Eight sets of African-American plaintiffs filed a mortgage-lending discrimination lawsuit against Flagstar Bank. Although the district court granted Flagstar summary judgment against three of these sets, the claims of the other five sets proceeded to trial. During jury selection, Flagstar excused the only African-American on the jury panel. But the district court recalled the juror after concluding that Flagstar's peremptory challenge was improperly exercised. The district court also decided a number of evidentiary issues against Flagstar during the trial. A verdict was subsequently returned in favor of Flagstar as to three of the five sets of plaintiffs, but the jury found that the remaining two sets of plaintiffs — the Edwardses and the Paschals — had proven by a preponderance of the evidence that Flagstar racially discriminated against them when they applied for mortgages. Flagstar filed a number of post-trial motions, all of which were denied by the district court. The bank now appeals the various adverse decisions that the district court rendered throughout the case. For the reasons set forth below, we AFFIRM the district court's judgment in favor of the Edwardses, REVERSE its judgment in favor of the Paschals, and REMAND the case with instructions to dismiss the Paschals' complaint.

The district court succinctly stated the general background of this case as follows:

This is a housing discrimination case (mortgage lending) brought under sections 805 and 818 of the Fair Housing Amendment Act of 1988, 42 U.S.C. § 3605 and § 3617, and corresponding regulations promulgated by the Department of Housing and Urban Development pertaining to mortgage lending, 24 C.F.R. §§ 100.120 to 100.130, as well as the Civil Rights Act of 1866 and 1870, 42 U.S.C. § 1981 and § 1982, and section 504 of Michigan's Elliot-Larsen Civil Rights Act, M.C.L. § 37.2504.

In general, plaintiffs claim that they were the subjects of racial discrimination in the manner in which defendant Flagstar Bank (Flagstar), a mortgage bank, handled their mortgage loan applications or in the way the terms and conditions of their mortgage loans were set.

Edwards v. Flagstar Bank, 109 F. Supp. 2d 691, 692-93 (E.D.Mich.2000).

Prior to trial, Flagstar filed a motion for summary judgment, arguing that the events underlying the claims of four sets of plaintiffs, including the Paschals, occurred outside of the applicable statute of limitations. Although the district court granted Flagstar's motion for partial summary judgment with regard to the three other sets of plaintiffs, it denied the motion with regard to the Paschals. Flagstar did not contend that the Edwardses' claims were barred by the statute of limitations.

Five sets of plaintiffs went to trial. They are described by the district court as follows:

Audra Carson (Carson) claimed she was the victim of racial discrimination in the manner in which her application for a mortgage loan was handled. She did not close with Flagstar.

Paquita Davis-Friday (Davis-Friday) claimed she was the victim of racial discrimination in the manner in which her application for a mortgage loan was handled. She did not close with Flagstar.

Heath Thomas (Thomas) claimed he was the victim of racial discrimination in the manner in which his application for a mortgage loan was handled. He did not close with Flagstar.

David Edwards and E. Stephanie Edwards, his wife, (the Edwards[es]) claimed they were the victims of racial discrimination in the manner in which their efforts to refinance their mortgage loan with Flagstar was handled. The Edwards[es] eventually obtained a new mortgage loan from Flagstar.

Gerald Paschal and Lisa Paschal (the Paschals) claimed they were the victims of racial discrimination in the manner in which their application for a mortgage loan was handled. The Paschals were approved for a mortgage loan by Flagstar on terms considerably less favorable than for which they applied and eventually obtained a mortgage loan elsewhere.

Edwards, 109 F.Supp.2d at 693.

Because the jury rendered a verdict in favor of the Edwardses and the Paschals, we will describe their claims in more detail. On March 11, 1994, the Edwardses applied for a 15-year mortgage from Flagstar at a fixed interest rate of 7.875%. Their application was denied on May 26, 1994. According to Flagstar, the Edwardses' application was denied because they had insufficient funds to close the loan, inadequate collateral, and Flagstar "do[es] not grant credit to any applicant on the terms and conditions [the Edwardses] requested." The Edwardses claim that the stated reasons were a pretext designed to mask racial discrimination.

