The Inclusive Communities Project Inc v. Texas Department of Housing and Community Affairs et al

Filing 271

MEMORANDUM OPINION finding that plaintiff has not proved a prima facie case on its disparate impact claim. (Ordered by Judge Sidney A Fitzwater on 8/26/2016) (Judge Sidney A Fitzwater)

IN THE UNITED STATES DISTRICT COURT
FOR THE NORTHERN DISTRICT OF TEXAS
DALLAS DIVISION
THE INCLUSIVE COMMUNITIES
PROJECT, INC.,
Plaintiff,
VS.
THE TEXAS DEPARTMENT OF
HOUSING AND COMMUNITY
AFFAIRS, et al.,
Defendants.

§
§
§
§
§ Civil Action No. 3:08-CV-0546-D
§
§
§
§
§
§
§

MEMORANDUM OPINION
On remand from the Supreme Court and the Fifth Circuit, and applying a materially
different (and more onerous) prima facie case burden of proof than the one applied originally,
the court must decide as trier of fact whether plaintiff The Inclusive Communities Project,
Inc. (“ICP”) has proved a prima facie case and may therefore proceed to the next sequential
step of its disparate impact claim under the Fair Housing Act, 42 U.S.C. § 3601, et seq.
(“FHA”). For the reasons that follow,1 the court finds that ICP has not proved a prima facie

1

Pursuant to Fed. R. Civ. P. 52(a)(1), the court sets out its findings of fact and
conclusions of law in this memorandum opinion. Although the court has carefully
considered the record, this memorandum opinion has been written to comply with the level
of detail required in this circuit for findings of fact and conclusions of law. See, e.g., Century
Marine Inc. v. United States, 153 F.3d 225, 231 (5th Cir. 1998) (discussing standards). The
court has not set out its findings and conclusions in punctilious detail, slavishly traced the
claims issue by issue and witness by witness, or indulged in exegetics, parsing or declaiming
every fact and each nuance and hypothesis. It has instead written a memorandum opinion
that contains findings and conclusions that provide a clear understanding of the basis for the
court’s decision. See id.

case, and it dismisses ICP’s disparate impact claim.
I
This is an action by ICP against defendants Texas Department of Housing and
Community Affairs and its Executive Director and board members in their official capacities
(collectively, “TDHCA”) challenging TDHCA’s allocation of Low Income Housing Tax
Credits (“LIHTC”) in the Dallas metropolitan area.2 ICP sought relief for intentional
discrimination based on race, in violation of the Equal Protection Clause of Fourteenth
Amendment (actionable under 42 U.S.C. § 1983) and 42 U.S.C. § 1982. Inclusive Cmtys.
Project, Inc. v. Tex. Dep’t of Hous. & Cmty. Affairs, 860 F.Supp.2d 312, 313 (N.D. Tex.
2012) (Fitzwater, C.J.) (“ICP III”), rev’d, 747 F.3d 275 (5th Cir. 2014), aff’d and remanded,
___ U.S. ___, 135 S.Ct. 2507 (2015). ICP also alleged that TDHCA was liable under §§
3604(a) and 3605(a) of the FHA on a disparate impact theory. Id. Following the entry of
partial summary judgment in ICP’s favor, and a bench trial, the court found in favor of ICP
on its FHA-based disparate impact claim and against ICP on all other claims. Id. at 314.
At the time the court first decided this case on the merits, the Fifth Circuit had not
“determined the legal standards that should be applied in disparate impact housing
discrimination cases.” Inclusive Cmtys. Project, Inc. v. Tex. Dep’t of Hous. & Cmty. Affairs,
747 F.3d 275, 280-81 (5th Cir. 2014) (“ICP V”), aff’d and remanded, ___ U.S. ___, 135
S.Ct. 2507 (2015). Absent controlling precedent from the Supreme Court or the Fifth

2

Intervenor Frazier Revitalization Inc.—who is aligned with TDHCA—has obtained
leave to intervene.
-2-

Circuit, this court essentially followed the approach of the Second Circuit in Huntington
Branch, NAACP v. Town of Huntington, 844 F.2d 926, 939 (2d Cir.), aff’d in part, 488 U.S.
15 (1988), but without engaging in a process of balancing the factors identified in
Metropolitan Housing Development Corp. v. Village of Arlington Heights, 558 F.2d 1283
(7th Cir. 1977). ICP III, 860 F.Supp.2d at 322 n.17. Before trial, the court granted partial
summary judgment holding that ICP had made a prima facie showing that TDHCA had
violated the FHA based on ICP’s disparate impact theory. Inclusive Cmtys. Project, Inc. v.
Tex. Dep’t of Hous. & Cmty. Affairs, 749 F.Supp.2d 486, 499-500 (N.D. Tex. 2010)
(Fitzwater, C.J.) (“ICP II”). At trial, the court decided ICP’s disparate impact claim on the
merits using as a starting point its summary judgment ruling that ICP had established a prima
facie case.
In ICP II the court held that ICP was entitled to summary
judgment establishing that it had made a prima facie showing of
disparate impact. . . . Because ICP has made this showing, the
burden has shifted to defendants to prove by a preponderance of
the evidence that their actions were in furtherance of a
legitimate governmental interest.
ICP III, 860 F.Supp.2d at 322. But as the court explains below, the court’s conclusion that
ICP had established a prima facie case beyond peradventure was reached without the benefit
of the Supreme Court’s decision in this case, and rested on understandings such as that
TDHCA “essentially [did] not contest ICP’s prima facie cases,” ICP II, 749 F.Supp.2d at
498, that “ICP’s prima facie burden [was] not a heavy one,” id. at 499, and that “ICP ha[d]
adduced evidence that [was] uncontested,” id. at 500.

