*895 With this appeal, we review the district court's dismissal with prejudice, pursuant to Rule 12(b)(6) of the Federal Rules of Civil Procedure, of Fair Housing Act claims - including claims of "disparate treatment" and "disparate impact" - asserted against the owners and management company of apartment complexes in the greater Dallas, Texas area that decline to participate in the federal "Section 8" Housing Choice Voucher Program. For the reasons stated herein, we affirm.

I.

The plaintiff, The Inclusive Communities Project ("ICP"), "is a fair housing focused nonprofit organization working with households seeking access to housing in predominately non-minority locations in the Dallas area." 1 In furtherance of its mission, ICP provides "counseling, financial assistance, and other services to Black or African American households participating in the [federal] Section 8 Housing Choice Voucher (HCV or voucher) Program administered by the Dallas Housing Authority (DHA)." According to ICP, its voucher clients seek assistance in finding and obtaining "dwelling units in safe and secure communities with higher median incomes, good schools, low poverty rates, and adequate public and private serve and facilities (high opportunity areas)."

The financial assistance offered by ICP may include the payment of landlord incentives or bonus payments (to encourage leasing to voucher participant households), application fees, and security deposits. ICP also offers landlords in higher opportunity areas the option of a contract with ICP as a guarantor for voucher households or with ICP as the sub-lessor for voucher households. ICP alleges that it proposed these alternative contractual arrangements in response to reasons stated by landlords and landlord associations for refusing to negotiate with or rent to voucher households.

ICP identifies Defendants-Appellees Legacy Multifamily North III, LLC ("Legacy"), CPF PC Riverwalk, L.L.C. ("Riverwalk"); HLI White Rock, L.L.C.

*896 ("White Rock"); and Brick Row Apartments, L.L.C. ("Brick Row") (collectively, "Owners") as owners of apartment complexes in the "higher opportunity" or "high opportunity" areas identified by ICP. Defendant-Appellee Lincoln Property Company (Lincoln) manages these complexes in addition to managing or owning and operating numerous other properties in "the Dallas metropolitan area." 2

ICP contends "its ability to assist its voucher clients in obtaining dwellings in high opportunity areas is obstructed by Defendants' discriminatory housing practices." ICP alleges that Lincoln has a general policy that it will not negotiate with, rent to, or otherwise make units available in "White non-Hispanic areas" to voucher households; moreover, Lincoln's written advertisements state that housing vouchers, Section 8 vouchers, and any government-subsidized rent programs are not accepted. According to ICP, the only apartment complexes for which Lincoln will negotiate with and rent to voucher households are those in predominately minority locations. These apartment complexes include complexes required by law or contract to not discriminate against voucher households based on their status as voucher program participants.

Lincoln's general "no vouchers" policy is applied at approximately 43 apartment complexes, located in majority white census tracts, that have at least some units available at rents payable under the voucher program. These complexes include the units owned by the Owners. ICP further contends that it has black voucher clients who are otherwise eligible under Lincoln's application criteria, and with whom ICP would have entered into subleases, but for Lincoln's policy against voucher tenants.

ICP alleges that it has attempted, on several occasions, to negotiate with Lincoln on behalf of voucher clients seeking rental units in properties that Lincoln manages and/or owns in majority white areas. The most recent requests, ICP reports, were letters that ICP sent to Lincoln, in 2015 and 2016, asking that it "reconsider" its policy of not accepting voucher families as tenants at the aforementioned apartment complexes. 3 According to ICP, neither Lincoln nor the Owners responded to ICP's request to negotiate and rent under the sublease/guarantor proposal. At least one Defendant-Appellee notes, however, that ICP alleges its transmittal of the letters but not their receipt. Nor is it clear when the Owners, as opposed to Lincoln, the manager, became aware of the letters and/or ICP's requests to discuss the "no vouchers" policy.

ICP asserts that the "no vouchers" policy forces voucher households in the Dallas metro area to seek housing in areas where vouchers are accepted, which are "racially concentrated [predominately minority] areas *897 of high poverty that are marked by substantially unequal conditions." Further, ICP contends, Lincoln's refusal to negotiate with or rent to voucher holders disparately impacts black households in the Dallas area. In short, ICP maintains that landlords who accept vouchers are disproportionately located in minority areas of Dallas, and property management companies located in non-minority areas disproportionately refuse vouchers. The waiting lists for the area voucher programs also are disproportionately black.

To support its disparate impact contentions, ICP references the most recent United States Department of Housing and Urban Development (HUD) "Picture of Subsidized Housing" reporting a total of 30,745 voucher households in the Dallas-Irving-Plano Metropolitan Division. According to ICP, 90% of those households are minorities, with the total breakdown being 81% black, 6% Hispanic, and 10% white non-Hispanic (white). Approximately 17,000 of the 30,745 voucher households in the Dallas-Irving-Plano Metropolitan Division participate in the program through the DHA, which has a voucher population that is 86% black and 6% white. The voucher households in the City of Dallas are 87% Black and 94% minority.

ICP likewise characterizes the voucher program in the Dallas metro area as racially segregated into predominantly minority census tracts. On average, voucher households in the Dallas metro area are located in 74% minority census tracts; voucher households in the City of Dallas are located in 88% minority and 33% poverty census tracts.

ICP also alleges the following facts regarding individual apartment complexes that the Defendants-Appellees own or manage:

• Park Central at Flower Mound Complex

• No Black renters in the "small census tract block group" containing this complex;

• 307 units in the complex; and

• Zero voucher households in the census tract containing this complex.

• McKinney Uptown Complex

• No Black renters in the "small census tract block group" containing this complex;

• 144 units in the complex; and

• No voucher households in the census tract containing this complex.

• Parkside at Legacy Complex

• Black renters are 14% of the 630 renter-occupied units in the "small census tract block group" containing this complex;

• 293 units in the complex; and

• No voucher households in the census tract containing this complex.

• White Rock Lake Apartment Villas

• Black renters are 11% of the 1,022 renter-occupied units in the "small census tract block group" containing this complex;

• 296 units in the complex; and

• No voucher household in the census tract containing this complex.

• Brick Row Apartments, LLC

• Black renters are 11% of the 532 renter-occupied units in the "small census tract block group" containing this complex;

• 500 units in the complex;

• 45 voucher households in the census tract containing this complex; and

• Majority of the voucher households in the census tract live in *898 single family or semi-detached structures.

Finally, ICP attaches city maps to its complaint showing that voucher households are concentrated in parts of Dallas where minorities live, with few voucher households in the parts of Dallas where non-minorities live.

Alleging it received no responses from Lincoln or the Owners to its latest letters, ICP filed a complaint on January 23, 2017, seeking declaratory and injunctive relief from the district court. Specifically, ICP seeks a declaration that Lincoln and the Owners have violated 42 U.S.C. § 3604 (a) and 42 U.S.C. § 1982 by declining to participate in the federal "Section 8" Housing Choice Voucher Program. ICP also seeks a permanent injunction compelling Lincoln and the Owners to accept Section 8 vouchers and requiring them to negotiate and contract with ICP under ICP's sublease/guarantor program.

In its complaint, ICP alleges a total of four claims. Two claims - disparate impact and disparate treatment - are asserted against all Defendants-Appellees (Lincoln and the Owners). Relative to disparate impact, ICP alleges that Defendants-Appellees' policy of declining to negotiate with or rent to voucher holders disparately impacts black households as evidenced by statistics establishing that more than 80% of the voucher holders in the Dallas area are black. 4 Relative to disparate treatment, ICP alleges that Defendants-Appellees' refusal to negotiate with or rent to ICP, pursuant to ICP's guarantor or sublease proposals, constitutes disparate treatment based on race and color, because ICP's voucher clients are predominantly black.

ICP also alleges two claims solely against Lincoln. The first concerns Lincoln's publication of its policy of refusing to "negotiate with or rent to voucher households" by including the following statements in advertisements placed with apartment locator services:

Our community is not authorized to accept housing vouchers.

Our community is not authorized to accept Section 8 housing.

Our community is not authorized to accept ANY government subsidized rent programs.

ICP maintain these advertisements "appeal to the stereotype that because voucher holders are Black, voucher tenants are undesirable as tenants ..." and, thus, perpetuate racial stereotypes in violation of 42 U.S.C. § 3604 (c).

The second claim against only Lincoln is for disparate treatment liability based on Lincoln's alleged refusal to negotiate with or rent to otherwise qualified voucher households in predominately white areas while, at the same time, negotiating with and renting to voucher holders in predominately minority areas. ICP argues Lincoln's conduct violates the disparate treatment standard of liability because the differing policies regarding vouchers are based on the race and color of the voucher holders.

In response to ICP's claims, Lincoln and the Owners filed motions to dismiss, pursuant to Rule 12(b)(6) of the Federal Rules of Civil Procedure, for failure to state claims upon which relief can be granted. The district court granted the motions in two orders dated July 13, 2017 (Brick Row's motion) and August 16, 2017 (the remaining motions). The district court entered final judgment on August 16, 2017. This appeal followed.

*899 II.

Appellate courts conduct a de novo review of a district court's dismissal of a complaint under Federal Rule of Civil Procedure 12(b)(6). See Clyce v. Butler, 876 F.3d 145 , 148 (5th Cir. 2017). We may affirm the district court's dismissal on any basis supported by the record. See, e.g., Torch Liquidating Tr. ex rel. Bridge Assocs., L.L.C. v. Stockstill, 561 F.3d 377 , 384 (5th Cir. 2009).

Rule 12(b)(6) authorizes the filing of motions to dismiss asserting, as a defense, a plaintiff's "failure to state a claim upon which relief can be granted." See Fed. R. Civ. P. 12(b)(6). Thus, claims may be dismissed under Rule 12(b)(6)"on the basis of a dispositive issue of law." Neitzke v. Williams , 490 U.S. 319 , 326, 109 S.Ct. 1827 , 104 L.Ed.2d 338 (1989). Dismissal under Rule 12(b)(6) also is warranted if the complaint does not contain sufficient factual matter, accepted as true, to "state a claim to relief that is plausible on its face." Ashcroft v. Iqbal, 556 U.S. 662 , 678, 129 S.Ct. 1937 , 173 L.Ed.2d 868 (2009) (quoting Bell Atl. Corp. v. Twombly , 550 U.S. 544 , 570, 127 S.Ct. 1955 , 167 L.Ed.2d 929 (2007) ). Where the well-pleaded facts of a complaint do not permit a court to infer more than the mere possibility of misconduct, the complaint has alleged - but it has not 'show[n]' - "that the pleader is entitled to relief." Iqbal , 556 U.S. at 678 , 129 S.Ct. 1937 (quoting Fed. Rule Civ. P. 8(a)(2) ). Thus, a complaint's allegations "must make relief plausible, not merely conceivable, when taken as true." United States ex rel. Grubbs v. Kanneganti , 565 F.3d 180 , 186 (5th Cir. 2009) ; see also Twombly , 550 U.S. at 555 , 127 S.Ct. 1955 ("Factual allegations must be enough to raise a right to relief above the speculative level ... on the assumption that all the allegations in the complaint are true (even if doubtful in fact).").