In an attempt to answer Flagstar's questions about various credit issues, the Edwardses met with a Flagstar loan officer on June 1, 1994. The Edwardses maintain that these alleged credit issues were old and inconsequential. Nevertheless, on September 16, 1994, Flagstar issued a second credit denial statement. The Edwardses claim that the reasons given for this denial were also pretextual.

After addressing additional questions about various credit issues, the Edwardses applied for a 30-year mortgage at a higher 9.5% interest rate. When this application was also denied, the Edwardses applied once again for a 30-year mortgage, but this time at a 10.875% interest rate. This final application was approved by Flagstar, and the Edwardses agreed to the terms. The Edwardses now claim that, throughout the entire process, Flagstar treated them as "second class citizens" due to their race.

In contrast, the Paschals were prequalified for a $75,000 mortgage at an adjustable 5.625% interest rate in May of 1992, despite having previously filed for bankruptcy. The Paschals were later notified by a loan officer that their mortgage had been tentatively approved. In August of 1992, however, Flagstar prepared a "Statement of Credit Denial, Termination, or Change" in which it informed the Paschals that Flagstar would approve only a $55,000 mortgage. Flagstar asserted that it was making this "counteroffer" because of the Paschals' other credit obligations, their prior bankruptcy, and the bank's inability to verify the Paschals' credit references. According to the Paschals, these reasons were a pretext intended to hide racial discrimination.

Flagstar's loan officer then took the Paschals' loan file to Mortgage Solutions, a/k/a Investaid Corporation, which preliminarily approved a $75,000 mortgage at a fixed 7.99% interest rate. Investaid, however, later withdrew this offer. According to the Paschals, "Flagstar pressured Investaid to not close on the Paschals' application...." The Paschals also allege that Flagstar fired its loan officer who took the Paschals' file to Investaid. On September 24, 1992, Investaid approved a $59,600 mortgage at 16.490%. After refusing this offer and withdrawing their application from Flagstar, the Paschals received a $68,700 mortgage, at an initial interest rate of 10.490%, from another lender.

The district court summarized the jury trial and subsequent verdicts as follows:

Plaintiffs' claims were tried to a jury in November of 1999. The jury returned a verdict in favor of Flagstar on the claims of Carson, Davis-Friday, and Thomas, and returned a verdict in favor of the Edwards[es] and Paschals. Particularly, the jury found the Paschals proved by a preponderance of the evidence that their race or color was one of the reasons why they did not obtain a mortgage loan from Flagstar. The jury awarded the Paschals $250,000 in compensatory damages and $325,000 in punitive damages. As to the Edwards[es], the jury found that they proved by a preponderance of the evidence that their race and color was one of the reasons why they did not obtain a mortgage loan on their initial application with Flagstar. The jury awarded the Edwards[es] $125,000 in compensatory damages.

Edwards, 109 F.Supp.2d at 693.

After the jury returned its verdict, Flagstar filed a motion for judgment as a matter of law pursuant to Rule 50(b) of the Federal Rules of Civil Procedure. It also filed alternative motions for a new trial pursuant to Rule 59 of the Federal Rules of Civil Procedure and for a remittitur. The district court denied all of Flagstar's motions. This timely appeal followed.

Flagstar contends that the district court erred in denying its motion for summary judgment, which was based on its argument that the Paschals' suit was barred by the applicable three-year statute of limitations established by 42 U.S.C. §§ 1981 and 1982 and the Michigan Elliott-Larsen Civil Rights Act. M.C.L. § 37.2504. This court has held, however, that "where summary judgment is denied and the movant subsequently loses after a full trial on the merits, the denial of summary judgment may not be appealed." Jarrett v. Epperly, 896 F.2d 1013, 1016 (6th Cir.1990). The policy behind this rule is based on the conclusion that the potential injustice of allowing the improper denial of a motion for summary judgment is outweighed by the injustice of "depriv[ing] a party of a jury verdict after the evidence was fully presented, on the basis of an appellate court's review of whether the pleadings and affidavits at the time of the summary judgment motion demonstrated the need for a trial." Id. at 1016 n. 1.