-3-

On appeal, the Fifth Circuit reversed and remanded. ICP V, 747 F.3d 275. It held that
this court had “correctly noted [that a] violation of the FHA can be shown either by proof of
intentional discrimination or by proof of disparate impact.” Id. at 280. But the Fifth Circuit
also noted that it had “not previously determined the legal standards that should be applied
in disparate impact housing discrimination cases,” id. at 280-81; that other circuits “have
applied multiple different legal standards to similar claims under the FHA,” id. at 281; and
that “after [this] court’s decision in this case, [the U.S. Department of Housing and Urban
Development (“HUD”)] issued regulations regarding disparate impact claims under the
FHA,” id. at 282. The panel adopted the burden-shifting approach of the new HUD
regulations for disparate impact claims brought under the FHA, id. at 282, and it remanded
the case for this court “to apply this legal standard to the facts in the first instance,” id. at
283.3
The Supreme Court then granted certiorari to decide whether disparate impact claims
are cognizable under the FHA. Affirming the judgment of the Fifth Circuit, the Supreme
Court held that disparate impact claims are cognizable. Tex. Dep’t of Hous. & Cmty. Affairs
v. Inclusive Cmtys. Project, Inc., ___ U.S. ___, 135 S.Ct. 2507 (2015) (“ICP VI”). Following
the Supreme Court’s remand to the Fifth Circuit, the panel then “remand[ed] this case . . . for
further proceeding consistent with [its] opinion and the opinion of the Supreme Court.”

3

The Fifth Circuit made clear that it was not requiring that this court “must retry the
case,” and that it was leaving “it to the sound discretion of [this] court to decide whether any
additional proceedings are necessary or appropriate.” ICP V, 747 F.3d at 283.
-4-

Inclusive Cmtys. Project, Inc. v. Tex. Dep’t of Hous. & Cmty. Affairs, 795 F.3d 509 (5th Cir.
2015) (per curiam).
This court, in turn, requested that the parties propose how the case should proceed on
remand. Inclusive Cmtys. Project, Inc. v. Tex. Dep’t of Hous. & Cmty. Affairs, 2015 WL
5916220, at *1 (N.D. Tex. Oct. 8, 2015) (Fitzwater, J.). ICP maintained that the briefing
should commence at step two of the three-part burden-shifting approach—that TDHCA4
should be required to address the justification element, followed by ICP’s addressing the less
discriminatory alternative element. Id. at *3. ICP “posit[ed] that the court should adhere to
its summary judgment decision that ICP ha[d] established the first element (prima facie
case).” Id. TDHCA and intervenor Frazier Revitalization Inc. contended that “the court
should reconsider whether ICP ha[d] pleaded and proved a prima facie case,” and that
“TDHCA should not be required to address the justification prong unless ICP meets its
obligation to plead and prove a prima facie case.” Id.
The court concluded “that the interests of justice and fundamental fairness require[d]
that [the court] first consider on the current (or a supplemented) trial record whether ICP
ha[d] established a prima facie case.” Id. This was so because, although the court had
granted partial summary judgment in ICP II holding that ICP had established a prima facie
case on its disparate impact claim, it had done so without the benefit of the Supreme Court’s
opinion; it had not given the prima facie requirement the same emphasis that the Supreme

4

The court uses TDHCA to refer to defendants collectively. ICP actually referred to
defendants.
-5-

Court had given it—for example, the court had stated that ICP’s prima facie burden was not
a heavy one, and it had concluded that ICP need only provide evidence that raised an
inference of discrimination; and the court arguably had “not analyze[d] ICP’s evidence
through the prism of the ‘robust causality requirement’ envisioned by the Supreme Court.”
Id.
The court also reasoned that TDHCA had not had the benefit of the Supreme Court’s
decision, and, therefore, “a pellucid sense of its importance in the equation of proving FHA
disparate impact liability.” Id. at *4. In ICP II the court pointed out that, in opposing ICP’s
summary judgment motion, defendants essentially had not contested ICP’s prima facie cases,
and ICP had established a prima facie case based on evidence that was “uncontested.” Id.
(quoting ICP II, 749 F.Supp.2d at 500). The court reasoned that TDHCA should therefore
be permitted to challenge ICP’s prima facie showing “based on a clearer understanding of
the requirements and consequences of ICP’s establishing a prima facie case.” Id.
Finally, the court concluded:
given the significant developments in this case on appeal, . . .
the interests of justice and fundamental fairness require not only
that ICP’s disparate impact claim be decided anew under the
burden-shifting regimen adopted by HUD and the Fifth Circuit,
but that the court start with whether ICP has established a prima
facie case.
Id.
The parties have submitted their prima facie case briefing and have supplemented the
record, and the court has heard oral argument.

-6-

II
ICP is a non-profit organization that seeks racial and socioeconomic integration in the
Dallas metropolitan area. To that end, ICP assists low-income, predominantly AfricanAmerican families who are eligible for the Dallas Housing Authority’s Section 8 Housing
Choice Voucher program (“Section 8”) in finding affordable housing in predominately
Caucasian, suburban neighborhoods. Because under the LIHTC program a development that
receives tax credits cannot refuse housing solely because a person is using a Section 8
voucher, it is important to ICP where the developments are located in the Dallas metropolitan
area.
Under § 42 of the Internal Revenue Code (“I.R.C.”), the government provides tax
credits (here, LIHTC) that a state (here, Texas) distributes to developers through a designated
state agency (here, TDHCA). See Tex. Gov’t Code Ann. § 2306.053(b)(10) (West 2008).
Developers apply to TDHCA for tax credits, which can be sold to finance construction of a
housing project. ICP alleges that TDHCA’s allocation of LIHTC in the Dallas metropolitan
area has a disparate impact on the location of low-income housing in the area. ICP
complains of TDHCA’s allocation of two different types of tax credits: the 4% credit, and
the 9% credit.
Nine percent tax credits are distributed on an annual cycle and are generally
oversubscribed. Certain federal and state laws dictate, at least in part, how TDHCA selects
applications for 9% tax credits. First, I.R.C. § 42 requires that the designated state agency
adopt a “Qualified Allocation Plan” (“QAP”) that prescribes the “selection criteria” for 9%
-7-

