City of Miami Gardens v. Wells Fargo & Co.

Case: 18-13152 Date Filed: 07/30/2019 Page: 1 of 49 [PUBLISH] IN THE UNITED STATES COURT OF APPEALS FOR THE ELEVENTH CIRCUIT ________________________ No. 18-13152 ________________________ D.C. Docket No. 1:14-cv-22203-FAM CITY OF MIAMI GARDENS, a Florida municipal corporation, Plaintiff-Appellant, versus WELLS FARGO & CO., WELLS FARGO BANK N.A., Defendants-Appellees. ________________________ Appeal from the United States District Court for the Southern District of Florida _______________________ (July 30, 2019) Before WILLIAM PRYOR, NEWSOM, and BRANCH, Circuit Judges. PER CURIAM: This appeal requires us to decide whether the summary-judgment standard applies in determining whether a plaintiff has standing when the district court has Case: 18-13152 Date Filed: 07/30/2019 Page: 2 of 49 limited discovery and the merits issues to be considered on summary judgment, and, if so, whether the plaintiff in this appeal introduced sufficient evidence of standing under that standard. The City of Miami Gardens filed a complaint against Wells Fargo & Co. and Wells Fargo Bank, N.A., alleging that they violated the Fair Housing Act, 42 U.S.C. §§ 3601–19, by steering black and Hispanic borrowers into higher-cost loans than similarly situated white borrowers. The district court bifurcated discovery, with the initial phase focused on whether the City could identify a violation that occurred within the two-year limitation period provided by the Act, id. § 3613(a)(1)(A). After initial discovery, the district court entered summary judgment in favor of Wells Fargo on the merits. The City challenges this ruling, but Wells Fargo argues that the district court should have dismissed the suit because the City failed to establish standing. Before oral argument, we asked the parties to address whether the City established standing under the standard ordinarily applicable at summary judgment and, if not, whether the limitations on the subject matter of discovery and summary judgment imposed by the district court mandate the application of a more lenient standard. We conclude that the ordinary standard applies and that the City has not established standing. We vacate and remand with instructions to dismiss for lack of subject- matter jurisdiction. 2 Case: 18-13152 Date Filed: 07/30/2019 Page: 3 of 49 I. BACKGROUND The City filed its initial complaint on June 13, 2014, alleging that between 2004 and 2008, Wells Fargo originated mortgage loans in “numerous geographic markets around the country” that violated the Fair Housing Act. The City did not allege that it had received such loans from Wells Fargo. Instead, the City asserted that Wells Fargo engaged in both redlining—the practice of denying credit to particular neighborhoods based on race—and reverse redlining—the practice of “flooding a minority community with exploitative loan products”— by “refusing to extend mortgage credit to minority borrowers . . . on equal terms as to non- minority borrowers” and “extending mortgage credit on predatory terms to minority borrowers in minority neighborhoods in Miami Gardens.” The district court dismissed the initial complaint without prejudice and instructed the City that any amended complaint would have to state “the exact violations of the Fair Housing Act” and “what specific predatory practices occurred in Miami Gardens and how minorities were allegedly targeted there.” The district court also determined that an amended complaint would need to “allege . . . the facts that confer standing to complain about private home foreclosures, the specific injury to the governmental entity, the precise number and dates of foreclosures, and the specific costs to the City of Miami Gardens.” To that end, the district court directed the City to detail “(1) how Miami Gardens is injured, (2) 3 Case: 18-13152 Date Filed: 07/30/2019 Page: 4 of 49 how that injury is traceable to the conduct of each Wells Fargo defendant, and (3) how the injury can be redressed with a favorable decision in this case.” The City twice amended its complaint. The amended complaint alleged that “African-Americans and Hispanics and residents of predominantly African-American and Hispanic neighborhoods in Miami Gardens . . . receive[d] mortgage loans from Wells Fargo that have materially less favorable terms than mortgage loans given by Wells Fargo to similarly situated whites and residents of predominantly white neighborhoods in Miami Gardens.” The complaint outlined a list of kinds of “predatory loans” that Wells Fargo allegedly “steered minorities into when they otherwise qualified for less expensive and less risky loans,” including high-cost loans (i.e., loans with an interest rate at least three percent above the Treasury rate prior to 2010 and one- and-a-half percent above the prime mortgage rate thereafter), subprime loans, interest-only loans, balloon-payment loans, loans with prepayment penalties, negative-amortization loans, no-documentation loans, higher-cost government loans, such as Federal Housing Administration and Veterans Affairs loans, home- equity line-of-credit loans, and adjustable-rate mortgage loans with “teaser rates” (loans in which the lifetime maximum rate is greater than the initial rate plus six percent). 4 Case: 18-13152 Date Filed: 07/30/2019 Page: 5 of 49 The amended complaint also addressed standing by alleging that loans issued to minority borrowers in Miami Gardens were more likely to go into default or foreclosure as a result of Wells Fargo’s alleged practice of steering those borrowers into higher-cost loans. These effects on the housing market in Miami Gardens allegedly caused the City to suffer “economic injury based upon reduced property tax revenues resulting from (a) the decreased value of the vacant properties themselves, and (b) the decreased value of properties surrounding the vacant properties.” Apart from the asserted impact on property-tax revenues, the foreclosures and defaults allegedly increased the “cost[s] of municipal services . . . to remedy blight and unsafe and dangerous conditions which exist at properties that were foreclosed as a result of Wells Fargo’s illegal lending practices.” The amended complaint also alleged that the City sustained non-economic injuries because Wells Fargo’s lending “impaired the City’s goals to assure that racial factors do not adversely affect the ability of any person to choose where to live in the City or . . . detract from the . . . benefits of living in an integrated society” and “adversely affected the City’s longstanding and active interest in promoting fair housing and securing the benefits of a stable racially non-discriminatory community.” The statute of limitations for claims under the Act requires a plaintiff to file suit “not later than 2 years after the occurrence or the termination of an alleged 5 Case: 18-13152 Date Filed: 07/30/2019 Page: 6 of 49 discriminatory housing practice.” 42 U.S.C. § 3613(a)(1)(A). Wells Fargo filed its first complaint on June 13, 2014, so for the complaint to be timely, an act of housing discrimination must have occurred on or after June 13, 2012. Although much of the amended complaint concerned subprime lending practices that ended before June 13, 2012, it also alleged that Wells Fargo “continued to issue predatory mortgage loans to minorities in Miami Gardens subsequent to June 13, 2012.” The alleged violations that occurred outside of the limitation period were actionable in principle under the continuing-violation doctrine of Havens Realty Corp. v. Coleman, 455 U.S. 363 (1982). Under that doctrine, if a plaintiff “challenges not just one incident of conduct violative of the Act, but an unlawful practice that continues into the limitations period, the complaint is timely” if the “last asserted occurrence of that practice” occurred within the limitation period. Id. at 380–81. Because the City could invoke the continuing-violation doctrine only if it could identify a loan violative of the Act that occurred within the limitation period, Wells Fargo moved to divide discovery into phases, with the initial phase focused on the threshold question whether the City could satisfy the statute of limitations. On January 19, 2016, the district court entered a scheduling order accepting Wells Fargo’s proposed bifurcation. The order instructed the parties “to complete initial discovery related to loans originated between June 13, 2012, and June 12, 2014,” 6 Case: 18-13152 Date Filed: 07/30/2019 Page: 7 of 49 by February 19, 2016, and imposed a deadline of February 29 to file “summary judgment motions on the statute of limitations issue.” On January 20, the City served requests for production of documents, requests for admission, and interrogatories. Five days later, Wells Fargo produced electronic data concerning 153 loans originated in Miami Gardens during the limitation period that included information about the characteristics of the borrowers and the details of the loans. Wells Fargo also produced “formal written policies” that the City requested. The City deposed three Wells Fargo officials, and Wells Fargo conducted two depositions, including a deposition of the City through its corporate representative. Fed. R. Civ. P. 