Flagstar acknowledges that the denial of summary judgment is generally not appealable "where summary judgment is denied and the movant subsequently loses after a full trial on the merits." Id. at 1016. It claims, however, that the underlying reason for this rule does not apply in this case because "Flagstar's motion was not denied because the trial court found a genuine issue of fact for trial. The trial court instead found that the Paschals' claims were timely under the continuing violations doctrine. There was no fact question for the jury to determine." Flagstar contends that the district court's denial of its motion for summary judgment "effectively granted partial summary judgment against Flagstar on its statute of limitations defense."

A district court's determination that a complaint was filed within the applicable statute of limitations is a conclusion of law. Tolbert v. State of Ohio Dept. of Trans., 172 F.3d 934, 938 (6th Cir.1999). As noted above, the district court denied Flagstar's motion for summary judgment because it found that the Paschals' "allegations are sufficient to allow the Paschals to invoke the continuing violation doctrine to toll the relevant statute of limitations." Flagstar preserved this issue by moving for judgment as a matter of law at the close of the evidence, predicated on "all the reasons known to the law." The district court denied this motion.

Although Flagstar failed to raise this specific issue when it moved for judgment as a matter of law after the jury rendered its verdict, this court has held that a litigant's failure to raise such a defense in a Rule 50(b) motion does not waive appellate review if an appeal of the denial of a motion of summary judgment on the same ground would involve review of a pure question of law. McPherson v. Kelsey, 125 F.3d 989, 995 (6th Cir.1997) (allowing appellate review of a motion for summary judgment on the ground of governmental immunity because, unlike Jarrett, the issue before the court was "purely one of law"). Such an exception is applicable in this case because the legal issue raised in Flagstar's motion for summary judgment does not require the resolution of any disputed facts.

Because the district court's decision is properly before us, we must determine whether the "continuing violations" doctrine is applicable to the Paschals' claims. This court has adopted a three-part inquiry for determining whether a continuing violation exists. "First, the defendant's wrongful conduct must continue after the precipitating event that began the pattern.... Second, injury to the plaintiff must continue to accrue after that event. Finally, further injury to the plaintiffs must have been avoidable if the defendant[] had at any time ceased [its] wrongful conduct." Tolbert, 172 F.3d at 940.

The district court found that Flagstar denied the Paschals' application for a loan more than three years before September 21, 1995, the date on which the Paschals filed their complaint. In considering whether the continuing violations doctrine was applicable, however, the district court focused on the Paschals' allegations that: (1) on September 29, 1992, the Paschals had written Flagstar a letter withdrawing their loan application on the grounds of unfair treatment, and (2) on November 4, 1992, Flagstar had granted the Paschals' request to withdraw their application in a "Statement of Credit Denial, Termination or Change." The district court held that "[t]hese allegations are sufficient to allow the Paschals to invoke the continuing violation doctrine to toll the relevant statute of limitations...."

Contrary to the district court's determination, we conclude that neither of these events constitutes a continuing violation. The first occurrence involves a letter the Paschals wrote to Flagstar. Because the passive receipt of a letter does not constitute "wrongful conduct," this event could not possibly satisfy the first prong of the abovementioned three-part inquiry. Tolbert, 172 F.3d at 940 (holding that "[p]assive inaction ... does not support a continuing violation theory"). The second occurrence was the granting of the Paschals' request to withdraw their application for a mortgage. Although this occurrence required action by Flagstar, it does not constitute a separate act of "wrongful conduct." At most, the granting of the Paschals' request was the continuing ill effect of a prior act of discrimination. Id. (noting that "[a] continuing violation is occasioned by continual unlawful acts, not continual ill effects from an original violation") (quotation marks omitted) (alteration in original).

The Paschals cite four other events in support of their claim that the complaint was filed within the applicable statute of limitations. First, the Paschals point out that on September 22, 1992, they "received preliminary approval from Investaid for a $75,000 mortgage at a 7.99% interest rate." (Emphasis added.) Second, the Paschals contend that "Flagstar forced Investaid to withdraw its offer to the Paschals." Third, the Paschals allege that, on September 24, 1992, "Investaid... changed the approval terms to the outrageous 16.490% interest rate...." (Emphasis added.) Fourth, the Paschals claim that they were discriminated against at a meeting on that date, during which a Flagstar loan officer allegedly told them about Investaid's latest loan offer.

