatic endeavors beneficial
to the constituency of the recipient as a whole should not be diverted to the
advocacy on behalf of a single person or subset of the constituency of the entity.

94

 RPS also separately argues that CFHC’s recovery in this case should be
offset by CFHC’s $13,000 recovery from the settlement with WinnResidential. [Dkt.
114 at 42] (citing SUMF 64).27 But, as Plaintiffs point out, this money was not “in
settlement of the administrative action,” but rather represented attorneys’ fees for
CFHC’s representation of the Arroyos, work that has not been claimed as
diversionary damages. [Dkt. 125-13 (Ex. 11 to Pls.’ Opp: Kemple Decl.) at ¶¶ 3-5.].
Therefore, it also does not offset any damages that RPS may owe for, particularly
for efforts which post-dated the services the award was made to compensate.
For the reasons given above, the Court denies RPS summary judgment as
to CFHC’s compensatory damages.
IV.

Conclusion
For the reasons stated above, the Court holds as follows:
Carmen Arroyo does have standing to bring claims under the FHA and under

CUTPA.

27

RPS does not cite any law regarding offsetting settlements, but it is a nuanced
question. To the question of whether non-settlign defendant’s damages should be
offset by a settlement, the Second Circuit applies a three-party standard to
determine: "First, if federal law is neither deficient nor inapplicable, it will apply.
Second, if federal law does not apply, state law does apply, unless, third, state law
would be inconsistent with the Constitution and laws of the United States." Restivo
v. Hessemann, 846 F.3d 547, 582 (2d Cir. 2017). Further, here, CFHC did not sue
RPS and WinnResidential in the same action, so there is less of an argument for
applying any settlement as an offset.
95

 The Court denies both parties summary judgment as to the FHA claims on
the basis of race and disability, and as to the CUTPA claims. These claims will
proceed to trial.
The Court grants RPS’s motion for summary judgment as to the FHA claim
for disparate claim on the basis of disability, the FHA claim for disparate treatment
on the basis of disability, and the FHA failure to accommodate claim. The Court
grants RPS’s motion for summary judgment as to Mr. Arroyo’s FCRA claims for
the period from April 26, 2016 to June 30, 2016, and after November 18, 2016. The
Court denies the parties’ motions for summary judgment as to Mr. Arroyo’s FCRA
claims for the period from June 30, 2016 to November 18, 2016, and also denies the
parties’ motions for summary judgment as to whether its actions were “willful”
under the FCRA for this period. These claims will proceed to trial.
The Court denies RPS summary judgment as to CFHC’s compensatory
damages, so their claims will proceed to trial.
SO ORDERED.
__________/s/______________
Hon. Vanessa L. Bryant
United States District of Connecticut

Dated: August 7, 2020 at Hartford, Connecticut

96