30(b)(6). On January 29, the City filed one discovery motion, but it did not challenge the restriction of the subject matter of discovery or Wells Fargo’s responses or document production. Instead, the motion requested “a 30 day extension of time to conduct discovery,” so that the deadline for the completion of initial discovery would fall on March 21, and the deadline to file motions for summary judgment would fall on March 31. The City also filed an unopposed motion for a “14 day extension of time” for summary-judgment briefing. At a hearing on the City’s motions held on February 25, the magistrate judge asked the parties whether there was “other discovery that needs to be done for [Wells Fargo] to file [a] motion for summary judgment or for [the City] to file [its] 7 Case: 18-13152 Date Filed: 07/30/2019 Page: 8 of 49 response.” The City requested data concerning loans originated outside Miami Gardens in Miami-Dade County. The magistrate judge granted the City’s request for an extension of the period for briefing but denied the City’s other requests. It concluded that “discovery was in large part completed within the time frame set by [the district court] and the discovery the plaintiff is now seeking should have been raised with opposing counsel and the Court earlier.” On March 14, Wells Fargo moved for summary judgment. After the City filed its opposition, the district court placed the case “in civil suspense” because the Supreme Court granted certiorari to review a decision of this Court in an appeal that arose from a similar suit brought against Bank of America and Wells Fargo by the City of Miami. See City of Miami v. Bank of Am. Corp. (City of Miami I), 800 F.3d 1262 (11th Cir. 2015), cert. granted, 136 S. Ct. 2545 (2016), vacated and remanded sub nom. Bank of Am. Corp. v. City of Miami (City of Miami II), 137 S. Ct. 1296 (2017). The district court granted leave to Wells Fargo to refile its motion for summary judgment after the Supreme Court’s decision. Wells Fargo refiled on May 31, 2017. On June 13—more than a year after the parties completed briefing on the initial motion for summary judgment and the hearing on the City’s initial request for expanded discovery—the City moved to “defer Wells Fargo’s [motion] due to the need to conduct additional discovery under Federal Rule of Civil Procedure 8 Case: 18-13152 Date Filed: 07/30/2019 Page: 9 of 49 56(d).” One month later, the district court held a hearing on the City’s motion to defer consideration of the motion for summary judgment. At the hearing, the City explained that it requested deferral because Wells Fargo had raised “new business necessity defenses” that were beyond the limited scope of the statute-of-limitations issue. But when the district court reiterated that the only merits issue to be considered on summary judgment was whether the City could satisfy the statute of limitations, the City stated that the only “discovery” it “would need” would be permission to introduce a supplemental expert report by Ian Ayres, a professor at the Yale Law School and the Yale School of Management. The district court granted that request and the report was admitted into evidence. Wells Fargo’s motion for summary judgment raised three principal arguments. First, it argued that the City was bound by the testimony of its representative who testified under Rule 30(b)(6), and because that testimony “conceded . . . that the City could not identify any ‘predatory’ or ‘discriminatory’ loans in the Limitations Period,” the City could not introduce new evidence of discriminatory lending to supplement the representative’s profession of ignorance. Second, the motion contended that the City had not presented sufficient evidence to support a reasonable inference that the 153 loans originated by Wells Fargo in Miami Gardens during the limitation period were unlawful under either a disparate-treatment or a disparate-impact theory of discrimination. And third, the 9 Case: 18-13152 Date Filed: 07/30/2019 Page: 10 of 49 motion argued that the City had not introduced sufficient evidence to establish standing because the undisputed evidence “reflect[ed] that none of the 153 loans Wells Fargo originated in Miami Gardens during the Limitations Period foreclosed.” With respect to standing, the City argued that because it was proceeding under a continuing-violation theory of liability, it had no duty to identify an injury causally attributable to a loan originated during the limitation period. The City also argued that “the loans issued during the statutory period [were] likely to injure the City in the same manner as the loans” that were identified in the City’s complaint “as part of the continuing violation from the pre-limitations period.” In support of this conjecture, the City pointed out that Ayres had “identified a loan” originated in the limitation period “that ha[d] already been delinquent since it was issued, whereas the lower cost loan issued to the similarly situated white borrower ha[d] not encountered similar problems.” The City’s response on the merits relied principally on Ayres’s reports, which identified two government-insured home-purchase loans, referenced by the parties as loans HC2 and HC6, that were allegedly more expensive after controlling for various factors than loan NHW8, which was issued to an allegedly similarly situated white borrower. Ayres opined that the cost differential between the loans was “consistent with the hypothesis” of race-based discrimination, but he 10 Case: 18-13152 Date Filed: 07/30/2019 Page: 11 of 49 acknowledged that Wells Fargo “also issued some loans to minority borrowers that were priced lower than loans made to non-Hispanic white borrowers with similar characteristics.” Ayres’s conclusions were partly at odds with those of Wells Fargo’s expert, Bernard Siskin, who argued in a rebuttal report that there were key differences between loans HC2 and HC6 on the one hand and loan NHW8 on the other. Loan NHW8 was originated during a two-week period in which Wells Fargo offered a promotional pricing discount on all loans. And the minorities who received loans HC2 and HC6 opted to receive a higher interest rate in exchange for more lender credits, which operate as rebates to offset closing costs. The borrowers of loans HC2 and HC6 received $8,000 and $1,877 in lender credits, respectively, whereas the borrower of loan NHW8 received only $479. The district court granted summary judgment to Wells Fargo on the merits. It declined to address the issue of standing and ruled that the City had failed to establish a genuine issue of material fact as to whether Wells Fargo engaged in disparate-impact or disparate-treatment discrimination within the limitation period. The district court accepted Wells Fargo’s argument that the City was “bound by the testimony of its Rule 30(b)(6) representative.” Because that representative “conceded during his deposition that the City could not identify any ‘predatory’ or ‘discriminatory’ loans in the limitations period” and “was unaware of any 11 Case: 18-13152 Date Filed: 07/30/2019 Page: 12 of 49 information providing a basis for the City’s allegation that borrowers were or may have been eligible for ‘more favorable and less expensive loans,’” the district court concluded that the City was not permitted to supplement that testimony with additional evidence of discrimination under Rule 30(b)(6). The district court ruled, in the alternative, that even if Rule 30(b)(6) did not bar the introduction of other evidence of discrimination, the City’s evidence was insufficient to support a prima facie case that disparate-impact or disparate- treatment discrimination occurred within the limitation period. The district court concluded that the City’s disparate-impact claim failed because the City identified only two loans issued to minorities that were purportedly more expensive than loans issued to similarly situated white borrowers, which was not enough “to show the policies produced statistically-imbalanced lending patterns.” The district court also concluded that the City failed to present any evidence of a causal connection between Wells Fargo’s lending policies and the cost disparity. As for the City’s disparate-treatment claim, the district court concluded that because “[t]he minority borrowers opted to receive a higher rate of interest in exchange for lender credits . . . to defray closing costs” while their nonminority comparator received de minimis lender credits and a promotional discount, the minority borrowers of the higher cost loans identified by the City were not similarly situated to their alleged comparator. 