Of these four events, the first, third, and fourth cannot possibly constitute a continuing violation. The first and third events clearly involve actions by Investaid, not Flagstar, and thus cannot be considered "wrongful conduct" by Flagstar. As for the fourth occurrence, the act of Flagstar allegedly telling the Paschals about Investaid's latest loan offer would at most be a continual ill effect of past discrimination by Flagstar, not wrongful conduct in and of itself.

The second alleged event — Flagstar "forcing" Investaid to withdraw its original loan offer — is the only one that could be construed to constitute wrongful conduct by Flagstar. Our inspection of the record, however, uncovered no proof to support the Paschals' allegation that Flagstar "forced" Investaid to withdraw its loan offer. The closest the Paschals come to providing evidence to corroborate their claim is the testimony of a former Flagstar employee, David Reed, Jr. But Reed had no personal knowledge of whether Flagstar tried to exert any influence on Investaid. He simply stated that he "believe[d]" that Flagstar "advised" Investaid to withdraw its loan offer. Moreover, the district court commented that Reed's testimony on this issue was hearsay, and the point was not pursued further by the Paschals' attorney. Rule 56 of the Federal Rules of Civil Procedure, however, requires more than "mere allegations" in support of a claim for a party to survive a motion for summary judgment. Instead, the party "must set forth specific facts showing that there is a genuine issue for trial." Fed.R.Civ.P. 56(e). The Paschals have thus failed to provide any specific facts in support of their claim that Flagstar "forced" Investaid to withdraw a loan offer.

In sum, none of these other four events support the Paschals' contention that any wrongful conduct by Flagstar took place within the three years before they filed their complaint. The Paschals' claim is thus barred by the statute of limitations, and the district court erred by denying Flagstar's motion for judgment as a matter of law with respect thereto.

Flagstar contends that the district court committed reversible error when it recalled an African-American juror whom Flagstar had removed through the use of a peremptory challenge. Specifically, Flagstar maintains that, after it articulated a race-neutral reason for challenging the juror, the district court improperly combined the second and third steps of the Batson test and erroneously placed the burden on Flagstar to establish that the peremptory challenge was not racially motivated. Batson v. Kentucky, 476 U.S. 79, 106 S. Ct. 1712, 90 L. Ed. 2d 69 (1986).

"In determining whether a party has committed a Batson violation in its exercise of a peremptory challenge, a district court must follow a three-step process." United States v. Mahan, 190 F.3d 416, 424 (6th Cir.1999). The district court must first determine whether "the opponent of a peremptory challenge has made out a prima facie case of racial discrimination." Purkett v. Elem, 514 U.S. 765, 767, 115 S. Ct. 1769, 131 L. Ed. 2d 834 (1995). Next, if the opponent of the peremptory challenge makes out a prima facie case, "the burden of production shifts to the proponent of the strike to come forward with a race-neutral explanation." Id. "[I]f a race-neutral explanation is tendered, the trial court must then decide whether the opponent of the strike has proved purposeful discrimination." Mahan, 190 F.3d at 424. At the third step of the analysis, "the district court has the responsibility to assess the [strike proponent's] credibility under all of the pertinent circumstances, and then to weigh the asserted justification against the strength of the [strike opponent's] prima facie case under the totality of circumstances." Id. at 425 (citation omitted).

"The district court's ruling on the ultimate question of discriminatory intent in the exercise of peremptory challenges is a finding of fact." Id. at 423. Although Flagstar acknowledges that this court generally reviews a district court's application of the Batson test for "clear error," it claims that the district court committed an error of law in applying the test and, therefore, we should review the court's determination de novo. In support of this argument, Flagstar cites United States v. McFerron, 163 F.3d 952, 955 (6th Cir.1998) (holding that the district court erred "by consolidating the second and third steps of the Batson analysis and imposing the burden of persuasion on [the proponent of the strike]").