"The plausibility standard is not akin to a 'probability requirement,' but it asks for more than a sheer possibility that a defendant has acted unlawfully." Id . Factual allegations that are "merely consistent with a defendant's liability, stop short of the line between possibility and plausibility of entitlement to relief," and thus are inadequate. Id. (internal quotations omitted). Accordingly, the requisite facial plausibility exists "when the plaintiff pleads factual content that allows the court to draw the reasonable inference that the defendant is liable for the misconduct alleged." Id . (emphasis added). "Determining whether a complaint states a plausible claim for relief" is "a context-specific task that requires the reviewing court to draw on its judicial experience and common sense." Iqbal , 556 U.S. at 679 , 129 S.Ct. 1937 (internal citations omitted). See also Robbins v. Oklahoma, 519 F.3d 1242 , 1248 (10th Cir. 2008) (degree of required specificity depends on context, i.e., the type of claim at issue).

In evaluating motions to dismiss filed under Rule 12(b)(6), the court "must accept all well-pleaded facts as true, and ... view them in the light most favorable to the plaintiff." Campbell v. Wells Fargo Bank, N.A., 781 F.2d 440 , 442 (5th. Cir.), cert. denied , 476 U.S. 1159 , 106 S.Ct. 2279 , 90 L.Ed.2d 721 (1986). Further, "[a]ll questions of fact and any ambiguities in the controlling substantive law must be resolved in the plaintiff's favor." Lewis v. Fresne , 252 F.3d 352 , 357 (5th Cir. 2001). On the other hand, courts "are not bound to accept as true a legal conclusion couched as a factual allegation." Papasan v. Allain , 478 U.S. 265 , 286, 106 S.Ct. 2932 , 92 L.Ed.2d 209 (1986) ; see also Iqbal , 556 U.S. at 678 , 129 S.Ct. 1937 ("tenet that a court must accept as true all of the allegations contained in a complaint is inapplicable to legal conclusions"). "Nor does a *900 complaint suffice if it tenders 'naked assertion[s]' devoid of 'further factual enhancement.' " Iqbal , 556 U.S. at 678 , 129 S.Ct. 1937 (quoting Twombly, 550 U.S. at 557 , 127 S.Ct. 1955 ); see also Christopher v. Harbury , 536 U.S. 403 , 416, 122 S.Ct. 2179 , 153 L.Ed.2d 413 (2002) (elements of a plaintiff's claim(s) "must be addressed by allegations in the complaint sufficient to give fair notice to a defendant").

In determining whether a plaintiff's claims survive a Rule 12(b)(6) motion to dismiss, the factual information to which the court addresses its inquiry is limited to (1) the facts set forth in the complaint, (2) documents attached to the complaint, and (3) matters of which judicial notice may be taken under Federal Rule of Evidence 201. See Norris v. Hearst Trust, 500 F.3d 454 , 461 n. 9 (5th Cir. 2007) ; R2 Invs. LDC v. Phillips, 401 F.3d 638 , 640 n. 2 (5th Cir. 2005). When a defendant attaches documents to its motion that are referenced in the complaint and are central to the plaintiff's claims, however, the court can also properly consider those documents. Causey v. Sewell Cadillac-Chevrolet, Inc ., 394 F.3d 285 , 288 (5th Cir. 2004) ; In re Katrina Canal Breaches Litig., 495 F.3d 191 , 205 (5th Cir. 2007). "In so attaching, the defendant merely assists the plaintiff in establishing the basis of the suit, and the court in making the elementary determination of whether a claim has been stated." Collins v. Morgan Stanley Dean Witter, 224 F.3d 496 , 499 (5th Cir. 2000).

III.

The federal Housing Choice Voucher Program pays rental subsidies to "aid[ ] low-income families in obtaining a decent place to live" and to "promot[e] economically mixed housing." 42 U.S.C. § 1437f(a). The voucher program is funded by HUD and administered by state and local public housing authorities (PHA's) in accordance with regulations promulgated by HUD. When a rent payment exceeds a specified percentage of a family's monthly income, the federal program pays the balance.

Landlord participation in the voucher program is voluntary under both federal and Texas state law. See 42 U.S.C. § 1437f ; 24 C.F.R. §§ 982.301 (b)(11), 982.302(a), 982.307 ; TEX. LOCAL GOV'T CODE § 250.007(a) ; TEX. GOV'T CODE. § 2306.269 ; Knapp v. Eagle Prop. Mgmt. Corp ., 54 F.3d 1272 , 1280 (7th Cir. 1995) ("Owner participation in the section 8 program is voluntary and non-participating owners routinely reject section 8 voucher holders."); Salute v. Stratford Greens Garden Apartments, 136 F.3d 293 , 300 (2d Cir. 1998) ("We think that the voluntariness provision of Section 8 reflects a congressional intent that the burdens of Section 8 participation are substantial enough that participation should not be forced on landlords, either as an accommodation to handicap or otherwise.").

Once admitted to the voucher program, program participants are responsible for finding a landlord in the private rental market willing to rent to them. 24 C.F.R. § 982.302 (a). Landlords who participate in the program are responsible for screening prospective tenants and reject them if screening reveals red flags in terms of paying rent and utility bills, caring for rental housing, respecting neighbors, criminal activity, and the like. Id. at § 982.307(a).

The Fair Housing Act (FHA), Title III of the Civil Rights Act of 1968, 42 U.S.C. § 3601 , et seq ., prohibits discrimination in the rental or sale of a dwelling based on certain protected characteristics, including race. See 42 U.S.C. § 3604 (a). That statute reflects "the policy of the United States to provide within constitutional limitations, for fair housing throughout *901 the United States." 42 U.S.C. § 3604 . Thus, the voluntary nature of landlord participation in the voucher program does not render it immune from liability if actionable discrimination under the FHA is established.

Pertinent here, § 3604(a) provides:

[I]t shall be unlawful to refuse to sell or rent after the making of a bona fide offer, or to refuse to negotiate for the sale or rental of, or otherwise make unavailable or deny, a dwelling to any person because of race, color, religion, sex, familial status, or national origin.

42 U.S.C. § 3604 (a). ICP's advertisement liability claim against Lincoln is governed by 42 U.S.C. § 3604 (c), which provides:

[I]t shall be unlawful to make, print, or publish, or cause to be made, printed, or published any notice, statement or advertisements, with respect to the sale or rental of a dwelling that indicates any preference, limitation, or discrimination based on race, color, religion, sex, handicap, familial status, or national origin, or an intention to make any such preference, limitation, or discrimination.

A. Disparate Impact Liability

In Texas Department of Housing & Community Affairs v. Inclusive Communities Project, Inc., --- U.S. ----, 135 S.Ct. 2507 , 192 L.Ed.2d 514 (2015) ( " ICP ") , the Supreme Court, construing 42 U.S.C. §§ 3604 (a) and 3605(a), determined that both disparate treatment claims (claims asserting "discriminatory intent or motive") and disparate impact claims ("claims asserting an unjustified, disproportionally adverse effect on minorities") are cognizable under the FHA. ICP, --- U.S. ----, 135 S.Ct. 2507 , 2513, 2518. 5 In recognizing the viability of disparate impact FHA claims, the Supreme Court emphasized, inter alia, the inclusion of the "results-oriented" phrase - "or otherwise make unavailable or deny" - in § 3604(a), reasoning that it "refers to the consequences of action rather than the actor's intent." Id. at 2518, 2525 . The Court also found "[r]ecognition of disparate-impact claims [to be] consistent with the FHA's central purpose ... [of] eradicat[ing] discriminatory practices within a sector of our Nation's economy." Id. at 2521 .

1. FHA Disparate Impact Liability: ICP's Test

To properly evaluate ICP's claims, we must first address, as a threshold matter, the applicable test for determining disparate impact claims asserted under the FHA. When ICP previously was before this court, we adopted HUD's burden-shifting approach for deciding disparate impact claims under the FHA. See 24 C.F.R. § 100.500 ; The Inclusive Communities Project, Inc. v. Texas Dep't of Hous. and Cmty. Affairs, 747 F.3d 275 , 282 (5th Cir. 2014), aff'd , --- U.S. ----, 135 S.Ct. 2507 , 192 L.Ed.2d 514 (2015). Under the HUD regulation, a plaintiff must first prove a prima facie case of discrimination by showing that the challenged practice causes a discriminatory effect. 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 of the defendant's substantial, legitimate, nondiscriminatory interests. Id. at § 100.500(c)(2). If the defendant meets its burden, the plaintiff *902 must then show that the defendant's interests "could be served by another practice that has a less discriminatory effect." Id. at § 100.500(c)(3).

Although it affirmed our decision, the Supreme Court never explicitly stated that it adopted the HUD regulation's framework. Because of this lack of clarity, debate exists regarding whether, in ICP , the Supreme Court adopted the regulation's approach or modified it. The Fourth Circuit has noted that "[t]he HUD regulation is similar to the framework the Supreme Court ultimately adopted in [ ICP ] , and indeed, some courts believe the Supreme Court implicitly adopted the HUD framework altogether." Reyes v. Waples Mobile Home Park Ltd. P'ship , 903 F.3d 415 , 424 n.4 (4th Cir. 2018) (citing Mhany Mgmt., Inc. v. Cty. of Nassau , 819 F.3d 581 , 618 (2d Cir. 2016) ("The Supreme Court implicitly adopted HUD's approach....")). The Fourth Circuit concluded, "[w]ithout deciding whether there are meaningful differences between the frameworks, ... the standard announced in [ ICP ], rather than the HUD regulation[,] controls our inquiry." Id.

We read the Supreme Court's opinion in ICP to undoubtedly announce a more demanding test than that set forth in the HUD regulation. As noted by a Minnesota district court: "the Supreme Court announced several 'safeguards' to incorporate into the burden-shifting framework to ensure that disparate impact liability does not 'displace valid governmental and private priorities.' " Crossroads Residents Organized for Stable & Secure Residencies v. MSP Crossroads Apartments LLC , No. 16-233, 2016 WL 3661146 , at *6 (D. Minn. 2016). "Those safeguards include a 'robust causality requirement' at the prima facie stage, and, after the burden shifts to the defendant, 'leeway to state and explain the valid interest served by [the defendant's] policies.' " Id. (quoting ICP , 135 S.Ct. at 2522-23 ). In contrast, the HUD regulation contains no "robust causation" requirement; rather it requires only a showing that "a challenged practice caused or predictably will cause a discriminatory effect." 24 C.F.R. § 100.500 (c)(1).