12 Case: 18-13152 Date Filed: 07/30/2019 Page: 13 of 49 II. STANDARD OF REVIEW We review questions of subject-matter jurisdiction de novo. United States v. Pavlenko, 921 F.3d 1286, 1289 (11th Cir. 2019). III. DISCUSSION Wells Fargo argues that the district court should have dismissed the City’s suit for lack of standing because “the undisputed evidence confirmed that none of the 153 loans originated by Wells Fargo [within the limitation period] foreclosed,” so the City could not have suffered an injury as a result of any of these loans. We agree that the City has not satisfied the injury or causation elements of standing, but not for the reason provided by Wells Fargo. Article III limits the subject-matter jurisdiction of the federal courts to “Cases” and “Controversies.” U.S. Const. art. III, § 2. “To have a case or controversy, a litigant must establish that he has standing, which must exist ‘throughout all stages of litigation.’” United States v. Amodeo, 916 F.3d 967, 971 (11th Cir. 2019) (quoting Hollingsworth v. Perry, 570 U.S. 693, 705 (2013)). “The federal courts are under an independent obligation to examine their own jurisdiction, and standing is perhaps the most important of the jurisdictional doctrines.” United States v. Hays, 515 U.S. 737, 742 (1995) (alteration adopted) (citation and internal quotation marks omitted). 13 Case: 18-13152 Date Filed: 07/30/2019 Page: 14 of 49 Article III standing has three elements. First, “the plaintiff must have suffered an injury in fact—an invasion of a legally protected interest which is (a) concrete and particularized, and (b) actual or imminent, not conjectural or hypothetical.” Lujan v. Defs. of Wildlife, 504 U.S. 555, 560 (1992) (citations and internal quotation marks omitted). Second, “there must be a causal connection between the injury and the conduct complained of—the injury has to be fairly traceable to the challenged action of the defendant, and not the result of the independent action of some third party not before the court.” Id. (alterations adopted) (citation and internal quotation marks omitted). Third, “it must be likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision.” Id. at 561 (internal quotation marks omitted). “The party invoking federal jurisdiction bears the burden of establishing these elements.” Id. “Since they are not mere pleading requirements but rather an indispensable part of the plaintiff’s case, each element must be supported in the same way as any other matter on which the plaintiff bears the burden of proof, i.e., with the manner and degree of evidence required at the successive stages of litigation.” Id. Although “[a]t the pleading stage, . . . factual allegations of injury resulting from the defendant’s conduct may suffice,” “[i]n response to a summary judgment motion, . . . the plaintiff can no longer rest on such ‘mere allegations,’ but must ‘set forth’ by affidavit or other evidence ‘specific facts,’ . . . which for 14 Case: 18-13152 Date Filed: 07/30/2019 Page: 15 of 49 purposes of the summary judgment motion will be taken as true.” Id. (quoting Fed. R. Civ. P. 56(e)). At the summary-judgment stage, the burden to establish standing is satisfied only if “‘affidavits or other submissions indicate that a genuine issue of material fact exists concerning standing.’” Bischoff v. Osceola County, 222 F.3d 874, 881 (11th Cir. 2000) (quoting Munoz-Mendoza v. Pierce, 711 F.2d 421, 425 (1st Cir. 1983)). Wells Fargo’s argument about the City’s standing rests on a flawed premise. Wells Fargo assumes that the City has standing only if it suffered an injury that was caused by a loan originated during the limitation period, but this assumption conflates the constitutional requirements of standing with the statutory requirement of timeliness. Article III requires, among other things, that the plaintiff establish an injury in fact and causation, Lujan, 504 U.S. at 560, but an injury need not occur as a result of conduct that occurred within the timeframe provided by the statute of limitations applicable to the plaintiff’s cause of action to satisfy those requirements. The City has standing so long as one of the loans challenged as discriminatory has caused or will cause the City to suffer a de facto injury redressable by favorable decision. Whether a complaint about that loan or loans would be timely is a separate issue. In its initial briefing on appeal, the City did not point to any evidence that it sustained an injury traceable to the conduct of Wells Fargo, but in response to our 15 Case: 18-13152 Date Filed: 07/30/2019 Page: 16 of 49 request that parties address standing at oral argument, the City relied on two pieces of evidence drawn from the summary-judgment record. First, the City asserted that one of the allegedly discriminatory loans identified in Ayres’s reports, HC2, has been delinquent since it was issued, but that the loan issued to that borrower’s purported nonminority comparator, NHW8, has not. The City speculated that loan HC2 will likely go into foreclosure and cause the City to suffer the kind of economic injuries asserted in the operative complaint. Second, the City pointed to ten loans identified in the complaint and an attached exhibit that were originated before the limitation period. According to the complaint, the value of the properties associated with these loans has declined since they entered foreclosure between 2008 and 2012. This evidence is insufficient to establish standing. The City’s evidence of a risk that loan HC2 will go into foreclosure at some point in the future does not satisfy the requirement that a threatened injury be “imminent, not conjectural or hypothetical.” Lujan, 504 U.S. at 560 (citation and internal quotation marks omitted). As the Supreme Court has explained, a “threatened injury must be certainly impending to constitute injury in fact,” and “[a]llegations of possible future injury are not sufficient.” Clapper v. Amnesty Int’l, 568 U.S. 398, 409 (2013) (citations and internal quotation marks omitted). The delinquency of a single loan does not establish a certainly impending risk that the City will lose 16 Case: 18-13152 Date Filed: 07/30/2019 Page: 17 of 49 property-tax revenues or be forced to increase municipal spending to remediate blight. Whether the delinquency on this loan will result in foreclosure, and whether that foreclosure will have any impact on property values, property-tax revenues, or municipal spending, are questions left entirely open by the evidence in the summary-judgment record. The evidence that loan HC2 may go into foreclosure also fails to satisfy the requirement of causation. The complaint concedes that “isolat[ing] the lost property value attributable to Wells Fargo foreclosures” would have required the use of a “statistical regression technique that focuses on effects on neighboring properties” “known as Hedonic regression,” which involves the “study[] [of] thousands of housing transactions.” The City never conducted any analysis of this kind and probably could not do so in the light of the paucity of allegedly discriminatory loans identified by the City. So even if we were to assume that loan HC2 will enter into foreclosure and that the value of the property associated with that loan will decline as a result, we would not be able to determine the extent to which any decline in the value of the property would be “fairly traceable to the challenged action[s] of the defendant.” Lujan, 504 U.S. at 560 (alterations adopted) (citations and internal quotation marks omitted). The complaint’s reference to ten loans that have gone into foreclosure also does the City no good. The exhibit attached to the complaint attests that the values 17 Case: 18-13152 Date Filed: 07/30/2019 Page: 18 of 49 of the properties associated with these loans have declined since they entered foreclosure, but the City did not produce any evidence of the effect of these foreclosures on property-tax revenues or municipal spending. Nor did the City present any evidence that these loans were issued on discriminatory terms or otherwise attempt to isolate the contribution of Wells Fargo’s actions, if any, to the decline in property value sustained by these properties, presumably because that would require the kind of hedonic-regression analysis that the City avers would require an analysis of thousands of housing transactions. So these loans are inadequate to establish that the City suffered an “actual or imminent” injury that was “fairly traceable to the challenged action of the defendant, and not the result of the independent action of some third party not before the court,” id. (alterations adopted) (citations and internal quotation marks omitted), under the standard applicable at summary judgment. The City contends that because the district court specified that it would entertain “summary judgment motions on the statute of limitations issue” in its January 19, 2016, scheduling order, it would be unfair to hold the City to its burden to establish a genuine dispute of material fact as to standing. But the legal effect of the scheduling order was not to bar the parties from raising jurisdictional issues on summary judgment. The order limited the merits issues to be considered and stayed discovery on unrelated matters. The order adopted a “[d]eadline to file 18 Case: 18-13152 Date Filed: 07/30/2019 Page: 19 of 49 partial summary judgment motions on the statute of limitations issue,” and stayed “discovery on all matters unrelated to Wells Fargo loans originated between June 13, 2012, and June 12, 2014 . . . until resolution of the parties’ partial summary judgment motions on the statute of limitations issue.” The order was silent on whether consideration of the City’s standing would be deferred until a later date. Although the order’s reference to “partial summary judgments on