Jury selection in this case began with 17 prospective jurors. Only one of these prospective jurors, Juror # 2, was African-American. During voir dire, Juror # 2 disclosed that she worked for the Michigan State Housing Development Authority (MSHDA). The district court asked her what type of work she did for the MSHDA. Juror # 2 explained that she was "the calculations clerk" and that she "process[ed] payments for low-income housing people" who receive Section 8 housing benefits. Flagstar did not request the district court to ask Juror # 2 any additional questions about the type of work that she did. After the district court eliminated several other jurors, both parties passed the jury for cause.

Flagstar then attempted to use one of its peremptory challenges to excuse Juror # 2. At a sidebar conference with the judge, the plaintiffs objected and pointed out that Juror # 2 was the only African-American juror on the panel. The following exchanges then took place at the sidebar:

The Court: Why did you excuse [Juror # 2]?

Mr. White: Your Honor, besides just using my peremptory, Your Honor, I believe her employment with MSHDA may give her some sort of information regarding this case because I believe Flagstar at some point in time —

The Court: You should have challenged her for cause.

...

Mr. White: Why would I have to have a reason on a peremptory?

The Court: There's a case, Batson, that gives you the right not to excuse them for discriminatory purposes, and you couldn't give me an answer.

Mr. Phifer: Other than —

The Court: You asked her no questions. That was a reason for cause. I find that the reason you have excused her is because of her race and it's an impermissible ground.

...

The Court: I don't want to look back. I'm accepting your representation. You had your opportunity during voir dire to inquire into what she did. She's a functionary at MSHDA. She collects payments from low-income people. You did not ask me to ask any follow-up questions. I believe that the Batson rule which prohibits discrimination in jury selection applies equally in civil cases as to criminal cases, and I find, based upon what you have told me, that you chose to challenge this juror, excuse her peremptorily because she is an African-American.

...

Mr. White: You wouldn't have excused the Black lady for cause for what I said, Your Honor.

The Court: ... The woman from MSHDA is the only African-American on this panel, and you have not given me a legitimate reason for exercising a peremptory challenge. The sole basis of your challenge is her employment, and you asked no questions about her employment during the course of the voir dire. Therefore, I draw the inference and make the finding.

Flagstar does not contest the district court's conclusion that the plaintiffs made out a prima facie case of racial discrimination. Instead, Flagstar focuses its challenge on the second and third steps of the Batson test. Flagstar contends that the district court erred by combining the second and third steps, thereby placing the burden on Flagstar to prove that its peremptory strike was not racially motivated.

Under the second step of the Batson test, "the burden of production shifts to the proponent of the strike to come forward with a race-neutral explanation." Purkett, 514 U.S. at 767, 115 S. Ct. 1769. When the district court asked Flagstar why it exercised a peremptory challenge on Juror # 2, Flagstar stated that it "believ[ed] her employment with MSHDA may give her some sort of information regarding this case...." Flagstar contends that the district court "summarily dismissed" this race-neutral explanation for the peremptory strike. Its contention is not supported by the record. Although the district court was not wholly satisfied with Flagstar's response, the court clearly stated that it was "accepting [Flagstar's] representation" as to the bank's race-neutral reason for the peremptory strike.

The district court, after accepting Flagstar's race-neutral explanation, then moved on to the final step of the Batson test. At the third step of the analysis, "the district court has the responsibility to assess the [strike proponent's] credibility under all of the pertinent circumstances, and then to weigh the asserted justification against the strength of the [strike opponent's] prima facie case under the totality of circumstances." Mahan, 190 F.3d at 425 (citation omitted). The district court articulated several reasons why, under the totality of the circumstances, it believed that the strength of the plaintiffs' prima facie case outweighed the credibility of Flagstar's asserted justification. According to the record, the district court primarily relied on the fact that Flagstar did not ask Juror # 2 any questions about her employment despite its purported concern about what she might have known of Flagstar through her employment at the MSHDA. The district court also pointed out that, based on the court's questioning of Juror # 2 during voir dire about her specific job duties, there was no reason to believe that she had any contact whatsoever with Flagstar.