tax credits. Id. at § 42(m)(1)(A)-(B). The QAP must include, inter alia, certain criteria and
preferences; otherwise, zero housing credit dollars will be provided. § 42(m)(1)(A)-(C).
Second, the Texas Government Code regulates how TDHCA administers the LIHTC
program: it requires that TDHCA adopt annually a QAP and corresponding manual, id. at
§ 2306.67022, and sets out how TDHCA should evaluate applications, id. at § 2306.6710(a).
First, TDHCA must “determine whether the application satisfies the threshold criteria” in the
QAP. Id. at § 2306.6710(a). Applications satisfying the “threshold criteria” are then
“score[d] and rank[ed]” by a point system that “prioritizes in descending order” ten listed
statutory criteria (also called “above-the-line criteria”), which directly affects TDHCA’s
discretion in creating “selection criteria” in each QAP. Id. at § 2306.6710(b). TDHCA is
also permitted to consider other criteria and preferences (also called “below-the-line”
criteria), but it “lacks discretionary authority to intersperse other factors into the ranking
system that will have greater points than” the statutory criteria. Tex. Att’y Gen. Op. No. GA0208, 2004 WL 1434796, at *6 (2004).
Once the TDHCA adopts a QAP, it submits it to the Governor, who can “approve,
reject, or modify and approve” it. Tex. Gov’t Code Ann. § 2306.6724(b)-(c) (West 2001).
The TDHCA staff then review applications, underwrite applications to determine their
financial feasibility and appropriate level of tax credits, and submit recommendations to the
TDHCA. § 2306.6710(b)(1)(A) & (d), § 2306.6724(e). The TDHCA reviews the staff
recommendations and issues final commitments in accordance with the QAP. Id. at §
2306.6724(e)-(f).
-8-

Unlike 9% tax credits, 4% tax credits are awarded according to non-competitive
criteria, available to applicants on a year-round basis, and generally are not oversubscribed.
In awarding a 4% tax credit, the TDHCA “reviews the application for threshold, eligibility
and then the development is underwritten.” P. Ex. 1 at 20; see also Tex. Gov’t Code Ann.
§ 2306.67021 (West 2001). Applications for the 4% tax credit are not subject to scoring
under the selection criteria and are not decided under a competitive scoring model.
III
The Supreme Court held in ICP VI that disparate impact claims are cognizable under
the FHA. ICP VI, 135 S. Ct. at 2514-15. The Court did not disturb the Fifth Circuit’s
adoption of the HUD burden-shifting regimen, but it prescribed several limitations on
disparate impact liability, and it questioned whether ICP can prove liability in this case. The
Court cautioned that disparate impact liability should not be established “based solely on a
showing of statistical disparity,” id. at 2522; and that it properly results in the removal of
“artificial, arbitrary, and unnecessary barriers,” but does not eschew “valid governmental
policies,” id.
As a result of the Fifth Circuit’s decision adopting the HUD regulations, and the
Supreme Court’s affirmance (without altering the burden-shifting approach), the following
proof regimen now applies to ICP’s disparate impact claim under the FHA:

-9-

We now adopt the burden-shifting approach found in 24 C.F.R.
§ 100.500 for claims of disparate impact under the FHA. See 24
C.F.R. § 100.500. First, a plaintiff must prove a prima facie
case of discrimination by showing that a challenged practice
causes a discriminatory effect, as defined by 24 C.F.R. §
100.500(a). 24 C.F.R. § 100.500(c)(1). If the plaintiff makes a
prima facie case, the defendant must then prove “that the
challenged practice is necessary to achieve one or more
substantial, legitimate, nondiscriminatory interests . . . .” Id. §
100.500(c)(2). If the defendant meets its burden, the plaintiff
must then show that the defendant’s interests “could be served
by another practice that has a less discriminatory effect.” Id. §
100.500(c)(3).
ICP V, 747 F.3d at 282.
In the first step of proving its prima facie case, ICP must “point to [TDHCA’s] policy
or policies causing that [statistical] disparity.” ICP VI, 135 S.Ct. at 2523. “[A] one-time
[housing] decision may not be a policy at all.” Id. at 2523. Importantly, “‘racial imbalance
. . . does not, without more, establish a prima facie case of disparate impact.’” Id. (quoting
Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 653 (1989)). This “robust causality
requirement . . . protects defendants from being held liable for racial disparities they did not
create.” Id. The Court noted that, in cases such as this one, “it may also be difficult to
establish causation because of the multiple factors that go into investment decisions about
where to construct or renovate housing units.” Id. at 2523-24. Accordingly, this court must
“examine with care whether a plaintiff has made out a prima facie case of disparate impact,
and prompt resolution of these cases is important.” Id. at 2523.
“A plaintiff who fails to allege facts at the pleading stage or produce statistical
evidence demonstrating a causal connection cannot make out a prima facie case of disparate
- 10 -

impact.” Id. The Court specifically noted Judge Jones’s specially concurring opinion in ICP
V, in which she observed that “if the ICP cannot show a causal connection between the
Department’s policy and a disparate impact—for instance, because federal law substantially
limits the Department’s discretion—that should result in dismissal of this case.” Id. at 2524
(citing ICP V, 747 F.3d at 283-84). The Court cautioned that, “[u]nlike the heartland of
disparate-impact suits targeting artificial barriers to housing, the underlying dispute in this
case involves a novel theory of liability,” id. at 2522, and that “[t]his case, on remand, may
be seen simply as an attempt to second-guess which of two reasonable approaches a housing
authority should follow in the sound exercise of its discretion in allocating tax credits for
low-income housing.” Id. at 2523.
IV
The heart of ICP’s complaint is the allocation of 9% tax credits for low-income
housing developments.5 ICP alleges that the policy or practice of permitting the TDHCA to
use its discretion in allocating 9% tax credits perpetuates racial segregation, in violation of
the FHA. ICP complains of the TDHCA’s exercise of discretion on the whole and in specific
ways. Because ICP must identify a specific policy or practice that causes a statistically
significant disparity in the location of low-income housing, see ICP VI, 135 S. Ct. 2507, the
court will separately address ICP’s challenges to TDHCA’s discretion on the whole and in