the statute of limitations issue” might be taken to suggest that the statute of limitations was the only issue to be considered on summary judgment, that interpretation would unreasonably impute to the district court a disregard of its basic obligation to “first satisfy itself of [its] own jurisdiction” before proceeding to resolve any merits issue. Amodeo, 916 F.3d at 971; see also Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 101 (1998) (rejecting the doctrine of “hypothetical jurisdiction,” under which a court may “resolve contested questions of law when its jurisdiction is in doubt”). Its later actions confirm that the district court did not intend to defer consideration of threshold issues other than the statute-of-limitations defense. The district court stayed proceedings on June 30, 2016—after Wells Fargo had filed a motion for summary judgment raising the issue of standing under Article III—until the Supreme Court resolved the issue of the “standing of cities” to sue mortgage originators “for alleged . . . discriminatory lending practices” in City of Miami II. 19 Case: 18-13152 Date Filed: 07/30/2019 Page: 20 of 49 True, the question addressed in City of Miami II was whether comparable allegations of discriminatory lending satisfied the requirement of “prudential standing,” 137 S. Ct. at 1303, not standing under Article III. But the decision to stay proceedings pending the resolution of that issue would have made little sense if the district court intended to impose a strict limit on the consideration of any and all issues other than the statute of limitations at summary judgment. Even on the supposition that the initial intended effect of the scheduling order was to defer consideration of standing, the parties had made standing a contested issue by the time the district court reviewed Wells Fargo’s motion for summary judgment. Wells Fargo’s initial motion for summary judgment prominently challenged the City’s standing under Article III. The motion advised the City that “at the summary judgment stage, the City must actually produce some evidence that the City, and not just the borrower, has Article III standing to sustain a claim under the Fair Housing Act,” and it reminded the City that it could “no longer rest on . . . mere allegations,” but instead had the burden to “set forth by affidavit or other evidence specific facts showing that it has suffered a cognizable injury-in-fact.” This motion was filed in March 2016 and refiled in late May 2017. The City’s response to the motion disputed Wells Fargo’s challenges to its standing using the same evidence that the City raised in support of standing at oral argument—Ayres’s conclusion that loan HC2 was delinquent and the ten loans that 20 Case: 18-13152 Date Filed: 07/30/2019 Page: 21 of 49 had entered into foreclosure mentioned in the complaint. So for more than two years before the district court entered summary judgment in late June 2018, both parties operated under the assumption that the City’s standing was in dispute and actively litigated that issue. To be sure, both this Court and the Supreme Court have determined that in limited circumstances, the absence of notice of the need to prove standing may mandate either the application of a more lenient standard or remand for further development of the record. In Church v. City of Huntsville, 30 F.3d 1332 (11th Cir. 1994), we applied the pleading standard in determining whether the plaintiffs had standing to seek a preliminary injunction because the defendant “did not question [the] plaintiff[s’] standing” and “the plaintiffs had only a few hours of hearing time to present their preliminary injunction case” to the district court and were “forced to limit their evidence to what they reasonably understood to be the contested issues.” Id. at 1336. And in Alabama Legislative Black Caucus v. Alabama, 135 S. Ct. 1257 (2015), the Supreme Court concluded that, under the circumstances at issue, the evidence in the record was “strong enough to lead the [plaintiff] reasonably to believe” that it satisfied a requirement of standing and the defendant failed to argue otherwise, so “elementary principles of procedural fairness” required the district court to provide notice and an opportunity to respond before deciding sua sponte that the plaintiff had not satisfied that requirement. Id. at 1269. 21 Case: 18-13152 Date Filed: 07/30/2019 Page: 22 of 49 Both Huntsville and Alabama Black Legislative Caucus circumscribe the power of a court to consider standing sua sponte without providing a plaintiff with notice and an opportunity to respond, but these precedents do not purport to speak to circumstances like those of this appeal, in which the opposing party raised the issue of standing. See Ala. Legislative Black Caucus, 135 S. Ct. at 1269 (limiting the district court’s authority to “act[] sua sponte” without first giving the plaintiff “an opportunity to provide evidence” that it satisfied the standing requirement at issue); Bischoff, 222 F.3d at 882 n.8 (explaining that Huntsville only applied a “more lenient standard of review because the standing issue was decided by the district court so early in the case and without any notice to plaintiffs that standing was at issue”). In Huntsville, our analysis turned on the notion that “[i]t might well be unfair . . . to impose a standing burden beyond the sufficiency of the . . . pleadings on a plaintiff seeking a preliminary injunction, unless the defendant puts the plaintiff on notice that standing is contested,” at least insofar as the plaintiff had little time to present his case. 30 F.3d at 1336. In contrast, more than two years elapsed between Wells Fargo’s filing of its motion for summary judgment and the order granting it, so the City had more than enough time to take any steps necessary to ensure that it would be able to prove standing. And Alabama Legislative Black Caucus assumed that special notice was only necessary in a circumstance in which the plaintiff reasonably believed that he had satisfied a 22 Case: 18-13152 Date Filed: 07/30/2019 Page: 23 of 49 requirement of standing and the defendant had not argued the contrary. 135 S. Ct. at 1269. But as we have explained, Wells Fargo actively contested the City’s proof of injury and causation, the very elements of standing that we have determined the City failed to establish. At oral argument, the City maintained that it would be unfair to apply the summary-judgment standard because the district court limited discovery to matters related to loans originated within the limitation period, but this contention fails too. Although “summary judgment should not be granted until the party opposing the motion has had an adequate opportunity for discovery,” we have made clear that “the party opposing the motion for summary judgment bears the burden of calling to the district court’s attention any outstanding discovery.” Snook v. Tr. Co. of Ga. Bank of Savannah, N.A., 859 F.2d 865, 870–71 (11th Cir. 1988). Failure to satisfy this burden is fatal to an argument that the district court granted summary judgment prematurely by failing to order or await the results of further discovery. See Urquilla-Diaz v. Kaplan Univ., 780 F.3d 1039, 1063–64 (11th Cir. 2015); Reflectone, Inc. v. Farrand Optical Co., 862 F.2d 841, 843–44 (11th Cir. 1989). By the same token, no unfairness occurs if a plaintiff fails to advise the district court of the need for further discovery to prove standing at summary judgment and a circuit court decides in the first instance that the plaintiff failed to establish standing. In either circumstance, the plaintiff has effectively consented to 23 Case: 18-13152 Date Filed: 07/30/2019 Page: 24 of 49 adjudication of the issues raised in the summary-judgment motion based on the existing record by failing to avail itself of the opportunity to seek further discovery. So the City’s argument could prompt the application of a more lenient standard in evaluating standing or remand for additional discovery only if the City satisfied its “burden of calling to the district court’s attention any outstanding discovery” on the issue of standing. Snook, 859 F.2d at 871. The City failed to satisfy that burden. The preferred vehicle for advising a district court of the need for further discovery is an affidavit or declaration submitted under Federal Rule of Civil Procedure 56(d). That Rule provides that “[i]f a nonmovant shows by affidavit or declaration that . . . it cannot present facts essential to justify its opposition” to summary judgment, “the court may (1) defer considering the motion or deny it; (2) allow time to obtain affidavits or declarations or to take discovery; or (3) issue any other appropriate order.” To invoke this Rule, a party “may not simply rely on vague assertions that additional discovery will produce needed, but unspecified facts,” but “must specifically demonstrate how postponement of a ruling on the motion will enable him, by discovery or other means, to rebut the movant’s showing of the absence of a genuine issue of fact.” Reflectone, 862 F.2d at 843 (citation and internal quotation marks omitted). 