Although the district court's application of the Batson test was less than ideal, in that the court prematurely stated its conclusion before fully articulating the Batson analysis, the district court did not commit an error of law in applying the test. We should refrain from overturning the district court's decision on this issue unless it clearly erred in concluding that Flagstar acted with discriminatory intent in excusing Juror # 2. Mahan, 190 F.3d at 423. The district court was in a much better position than we are to determine the credibility of Flagstar's attorneys. After reviewing the record, we are not "left with the definite and firm conviction" that the district court erred in concluding that Flagstar exhibited discriminatory intent in peremptorily striking the sole African-American juror. We therefore hold that the district court did not commit reversible error when it recalled Juror # 2.

Flagstar contends that the district court erred in: (1) allowing into evidence testing conducted by the Fair Housing Center of Metropolitan Detroit (FHC), which Flagstar argued had extremely low, if any, probative value, and a substantial prejudicial effect; (2) preventing Flagstar from offering into evidence "four loan files of minority applicants who were extended mortgages despite deficiencies in their credit applications or credit histories"; and (3) preventing Flagstar from offering into evidence Community Reinvestment Act performance evaluations conducted by the Office of Thrift Supervision. It maintains that the plaintiffs' testing evidence should have been excluded, while its own evidence should have been admitted, all pursuant to Rules 401, 402, and 403 of the Federal Rules of Evidence.

We will not disturb a district court's evidentiary rulings unless we find an abuse of discretion. Trepel v. Roadway Express, Inc., 194 F.3d 708, 716 (6th Cir. 1999). "In reviewing the trial court's decision for an abuse of discretion, the appellate court must view the evidence in the light most favorable to its proponent, giving the evidence its maximum reasonable probative force and its minimum reasonable prejudicial value." United States v. Schrock, 855 F.2d 327, 333 (6th Cir.1988) (quotation marks omitted). An abuse of discretion occurs when we are left with the "definite and firm conviction that the court below committed a clear error of judgment in the conclusion it reached upon a weighing of the relevant factors" or "where [the trial court] improperly applies the law or uses an erroneous legal standard." Huey v. Stine, 230 F.3d 226, 228 (6th Cir.2000) (citations and internal quotation marks omitted).

Under Rule 402, "[e]vidence which is not relevant is not admissible." Rule 401 defines relevant evidence as "evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence." Even relevant evidence, however, may be properly excluded if "its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence." Fed.R.Evid. 403.

Flagstar contends that the testing evidence should have been excluded pursuant to Rule 402 because it "had no probative value on the issue of whether Flagstar discriminated against the plaintiffs in the consideration of their mortgage applications." In the alternative, Flagstar argues that the testing evidence should have been excluded under Rule 403 because it focused on the racially inflammatory conduct of Chet Allen, a Flagstar loan officer who was not involved in the processing of the plaintiffs' applications, and, therefore, the low probative value of the evidence was substantially outweighed by the danger of unfair prejudice.

Although the FHC sent ten pairs of testers to Flagstar between July 16, 1994 and March 2, 1995, the plaintiffs offered the results of only three matched-pair tests into evidence. The pairs of testers were named as follows: (1) Selman/Taylor, (2) Jordan/Kisch, and (3) Cox/Chapp. In each case, the first name listed is that of an African-American tester and the second name listed is that of a Caucasian tester. Although the district court preliminarily decided to admit into evidence the results of all three pairs of testers, it later excluded a portion of the evidence related to the Cox/Chapp test as substantially more prejudicial than probative.

Clifford Schrupp, the Director of the FHC, gave background information regarding the testing conducted by the three pairs of testers. During his direct testimony, Schrupp stated that the FHC decided to perform the testing after receiving several complaints, including one from the Paschals, about Flagstar. Schrupp noted that the FHC "incorporated some of the elements of the various complaints that we had, in particular the Paschals, into the assignments that we then gave to the testers." He also explained why the testers, unlike the plaintiffs, did not fill out loan applications:

Q. And in terms of the mortgage testers, do the testers fill out loan applications?

A. No, they do not.

Q. In a rental test do the testers fill out applications for a rental?

A. It's not a routine or normal thing for testers to do, but there have been instances where that has happened.

Q. How come — can you explain why applications are not submitted when you do testing of financial institutions?

A. An application normally for a financial institution runs 300, 350 dollars, sometimes more. [The FHC is] not funded to be able to put those kinds of investments into, into filling out the application.