5

The court addresses ICP’s challenges to the 4% program and the 9% program
separately because the allocation of each type of credits is done separately and the programs
are administered in different ways.
- 11 -

specific contexts.
A
The court first decides whether TDHCA’s exercise of discretion in deciding how to
allocate 9% tax credits is a specific, facially neutral policy sufficient to establish a prima
facie case of disparate impact liability.
1
ICP alleges that the exercise of TDHCA’s discretion in the 9% tax credit program has
caused the exclusion of all but 6% of the LIHTC units from Caucasian areas in the City of
Dallas and all but 26% of the LIHTC units from Caucasian areas in the Dallas metropolitan
area. Between 1999 and 2008, applications for 9% tax credits for units located in minority
tracts had a 41% approval rate, while applications for units located in Caucasian tracts had
a 21% approval rate. As evidence of disparate impact, ICP points to TDHCA’s decisions to
deny or approve applications for specific applications for 9% tax credits between 1999 and
2008. ICP alleges that the use of TDHCA discretion in the aggregate resulted in an approval
rate for units located in Caucasian areas of nearly half the approval rate for units located in
minority areas.
TDHCA responds that ICP has not identified a specific, facially-neutral policy that
caused a racially disparate impact sufficient to establish a prima facie case under the FHA.
It argues that discretion itself is not a policy; ICP has not accounted for the actions, policies,
or preferences of third parties—such as Congress, the Texas Legislature, developers, and
local communities—that impact the location of LIHTC units; ICP relies on several specific
- 12 -

exercises of discretion—such as the use of set-asides for elderly units—but does not allege
any facts demonstrating a statistical disparity caused by that use of discretion; and ICP
challenges statewide policies but claims only localized statistical disparity. TDHCA also
maintains that ICP is not actually challenging TDHCA’s use of discretion; rather, ICP is
seeking to require TDHCA to use its discretion in a specific way, that is, to provide
desegregated housing opportunities for ICP’s clients and to remedy historical housing
segregation.
ICP replies that discretion can be a policy sufficient to state a disparate impact claim,
and that federal laws establishing a preference for providing low-income housing in minority
areas did not cause the statistical disparity in low-income housing.
2
To prove a disparate impact claim, the plaintiff must first identify a facially-neutral
policy that has resulted in the disparate impact. See ICP VI, 135 S.Ct. at 2522-24; see also
City of Miami v. Bank of Am. Corp., 2016 WL 1072488, at *4-5 (S.D. Fla. Mar. 17, 2016)
(applying ICP VI).6 As the Supreme Court explained, disparate impact claims seek to
remove impermissible barriers; therefore, ICP must identify a specific policy that has created
barriers to fair housing. See City of Los Angeles v. Wells Fargo & Co., 2015 WL 4398858,
at *8 (C.D. Cal. July 17, 2015) (applying ICP VI). ICP must affirmatively identify a specific

6

As the Supreme Court explained, disparate impact cases “interpreting Title VII and
the ADEA provide essential background and instruction” in this case. ICP VI, 135 S.Ct. at
2518.
- 13 -

policy that produced a disparate impact, rather than point to a lack of policy that caused it.
See id. Nor is it sufficient for ICP to point to a generalized policy that leads to a disparate
impact. Meacham v. Knolls Atomic Power Lab., 554 U.S. 84, 100 (2008) (quoting Smith v.
City of Jackson, 544 U.S. 228, 241 (2005)). This is not a “trivial burden.” Id.
The offending policy or practice must be specifically identified for at least two
reasons. First, the court cannot evaluate whether the offending policy caused a statistically
significant disparity without knowing the offending policy with precision. Meacham, 554
U.S. at 100 (explaining that plaintiff must specifically identify the offending policy so that
the court can ensure that employers are not liable for “the myriad of innocent causes that may
lead to statistical imbalances.”); Hanrahan v. Blank Rome LLP, 142 F.Supp.3d 349, 354
(E.D. Pa. 2015) (dismissing disparate impact claim where plaintiff failed to identify and
isolate a specific employment policy). Especially where the defendant has combined
discretionary or subjective decision-making with objective criteria, “the plaintiff is in our
view responsible for isolating and identifying the specific . . . practices that are allegedly
responsible for any observed statistical disparities.” Wards Cove, 490 U.S. at 656; see also
Anderson v. Douglas & Lomason Co., 26 F.3d 1277, 1283-85 (5th Cir. 1994). If the plaintiff
does not identify a specific policy or practice, the court cannot determine whether that policy
or practice has created a barrier to fair housing or has resulted in a statistical disparity. See
Wells Fargo, 2015 WL 4398858, at *8 (citing ICP VI, 135 S.Ct. at 2523). In Anderson, an
employment discrimination case, the plaintiff asserted that the cumulative effects of the
defendant’s employment practices caused a racial disparity in promotions. Anderson, 26
- 14 -

F.3d at 1284. The Fifth Circuit affirmed the dismissal of the plaintiff’s disparate treatment
impact claim and noted that the plaintiff had failed to identify a specific policy that caused
the racial disparity. Id.
Likewise, ICP has failed to point to a specific, facially neutral policy that purportedly
caused a racially disparate impact. By relying simply on TDHCA’s exercise of discretion
in awarding tax credits, ICP has not isolated and identified the specific practice that caused
the disparity in the location of low-income housing. Like the plaintiff in Anderson, ICP has
pointed to the “cumulative effects” of TDHCA’s decision-making process over a multi-year
period. ICP cannot rely on this generalized policy of discretion to prove disparate impact.
Identification of a specific policy or practice is also essential for determining how to
remedy the disparate impact. Although the court need not now decide what remedy, if any,
is required in this case, consideration of the potential remedies for disparate impact claims
demonstrates what kinds of policies are offending. The Supreme Court requires that
“[r]emedial orders in disparate-impact cases . . . concentrate on the elimination of the
offending practice, and courts should strive to design race-neutral remedies.” ICP VI, 135
S.Ct. at 2512. The Court cautioned that lower courts should be careful not to “impose racial
targets or quotas,” because doing so “might raise difficult constitutional questions.” Id. In
sum, any remedy must focus on removing the offending practices and must be race-neutral.
ICP has not identified any offending policy for which there exists a constitutionallysound remedy. To remedy disparate impact, the court must craft a race-neutral remedy that
removes the offending practice. For example, if ICP complained that TDHCA’s use of bonus
- 15 -