A careful review of the Supreme Court's analysis in ICP , moreover, reveals its modification of HUD's test to be both purposeful and significant. Indeed, the Court emphasizes:

[D]isparate-impact liability has always been properly limited in key respects that avoid the serious constitutional questions that might arise under the FHA, for instance, if such liability were imposed based solely on a showing of a statistical disparity.

* * *

[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. A robust causality requirement ensures that "[r]acial imbalance ... does not, without more, establish a prima facie case of disparate impact" and thus protects defendants from being held liable for racial disparities they did not create. Wards Cove Packing Co. v. Atonio , 490 U.S. 642 , 653, 109 S.Ct. 2115 , 104 L.Ed.2d 733 (1989), superseded by statute on other grounds , 42 U.S.C. § 2000e-2(k). Without adequate safeguards at the prima facie stage , disparate-impact liability might cause race to be used and considered in a pervasive way and would almost inexorably lead governmental or private entities to use numerical quotas and serious constitutional questions then could arise.

* * *

Courts must therefore examine with care whether a plaintiff has made out a prima facie case of disparate impact and *903 prompt resolution of these cases is important. 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 impact.

ICP, 135 S.Ct. at 2522-23 (internal quotation marks omitted) (emphasis added).

Other statements by the Court, regarding a defendant's competing interests, dispel any remaining doubt as to the limited nature of the disparate impact claim that exists under the FHA. Indeed, citing HUD's then recent rulemaking, the Court emphasized that disparate-impact liability "does not mandate that affordable housing be located in neighborhoods with any particular characteristic." ICP, 135 S.Ct. at 2523 (citing 78 Fed. Reg. 11476 ) (emphasis added). Likewise, "[t]he FHA does not decree a particular vision of urban development." Id. (emphasis added). Rather, "entrepreneurs must be given latitude to consider market factors." Id. The Court additionally cautions:

[A] plaintiff bringing a disparate-impact claim challenges practices that have a "disproportionately adverse effect on minorities" and are otherwise unjustified by a legitimate rationale. Ricci v. DeStefano, 557 U.S. 557 , 577, 129 S.Ct. 2658 , 174 L.Ed.2d 490 (2009) (internal quotation marks omitted).

* * *

An important and appropriate means of ensuring that disparate-impact liability is properly limited is to give housing authorities and private developers leeway to state and explain the valid interest served by their policies. Just as an employer may maintain a workplace requirement that causes a disparate impact if that requirement is a reasonable measurement of job performance, ... so too must housing authorities and private developers be allowed to maintain a policy if they can prove it is necessary to achieve a valid interest.

* * *

Governmental or private policies are not contrary to the disparate impact requirement, unless they are "artificial, arbitrary and unnecessary barriers." Difficult questions might arise if disparate-impact liability under the FHA caused race to be used and considered in a pervasive and explicit manner to justify governmental or private actions that, in fact, tend to perpetuate race-based considerations rather than move beyond them. Courts should avoid interpreting disparate-impact liability to be so expansive as to inject racial considerations into every housing decision.

Id. at 2513, 2522-24 (quoting Griggs v. Duke Power Co., 401 U.S. 424 , 431, 91 S.Ct. 849 , 28 L.Ed.2d 158 (1971) ).

Given the foregoing, we are convinced the Supreme Court's language in ICP is stricter than the regulation itself. Accordingly, as noted by the Fourth Circuit, we are bound to apply the stricter version of the burden-shifting analysis. Reyes, 903 F.3d at 424 . 6

2. Disparate Impact: Four Views of "Robust Causation"

Although the Supreme Court's opinion in ICP established "robust causation" as a key element of the plaintiff's prima facie burden in a disparate impact case, the Court did not clearly delineate its meaning or requirements. Nor are we aware of any *904 post- ICP Supreme Court or Fifth Circuit decisions clarifying the standard. However, decisions from three other circuits - the Fourth, Eighth and Eleventh - have considered its application, yielding opinions reflecting varying views of the prerequisites.

The first view is provided by Ellis v. City of Minneapolis , in which the Eighth Circuit construed ICP to require that a plaintiff's allegations "point to an 'artificial, arbitrary, and unnecessary' policy causing the problematic disparity," in order to establish a prima facie disparate impact case. Ellis v. City of Minneapolis , 860 F.3d 1106 , 1114 (8th Cir. 2017) (quoting ICP , 135 S.Ct. at 2524 ). In Ellis, the plaintiffs, low-income housing landlords, alleged that the city was targeting their properties with inspections, issuing citations for code violations that did not exist, and threatening to revoke their rental licenses. Id . at 1108-09. As a result, the plaintiffs argued, the city's actions displaced hundreds of FHA-protected individuals from their homes. Id. at 1109. In denying the claim, the Eighth Circuit explained that the plaintiffs' complaint "must still allege facts plausibly demonstrating that the housing-code standards complained of are arbitrary and unnecessary under the FHA." Id . at 1112. The Ellis complaint fell short, however, because it "suggest[ed] no more than disagreement between the [plaintiffs] and the City on the extent of deficiencies based on reasonable housing-code provisions." Id . at 1113. Furthermore, "[t]o the extent their complaint mentions specific housing-code provisions" it lacked "factually supported allegations that those provisions are arbitrary or unnecessary to health and safety." Id . at 1112.

The second view is provided by the Fourth Circuit's majority opinion in Reyes, in which "understanding [the] robust causality requirement [was] at the crux of th[e] appeal." Reyes, 903 F.3d at 425 . In Reyes , a mobile home park began enforcing a previously unenforced policy requiring all adult occupants to provide documentation showing that they were legally present in the United States in order to renew their leases, or face eviction. Id. at 419-20, 428 . 7 The plaintiffs alleged that this policy disproportionately affected Latino families because Latinos comprised 64.6% of the undocumented immigrant population in Virginia and are "ten times more likely than non-Latinos to be adversely affected by the Policy, as undocumented immigrants constitute 36.4% of the Latino population compared with only 3.6% of the non-Latino population." Id . at 428.

Noting "statistical disparities must be sufficiently substantial that they raise [the necessary] inference of causation," the Fourth Circuit majority concluded the plaintiffs had properly stated a prima facie disparate impact case by alleging that the defendant's first-time enforcement of a previously unenforced policy (except as to the leaseholder) "caused a disproportionate number of Latinos to face eviction from the Park compared to the number of non-Latinos who faced eviction." Id. at 425, 428-29 (quoting Watson v. Fort Worth Bank & Tr., 487 U.S. 977 , 994-95, 108 S.Ct. 2777 , 101 L.Ed.2d 827 (1988) ). The majority concluded the statistical evidence that the plaintiffs provided "satisfied the robust causality requirement" when considered in the context of the newly enforced policy.

The third construction of "robust causation" is provided by Judge Keenan's dissenting *905 opinion in Reyes. In Judge Keenan's view, the plaintiffs had not met this requirement. Id. at *434-35. Rather, Judge Keenan reasoned:

In my view, the plaintiffs have not adequately alleged that the defendants' policy caused the statistical disparity that they challenge. The plaintiffs rest their claim of causality on statistics showing that Latinos constitute the majority of undocumented aliens in the geographic area of the park, and thus that Latinos are disproportionately impacted by a policy targeting undocumented aliens. Despite this statistical imbalance, however, all occupants of the park must comply with the policy addressing their immigration status, irrespective whether they are Latino. Not all Latinos are impacted negatively by the policy, nor are Latino undocumented aliens impacted more harshly than non-Latino undocumented aliens. Accordingly, I would conclude that the defendants' policy disproportionately impacts Latinos not because they are Latino, but because Latinos are the predominant sub-group of undocumented aliens in a specific geographical area.

Although Latinos constitute the majority of the undocumented population in the park, at different times and in different locales, the disparate impact might have been on immigrant populations from many other parts of the world. Such geographical happenstance cannot give rise to liability against an entity not responsible for the geographical distribution. Nor does linking disparate impact liability to the coincidental location of certain undocumented aliens further the aim of the FHA to avoid "perpetuating segregation." Inclusive Cmtys., 135 S.Ct. at 2522 . Thus, because the defendants' policy has not caused Latinos to be the dominant group of undocumented aliens in the park, the policy has not "caused" a disparate impact on Latinos.

Reyes, 903 F.3d at 434-35 (Keenan, J., dissenting) (emphasis added).

Thus, in the Reyes majority's view, that the policy impacted Latinos more than non-Latinos was enough to show robust causation. In Judge Keenan's dissenting view, however, robust causation was not satisfied by pre-existing conditions (Latinos status as the predominant sub-group of undocumented aliens) not brought about by the challenged policy.

The fourth view of robust causation is provided by the Eleventh's Circuit's unpublished per curiam opinion in Oviedo Town Ctr, II, L.L.P. v. City of Oviedo, Florida , 759 Fed. Appx. 828 , 833-35, No. 17-14254, 2018 WL 6822693 , *4 (Dec. 28, 2018), which describes ICP as "promulgat[ing] detailed causation requirements as a means of cabining disparate impact liability." (Emphasis added.) Specifically, citing the Supreme Court's instruction to "avoid interpreting disparate impact liability to be so expansive to inject racial considerations into every housing decisions," the Eleventh Circuit concluded: "The Supreme Court's solution was to impose '[a] robust causality requirement ensur[ing] that [r]acial imbalance ... does not, without more, establish a prima facie disparate impact.' " Id. (quoting ICP , 135 S.Ct. at 2523 ) (internal citations omitted). Otherwise, "[i]f a disparate impact claim could be found on nothing more than a showing that a policy impacted more members of a protected class that nonmembers of protected classes, disparate impact liability undeniably would overburden cities and developers." Id. In Oviedo , no prima facie case was established, the court of appeals reasoned, because the submitted data "[did] not establish a disparate impact let alone any causal connection" with the policy at issue. Id. at 835, at *5.

*906 3. FHA Disparate Impact Liability: Application

In the instant matter, the district court found ICP had not adequately alleged facts demonstrating the necessary causation. Although acknowledging that ICP had shown "a possible statistical imbalance with the amount of voucher households in the census tract," the district court concluded that ICP had not provided facts linking the "no vouchers" policy to the "possible statistical disparity." Further, the court found ICP's statistical information and arguments "conclusory rather than descriptive of how [the defendants'] policy actually caused a disparate impact."