24 Case: 18-13152 Date Filed: 07/30/2019 Page: 25 of 49 The City filed a declaration under Rule 56 at one point in the litigation, but it did not mention standing, and the City later retracted the declaration and opposed Wells Fargo’s motion on the merits. The City responded to Wells Fargo’s refiled motion for summary judgment with a motion to strike or, in the alternative, to stay or deny the motion to allow for additional discovery under Rule 56(d). The motion argued that Wells Fargo had “attempt[ed] to turn a partial motion for summary judgment limited to the issue of the statute of limitations, into a fully briefed motion for summary judgment that would decide the entire case,” but neither the motion itself nor the accompanying declaration mentioned the need for additional discovery to produce evidence of standing or otherwise clarify the basis of its objection to Wells Fargo’s motion. At a hearing conducted on July 20, 2017, the City explained that its motion and declaration were prompted by concerns that Wells Fargo’s motion had raised “new business necessity defenses.” The City conceded that if the only merits issue was “whether there were loans issued in that two-year time period where a minority was treated disparately and adversely relative to a similarly situated white,” it “ha[d] the data” to prevail, but it did not mention the need for additional discovery to support standing. When the district court intimated that the only merits issue under consideration was the statute of limitations and asked the City if it nonetheless needed more discovery, the City said that the only “discovery” it 25 Case: 18-13152 Date Filed: 07/30/2019 Page: 26 of 49 needed was for the district court to admit “the supplemental declaration of Dr. Ayres.” After the district court admitted the report into evidence, the City changed course and opposed the motion for summary judgment on the merits. We have no difficulty concluding that the City failed to “specifically demonstrate how postponement of a ruling on the motion” would enable it, “by discovery or other means, to rebut the movant’s showing of the absence of a genuine issue of fact.” Reflectone, 862 F.2d at 843 (citation and internal quotation marks omitted). The City’s declaration under Rule 56(d) was not sufficient for this purpose because it never even mentioned the need for further discovery to support standing. And although we have held that in limited circumstances, “the interests of justice will sometimes require a district court to postpone its ruling on a motion for summary judgment even though the technical requirements” of Rule 56(d) “have not been met,” we have limited that exception to the requirement to comply with Rule 56(d) to circumstances in which “the nonmovant properly apprised the district court of the outstanding discovery request” through an equivalent form of notice. Snook, 859 F.2d at 871; see also Reflectone, 862 F.2d at 844 (holding that the nonmovant was not entitled to invoke the exception because it “did not even make any motion to compel discovery” and “did not raise the issue anywhere in its papers opposing summary judgment”). The City’s later remarks to the district court did not suggest that its request for further discovery involved the need for 26 Case: 18-13152 Date Filed: 07/30/2019 Page: 27 of 49 additional information to establish standing, so this exception cannot apply. By withdrawing its opposition under Rule 56(d) and opposing the motion for summary judgment on the merits, the City acceded to the entry of judgment on any issue raised in the motion for summary judgment based on the existing record. And even on appeal, the City has failed to provide us with any explanation of how further discovery would have enabled it to establish standing. The City failed to satisfy its “burden of calling to the district court’s attention any outstanding discovery” that might have been necessary to support its standing, Reflectone, 862 F.2d at 844, so we cannot conclude that it would be unfair to the City to require it to establish standing under the standard ordinarily applicable at summary judgment. Because we have determined that the City has not satisfied that standard, we conclude that the City has not established “that a genuine issue of material fact exists concerning standing.” Bischoff, 222 F.3d at 881 (citation and internal quotation marks omitted). The district court should have dismissed the action for lack of standing. The parties briefed the issue at summary judgment, and it was clear that the City had no evidence of injury or causation. IV. CONCLUSION We VACATE the summary judgment in favor of Wells Fargo and REMAND with instructions to dismiss for lack of subject-matter jurisdiction. 27 Case: 18-13152 Date Filed: 07/30/2019 Page: 28 of 49 WILLIAM PRYOR, Circuit Judge, joined by BRANCH, Circuit Judge, concurring: Despite our earlier decision about the sufficiency of the pleadings in a related case, see City of Miami v. Wells Fargo & Co., 923 F.3d 1260 (11th Cir. 2019), it would be difficult to overstate how misguided this litigation has proved to be. For example, even if we had jurisdiction to decide the merits of this appeal, we would have to agree with the district court that Wells Fargo is entitled to summary judgment. To explain why, I recount below the City’s evidence of discrimination and then explain that the City failed to create a genuine dispute of material fact with respect to its disparate-treatment claim and that the City abandoned any challenge to the summary judgment against its disparate-impact claim. A. The City’s Evidence of Discrimination. The City’s principal evidence of discrimination was a pair of reports prepared by its expert, Ian Ayres. The first report concluded that “Wells Fargo issued loans to minority borrowers in Miami Gardens between June 13, 2012 and June 12, 2014 . . . that [were] more expensive or riskier than loans issued to non- Hispanic white borrowers with similar characteristics in Miami Gardens.” Ayres reached this conclusion by conducting a matched-pair analysis using data on 153 first-lien mortgages originated by Wells Fargo between those dates. To identify “high-cost loans,” Ayres relied on the standards adopted by the Federal Financial Institutions Examination Council under the Home Mortgage 28 Case: 18-13152 Date Filed: 07/30/2019 Page: 29 of 49 Disclosure Act, 12 U.S.C. §§ 2801–11. Under the then-applicable regulation,12 C.F.R. pt. 203, app. A(I)(G)(1)(a) (2016), rescinded by Home Mortgage Disclosure, 82 Fed. Reg. 60673 (Dec. 22, 2017), a lender was required to report the “rate spread” for a loan if the spread was equal to or greater than 1.5 percentage points for a first-lien loan. As Ayres explained, the “rate spread for a loan origination is the spread between the Annual Percentage Rate (APR) and a survey- based estimate of [Annual Percentage Rates] offered on originated prime mortgage loans of a comparable amortization type, interest rate lock-in date, fixed term (loan maturity) or variable term (initial-fixed rate period), and lien status.” “The survey- based estimates are referred to as the ‘average prime offer rate’ . . . .” Using the regulatory threshold, Ayres classified a loan as high-cost if its rate spread was equal to or greater than 1.5 percentage points. By this standard, Ayres identified “seven High-Cost Loans in Wells Fargo’s data, six of which were made to African- American borrowers and one of which was made to [a] Hispanic borrower.” No rate-spread reportable loans in the dataset were made to non-Hispanic white borrowers. From there, Ayres attempted to determine “whether a High-Cost Loan was issued to a minority borrower whereas a non-Hispanic white borrower with similar characteristics did not receive a High-Cost Loan.” Because the “rate spread already accounts for differences in the date of the loan’s rate lock, the length of the loan 29 Case: 18-13152 Date Filed: 07/30/2019 Page: 30 of 49 term, and whether the loan was a fixed-rate or [adjustable-rate mortgage] loan,” Ayres focused “only on those core underwriting differences not accounted for in the rate spread, such as occupancy status, [credit] score [as determined through the Fair Isaac Corporation’s model], the loan-to-value ratio (LTV), debt-to-income ratio (DTI), and the underwriting history of bankruptcy, foreclosures, charge-offs, collections, late payments, delinquencies, judgments, and public records on the borrower’s credit report.” After controlling for these variables, Ayres ultimately identified two high- cost loans issued to minority borrowers—labeled HC2 and HC6—that had a greater rate spread than a loan issued to a non-Hispanic white, NHW8. HC2 is a loan issued to a Hispanic borrower. The credit score for this borrower is 712, the borrower’s loan-to-value ratio is 98 percent, and the borrower’s debt-to-income ratio is 47 percent. HC6 is a loan to an African-American borrower with a credit score of 741, a loan-to-value ratio of 98 percent, and a debt-to-income ratio of 43 percent. NHW8, in contrast, is a loan issued to a non-Hispanic white borrower with a credit score of 702, a loan-to-value ratio of 98 percent, and a debt-to-income ratio of 41 percent. The rate spread for loan HC2 is 2.03 percent and its Annual Percentage Rate is 5.58 percent. The rate spread of loan HC6 is 1.58 percent and its Annual Percentage Rate is 6.00 percent. But the rate spread of loan NHW8 is 1.12 percent and its Annual Percentage Rate is 5.32 percent. 