points for certain types of housing developments had a disparate impact, the court could
order a remedy that adjusted how the TDHCA awarded such bonus points. Here, ICP has
not identified any barriers to housing that the court can remove. Although ICP complains
of TDHCA’s exercise of discretion in housing decisions, it does not ask the court to prohibit
TDHCA from using its discretion; rather, it asks the court to require that TDHCA exercise
its discretion in a specific way: to desegregate housing. For example, ICP implies that
TDHCA should be required to award bonus points for developments that provide
opportunities for desegregation and to subtract points for developments in areas of slum and
blight.7 Because ICP has not sufficiently identified a specific, facially-neutral policy that has
caused a statistical disparity, the court cannot fashion a remedy that removes that policy.
B
ICP’s disparate impact claim must be dismissed for an additional reason: regardless
of the label ICP places on its claim, it is actually complaining about disparate treatment, not
disparate impact.8

The purpose of disparate impact liability is to root out a facially neutral

policy that has an unintended discriminatory result. Phillips v. Cohen, 400 F.3d 388, 398
(6th Cir. 2005) (citing Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 986 (1988)). But

7

Additionally, in response to TDHCA’s proposed remedial plan, ICP argued that
TDHCA “can and should use its discretion to remedy the violation” to “accomplish the
remedial purpose of the plan.” ICP 6/18/12 Resp. to Ds. Proposed Remedial Plan at 10.
8

ICP originally asserted claims for disparate treatment and disparate impact, both
rooted in the TDHCA’s allocation of tax credits for low-income housing. The court ruled
against ICP on its disparate treatment claim. ICP III, 860 F.Supp.2d at 314.
- 16 -

a claim for intentional discrimination is evaluated under the disparate treatment framework,
which requires a showing of targeted discrimination. Wards Cove, 490 U.S. at 666. Where
the plaintiff establishes that a subjective policy, such as the use of discretion, has been used
to achieve a racial disparity, the plaintiff has shown disparate treatment. See, e.g., Johnson
v. Metro. Gov’t of Nashville & Davidson Cnty., 2008 WL 3163531, at *4-6 (M.D. Tenn.
Aug. 4, 2008) (citing Phillips and Wards Cove). On the other hand, where the plaintiff
establishes that the existence of the policy itself, rather than how the policy is applied,
resulted in a racial disparity, the plaintiff has shown disparate impact. Id.
These examples are illustrative. In Johnson the plaintiffs alleged that the defendantemployer applied a promotion method that had an adverse effect on white male employees.
Id.

Rather than challenge the method itself, plaintiffs challenged the results of the

employer’s application of the method to them, and argued that the method was used to avoid
promoting white male employees. Id. The court held that plaintiffs’ claim was one for
disparate treatment because they argued that the method was used to intentionally
discriminate against them, even if they did not label their claim as such.
Similarly, in Reminder v. Roadway Express, Inc., 215 Fed. Appx. 481 (6th Cir. 2007)
(per curiam), the plaintiff asserted a disparate impact claim, arguing that his employer’s
method of downsizing resulted in a disproportionate termination of employees over age 40.
Id. at 485.

The Sixth Circuit held that the plaintiff “clearly” asserted a disparate

treatment—not disparate impact—claim despite its label because the plaintiff did not identify
a specific neutral policy but rather the impact of the downsizing methods that targeted
- 17 -

persons over age 40. Id.
ICP likewise complains of disparate treatment. Although ICP labels its claim as one
for disparate impact, a fair reading of ICP’s arguments reveals that it is complaining about
the results of TDHCA’s discretion, not the existence of the discretion. See, e.g., ICP 2/25/16
Prima Facie Reply Br. at 10 (“This longstanding result [of TDHCA’s decisions] is consistent
with the use of a race based cap on the percentage of LIHTC family units to be allowed in
Caucasian areas.”); at 15-16 (arguing that TDHCA’s stated reasons for its decisions are
pretextual). If ICP were challenging the existence of TDHCA’s discretion rather than how
the discretion is used, ICP would seek to enjoin that discretion and to mandate a points-only
system or another wholly objective method of awarding tax credits. Instead, ICP maintains
that TDHCA’s exercise of discretion should be the means to achieve a specific end: to
provide increased opportunities for desegregated low-income housing. Because ICP is
actually complaining of the results of TDHCA’s exercise of discretion rather than of the
existence of that discretion, ICP has not established a claim for disparate impact.
C
Even assuming arguendo that TDHCA’s use of discretion is a specific, facially neutral
policy, ICP has not established that the exercise of that discretion caused the disparity in the
location of low-income housing.
1
“A disparate-impact claim relying on a statistical disparity must fail if the plaintiff
cannot point to a defendant’s policy or policies causing that disparity.” ICP VI, 135 S.Ct. at
- 18 -

2523. This causality requirement is “robust” to ensure that “defendants do not resort to the
use of racial quotas” to avoid liability for statistical disparities. Id. The plaintiff must
demonstrate that the statistical disparity is caused by the defendant’s policy or policies, rather
than by other factors. Id. (citing CFR § 100.500(c)(1) (2014)). Accordingly, to establish a
prima facie case of disparate impact, ICP must prove facts that plausibly demonstrate a
causal link between the challenged policy or practice and a statistical disparity. Id.; see also
Ellis v. City of Minneapolis, 2015 WL 5009341, at *10 (D. Minn. Aug. 24, 2015) (applying
ICP VI ). The use of discretion “should itself raise no inference of discriminatory conduct.”
Johnson v. Uncle Ben’s, Inc., 965 F.2d 1363, 1369 (5th Cir. 1992) (citing Watson, 487 U.S.
989).
2
ICP has not proved that TDHCA’s policy of allowing discretion in allocating 9 % tax
credits caused a statistically-significant disparity. ICP has established that (1) TDHCA has
discretion in how it allocates 9% tax credits and (2) as of 2008, 26% of LIHTC units in the
Dallas metropolitan area are located in Caucasian areas and 74% are located in minority
areas. ICP has not proved that it was TDHCA’s exercise of discretion—and not something
else—that caused the difference in the number of LIHTC units in each area. Put differently,
ICP has not demonstrated that, had TDHCA not been permitted to exercise any discretion
when allocating 9% tax credits, and had those decisions been made strictly by the points
system, there would be no, or significantly less, disparity in the location of LIHTC units.
Nor has ICP proved what the distribution of LIHTC units or the percentages of
- 19 -