The district court additionally determined that "[e]ven if Plaintiff ICP met its burden to establish a prima facie showing of disparate impact, Plaintiff ICP does not establish a disparate impact claim" because of the burden-shifting framework. Specifically, the district court identified business concerns referenced in ICP's complaint, such as increased costs, administrative delays, and other financial risks, as legitimate business reasons for not participating in the voucher program. 8

Proceeding to the third step of the burden-shifting framework, the district court rejected the less discriminatory alternatives proposed by ICP, such as "the incentive payments, Sublease Program, and Third Party Guarantor Program to alleviate Defendants' anticipated business concerns." The court reasoned: "ICP's proposals, while laudable, do not show how or if the proposed programs have performed, or if Plaintiff ICP can financially support the programs." Thus, if ICP's programs were not successfully executed, Lincoln and the Owners "could experience financial harm."

Considering the instant record, we find no error in the district court's determination that the allegations of ICP's complaint regarding Lincoln's and the Owners' "no vouchers" policies fail to allege facts sufficient to provide the robust causation necessary for an actionable disparate impact claim. Moreover, we find this conclusion to be warranted under any of the analyses of robust causation discussed above, i.e. that of the Eighth Circuit in Ellis , the Fourth Circuit majority in Reyes, Judge Keenan's dissent in Reyes , or the Eleventh Circuit's per curiam in Oviedo.

Focusing first on the Reyes majority, we note the opinion arguably could be understood to support a finding of robust causation any time that a defendant's policy impacts a protected class more than others. Nevertheless, absent a contrary ruling by the Fourth Circuit, we believe a narrower construction of the opinion is warranted, given the stringent framework outlined by the Supreme Court in ICP for evaluating disparate impact claims. Thus, we find it significant that the disproportionate impact upon Latinos that the Reyes majority held satisfied robust causation was the consequence of a change in the defendant's enforcement of its policy that increased the number of Latinos facing eviction from the park than before. And, as previously stated, the Reyes dissent reasons that "geographical happenstance cannot give rise to liability against an entity not responsible for the geographical distribution." 903 F.3d at 434 . Thus, because the park's policy had not caused Latinos to be the dominant group of undocumented aliens in the park, Judge Keenan, dissenting, found robust causation lacking in Reyes.

*907 The logic of both the majority and dissenting opinions in Reyes , as well as the Eleventh Circuit's per curiam in Oviedo , likewise applies here. Neither the aforementioned "city-level data" nor the "census-level data" cited by ICP supports an inference that the implementation of Defendants-Appellees' blanket "no vouchers" policy, or any change therein, caused black persons to be the dominant group of voucher holders in the Dallas metro area (or any of the other census areas discussed by ICP). Similarly, ICP alleges no facts supporting a reasonable inference that Defendants-Appellees bear any responsibility for the geographic distribution of minorities throughout the Dallas area prior to the implementation of the "no vouchers" policy. Indeed, ICP pleads no facts showing Dallas's racial composition before the Defendants-Appellees implemented their "no vouchers" policy or how that composition has changed, if at all, since the policy was implemented.

Thus, as the district court noted, none of these factual allegations "show or infer that Defendants-Appellees' policy diminished the amount of rental opportunities for African American or Black prospective tenants previously available before Defendants' policy was implemented." (Emphasis added.) Accordingly, it is entirely speculative whether the "no vouchers" policy, as opposed to some other factor, not attributable to Defendants-Appellees, caused there to be less minority habitation in individual census tracts after the policy was implemented. Without that information, any landlord who did not accept vouchers would be vulnerable to a disparate impact challenge any time a less than statistically proportionate minority population lived in that landlord's census tract. 9 Because "disparate-impact liability has always been properly limited," see ICP , 135 S.Ct. at 2522 , that cannot be the correct result.

Finally, in the Eighth Circuit's view, the "no vouchers" policy, even if causing a "problematic disparity," does not state an actionable FHA disparate treatment claim unless the policy is "artificial, arbitrary, and unnecessary." Ellis , 860 F.3d at 1112-1114 . A private entity's choice to opt out of participation in a government program that is voluntary under both federal and Texas law cannot be artificial, arbitrary, and unnecessary absent the existence of pertinent, contrary factual allegations sufficiently rendering a plaintiff's claimed entitlement to disparate impact relief plausible, rather than merely conceivable or speculative. As we have explained, on the record before us, we find none.

The dissenting opinion objects to our treatment of "robust causation," contending that, in ICP , the Supreme Court simply "made clear that the plaintiff must identify an offending policy in order to establish a prima facie case" and, as evidenced by its citation to Wards Cove , confirmed *908 that standards for disparate impact employment discrimination claims likewise apply to FHA disparate impact claims. We respectfully disagree.

To the contrary, in ICP , the Supreme Court stressed the need for both a policy attributable to the defendant and the requisite causal connection, clarifying that a robust causality requirement "protects defendants from being held liable for racial disparities they did not create." ICP , 135 S. Ct. at 2523 (emphasis added) (quoting Wards Cove , 490 U.S. at 653, 109 S.Ct. 2115 ). Indeed, the Court specifically stated: "If a statistical discrepancy is caused by factors other than the defendant's policy , a plaintiff cannot establish a prima facie case, and there is no liability." ICP , 135 S.Ct. at 2514 (emphasis added). The Supreme Court's previous analysis in Wards Cove , moreover, further supports this point:

[A plaintiff] will also have to demonstrate that the disparity [at issue] is the result of one or more of the employment practices [under attack], specifically showing that each challenged practice has a significantly disparate impact on employment opportunities for whites and nonwhites. To hold otherwise would result in [a defendant] being held liable for the "myriad of innocent causes that may lead to statistical imbalances in the composition of their work force."

Wards Cove, 490 U.S. at 657, 109 S.Ct. 2115 (quoting Watson v. Fort Worth Bank & Trust , 487 U.S. 977 , 992, 108 S.Ct. 2777 , 101 L.Ed.2d 827 (1988) (emphasis added)).

Lastly, the dissent looks to the Second Circuit's opinion in Huntington Branch, NAACP v. Town of Huntington, 844 F.2d 926 (2nd Cir. 1988), to support its contrary position regarding ICP's disparate impact claim, essentially arguing that we and the district court overlooked ICP's assertions of harm to the community by perpetuation of segregation. Huntington Branch, however, is materially distinguishable. Importantly, like the other decisions characterized by the Supreme Court as "resid[ing] at the heartland of disparate-impact liability," see ICP, 135 S.Ct. at 2522 , 10 Huntington Branch addressed a public defendant's prohibitory enforcement of a facially neutral zoning ordinance in such a manner that restricted multi-family housing to a small predominantly minority area of the city. Thus, the Second Circuit simply employed the FHA to remove indefensible government policies that operated to perpetuate segregation by unreasonably restricting private construction of multi-family housing that would increase affordable housing options for minorities. Significantly, Huntington Branch did not impose affirmative housing obligations on private actors. 11

*909 To adopt the dissent's position would effectively mandate a landlord's participation in the voucher program any time the racial makeup of multi-family rental complex does not match the demographics of a nearby metropolitan area. That result, however, would be contrary to the cautionary standards that the Supreme Court has declared to be necessary both in evaluating a prima facie case and in ordering any remedial action:

Were standards for proceeding with disparate-impact suits not to incorporate at least the safeguards discussed here, then disparate-impact liability might displace valid governmental and private priorities, rather than solely "remov[ing] ... artificial, arbitrary, and unnecessary barriers." Griggs, 401 U.S. at 431 , 91 S.Ct. 849 . And that, in turn, would set our Nation back in its quest to reduce the salience of race in our social and economic system.

* * *

Remedial orders in disparate-impact cases should concentrate on the elimination of the offending practice that "arbitrar[ily] ... operate[s] invidiously to discriminate on the basis of rac[e]." If additional measures are adopted, courts should strive to design them to eliminate racial disparities through race-neutral means. Remedial orders that impose racial targets or quotas might raise more difficult constitutional questions.

ICP , 135 S.Ct. at 2524 (internal citations omitted). In any event, if such a burdensome and extreme mandate were to be attempted, it should be expressly legislated by Congress, not this court. Accordingly, we affirm the district court's rejection of ICP's disparate impact claim-whether it is viewed as one alleging an adverse impact on a particular minority group or, as discussed in Huntington Branch, one asserting "harm to the community generally by the perpetuation of segregation." 844 F.2d at 937 .

B. Disparate Treatment Liability: Lincoln and Owners

The district court concluded ICP's disparate treatment claims asserted against all Defendants-Appellees were essentially "mislabeled" disparate impact claims that likewise should be dismissed. Specifically, the district court stated:

Defendants refuse[d] to rent to or negotiate with Section 8 voucher holders 12 regardless of race or color. There are no allegations made against Defendants' subjective application of their policy to Section 8 voucher holders. Plaintiff ICP's issue is with the existence of Defendants' policy, which is indicative of disparate impact rather than disparate treatment.

Accordingly, having rejected ICP's disparate impact claims, the district court likewise dismissed ICP's disparate treatment claims.

"Disparate treatment" is "deliberate discrimination." Munoz v. Orr , 200 F.3d 291 , 299 (5th Cir. 2000). It refers to treating some people "less favorably than others because of a protected trait." Ricci v. DeStefano , 557 U.S. 557 , 577, 129 S.Ct. 2658 , 174 L.Ed.2d 490 (2009) (internal citations omitted). Such discrimination is shown by evidence of discriminatory action *910 or by inferences from the fact of differences in treatment. L & F Homes & Dev., L.L.C. v. City of Gulfport, Miss., 538 F. App'x 395 , 401 (5th Cir. 2013) (unpub.) (internal citations omitted). With discriminatory treatment claims, there can be no liability without a finding that the protected trait ( e.g. , race) motivated the challenged action. Greater New Orleans Fair Hous. Action Ctr., Inc. v. Hotard , 275 F.Supp.3d 776 , 786 (E.D. La. 2017) (citing Simms v. First Gibraltar Bank , 83 F.3d 1546 , 1556 (5th Cir. 1996) (evidence must create reasonable inference race was significant motivating factor); Woods-Drake v. Lundy, 667 F.2d 1198 , 1202 (5th Cir. 1982) ("Plaintiff need only prove that race must have been a significant factor in defendant's dealings)).