30 Case: 18-13152 Date Filed: 07/30/2019 Page: 31 of 49 Based on his definition of relative loan cost in terms of the rate-spread differential, Ayres determined that loans HC2 and HC6 “were priced higher . . . than Loan NHW8 that was originated to a non-Hispanic white borrower with similar (and in some cases, riskier) characteristics.” He explained that the borrowers of loans HC2 and HC6 “had higher [credit] scores, the same occupancy status, the same [loan-to-value ratios], the same underwriting history of bankruptcy, foreclosures, charge-offs, collections, late payments, delinquencies, judgments, and public records on the borrower’s credit report, but had slightly higher debt-to-income ratios than the Loan NHW8 borrower.” Ayres concluded that the evidence was “consistent with the hypothesis that Wells Fargo issued more expensive loans to minority borrowers than non-Hispanic white borrowers with similar characteristics even after controlling for plausible and generally accepted business justifications.” In his rebuttal report, Bernard Siskin, the expert for Wells Fargo, posited two alternative explanations of the rate-spread discrepancy. First, he explained that the borrowers of loans HC2 and HC6 “chose a higher note rate in exchange for significant lender credits to be used at settlement to pay closing costs,” but the borrower on loan NHW8 “received only de minimis lender credits.” The borrower on loan HC2 opted for $8,000 in lender credits—representing 7.34 percent of the loan amount—and the borrower on HC6 opted for $1,878 in lender credits—2.12 31 Case: 18-13152 Date Filed: 07/30/2019 Page: 32 of 49 percent of the loan amount—but the borrower on loan NHW8 opted for only $479 in lender credits—equal to 0.38 percent of the loan amount. Second, Siskin argued that Ayres “fail[ed] to account for the fact that the white borrower received a loan during the two-week period when a promotional pricing discount was applied to all conventional and government purchase loans.” As a result of its origination date, NHW8 “received a 50 basis points pricing discount.” Based on data provided by an official of Wells Fargo named Jill Hunt, Siskin calculated the hypothetical rate spread and Annual Percentage Rate on each of these loans after controlling for the effects of lender credits and the promotional discount. Hunt attested that the note rate of NHW8 would have been 0.125 percent higher if the pricing discount had not been applied, which led Siskin to conclude that zeroing that discount would yield an Annual Percentage Rate of 5.4431 percent and a rate spread of 1.25 percent. Hunt also stated that if the borrowers on loans HC2 and HC6 had elected not to receive lender credits, the note rate on the loans would have been 3.25 and 4.125 percent, respectively. Based on these numbers, Siskin calculated that, “[a]ssuming the lender credits were all applied to fees included in the [Annual Percentage Rate] computation and the calculation of the [Annual Percentage Rate] did not include the fees paid with lender credit,” zeroing the lender credits elected by HC2 would result in an Annual Percentage Rate of 4.71 percent and a rate spread of 1.16. With respect to HC6, the same 32 Case: 18-13152 Date Filed: 07/30/2019 Page: 33 of 49 calculation yielded an Annual Percentage Rate of 5.86 percent and a rate spread of 1.44 percent. So under the analysis conducted by Ayres, HC2 has an Annual Percentage Rate of 5.58 percent and a rate spread of 2.03 percent, HC6 has an Annual Percentage Rate of 6.00 percent and a rate spread of 1.58 percent, and NHW8 has an Annual Percentage Rate of 5.32 percent and a rate spread of 1.12 percent. But eliminating the promotional discount and zeroing the lender credits on loans HC2 and HC6 yields an Annual Percentage Rate of 4.71 percent and a rate spread of 1.16 for loan HC2, an Annual Percentage Rate of Annual Percentage Rate of 5.86 percent and a rate spread of 1.44 percent for loan HC6, and an Annual Percentage Rate of 5.4431 percent and a rate spread of 1.25 percent for loan NHW8. Under Siskin’s analysis, loan HC2 is slightly cheaper than NHW8 by 0.09 percent and HC6 is more expensive than NHW8 by 0.19 percent. In his supplemental report, Ayres argued that Siskin’s calculations were distorted by his apparent failure to zero the lender credits received by NHW8. As Ayres explained, Siskin did not “explicitly state whether his calculation of the hypothetical rate spread for NHW8 includes the actual lender credits or whether he assumes a hypothetical lender credit of zero.” Instead, Siskin’s report only “provide[d] the hypothetical note rate and [Annual Percentage Rate] that would have been offered on [HC2 and HC6] if no lender credit had been provided or no 33 Case: 18-13152 Date Filed: 07/30/2019 Page: 34 of 49 promotional pricing discount had been offered,” and failed to “provide[] the note rate that would have been offered for loan NHW8 if no lender credit had been provided to that borrower.” Despite this alleged insufficiency, Ayres constructed hypothetical comparisons of the rate spreads of loans HC2, HC6, and NHW8 on the assumption that “NHW8’s lender credits remain $479 in Dr. Siskin’s hypothetical rate spread calculation.” Ayres “attempted to replicate Dr. Siskin’s calculations under the incomplete hypothetical scenario in which HC6 received no lender credit (and their note rates adjusted accordingly to the note rates specified by Ms. Hunt), NHW8 continued to receive a lender credit, and neither loan received the 50 basis point promotional discount allegedly given to loan NHW8.” He applied the same procedure to develop a comparison of loans HC2 and NHW8. But Ayres departed from Siskin’s method in one key respect. Ayres faulted Siskin for failing to “control for the difference in [Federal Housing Administration] Mortgage Insurance Premiums (“MIP”) policies that were in place at the times HC2 and NHW8 were originated.” Because “HC2 was originated in November 2012, whereas NHW8 was originated in December 2013” and “the government increased the cost and duration of [mortgage insurance premiums] in April 2013 and June 2013,” Ayres projected that “adjusting for the differences in [mortgage insurance premiums] policies would serve to increase the difference between the rate spreads 34 Case: 18-13152 Date Filed: 07/30/2019 Page: 35 of 49 of HC2 and NHW8.” So Ayres attempted to control for the difference in mortgage- insurance premium costs. Ayres’s analysis yielded somewhat different results from Siskin’s. Ayres calculated the Annual Percentage Rates of HC2, HC6, and NHW8 as 5.4118 percent, 5.9560 percent, and 5.4448 percent, respectively. He deduced a rate spread of 1.86 percent for loan HC2, 1.54 percent for loan HC6, and 1.25 percent for loan NHW8. So although Ayres agreed with Siskin’s conclusion that the hypothetical rate spread of NHW8 would equal 1.25 percent, Ayres’s estimates of the rate spreads for HC2 and HC6 were higher than Siskin’s estimates of 1.16 percent and 1.44 percent. Under Ayres’s projections, HC2 is more expensive than NHW8 by 0.61 percent and HC6 is more expensive than NHW8 by 0.29 percent. The results of each expert analysis are replicated in the following table: 35 Case: 18-13152 Date Filed: 07/30/2019 Page: 36 of 49 Actual and Projected APR & Rate Spreads for Loans HC2, HC6, & NHW8 Actual Siskin Hypothetical Ayres Hypothetical Assuming $0 Assuming 2013 Lender Credits for MIP Policies, $0 HC2 & HC6 & No Lender Credits for Promotional HC2 & HC6 & No Discount Promotional Discount HC2 APR APR APR 5.5816% 4.71% 5.4118% Rate Spread Rate Spread Rate Spread 2.03% 1.16% 1.86% HC6 APR APR APR 5.9978% 5.86% 5.9560% Rate Spread Rate Spread Rate Spread 1.58% 1.44% 1.54% NHW8 APR APR APR 5.3181% 5.4431% 5.4448% Rate Spread Rate Spread Rate Spread 1.12% 1.25% 1.25% 36 Case: 18-13152 Date Filed: 07/30/2019 Page: 37 of 49 The City’s only other evidence of discrimination was the declaration of a former Wells Fargo loan officer named Alvaro Orozco who worked for Wells Fargo for a “very short period of time in 2010,” before the limitation period began. Orozco attested that when he worked for Wells Fargo, his manager told him “to push borrowers into certain types of loans” that were more expensive than other loans for which he believed borrowers might be eligible. Orozco also asserted that “Wells Fargo’s desire to sell government loans . . . hit African-American and Hispanic borrowers the hardest,” and he conjectured that “if African-American or Hispanic borrowers in a community received loans with higher rate spreads than similarly situated non-Hispanic Caucasian borrowers, that result would be consistent with a bank’s decision to target African-American or Hispanic borrowers for more expensive mortgage loans.” B. The City Failed to Create a Genuine Dispute of Material Fact with Respect to Its Disparate-Treatment Claim. “Disparate treatment claims require proof of discriminatory intent either through direct or circumstantial evidence.” Equal Emp’t Opportunity Comm’n v. Joe’s Stone Crab, Inc., 220 F.3d 1263, 1286 (11th Cir. 2000). Proof of