approved 9% applications would have been had TDHCA awarded credits based strictly on
points—had it exercised no discretion. It is therefore possible that, based strictly on points,
the statistical disparity would be the same, lesser, or even greater. For example, ICP adduced
evidence that, in 1999, TDHCA awarded credits to the Rosemont at Arlington Park
development, a family development with 100 units in a predominately minority area of
Dallas, with 92 points. ICP 1/7/16 Prima Facie Br. App. at 728. The highest scoring
application in 1999, with 95 points, was the Columbia Valley Townhomes, a family
development with 160 units in a predominately minority area of Dallas. Id. Accordingly,
had the TDHCA awarded credits based strictly on points, it would have created a greater
disparity in the number of LIHTC units in minority areas and in Caucasian areas: it would
have awarded credits to 160 units in a minority area instead of 100 units. In 2005, the
TDHCA’s awards would have been the same had it awarded credits based solely on points.
Id. at 738. And as TDHCA explained at oral argument, from 2004 to 2008, TDHCA
approved 12 applications for tax credits for low-income housing in minority areas and 12
applications for tax credits for low-income housing in Caucasian areas.
Nor has ICP accounted for other potential causes of the statistical disparity. TDHCA
argues that “ICP does not account for the actions, policies, or preferences of third parties like
Congress, the Texas Legislature, developers, and local communities that impact the location
of LIHTC units.” Ds. 2/4/16 Prima Facie Resp. Br. at 19. It points to other potential causes
of the statistical disparity: the preference under federal law for placement of LIHTC
properties in low-income communities and the corresponding “basis boost” received by
- 20 -

applications for tax credits for developments in low-income communities; the preference
under state law for financial feasibility and community participation; developers’ decisions
regarding the locations of housing projects; and the decisions and preferences of local
governments regarding zoning and approval of specific projects.
ICP replies that the existing statistical disparity cannot be caused by the federal
preference for placement of LIHTC properties in low-income communities because the
statistical disparity predated the federal preference.

And according to ICP, “[t]he

uncontested facts rule out the possibility that either third parties or other factors are causing
the perpetuation of segregation.” ICP 2/25/16 Prima Facie Reply Br. at 18. ICP offers the
example of two applications for 9% tax credits for housing projects in Caucasian tracts that
were eligible for credits but denied. ICP reasons that, because these two projects were not
eligible for the federal “basis boost,” TDHCA’s exercise of discretion was the only cause for
the denial of the applications.
ICP has not proved that TDHCA’s exercise of discretion—and not other
factors—caused the statistical disparity. The Supreme Court recognized in ICP VI “[i]t may
be difficult [for ICP] to establish causation because of the multiple factors that go into
investment decisions about where to construct or renovate housing units.” ICP VI, 135 S.Ct.
at 2524. Therefore, “if the ICP cannot show a causal connection between [TDHCA’s] policy
and a disparate impact—for instance, because federal law substantially limits [TDHCA’s]
discretion—that should result in dismissal of this case.” Id. (citing Jones, J., specially
concurring in ICP V, 747 F.3d at 283-84). ICP has failed to demonstrate that local zoning
- 21 -

rules, community preferences, or developers’ choices did not contribute to the statistical
disparity. For example, TDHCA asserted at oral argument, and ICP does not refute, that no
developers have applied for credits for developments in predominately Caucasian areas in
the City of Dallas since 1995. Accordingly, TDHCA has neither granted nor denied any such
application, and it is not TDHCA’s discretion that has caused the lower number of LIHTC
units in the City of Dallas.
ICP also attempts to causally link to a cumulative statistical disparity TDHCA’s
exercise of discretion in a case-by-case review during an annual-cycle over a multi-year
period. The facts belie ICP’s position. Each year, TDHCA decides whether to approve or
deny applications for 9% tax credits submitted that year. Because there is a fixed number
of credits that TDHCA can award, it must compare the applications during each annual cycle
and decide which ones to award. TDHCA does not, however, compare applications
submitted in one year to applications submitted in a different year. In other words,
TDHCA’s approval of an application in 2000 should have no effect on its denial of an
application in 2003, and vice versa. Accordingly, ICP’s reliance on a statistical disparity that
has accumulated over a multi-year period is misplaced; the disparity cannot be causally
linked to the exercise of discretion in a single year.
Finally, ICP acknowledges that the discretion exercised by TDHCA has changed from
year to year. In other words, the policy to which ICP attributes a cumulative disparity has
in fact changed during that time period. For example, from 1995 to 2011, the QAP permitted
TDHCA to award forward commitments. But TDHCA has not been authorized to award
- 22 -

forward commitments since 2011. The point system set forth in the QAP has also changed
from year to year. Because the challenged policy has changed from year to year in each
QAP, ICP cannot point to a specific policy that caused an aggregate statistical disparity over
a multi-year period. See, e.g., City of Miami, 2016 WL 1072488, at *5 & n.3 (holding that
plaintiff failed to demonstrate disparate impact where it “lump[ed] together eight years of
data” over which time period defendant’s policies had changed).
D
In addition to challenging TDHCA’s general policy of exercising discretion in
awarding tax credits, ICP also challenges several specific ways in which TDHCA exercises
its discretion. For the reasons explained supra at § IV(A)(2), the court holds that TDHCA’s
use of discretion is not a specific, facially neutral policy that can support a claim for disparate
impact. But assuming arguendo that ICP can prove a disparate impact claim based on a
specific exercise of discretion, the court must still decide whether ICP has proved that the
exercise of discretion has caused a statistically significant disparity in the approval of 9% tax
credits. ICP relies on the following specific exercises of discretion: (1) “discretionary
forward commitments in the 9% program”; (2) “other exercises of discretion in taking factors
other than point totals into account in the 9% program”; (3) “discretion in the scoring of 9%
selection criteria”; (4) “the discretionary decision to eliminate taking into account in
individual allocation decisions whether the project would provide desegregated housing
opportunities in connection with court ordered desegregation remedies”; (5) “using discretion
to allocate 9% tax credits for units restricted to elderly only residents”; (6) “discretion in
- 23 -