We also found — we began doing tests of financial institutions in 1984 under a contract that we had with the City of Pontiac, and we found out in that series of tests that we did in 1984 of financial institutions in Pontiac that we could get quite a bit of information about how people are treated and may be treated differently by loan officers before we get to the point of having to fill out a loan application. And so we were satisfied that we could identify quite a bit of information about how somebody is being treated in that early stage, and so for those two reasons we haven't done loan applications or paid — asked our testers to fill out loan applications themselves.

Schrupp testified that the three pairs of testers experienced differences in treatment that were potentially attributable to racial discrimination. First, although Selman and Taylor presented similar financial qualifications and each told the Flagstar loan officer that they wanted a house in the Detroit area, they were directed to different neighborhoods. The loan officer suggested that Selman, the African-American tester, should look for a home in a neighborhood, Rosedale Park, that is approximately 75 percent African-American. Taylor, the Caucasian tester, was directed to a neighborhood that is approximately 70 percent Caucasian. Furthermore, the loan officer gave the African-American tester the name of an African-American real estate agent and the Caucasian tester the name of a Caucasian real estate agent.

Second, Jordan and Kisch presented similar financial qualifications, and each told the Flagstar loan officer that they wanted a house in Ann Arbor, a predominantly Caucasian community. The loan officer, however, encouraged Jordan, the African-American tester, to consider instead Ypsilanti, which has approximately a 50 percent African-American population. No such suggestion was made to the Caucasian tester.

Finally, Cox and Chapp attempted to set up appointments to meet with Chet Allen, the loan officer who was involved in the Selman/Taylor testing. Cox, the African-American tester, visited Flagstar and asked the receptionist if he could meet with Allen. Because Allen was not in, the receptionist asked Cox to leave his name and phone number. Allen called Cox later that day and indicated that he could pre-qualify Cox for a loan over the phone. When Chapp, the Caucasian tester contacted Allen, Allen suggested that he should set up an appointment for a meeting to get more information about a loan. Schrupp testified that the different treatment suggested the possibility of racial bias.

Despite Flagstar's claim to the contrary, we do not believe that the district court abused its discretion in finding that the testing evidence was relevant. The plaintiffs have pointed out that the testers were given some of the same financial attributes that the plaintiffs presented to Flagstar. For example, the Jordan/Kisch test involved a Caucasian tester who, like Thomas, worked for a temporary employment agency. The Selman/Taylor and the Jordan/Kisch tests also involved Caucasian testers who, like the Paschals, had a history of bankruptcy. Overall, their experiences suggest that Flagstar might treat African-American and Caucasian customers differently, at least during the preapplication stage.

As Flagstar points out, however, Schrupp admitted that the testing was "inconclusive" on the question of whether Flagstar was less likely to grant a mortgage to an African-American than it was to grant a mortgage to a Caucasian. The fact that this evidence is inconclusive as to the resolution of the ultimate question involved in this trial does not, however, mean that it is irrelevant. If, for instance, the results of the testing had shown that the testing uncovered no evidence of discriminatory treatment in the preapplication stage, Flagstar would likely be arguing that the evidence was relevant to the question of whether it had discriminated against the plaintiffs. The testing evidence, though not strong, possesses a "tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence." Fed.R.Evid. 401. We therefore find no abuse of discretion in the district court's determination that the testing evidence was relevant.

This brings us to the question of whether the district court abused its discretion by failing to exclude the testing evidence under Rule 403. Flagstar contends that the testing evidence should have been excluded because it focused on the potentially inflammatory conduct of Chet Allen, a Flagstar loan officer who was not involved in processing the plaintiffs' mortgage applications. The bank therefore maintains that the low probative value of the testing evidence was substantially outweighed by the danger of unfair prejudice. But "[u]nfair prejudice does not mean the damage to a defendant's case that results from the legitimate probative force of the evidence; rather it refers to evidence which tends to suggest a decision on an improper basis." United States v. Bonds, 12 F.3d 540, 567 (6th Cir.1993) (citation and internal quotation marks omitted).