Although we find dismissal of the disparate treatment claims asserted collectively against Defendants-Appellees to have been warranted, the district court's finding that they were "mislabeled" was not, given ICP's contention that the true rationale for the facially neutral "no vouchers" policies is the race of the voucher tenants, not the means (vouchers) by which rent is paid. As argued by the amicus curiae , the Lawyers' Committee for Civil Rights Under Law, so long as the requisite discriminatory intent is present, a seemingly race-neutral policy can give rise to actionable disparate treatment. See, e.g., ICP, 135 S.Ct. at 2513, 2518 (distinguishing between the discriminatory intent or motive required for disparate treatment liability and the discriminatory effect or consequence required for a disparate impact liability); Greater New Orleans Fair Housing Action Center v. St. Bernard Parish , 641 F.Supp.2d 563 (E.D.La. 2009) (parish council's moratorium against construction of multi-family housing gave rise to both disparate intent and impact FHA discrimination violations). Such is often the case in employment discrimination cases where the proffered (neutral) rationale for an adverse employment action allegedly is pretext for discrimination.

Further, that certain facts may be cited in support of both a disparate treatment and disparate impact claim does not automatically cause one claim to supersede the other. Cf. Vill. of Arlington Heights v. Metro. Hous. Dev. Corp ., 429 U.S. 252 , 265, 97 S.Ct. 555 , 50 L.Ed.2d 450 (1977) (Disproportionate impact of facially neutral legislation is not the "sole touchstone" of racially discriminatory purpose but "is not irrelevant" and "may provide an important starting point"). In fact, in ICP , the Supreme Court notes: "Recognition of disparate-impact liability under the FHA also plays a role in uncovering discriminatory intent: It permits plaintiffs to counteract unconscious prejudices and disguised animus that escape easy classification as disparate treatment." 135 S.Ct. at 2513 .

In the absence of direct evidence, claims of disparate treatment are evaluated utilizing the burden-shifting evidentiary standard established for discrimination cases based on circumstantial evidence. Petrello v. Prucka , 484 Fed. Appx. 939 , 942 (5th Cir. 2012) ((citing Lindsay v. Yates, 498 F.3d 434 , 438-39 (6th Cir. 2007) )); Mitchell v. Shane , 350 F.3d 39 , 47 (2d Cir. 2003) ; Cox v. Phase III, Invs. No. CIV.A. H-12-3500, 2013 WL 3110218 , at *8 (S.D. Tex. June 14, 2013). Thus, to state a claim for disparate treatment under § 3604(a), the plaintiff must allege facts supporting a prima facie case of (1) membership in protected class, (2) that the plaintiff applied and was qualified to rent or purchase housing; (3) that the plaintiff was rejected, and (4) that the housing thereafter remained open to similarly situated applicants after the plaintiff was rejected. Petrello, 484 Fed. Appx. at 942 ;

*911 Graoch Assocs. #33, LP v. Louisville/Jefferson Cty. Metro , 508 F.3d 366 , 371 (6th Cir. 1996). If a prima facie case is alleged, a defendant may offer a legitimate, non-discriminatory reason for the rejection. The burden then shifts back to the plaintiff to rebut the reason offered by the defendant by showing it is a pretext for discrimination.

In this instance, the vague and conclusory allegations of disparate treatment that ICP asserts collectively against Defendants-Appellees are legally insufficient to support a reasonable inference of intentional race discrimination. In short, ICP essentially asks the panel to automatically view a "no voucher tenants" policy as synonymous with a "no black tenants" policy without providing adequate (well-pleaded) factual support for that linkage (as opposed to conclusory statements and assertions based on speculation, assumptions, and stereotypes). Defendants-Appellees' presumed awareness that the voucher population in the Dallas metro area is disproportionately black cannot alone be enough.

The same is true of Defendants-Appellees' alleged failure to respond to ICP's proposed financial incentive (one month's rent) and use of sublease and guarantor provisions purportedly sometimes utilized under other circumstances, e.g ., students subsidized by parents, first-time renters, and renters with low credit scores, and corporations subleasing to employees. Such conclusory allegations are at most "merely consistent with [ ] liability," and lack the factual support necessary to support a reasonable , rather than speculative, inference of intentional discrimination. For instance, ICP includes no facts supporting the claimed general existence of an otherwise qualified pool of voucher recipient applicants for Defendants-Appellees' properties. Nor is it plausible, based solely on ICP's conclusory assertions, that the proposed subleases and guarantees render ICP and its government voucher beneficiaries sufficiently similarly situated to business entities subleasing rental units to their employees, and credit-worthy parents serving as guarantors for students, such that Defendants-Appellees' lack of response is indicative of intentional race discrimination.

C. Disparate Treatment Liability: Lincoln

The disparate treatment asserted only against Lincoln differs from that asserted collectively against all Defendants-Appellees. Specifically, ICP contends Lincoln's alleged refusal to negotiate with or rent to otherwise qualified voucher households in majority white areas while, at the same time, negotiating with and renting to voucher holders in majority minority areas, evidences intentional race discrimination for purposes of the disparate treatment standard of liability under 42 U.S.C. § 3604 (a) and 42 U.S.C. § 1982 . 13

Although ICP's complaint and briefs arguably intimate that Lincoln accepts vouchers at certain other properties located in predominately minority areas (as opposed to majority white areas), the pertinent allegations are insufficiently clear to provide the necessary certainty. Even after oral argument, it is unclear whether Lincoln's acceptance of vouchers is alleged to occur at properties in minority areas other than those for which voucher acceptance is mandated by law (e.g., in exchange for low-income tax credits) or contract, and thus is not the subject of Lincoln's discretion.

*912 On such a bare showing, we find no error in the district court's dismissal of the claim.

D. Advertising Liability: Lincoln

As its fourth claim under the FHA, ICP alleges Lincoln's advertisements violate 42 U.S.C. § 3604 (c), which "prohibits advertisements for rental dwellings that show preference or discriminate based on race or color." 42 U.S.C. § 3604 (c). Specifically, ICP contends Lincoln's statements of its "no vouchers" policy in the advertisements "appeal to the stereotype that because voucher tenants are Black, voucher tenants are undesirable as tenants and that the exclusion of voucher households makes the complex a more desirable place for White non-Hispanic tenants to live." Additionally, ICP contends "[t]he advertising injures ICP by perpetuating the stereotype that Black voucher households are inferior and undesirable." The advertisements about which ICP complains include the following three statements:

Our community is not authorized to accept housing vouchers.

Our community is not authorized to accept Section 8 housing.

Our community is not authorized to accept ANY government subsidized rent programs.

The district court rejected this claim, finding that Lincoln's advertisements "do not involve race or show any sort of racial preference." In making this determination, the district court concluded the advertisements do not "suggest[ ] to an 'ordinary reader' that a particular race is preferred or not preferred," and "[o]ne race is not synonymous with the words 'Section 8 housing', 'government subsidized rent program,' or 'housing voucher.' " Citing Miami Valley Fair Housing Center, Inc. v. Connor Group., 725 F.3d 571 , 577 (6th Cir. 2013) for the proposition that " '[a]n ordinary reader' is neither the most suspicious nor the most insensitive person in our society," the district court concluded ICP had failed to convince it that "an ordinary reader would automatically equate Section 8 housing with [black] applicants."

On appeal, ICP emphasizes that Lincoln utilizes these advertisements only in majority white census tracts, and questions the necessity of using three different statements to state what ICP considers to be the same single message, i.e. that vouchers are not accepted in the rental community. The advertisement, however, contains no explicit reference to race; rather, it simply states Lincoln's policies regarding the acceptance of vouchers or other government rent subsidies. And, while an ordinary reader might think one "no acceptance" statement is adequate, or question whether all three statements are necessary, the supposition that that an ordinary reader would infer a racial preference from them is entirely speculative and unwarranted. Indeed, including all three explicit statements likely lessens confusion and streamlines the rental process for prospective tenants and landlords. Furthermore, it is entirely logical that these advertisements would be utilized only where vouchers are not accepted. Accordingly, on the record before us, we likewise find no error in the district court's dismissal of this claim.

IV.

Based on the foregoing, we find the district court properly concluded that ICP's allegations fail to state claims upon which relief legally can be granted. Accordingly, the district court's judgment of DISMISSAL WITH PREJUDICE is AFFIRMED.

W. EUGENE DAVIS, Circuit Judge, concurring in part and dissenting in part:

While I concur in the majority's decision to affirm the dismissal of ICP's disparate-treatment *913 and discriminatory advertising claims, I strongly dissent from the majority's decision to affirm the dismissal of ICP's disparate-impact claim. The question presented is whether ICP has alleged a plausible disparate-impact claim under the Fair Housing Act ("FHA"). Before that question can be answered, the type of disparate-impact claim asserted must be identified. As ICP contends, and as explained below, there are two different types of disparate-impact claims that may be asserted under the FHA, requiring different elements for a prima facie case. The first type of disparate-impact claim is like that recognized traditionally in the employment discrimination context; i.e. , the defendant's facially-neutral policy or practice has a disproportionately adverse effect on a protected group. The second type of disparate-impact claim, not found in the employment discrimination context, is that the defendant's policy or practice harms the community in general by perpetuating segregation.

ICP asserts both types of disparate-impact claims in its complaint. In analyzing those claims under Rule 12(b)(6), however, the district court erred by failing to distinguish these two claims, mixing up their prima facie elements, and examining the alleged statistical information under its distorted framework. The majority compounds the district court's error by not only failing to differentiate the claims and their prima facie elements, but also, in defining "robust causation," creating new prima facie elements simply not found in any controlling precedent. Moreover, the majority's interpretation of "robust causation" threatens to eviscerate disparate-impact claims under the FHA altogether.

In my view, and for the following reasons, ICP has set forth a plausible disparate-impact claim under the FHA in this case. The district court's decision dismissing the claim pursuant to Rule 12(b)(6) should be reversed, and ICP should be allowed to proceed.

I.

ICP asserts that Defendants' facially neutral "no vouchers" policy excludes a disproportionately Black population from housing at Defendants' properties, in violation of the FHA. As support for its disparate-impact claim, ICP alleged that the renter population in the Dallas area is such that Defendants'' "no vouchers" policy has a greater adverse impact on the Black renter population than the White renter population. Specifically, ICP offered statistical information indicating that the group of renter households that the policy excludes from Defendants' apartment complexes (voucher households in the Dallas area) is predominantly Black, and the group of renter households that the policy does not exclude (non-voucher households in the Dallas area) is predominantly White. This statistical information, accepted as true at the Rule 12(b)(6) stage, 1 makes plausible that, while neutral on its face, Defendants' "no vouchers" policy operates to exclude more Black renters from housing at Defendants' properties than White renters and, thus, has a discriminatory effect in violation of the FHA.