selecting groups of applications for guaranteed allocations through set asides”; and (7) “the
exercise of discretion to approve rather than disapprove projects that would be located in
areas marked by conditions of slum and blight, high crime rates, noxious industrial uses,
adverse environmental conditions, and concentrations of LIHTC units.” ICP 1/7/16 Prima
Facie Br. at 2.
1
ICP complains of TDHCA’s discretion to award forward commitments to applications
seeking 9% tax credits. According to ICP, each QAP from 1995 to 2011 permitted TDHCA
to award any eligible 9% tax credit application an allocation of credits from the following
year’s allocation rather than from the current year’s allocation. For example, TDHCA could
deny an application for tax credits in 1999 but commit to award the applicant tax credits in
2000. ICP alleges that, although no QAP from 2012 to the present expressly gives TDHCA
discretion to award forward commitments, pertinent regulations governing the award of tax
credits permit TDHCA to exercise such discretion today.
TDHCA responds that it used its discretion to award forward commitments on a caseby-case basis to award credits for projects that were “uniquely deserving” but that did not
have enough points to be awarded credits in the application year; the current QAP does not
allow for forward commitments, and ICP’s complaint is moot; ICP has not identified a policy
of awarding forward commitments sufficient to state a prima facie case; and ICP has not
alleged facts demonstrating a statistical imbalance caused by the approval or denial of
forward commitments.
- 24 -

ICP has not satisfied the robust causality requirement. Although ICP has proved that
TDHCA gave forward commitments to several applications for 9% tax credits, ICP has not
proved a statistical disparity caused by the forward commitments. For example, ICP has not
proved how many applications were awarded forward commitments that would not otherwise
have been awarded tax credits in their application years or in the following year, or what
applications would have been awarded had the applications been denied instead of
forwarded. ICP has not proved how the statistical disparity would have lessened if TDHCA
had no discretion to award forward commitments.9
2
ICP also complains of TDHCA’s discretion to take into account factors other than
point totals, and its discretion in scoring applications for tax credits. TDHCA responds that
ICP has not isolated and identified a specific policy causing a disparity, as required by the
Supreme Court, and that this court therefore cannot craft a remedial order eliminating the
offending policy. TDHCA also argues that what ICP is actually seeking is for the court to
engineer a QAP so as to remedy racial imbalance in housing and to achieve ICP’s ideal
distribution of LIHTC units, and that the Supreme Court has cautioned that such remedies
raise serious constitutional concerns.
ICP has not proved that TDHCA’s exercise of discretion in scoring or in taking other
factors into account has caused a statistically-significant disparity. As discussed supra at §

9

TDHCA also argues that, because the QAP no longer grants TDHCA discretion to
award forward commitments, there is nothing for the court to remedy. The court agrees.
- 25 -

IV(A)(2), ICP has not demonstrated a specific policy that the court could remedy, or that
TDHCA’s discretion to consider factors other than points has caused a statisticallysignificant disparity.
3
ICP complains of TDHCA’s decision not to consider whether a proposed development
would provide desegregated housing opportunities. According to ICP, until 1994, TDHCA
awarded additional points for developments located in Caucasian areas. After 1994, TDHCA
could in its discretion award additional points for desegregated developments, but chose not
to do so.
TDHCA responds that ICP has failed to allege any facts demonstrating that, had
TDHCA awarded additional points to applications for developments in Caucasian areas,
those applications would have been approved, or that approval of such applications would
have lessened the statistical disparity.
ICP has failed to prove that TDHCA’s decision not to award additional points for
projects that provide desegregated housing caused any statistical disparity. In addition, ICP
does not seek any constitutionally-sound remedy—i.e., the “elimination of the offending
practice.” ICP VI, 135 S.Ct. at 2524. ICP does not seek to remove a specific barrier to fair
housing, but to impose on TDHCA the requirement that it award additional points for
housing that furthers desegregation. The Supreme Court has made clear that such remedies
are constitutionally suspect, and the court declines to impose such remedy. See id.; see also
Wells Fargo, 2015 WL 4398858, at *8 (holding that absence of policy cannot form basis for
- 26 -

disparate impact claim because there is nothing to remedy).
4
ICP challenges TDHCA’s discretion to allocate tax credits to units for elderly-only
residents and the use of set-asides for elderly-only units. TDHCA has the discretion to “set
aside” certain types of applications and to score these applications in their subset, rather than
against the general application pool. From 2001 to 2003, TDHCA set aside a certain number
of 9% tax credits for units for elderly residents. ICP complains that TDHCA did not
establish a set-aside for applications for units located in Caucasian areas, and it challenges
TDHCA’s use of elderly-only set-asides.
TDHCA responds that ICP does not explain how TDHCA’s decision to set aside a
certain number of tax credits for elderly-only developments caused the placement of more
LIHTC units in minority areas than in Caucasian areas, and that ICP has not proved that, had
TDHCA not set aside any tax credits for elderly-only developments, the racial disparity
would have been lessened.
ICP has not proved that TDHCA’s use of set-asides or other allocation of LIHTC for
elderly-only units has caused a statistically-significant disparity, and it has not proved how
tax credits for elderly-only units in Caucasian areas contributes to racial disparity in housing.
To the extent ICP asks the court to require that TDHCA implement a new set-aside for
applications for family units in Caucasian areas, such a remedy is not constitutionally sound,
does not eliminate an offending practice, and may result in the use of racial quotas. See ICP
VI, 135 S.Ct. at 2524.
- 27 -