intent by circumstantial evidence relies on the burden-shifting framework of McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973). See Sec’y, U.S. Dep’t of Hous. & Urban Dev. v. Blackwell, 908 F.2d 864, 870 (11th Cir. 1990) (holding that the “test developed in McDonnell Douglas” governs suits brought under the Fair Housing 37 Case: 18-13152 Date Filed: 07/30/2019 Page: 38 of 49 Act). Under this framework, “the plaintiff bears the initial burden of establishing a prima facie case of discrimination.” Lewis v. City of Union City, 918 F.3d 1213, 1217 (11th Cir. 2019) (en banc). “If the plaintiff succeeds in making out a prima facie case, the burden shifts to the defendant to articulate a legitimate, nondiscriminatory reason for its actions.” Id. at 1221. “[S]hould the defendant carry its burden, the plaintiff must then demonstrate that the defendant’s proffered reason was merely a pretext for discrimination.” Id. This burden “merges with the ultimate burden of persuading the court that she has been the victim of intentional discrimination.” Id. (alterations omitted) (quoting Tex. Dep’t of Cmty. Affairs v. Burdine, 450 U.S. 248, 256 (1981)). The “elements of a prima facie case are flexible and should be tailored . . . to differing factual circumstances.” Fitzpatrick v. City of Atlanta, 2 F.3d 1112, 1123 (11th Cir. 1993) (citation and internal quotation marks omitted). In this appeal, a prima facie case of intentional discrimination required proof that (1) the borrower was a member of a protected class, (2) the borrower applied for and was qualified to receive loans from the defendant, and (3) the loan was offered on less favorable terms than a loan offered to a similarly situated person who was not a member of the borrower’s class. Cf. McDonnell Douglas, 411 U.S. at 802. A plaintiff and a comparator are “similarly situated” under McDonnell Douglas if they are 38 Case: 18-13152 Date Filed: 07/30/2019 Page: 39 of 49 “similarly situated in all material respects.” Lewis, 918 F.3d at 1226 (internal quotation marks omitted). The district court ruled that the City failed to establish a prima facie case because the borrowers of HC2 and HC6 are not similarly situated to the borrower of NHW8. It concluded that the loans were “‘apples and oranges’ that cannot be compared” because “the borrowers elected different structures to either finance closing costs over time or pay them at the outset” and NHW8 received a promotional discount. The district court refused “to consider Dr. Ayres’s efforts to extrapolate what the [Annual Percentage Rate] would be on HC2 and HC6 without the lender credits” because the credits were “simply a term of the loan that [the] Court cannot ignore.” In the alternative, the district court ruled that even if the City could establish a prima facie case, it failed to establish pretext because Ayres’s “method of comparison also reveal[ed] situations in which Wells Fargo originated loans to minority borrowers that were less expensive than loans issued to white borrowers,” undermining any inference that the difference in loan cost posited by Ayres was caused by an intent to discriminate. The City argues that the district court erred by refusing to credit Ayres’s calculation of the Annual Percentage Rate and rate spread for each loan after controlling for lender credits and the promotional discount because it is possible to prove a prima facie case of discrimination in this context merely by establishing 39 Case: 18-13152 Date Filed: 07/30/2019 Page: 40 of 49 that “one more expensive or riskier loan [was] given to a minority borrower.” Interpreted charitably, the City’s argument is that the discount and lender credits had an ascertainable impact on the bottom-line cost of the loans in question, so it was possible, using Ayres’s methodology, to control for the effect of those differences on the rate spread of each loan and determine whether the loans issued to minority borrowers were more costly than NHW8. In the City’s view, the continued existence of a cost disparity between loans HC2 and HC6 and loan NHW8 after controlling for lender credits and the discount supports a reasonable inference that the most likely explanation of the residual cost difference is the race or ethnicity of the borrowers, which suffices to establish a prima facie case. See Furnco Const. Corp. v. Waters, 438 U.S. 567, 577 (1978) (“A prima facie case under McDonnell Douglas” must establish that the challenged acts, “if otherwise unexplained, are more likely than not based on the consideration of impermissible factors.”). This Circuit has never held that a plaintiff can establish that individuals are similarly situated by reductively analyzing apparent differences between them in terms of a common metric of comparison, but even if we assume that a plaintiff can do so, we should nevertheless conclude that the City failed to establish an inference of discriminatory intent. Wells Fargo volunteered “legitimate, nondiscriminatory reason[s] for its actions,” Lewis, 918 F.3d at 1221, namely (1) 40 Case: 18-13152 Date Filed: 07/30/2019 Page: 41 of 49 the difference in lender credits, and (2) the availability of the promotional discount, so “the inquiry proceeds to a new level of specificity” at which “the plaintiff must show the . . . proffered reason[s] to be a pretext for unlawful discrimination.” Smith v. Lockheed-Martin Corp., 644 F.3d 1321, 1326 (11th Cir. 2011) (citation and internal quotation marks omitted). To establish pretext, the plaintiff must produce evidence sufficient to support a reasonable inference “that a discriminatory reason more likely motivated” the defendant or that the defendant’s “proffered explanation is unworthy of credence.” Burdine, 450 U.S. at 256. But under either avenue of proof, the ultimate question is “whether the evidence . . . yields the reasonable inference that the [defendant] engaged in the alleged discrimination.” Smith, 644 F.3d at 1326. The City failed to establish a reasonable inference that a discriminatory motive accounted for the cost differential between loans HC2 and HC6 and loan NHW8. Even if the lender-credits and promotional-discount explanations failed to account for the totality of the cost difference between HC2 and HC6 on the one hand and NHW8 on the other, it is undisputed that Wells Fargo also issued two loans to minority borrowers similarly situated to the borrower on loan NHW8 that were less expensive than NHW8. These loans, ML1 and ML2, were Federal Housing Administration purchase loans issued to minority borrowers who had, respectively, credit scores of 671 and 693, loan-to-value ratios of 98 percent, and 41 Case: 18-13152 Date Filed: 07/30/2019 Page: 42 of 49 debt-to-income ratios of 46.3 percent and 41.1 percent. The borrower of NHW8 had a credit score of 702, a loan-to-value ratio of 98 percent, and a debt-to-income ratio of 41 percent. Although the underwriting characteristics of the borrower of NHW8 are similar to those of the borrowers of ML1 and ML2 under Ayres’s criteria, the rate spreads of ML1 and ML2 are 0.74 percent and 0.94 percent while the rate spread of NHW8 is 1.12 percent. As the district court correctly ruled, this evidence precludes any inference “that a discriminatory reason more likely motivated” Wells Fargo. Burdine, 450 U.S. at 256. Apart from the declaration of Orozco—which, as discussed below, provides no support for the City’s position—the City’s case for disparate treatment is based entirely on the theory that one can rationally infer that intentional discrimination explains the residual cost discrepancy between loans HC2 and HC6 and loan NHW8. But this theory of intentional discrimination cannot account for the existence of nonminority borrowers who received more costly loans than similarly situated minorities. If Wells Fargo priced membership in a minority race or ethnicity into its loans, one would expect that minority borrowers would be systematically charged more than non-Hispanic white borrowers. But the evidence does not bear out that prediction. Indeed, the City’s theory of intentional discrimination is less accurate than a competing hypothesis of random variation in pricing because that explanation would at least potentially account for the 42 Case: 18-13152 Date Filed: 07/30/2019 Page: 43 of 49 existence of loans both more and less favorable to minorities. The City’s theory renders the existence of the former class of loans inexplicable. The City does not attempt to establish pretext by arguing that Wells Fargo’s “proffered explanation[s] [are] unworthy of credence,” id., and with good reason. True, Ayres’s hypothetical calculations of the rate spreads of HC2, HC6, and NHW8, may allow a reasonable inference that the lender credits and discount fail to explain the entirety of the cost discrepancy, as Wells Fargo maintained. But although “a plaintiff’s prima facie case, combined with sufficient evidence to find that the [defendant’s] asserted justification is false, may permit the trier of fact to conclude that the [defendant] unlawfully discriminated,” that proof will not “always be adequate to sustain a jury’s finding of liability.” Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 148 (2000). The issue on summary judgment is whether a genuine dispute of material fact exists, and there are “instances where, although the plaintiff has established a prima facie case and set forth sufficient evidence to reject the defendant’s explanation, no rational factfinder could conclude that the action was discriminatory.” Id. As I have explained, this appeal is one of those instances because of the “uncontroverted independent evidence” that Wells Fargo’s lending behavior produced unexplained cost discrepancies favorable to minority borrowers as well as one favorable to a nonminority borrower. Id. 43 Case: 18-13152 Date Filed: 07/30/2019 Page: 44 of 49 The City also argues that the district court should have considered Orozco’s affidavit and the testimony of a Wells Fargo executive named Mary Woodward, but this evidence does nothing to improve the City’s position. Orozco attested that his manager told him “to push borrowers into certain types of loans,” such as Federal Housing Administration loans, instead of other loans that might be cheaper. He also opined that “if African-American or Hispanic borrowers in a community received loans with higher rate spreads than similarly situated non- Hispanic Caucasian borrowers, that result would be consistent with a bank’s decision to target African-American or Hispanic borrowers for more expensive mortgage loans.” But Orozco did not provide any reason to believe that Wells Fargo “targeted” African-American or Hispanic borrowers for more expensive loans any more than they targeted members of other racial or ethnic groups. Indeed, he admitted that he was instructed to push borrowers into more expensive loans not because of their race, but “because these loans made more money for the bank and were easier to sell on the secondary market.” So his affidavit provides no basis for an inference of intent. Woodward testified only that she was unaware of any analysis prepared by Wells Fargo’s Internal Audit Department or any other department of the bank concerning allegations of violations of fair-lending laws and did not know of any reports, memoranda, or other written documents regarding the results of internal 44 Case: 18-13152 Date Filed: 07/30/2019 Page: 45 of 49 investigations into compliance with such laws. The ignorance of a single Wells Fargo executive about whether the bank had conducted any internal investigation into its compliance with fair-lending laws does not support an inference of discriminatory intent at all, so this testimony adds nothing to the City’s case. At bottom, even if one were to consider all of its evidence, the City failed to establish a genuine issue of material fact. C. The City Abandoned Any Challenge to the District Court’s Ruling on its Disparate-Impact Claim. Disparate-impact liability under the Fair Housing Act requires proof that a policy or practice of the defendant has “a ‘disproportionately adverse effect on minorities,’” Tex. Dep’t of Hous. & Cmty. Affairs v. Inclusive Cmtys. Project, 135 S. Ct. 2507, 2513 (2015) (quoting Ricci v. DeStefano, 557 U.S. 557, 577 (2009)), for which a prima facie case has three distinct elements. First, a prima facie case requires “the identification of a specific, facially-neutral . . . practice” or policy. Joe’s Stone Crab, 220 F.3d at 1268; see also Inclusive Cmtys., 135 S. Ct. at 2523 (holding that “a disparate-impact claim” under the Fair Housing Act “must fail if the plaintiff cannot point to a . . . policy or policies”). Second, the plaintiff must establish the existence of a “significant statistical disparity” between the effects of the challenged policy or practice on minorities and non-minorities. Joe’s Stone Crab, 220 F.3d at 1274. Third, in the light of the “serious constitutional questions that might arise . . . if such liability were imposed based solely on a showing of 45 Case: 18-13152 Date Filed: 07/30/2019 Page: 46 of 49 statistical disparity,” a plaintiff proceeding on a disparate-impact theory must also establish a “robust causality” connecting the challenged policy and the statistical disparity. Inclusive Cmtys., 135 S. Ct. at 2512. “A plaintiff who fails to . . . produce statistical evidence demonstrating a causal connection cannot make out a prima facie case of disparate impact.” Id. at 2523. The district court ruled that the City failed to produce sufficient evidence with respect to the statistical-disparity and causation elements of its claim. The district court interpreted the City’s claim as a challenge to (1) Wells Fargo’s Product Validation Process, which “examines borrowers to determine if they are eligible for less expensive loans,” and (2) Wells Fargo’s “practice of allowing lender credits on certain [Federal Housing Authority] loans,” which purportedly was “a vehicle for differential pricing.” The City’s principal evidence of disproportionate effect was Ayres’s reports, which identified two loans issued to minorities that allegedly were more expensive than loans issued to a similarly situated white borrower. The district court rejected the City’s contention that these loans supported an inference of a disproportionate adverse impact on minority borrowers because “[t]wo loans, even assuming they were more expensive, is insufficient record evidence to show the policies produced statistically-imbalanced lending patterns.” The district court also ruled that the City failed to produce any 46 Case: 18-13152 Date Filed: 07/30/2019 Page: 47 of 49 “evidence of the robust causation needed to show the polic[ies] caused the statistical disparity.” The City argues that it was error for the district court to require evidence of a statistical disparity because its burden was only to “identify at least one loan in the [limitation] period that exemplifies the discriminatory practice pleaded by the City,” but the City does not so much as attempt to challenge the district court’s alternative ruling on the causality element. So the City abandoned any challenge to the district court’s ruling on its disparate-impact claim. As we have explained, “[w]hen an appellant fails to challenge properly on appeal one of the grounds on which the district court based its judgment, he is deemed to have abandoned any challenge of that ground, and it follows that the judgment is due to be affirmed.” Sapuppo v. Allstate Floridian Ins. Co., 739 F.3d 678, 680 (11th Cir. 2014). The district court was also right to conclude that the City produced no evidence of causation. Even if one grants the City’s tendentious assumption that the two loans identified by Ayres suffice to establish “a disproportionately adverse effect on minorities,” Inclusive Cmtys., 135 S. Ct. at 2513 (citation and internal quotation marks omitted), the City never pointed to any evidence that even suggests that Wells Fargo’s policies caused this disparity in loan cost. For all we can infer from the evidence, the putative divergence in cost is attributable to ad hoc decisions, rounding errors, small differences between the borrowers, or factors not 47 Case: 18-13152 Date Filed: 07/30/2019 Page: 48 of 49 accounted for in Ayres’s analysis. So even if the City had not abandoned its disparate-impact claim, its failure to come forward with anything more than groundless speculation that a Wells Fargo policy must account for the cost discrepancy is fatal to its claim. Even if one ignores these glaring problems with the City’s position and considers the merits of its challenge to the district court’s ruling on the statistical- disparity element, the City comes up short. The City faults the district court for concluding that the two loans identified by Ayres failed to establish a violation on a disparate-impact theory. The City argues that under the continuing-violation doctrine its only “task [was] to identify at least one loan in the [limitation] period that exemplifies the discriminatory practice pleaded by the City.” But to invoke the continuing-violation doctrine, a plaintiff must establish that a violation of the Act occurred in the limitation period. See Hipp v. Liberty Nat’l Life Ins. Co., 252 F.3d 1208, 1221 (11th Cir. 2001). And under a disparate-impact theory of liability, proof of a violation requires the plaintiff to establish that the challenged policy produced a “significant statistical disparity,” Joe’s Stone Crab, 220 F.3d at 1274; see also Ricci, 557 U.S. at 587 (“[A] prima facie case of disparate-impact liability” is “essentially, a threshold showing of a significant statistical disparity.”). The City failed to present any evidence of a statistical correlation between the race of a borrower and the cost of the loan Wells Fargo would issue to him 48 Case: 18-13152 Date Filed: 07/30/2019 Page: 49 of 49 under its existing policies. Ayres never conducted a statistical analysis of whether Wells Fargo’s lending practices disproportionately impacted minorities. Indeed, he stated that he would forgo any attempt to analyze the “disparate impact of Wells Fargo’s mortgage lending,” but would “prepare a detailed analysis” if the case survived the summary-judgment stage. So even if the City had not abandoned its disparate-impact claim or failed to produce any evidence of causation, the City still would have failed to create a genuine issue of material fact with respect to this claim. 49