ICP also asserts that Defendants' "no vouchers" policy has the discriminatory effect of perpetuating racial segregation in the Dallas area by excluding predominantly Black voucher households from predominantly White census tracts. However, because *914 ICP's traditional disparate-impact claim is more straightforward, and its segregative-effect claim is based on census-tract data the majority finds problematic, the analysis that follows is limited to ICP's traditional disparate-impact claim, which alone provides a basis for reversing the district court's Rule 12(b)(6) dismissal. The majority's decision today reflects a fundamental misunderstanding of disparate-impact liability under the FHA, rendering it unable to discern the significance of the statistical data ICP alleged in support of its traditional disparate-impact claim. Discussion of the development of disparate-impact liability under Supreme Court jurisprudence, therefore, is in order. Moreover, analysis of the Supreme Court's precedents regarding disparate-impact liability is key to understanding what the Supreme Court meant by "robust causation" in its decision in Texas Department of Housing & Community Affairs v. Inclusive Communities Project, Inc., --- U.S. ----, 135 S.Ct. 2507 , 192 L.Ed.2d 514 (2015) ( ICP ), which governs this case.

II.

A. Overview of Disparate-Impact Liability Under Supreme Court Jurisprudence.

"[A] plaintiff bringing a disparate-impact claim challenges practices that have a 'disproportionately adverse effect on minorities' and are otherwise unjustified by a legitimate rationale." ICP, 135 S.Ct. at 2513 (citation omitted). In ICP , the Supreme Court determined, as a matter of first impression, that disparate-impact claims are cognizable under the FHA. In doing so, the Court relied on its prior precedents recognizing that such claims could be asserted in the employment discrimination context.

1. Griggs v. Duke Power

As described in ICP , the Supreme Court first addressed disparate-impact liability in Griggs v. Duke Power Co. , 401 U.S. 424 , 91 S.Ct. 849 , 28 L.Ed.2d 158 (1971), where the plaintiffs asserted claims of racial discrimination in employment under Title VII of the Civil Rights Act of 1964. See ICP , 135 S.Ct. at 2516-20 . In Griggs , the defendant-employer, a North Carolina power company, adopted a policy requiring a high school education or the passing of a standardized general intelligence test as a condition of employment in the company. 401 U.S. at 426-28 , 91 S.Ct. 849 . However, the evidence showed that White people "register[ed] far better on [these] requirements than" Black people. Id. at 430 , 91 S.Ct. 849 (citation omitted). In discussing the reasons why Whites fared better, the Supreme Court noted: "In North Carolina, 1960 census statistics show[ed] that, while 34% of [W]hite males had completed high school, only 12% of [Black] males had done so." Id. at 430 n.6, 91 S.Ct. 849 . Moreover, with respect to the standardized tests required by the defendant-employer, the Court noted evidence that "58% of [W]hites pass[ed] the tests, as compared with only 6% of [B]lacks." Id.

The plaintiffs in Griggs argued that "because the two requirements operated to render ineligible a markedly disproportionate number of [Black people], they were unlawful under Title VII unless shown to be job related." Id. at 429 , 91 S.Ct. 849 . The Court agreed with the plaintiffs and determined that Title VII "proscribe[d] not only overt discrimination but also practices that are fair in form, but discriminatory in operation. The touchstone is business necessity." Id. at 431 , 91 S.Ct. 849 . Emphasizing that Title VII condemned discriminatory preference for any group, whether minority or majority, the Court stated: "What is required by Congress *915 is the removal of artificial, arbitrary, and unnecessary barriers to employment when the barriers operate invidiously to discriminate on the basis of racial or other impermissible classification." Id. The Court held that "[i]f an employment practice which operates to exclude [Black people] cannot be shown to be related to job performance, the practice is prohibited." Id. Because the defendant-employer was unable to show that its requirements of a high school education and the passing of standardized intelligence tests were related to job performance, the Court held that the requirements were unlawful under Title VII. Id. at 433-35 , 91 S.Ct. 849 .

2. Wards Cove v. Atonio

Approximately eighteen years after deciding Griggs , the Supreme Court issued another important decision regarding disparate-impact employment discrimination claims. In Wards Cove Packing Co. v. Atonio , 490 U.S. 642 , 109 S.Ct. 2115 , 104 L.Ed.2d 733 (1989) ( Wards Cove ), the plaintiffs were employees of a company that operated salmon canneries in Alaska. The plaintiffs sued their employer under Title VII based on statistics showing that a high percentage of the skilled positions (noncannery jobs), which garnered higher pay, were filled by predominantly White employees, while the unskilled positions (cannery positions) with lower pay were filled by predominantly minority employees. Wards Cove , 490 U.S. at 647, 109 S.Ct. 2115 . The plaintiffs alleged that a variety of the employer's hiring and promotion practices, such as "nepotism, a rehiring preference, lack of objective hiring criteria, separate hiring channels, and a practice of not promoting within," caused the racial stratification of the work force. Id. at 647-48 , 109 S.Ct. 2115 . The Ninth Circuit determined that the plaintiffs "had made out a prima facie case of disparate impact ... rel[ying] solely on respondents' statistics showing a high percentage of [minority] workers in the cannery jobs and a low percentage of such workers in the noncannery positions." Id. at 650 , 109 S.Ct. 2115 (footnote omitted).

The Supreme Court reversed. It held that the Ninth Circuit's ruling "misapprehend[ed] [the Court's] precedents and the purposes of Title VII." Id. As the Court explained, under the Ninth Circuit's theory, "simply because [minorities] comprised 52% of the cannery workers at the cannery in question, [plaintiffs] would be successful in establishing a prima facie case of racial discrimination under Title VII." Id. at 652 , 109 S.Ct. 2115 (citation omitted). The Court further stated that the Ninth Circuit's "theory, at the very least, would mean that any employer who had a segment of his work force that was-for some reason-racially imbalanced could be haled into court and forced to ... defend[ ] the 'business necessity' of the methods used to select the other members of his work force." Id. The Court noted that such a theory would leave employers little choice but to adopt racial quotas, which was "far from the intent of Title VII." Id. (citation omitted).

The Court then turned to "the question of causation in a disparate-impact case" and the "plaintiff's burden in establishing a prima facie case." Id. at 656 , 109 S.Ct. 2115 . The Court noted that, as stated in its prior cases, the first step in establishing a prima facie case of disparate impact is that the plaintiff must "identify[ ] the specific employment practice that is challenged." Id. at 655 , 109 S.Ct. 2115 (quoting Watson v. Fort Worth Bank & Trust , 487 U.S. 977 , 994, 108 S.Ct. 2777 , 101 L.Ed.2d 827 (1988) ). Second, the plaintiff "must demonstrate that it is the application of a specific or particular employment practice that has created the disparate impact under attack." Id. at 657, 109 S.Ct. 2115 . In proving *916 such causation, "the plaintiff must offer statistical evidence of a kind and degree sufficient to show that the practice in question has caused the exclusion of applicants for jobs or promotions because of their membership in a protected group." Watson , 487 U.S. at 994 , 108 S.Ct. 2777 . The "statistical disparities must be sufficiently substantial that they raise such an inference of causation." Id. at 995 , 108 S.Ct. 2777 .

As set forth above, Griggs and Wards Cove explained that disparate-impact claims under Title VII involve challenges to facially neutral policies or practices that, in operation, have discriminatory effects on minorities with respect to employment. The cases also provided the standards for establishing a prima facie case of disparate impact. Specifically, a plaintiff must first identify the policy or practice being challenged, and next prove through statistical evidence that application of the policy or practice causes a disproportionate adverse effect on minorities with respect to employment. As detailed below, the ICP Court relied on these precedents in holding that the FHA encompasses disparate-impact claims and in indicating the standards that should apply to such claims.

B. Development of Disparate-Impact Claims under the FHA in the Courts of Appeals.

Although the Supreme Court had not held disparate-impact claims cognizable under the FHA prior to its 2015 decision in ICP , all of the Courts of Appeals to have addressed the question-including this court-had concluded that the FHA encompassed such claims. See ICP , 135 S.Ct. at 2519 (listing appellate court decisions). The Courts of Appeals determined that the FHA encompassed not only disparate-impact claims of the kind traditionally recognized in employment discrimination cases, but also "segregative-effect" claims. As stated above, ICP asserts both types of claims in its complaint herein. Understanding the difference between these claims and the type of statistical evidence used to support them is key to determining whether ICP has alleged a plausible disparate-impact claim under the FHA in this case.

One of the most significant appellate court cases, described by the ICP Court as "resid[ing] at the heartland of disparate-impact liability [under the FHA]," ICP , 135 S.Ct. at 2522 , is the Second Circuit's decision in Huntington Branch, NAACP v. Town of Huntington , 844 F.2d 926 (2d Cir. 1988). In that case, the plaintiffs sued the Town of Huntington and members of its board to challenge a zoning ordinance that restricted private construction of multi-family housing to a small urban renewal area that was 52% minority. Id. at 928 . The plaintiffs sought to construct an integrated, multi-family subsidized apartment complex in a part of the town that was virtually all White. Id. When the town refused to rezone the area where plaintiffs wished to build the property, the plaintiffs sued under the FHA to compel the town to amend its ordinance and rezone the area to permit construction. Id.

In a matter of first impression, the Second Circuit held that disparate-impact claims were cognizable under the FHA. The court stated: "[J]ust as the Supreme Court held that Title VII is violated by a showing of discriminatory effect, we hold that a [FHA] violation can be established without proof of discriminatory intent." Id. at 935 (citing Griggs , 401 U.S. at 429-36 , 91 S.Ct. 849 ). Furthermore, the court determined that the FHA allowed for two types of discriminatory effect claims: (1) "adverse impact on a particular minority group," akin to the disparate-impact claims traditionally recognized in the employment discrimination context, and (2) "harm to *917 the community generally by perpetuation of segregation," also known as "segregative-effect" claims. Id. at 937 (citation omitted). The court stated that "recognizing this second form of effect advance[d] the principal purpose of [the FHA] to promote open, integrated residential housing patterns." Id. (citations omitted).

The plaintiffs in Huntington Branch asserted both types of claims under the FHA. As to their traditional disparate-impact claim, the plaintiffs argued that the town's refusal to amend its restrictive zoning ordinance to permit construction of low-cost housing on the desired site had a substantial adverse impact on minorities with respect to the availability of housing. Id. at 938. In support of their claim, the plaintiffs submitted the following statistical information: (1) minorities constituted a far greater percentage of the current occupants of subsidized rental projects compared to their percentage in the town's population; (2) 60% of Section 8 voucher holders consisted of minorities; (3) 61% of those on the waiting list for Section 8 vouchers consisted of minorities; and (4) 7% of all Huntington families needed subsidized housing, while 24% of Black families needed such housing. Id. Based on these statistics, the court concluded that the town's failure to amend its restrictive ordinance and rezone the land to permit the construction of subsidized housing "had a substantial adverse impact on minorities." Id. (footnote omitted).