5
Finally, ICP maintains that TDHCA has used its discretion to approve projects located
in areas of slum and blight, with high crime rates, adverse environmental conditions, and
where there is a high concentration of LIHTC units. TDHCA responds that ICP’s concerns
are exaggerated, and that ICP has not demonstrated that the decision to approve projects in
certain areas causes a statistically-significant disparity.
ICP has not established that TDHCA’s approval of projects in areas of slum and blight
caused a racially disparate impact, and ICP does not seek a constitutionally-permissible
remedy. ICP does not seek to restrict TDHCA’s discretion or to remove a challenged policy;
instead, ICP asks the court to preclude the TDHCA from granting tax credits for
developments in areas with certain conditions. For the reasons explained supra at §
IV(A)(2), the court declines to grant such remedy.
V
The court turns next to the question whether ICP has proved a prima facie case of
disparate impact with respect to the allocation of 4% tax credits.
A
ICP contends that TDHCA exercised its discretion to implement specific discretionary
policies, practices, and decisions affecting the allocation of 4% tax credits, and that the
exercise of discretion caused the perpetuation of racial segregation. According to ICP,
TDHCA consistently used its discretion to approve applications for 4% tax credits for units
in minority areas that maintained racial disparity across the Dallas metropolitan area, already
- 28 -

had a high concentration of LIHTC units, and were plagued by slum and blight. ICP
complains that, once in 1999 and once in 2003, the TDHCA exercised its discretion and
denied applications for 4% tax credits for projects located in Caucasian tracts in the Dallas
metropolitan area for reasons that were not applied to applications for 4% tax credits for
similar projects located in minority areas. ICP maintains that TDHCA’s exercise of
discretion resulted in the approval of 4% tax credits for 4,896 of the 7,283 units in minority
areas in the City of Dallas.
TDHCA responds that discretion in and of itself is not a specific policy sufficient to
prove disparate impact based on a violation of the FHA; each decision to approve or deny
a tax credit is the result of numerous individual policy decisions, many of which are outside
TDHCA’s control; the 4% tax credit program is not oversubscribed, and ICP has not alleged
that TDHCA approved a 4% tax credit for a project in a minority area that required it to deny
a 4% tax credit for a project in a Caucasian area; ICP only identifies two projects for which
TDHCA denied applications for 4% tax credits over a span of 17 years; and ICP has not
acknowledged that, unless developers seek 4% tax credits for projects in Caucasian areas,
TDHCA cannot allocate such credits.
ICP replies that TDHCA perpetuated racial segregation by using its discretion to
approve applications for 4% tax credits in minority areas with undesirable site conditions,
and by denying two applications for 4% tax credits for projects in Caucasian areas.

- 29 -

B
Even assuming arguendo that ICP has proved that TDHCA had a facially neutral
policy or practice in allocating 4% tax credits—the use of discretion in the allocation of 4%
tax credits—ICP has not proved that TDHCA’s exercise of discretion caused a statisticallysignificant disparity in the location of low-income housing.
ICP maintains that TDHCA in its discretion denied the applications of two projects
for 4% credits, even though the projects met the stated criteria for approval. In 1999,
TDHCA rejected an application for 4% credits for the Treymore at Duck Creek, a project in
a Caucasian tract in Garland, Texas. TDHCA’s stated reasons for the denial were the
“desires of the elected officials of the City of Garland,” i.e., that Garland had sufficient lowincome housing and did not need additional units. ICP casts doubt on the TDHCA’s stated
reason and alleges that the following year, TDHCA approved an application for 4% tax
credits for a low-income housing development in a minority tract in Garland.
In 2003, TDHCA exercised its discretion to deny an application for 4% tax credits for
the Primrose at Stonebrook, a housing development in Frisco, Texas. The Stonebrook was
a proposed development of three-bedroom units located in a Caucasian area in Frisco, and
TDHCA denied the project because it lacked diversity in the type of units. ICP argues that
TDHCA’s proffered reason is pretextual because TDHCA waived unit-mix requirements for
other applications and approved other developments containing only three- or four-bedroom
units.
ICP has not satisfied the robust causality requirement of ICP VI. See ICP VI, 135
- 30 -

S.Ct. at 2523. ICP states, without explanation or supporting evidence, that TDHCA’s
discretion in awarding 4% credits has resulted in the approval of 4,896 of the 7,283 units for
4% tax credits located in minority areas in the Dallas metropolitan area. Although ICP states
that “TDHCA did consistently use its discretion to approve 4% application[s] for units in
minority areas that maintained the racial segregation,” ICP 1/7/16 Prima Facie Br. at 42, ICP
does not explain how TDHCA’s exercise of discretion in denying two applications over a 17year period has maintained racial segregation.
ICP has not established that TDHCA’s denial of two applications over a 17-year
period is a “policy” that can cause a disparate impact. See, e.g., ICP VI, 135 S.Ct. at 2523
(noting that a “one-time decision may not be a policy at all”). Nor has ICP proved that
TDHCA’s denial of two applications for 4% tax credits over that time period has caused any
statistically-significant disparity in the location of low-income housing.
ICP has also failed to prove that factors other than TDHCA’s discretion—such as
developers’ preferences for building projects in certain areas, local and state laws, and the
expressed needs of communities and local officials—have not caused or contributed to any
disparity in the location of housing developments receiving 4% tax credits. Because TDHCA
has approved the vast majority of applications for 4% tax credits since 1999, any disparity
is likely caused by factors other than TDHCA’s exercise of discretion.

- 31 -

*

*

*

For the reasons explained, the court finds and concludes that ICP has failed to prove
a prima facie case of discrimination by showing that a challenged practice caused a
discriminatory effect, as defined by 24 C.F.R. § 100.500(a). Accordingly, the court
dismisses ICP’s disparate impact claim and enters judgment today in favor of defendants.
August 26, 2016.

_________________________________
SIDNEY A. FITZWATER
UNITED STATES DISTRICT JUDGE

- 32 -