As to their segregative-effect claim, the plaintiffs asserted that allowing for construction of subsidized housing, which had a goal of housing 25% minorities, "would begin desegregating a neighborhood which [was at that time] 98% [W]hite." Id. at 937. Moreover, by refusing to permit construction of the project outside the urban renewal area, which already had a high concentration of minorities, the town "reinforced racial segregation in housing." Id. The evidence relating to the plaintiffs' segregative-effect claim included the following census-tract data: (1) 43% of the total Black population lived in four census tracts in one neighborhood; (2) 27% of the total black population lived in two census tracts; (3) outside of those two neighborhoods, the town's population was overwhelmingly White; and (4) of the forty-eight census tracts in the town, thirty contained Black populations of less than 1%. Id. at 929. Based on these statistics, the court determined the plaintiffs proved that the town "significantly perpetuated segregation" by refusing to amend its ordinance to permit construction of subsidized housing. Id. at 938. 2

Thus, Huntington Branch established that the FHA encompasses two types of discriminatory effect claims: (1) claims alleging a disparate impact on minorities with respect to the availability of housing and (2) claims alleging perpetuation of segregation. Huntington Branch also detailed the types of statistical data that can be used to prove a prima facie case of traditional disparate impact under the FHA and a prima facie case of segregative effect under the FHA. Specifically, the data for a disparate-impact claim should indicate that the group affected adversely by the challenged *918 policy is predominantly comprised of minorities, while the group unaffected is predominantly White. The data for a segregative-effect claim, which may include census-tract data, should indicate the existing locations of the predominantly White and predominantly minority areas of the town or city and how the policy or ordinance being challenged will continue the existing segregation.

C. The Supreme Court's Decision in ICP .

In ICP , the Supreme Court held that disparate-impact claims are cognizable under the FHA. The Court determined that its prior precedents in the employment discrimination arena "instruct[ed] that antidiscrimination laws must be construed to encompass disparate-impact claims when their text refers to the consequences of actions and not just to the mindset of actors, and where that interpretation is consistent with statutory purpose." ICP , 135 S.Ct. at 2518 . The Court determined that the FHA contained "results-oriented language counsel[ing] in favor of recognizing disparate-impact liability." Id. (citation omitted). It also concluded that such liability was "supported by amendments to the FHA that Congress enacted in 1988," which indicated that "Congress [had] accepted and ratified the unanimous holdings of the Courts of Appeals finding disparate-impact liability [under the FHA]." Id. at 2519 .

The Court additionally determined that disparate-impact claims are consistent with the FHA's central purpose, which like Title VII and the Age Discrimination in Employment Act, "was enacted to eradicate discriminatory practices within a sector of our Nation's economy." Id. at 2521 (citations omitted). The Court explained that "[t]hese unlawful practices include zoning laws and other housing restrictions that function unfairly to exclude minorities from certain neighborhoods without any sufficient justification." Id. at 2521-22 . The Court further stated that "[s]uits targeting such practices reside at the heartland of disparate-impact liability." Id. at 2522 (citing, inter alia , Huntington , 488 U.S. at 16-18, 109 S.Ct. 276 ).

Although the Court held that disparate-impact claims are cognizable under the FHA, it noted that "disparate-impact liability has always been properly limited." Id. The Court explained, echoing the principles set forth in Griggs and Wards Cove , that such liability is not "imposed based solely on a showing of a statistical disparity" and that it "mandates the 'removal of artificial, arbitrary, and unnecessary barriers.' " Id. (quoting Griggs, 401 U.S. at 431 , 91 S.Ct. 849 ). In a nod to the appellate court decisions recognizing segregative-effect claims, the Court noted that the FHA aims to ensure that the priorities of housing authorities, private developers, and landlords "can be achieved without arbitrarily creating discriminatory effects or perpetuating segregation." Id.

Furthermore, the Court emphasized that the plaintiff asserting a disparate-impact claim under the FHA must be able, as part of the prima facie case, to "point to a defendant's policy or policies causing" the disparity alleged. Id. at 2523. Citing Wards Cove , the Court described this causality requirement as "robust" and explained that such a requirement "protect[ed] defendants from being held liable for racial disparities they did not create." Id. (citing Wards Cove , 490 U.S. at 653, 109 S.Ct. 2115 ). The Court stated that such "safeguards at the prima facie stage" were needed to prevent governmental or private entities from believing numerical quotas were required, which could raise "serious constitutional issues." Id. (citations omitted).

*919 Turning to the specific allegations of the complaint in that case, however, the Court noted that ICP's disparate-impact claim was based on a "novel theory of liability." Id. at 2522 (citing Stacey E. Seicshnaydre, Is Disparate Impact Having Any Impact? An Appellate Analysis of Forty Years of Disparate Impact Claims Under the Fair Housing Act , 63 AM. U. L. REV. 357, 360-63 (2013) ). Specifically, ICP alleged that the Texas Department of Housing and Community Affairs ("Department") "caused continued segregated housing patterns by its disproportionate allocation of [low-income housing] tax credits." Id. at 2514. ICP contended that the Department "grant[ed] too many credits for housing in predominantly [B]lack inner-city areas and too few in predominantly [W]hite suburban neighborhoods." Id.

In support of its assertions, ICP submitted statistical evidence showing the location of low-income housing throughout the city of Dallas. 3 Id. However, much like the plaintiffs in Wards Cove, ICP did not identify a specific Department policy or practice causing the segregated housing patterns. Foretelling the district court's ultimate decision on remand, the Court noted that ICP's claim "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." ICP , 135 S.Ct. at 2522 .

On remand, the district court determined, as predicted by the Supreme Court, that ICP had failed to identify a specific, facially neutral policy that caused the disparity in the location of low-income housing. See Inclusive Cmtys. Project, Inc. v. Tex. Dep't of Hous. & Cmty. Affairs , No. 3:08-CV-0546-D, 2016 WL 4494322 , *6 (N.D. Tex. Aug. 26, 2016). Instead, ICP relied simply on the "cumulative effects" of the Department's decision-making process over a multi-year period. Id. The district court held that ICP could not rely on this generalized policy of discretion to prove disparate impact under the FHA and determined that ICP, consequently, failed to prove a prima facie case of disparate impact under the FHA. Id. at *7-8.

III.

Unlike its disparate-impact claim in ICP , in the instant matter, ICP clearly challenges a policy-Defendants' policy of not renting to voucher holders-and asserts that this policy causes a disparate impact on Black persons with respect to housing. Specifically, ICP contends that Defendants' policy causes the exclusion of the predominantly Black voucher population in the Dallas area from Defendants' apartment complexes, while the disproportionately White population without vouchers is not excluded by the policy.

ICP alleges that the voucher population in the Dallas area, the group affected by Defendants' policy, is over 80% Black and 10% or less White. The non-voucher population in the Dallas area, the group unaffected by Defendants' policy, is alleged to be 19% Black and 53% White. Therefore, ICP contends that when Defendants apply their "no vouchers" policy, the Black renter population is disproportionately and adversely *920 affected in violation of the FHA. The statistical information alleged by ICP is similar to the evidence offered by the plaintiffs in Huntington Branch , which evidence indicated, inter alia , that 60% of the Section 8 voucher holders consisted of minorities. In this matter, the percentage is alleged to be even higher-over 80% of the voucher holders are Black.

Moreover, district court cases addressing similar challenges to landlords' "no vouchers" policies with similar statistical evidence have determined that such claims demonstrate a disproportionate impact on Black persons in violation of the FHA. See, e.g., Nat'l Fair Hous. All. v. Travelers Indem. Co. , 261 F.Supp.3d 20 (D.D.C. 2017) (holding that plaintiffs showed "robust causation" under ICP in disparate-impact challenge to insurance company's policy of refusing to provide habitational insurance policies to landlords who rent to Section 8 tenants, where voucher population was 92% Black and 81.5% women); Bronson v. Crestwood Lake Section 1 Holding Corp. , 724 F.Supp. 148 , 154 (S.D.N.Y. 1989) (finding that disparate impact of landlord's "no Section 8" policy on minority persons in comparison to White applicants was "not surprising," given that 82.6% of Section 8 voucher holders in the City of Yonkers were minority persons). 4

Here, as ICP contends, the alleged causal connection between Defendants' policy and the resulting disparate impact on the Black renter population is direct and robust. 5 Specifically, Defendants' "no vouchers" policy causes the total exclusion of voucher households from Defendants' apartment complexes. Taking ICP's statistical information as true under Rule 12, because the voucher households in the Dallas area are disproportionately Black and the non-voucher households are disproportionately White, ICP's claim that application of Defendants' "no vouchers" policy results in a prohibited discriminatory effect under the FHA is plausible.

IV.

In determining that ICP failed to state a disparate-impact claim, the district court did not distinguish ICP's traditional disparate-impact claim from its segregative-effect claim. Furthermore, the district court mixed up the prima facie elements of the different claims and examined the alleged statistical information under its distorted framework. For example, the district court *921 imposed the burden on ICP to show that the challenged policy caused the segregation existing in the census-tract data. However, as detailed above, a segregative-effect claim does not require the plaintiff to prove that the challenged policy caused the initial segregation, but that the policy will perpetuate it. Moreover, the district court also imposed the burden on ICP to show that Defendants' "no vouchers" policy "diminished the amount of rental opportunities for African American or Black prospective tenants previously available before Defendants' policy was implemented." The prima facie elements of a traditional disparate-impact claim under the FHA do not require ICP to prove a "before-and-after" effect of the policy.

The majority, however, somehow affirms the district court through its strained reading of what the Supreme Court meant by "robust causation" in ICP . Tellingly, in its quest to find the meaning of ICP 's"robust causation" requirement, the majority does not delve into the actual ICP decision, which provides the answer. As discussed above, the Supreme Court's discussion of "robust causation" was prompted by ICP's failure in that case to identify a policy or practice causing the alleged disparate impact on minorities. See ICP , 135 S.Ct. at 2523 . The Court's discussion made clear that the plaintiff first must identify an offending policy in order to establish a prima facie case in an FHA disparate-impact claim.

If there were any doubt as to what the Supreme Court meant by "robust causation," its citation to Wards Cove clarified that the Supreme Court was indicating that the standards it had previously set forth for disparate-impact claims in the employment discrimination context should be applied equally to the newly-recognized disparate-impact claims under the FHA. See id. In fact, in Reyes v. Waples Mobile Home Park Ltd. Partnership, upon which the majority relies, the Fourth Circuit similarly recognized that the Supreme Court in " Inclusive Communities cited to Wards Cove in explaining the robust causality requirement." 903 F.3d 415 , 426 n.6 (4th Cir. 2018). The Supreme Court made no indication that, by describing the causation needed for a disparate-impact claim as "robust," it was ushering in a new requirement for disparate-impact claims under the FHA not required for disparate-impact claims in the employment discrimination context.

The majority, instead, interprets "robust causation" to require the plaintiff in a disparate-impact claim under the FHA to establish that the challenged policy was previously unenforced or that the challenged policy caused a "pre-existing condition." There is no precedent requiring such elements as part of the plaintiff's prima facie case in either type of disparate-impact claim under the FHA. The majority's belief that "robust causation" requires a previously unenforced policy is based on its admitted "narrower construction" of the Fourth Circuit's decision in Reyes , which decision nowhere mentions such a requirement as part of the plaintiff's prima facie case. 6 See 903 F.3d at 415-33 . Moreover, the Reyes decision supports ICP's traditional disparate-impact claim here. The Fourth Circuit determined that the plaintiffs "satisfied the robust causality requirement" by offering statistical evidence that the challenged policy "was likely to cause *922 Latino tenants at [defendant's property] to be disproportionately subject to eviction compared to non-Latinos at [defendant's property]." Id. at 429 . In this matter, ICP has offered statistical information making plausible that Defendants' "no vouchers" policy excludes more Black renters from housing at Defendants' properties than White renters.

The majority's application of a "pre-existing condition" requirement is taken from the dissenting opinion in Reyes. The majority applies this requirement to ICP's segregative-effect claim. See ante at pp. 906-07. In doing so, like the district court, the majority fails to recognize that a segregative-effect claim requires a plaintiff to show existing patterns of segregation and that the challenged policy will perpetuate that segregation. Such a claim does not require that the policy caused the initial segregation. See Huntington Branch , 844 F.2d at 938 . The majority goes even further and applies the same flawed reasoning to ICP's traditional disparate-impact claim-the majority imposes the burden on ICP to show that Defendants' "blanket 'no vouchers' policy, or any change therein, caused black persons to be the dominant group of voucher holders in the Dallas area." See ante at p. 907. Imposing the burden on ICP to make such a showing is akin to requiring the plaintiffs in Griggs to show that their employer's policy caused Black persons not to have a high school education. Such a requirement turns disparate-impact liability on its head because it would compel the plaintiff to establish that the offending policy not only had a disparate impact on a protected group, but that somehow the policy also created the characteristics making the protected group susceptible to the disparate impact.

Furthermore, the fact that the Section 8 program is a voluntary government program does not foreclose a finding that Defendants' "no vouchers" policy is an "artificial, arbitrary, and unnecessary" barrier to housing under the FHA. 7 Contrary to the majority's suggestion, the Eighth Circuit's decision in Ellis v. City of Minneapolis , 860 F.3d 1106 (8th Cir. 2017), does not support dismissal of ICP's disparate-impact claim on this basis. In Ellis , the plaintiffs alleged that the City's heightened enforcement of its housing code at for-profit, low-income rental housing had a disparate impact on the availability of housing for individuals protected by the FHA. 860 F.3d at 1107 . However, the plaintiffs "mount[ed] no serious challenge to the housing code itself." Id. at 1112 . Additionally, the plaintiffs alleged that the City had "adopted a policy to discourage rental housing and effected such a policy through deliberate or negligent misapplication of the housing code." Id. But, the plaintiffs did "not plead[ ] sufficient facts to plausibly support the existence of such a policy." Id. at 1114 . In contrast, in this matter, the existence of Defendants' "no vouchers" policy is undisputed-in fact, as described above, the policy is advertised. Furthermore, ICP specifically alleges in its complaint that Defendants' policy is an "artificial, arbitrary, and unnecessary barrier" to housing that does not achieve any legitimate interest. ICP contends that although *923 landlord trade associations and other landlords have asserted various interests or concerns upon which their refusal to negotiate with or rent to voucher holders is allegedly based, those interests or concerns do not require the "no vouchers" policy. In sum, the inadequacies of the plaintiffs' alleged disparate-impact claim in Ellis are not present in this matter.

The Eleventh Circuit's unpublished, per curiam opinion in Oviedo Town Center II, L.L.L.P. v. City of Oviedo, Florida 759 F. App'x 828 , No. 17-14254, 2018 WL 6822693 (11th Cir. Dec. 28, 2018) ( Oviedo ), similarly involved inadequacies in an alleged disparate-impact claim. However, the legal principles discussed in Oviedo actually support ICP's disparate-impact claim. In that case, several real estate developers alleged that rate increases in water and sewage services adopted by the City of Oviedo caused a disparate impact on minorities, in violation of the FHA. Id. at 830-31, at *1. The district court dismissed the claim on summary judgment; the Eleventh Circuit affirmed, determining that the appellants "failed to establish a prima facie case of disparate impact under the [FHA]." Id. The court noted the Supreme Court's decision in ICP , and stated that even before the ICP decision, it "had arrived at similar conclusions ... about the need for a relevant statistical showing in order to support a disparate-impact claim under the FHA." Id. at 834, at *4 (citing Schwarz v. City of Treasure Island , 544 F.3d 1201 (11th Cir. 2008) ).

Specifically, the Eleventh Circuit had previously held in Schwarz that "in order to show disparate impact, the plaintiff must provide evidence comparing members of the protected class affected by the [policy] with nonmembers affected by the [policy]." Id. (citing Schwarz , 544 F.3d at 1217 ). The court explained: "If the percentage of members of the protected class ... affected was higher than the percentage of nonmembers impacted, this disproportionality could form the basis for a prima facie case of disparate impact." Id. However, in Schwarz , the plaintiff presented " no comparative data at all," and consequently the plaintiff failed to establish a prima facie case. Id. (citing Schwarz , 544 F.3d at 1217-18 ). The court determined that the plaintiffs' claim in Oviedo was similarly deficient. The only statistical data submitted by the plaintiffs showed that more racial minorities lived in the plaintiffs' low-income housing complex than lived in the rest of the City of Oviedo. Id. at 835-36, at *5. The court determined that such information "[did] not establish a disparate impact, let alone any causal connection between the [policy] and the disparate impact." Id. The court further stated that relying on such inadequate statistical data "to make a prima facie showing" was "precisely the circumstances the [Supreme] Court sought to avoid in Inclusive Communities . " 8 Id.

In this case, the statistical data submitted by ICP would satisfy the standards set forth by the Eleventh Circuit in Oviedo and Schwarz . Specifically, ICP included comparative data regarding the makeup of voucher households in the Dallas area (the group which is excluded by Defendants' "no vouchers" policy) versus non-voucher households (the group unaffected by the policy) in the Dallas area. As explained *924 above, this statistical information makes plausible that Defendants' "no vouchers" policy operates to exclude more Black renters from housing at Defendants' properties than White renters. Under the Eleventh Circuit's jurisprudence, this disproportionality forms the basis for a prima facie case of disparate impact. Thus, in actuality, Oviedo shows that the majority's analysis of ICP's claim is misguided.

Finally, the majority submits that the relief ICP seeks-elimination of Defendants' "no vouchers" policy-would impose a "burdensome and extreme mandate" on Defendants that only Congress could grant through express legislation. However, Congress has had the opportunity to exempt landlords from disparate-impact liability under the FHA for discriminating against potential tenants based on their status as voucher holders, but Congress has not done so. As described in ICP , the FHA contains some exemptions from disparate-impact liability. 135 S.Ct. at 2520-21 . For instance, the FHA does not prohibit discrimination against a person "because such person has been convicted by any court of competent jurisdiction of the illegal manufacture or distribution of a controlled substance." See 42 U.S.C. § 3607 (b)(4). And a landlord may impose reasonable restrictions regarding the maximum number of occupants permitted to occupy a dwelling. See id. § 3607(b)(1). A landlord's refusal to negotiate or rent to a tenant based on his or her status as a voucher holder, however, is not listed as an exemption. Therefore, under ICP , if such a policy results in a disparate impact on a protected group, then the policy may be subject to challenge under the FHA. 9 The majority cannot carve out this exemption on Congress' behalf.

* * *

Despite ICP's clear allegations that, based on statistical evidence, Black persons are disproportionately and adversely impacted by Defendants' "no vouchers" policy, the majority says the complaint fails to state a claim. This conclusion is based on the majority's view of the pleading and proof necessary to establish "robust causation" between Defendants' policy and the effect of that policy.

To sum up, the majority offers two primary reasons for its position. First, based on the dissent in the Fourth Circuit's Reyes opinion, the majority concludes that robust causation is not alleged because ICP failed to state that Defendants were responsible for the fact that Black renters in Dallas hold a disproportionate number of vouchers for low-income housing. Since obtaining a voucher requires a showing of limited financial resources, presumably the majority would require ICP to show why a disproportionate number of Black renters in Dallas had more limited financial resources than White renters and that defendants were responsible for this fact. The majority states: "Neither the aforementioned 'city-level data' nor the 'census-level data' cited by ICP supports an inference that the implementation of Defendants-Appellees' blanket 'no vouchers' policy, or any change therein, caused black persons to be the dominant group of voucher holders in the Dallas metro area (or any of the other census areas discussed by ICP)."

To require such proof for a plaintiff to establish causation would render disparate-impact liability under the FHA a dead letter. This was certainly not the model of a disparate-impact case adopted by the Supreme Court in Griggs under Title VII.

*925 The Court in that case held that the employer's policy of requiring an applicant for employment to have a high school education and to pass a standardized general intelligence test had an adverse impact on Black applicants. The Court did not require proof that the employer was responsible for the disproportionately lower educational levels and test scores for Blacks as compared to Whites.

Second, the majority contends that ICP is required to show that Defendants' "no vouchers" policy was "previously unenforced" or " newly enforced." In doing so, the majority misconstrues the carefully reasoned Reyes opinion. The Fourth Circuit in Reyes held that the plaintiffs stated a claim for disparate impact under the FHA when they alleged that the landlord's policy requiring all adult occupants to provide documentation evidencing legal status in the United States in order to renew their leases had an adverse impact on Latinos. This policy was challenged after the landlord began enforcing it not only against the leaseholder, but also against all occupants over the age of eighteen. The majority contends that the fact that this was a "change" in enforcement of a policy distinguishes it from today's case, where the challenged policy had always been enforced. No explanation is given why the consequences of a changed discriminatory policy should be viewed differently from a discriminatory policy implemented from the outset. Enforcement of a policy is always required in a disparate-impact claim; if a policy is not enforced, it has no impact.

For these reasons, I respectfully dissent.