Flagstar argues that the testimony regarding Chet Allen was unfairly prejudicial because it tended to suggest a decision against the bank based on the fact that it employed someone who allegedly had racist views. During Schrupp's testimony regarding the Cox/Chapp test, the following evidence was heard by the jury:

Q. Was there anything in the narrative section, Page 2035 through 2039, that raised concerns when you reviewed the testing?

A. There were several things in the narrative section, yes.

Q. And what were those concerns that you had?

A. Well, in the narrative section, Page 2035 near the bottom of the page, there is a discussion reported by Mr. Chapp that he had told [loan officer Chet Allen] that he was looking [for a house] in Farmington. He had been asked by the loan officer where he was looking for housing. He said Farmington, but the prices were too high for him in Farmington. He had previously indicated that he was staying with his son in Southfield, and apparently at that point the loan officer said you don't want Southfield or east Dearborn, [there are] too many Arabs. Instead he suggested Royal Oak, Pleasant Ridge, Huntington Woods.

...

Q. I believe you can continue on 2036, the second page of the narrative test form. Just very, very briefly were there any other items that raised concerns? You don't have to read it, but just hit the highlights of it.

A. There was a reference to east English Village again, the area where the loan officer lived. He indicated to the tester that it was a nice area, but he would not want his son to go to school there but instead recommended that his son go to schools in the Grosse Pointe area.

Q. And I take it there were other statements made?

A. Yeah, there were other references in the report form to the Dearborn area and to the fact that Arabs live in Dearborn and other positive references to other communities and in fact references to several real estate agents who might be of assistance to Mr. Chapp.

Q. Now, are you familiar with the racial composition or approximate racial composition of the various communities identified by the loan officer?

A. Generally, yes.

At a side bar conference, the district court described the problem with this line of questioning: "You've now got a situation where you've got, in the most extreme terms, a racist loan officer. Well, you can't extrapolate from a racist loan officer institutional racism on the part of the institution. You can't do that. That's not probative of an institutional animus." To the extent that such testimony suggested to the jury an improper basis for its verdict, the evidence was unfairly prejudicial.

The question under Rule 403, however, is not simply whether the evidence was unfairly prejudicial. Instead, the testing evidence was subject to exclusion only if its probative value was "substantially outweighed by the danger of unfair prejudice." Although admittedly a close question, we are not left with the "definite and firm conviction that the trial court ... committed a clear error of judgment" in concluding that the probative value of the testing evidence was not substantially outweighed by the danger of unfair prejudice. This is particularly true in light of the fact that "[i]n reviewing the trial court's decision for an abuse of discretion, the appellate court must view the evidence in the light most favorable to its proponent, giving the evidence its maximum reasonable probative force and its minimum reasonable prejudicial value." United States v. Schrock, 855 F.2d 327, 333 (6th Cir.1988) (quotation marks omitted).

Our conclusion is buttressed by the fact that the district court did exclude two of the exhibits related to the potentially inflammatory Cox/Chapp test. One of these exhibits included test report forms filled out by Chapp and Ratkowski, another Caucasian tester who posed as Chapp's wife, relating to a fair-lending test that they conducted on March 2, 1995. This exhibit described racist statements allegedly made by Chet Allen during the test. The other exhibit was the transcript of a recording that Chapp and Ratkowski made of the same fair-lending test. These exhibits were excluded by the district court because it found that the danger of unfair prejudice substantially outweighed whatever probative value the exhibits might have. This decision demonstrates that the district court was cognizant of the potentially prejudicial impact of the testimony and that it attempted to draw the appropriate balance under Rule 403. The district court's exclusion of the evidence of testing conducted on Flagstar's lending practices in Grand Rapids, Michigan further points out that it was aware of potential limitations on the probative value of testing evidence.

In addition, the evidence regarding Allen was most explicit regarding his animus against Arab-Americans, whereas this case involved alleged discrimination against African-Americans. We also note that testing evidence in the context of rental discrimination has long been held admissible under Rule 403 despite the fact that it has some of the same potential limitations on its probative value as in the lending context. Richardson v. Howard, 712 F.2d 319, 321-22 (7th Cir.1983). Finally, we note that the jury returned a verdict in favor of Flagstar as to the claims regarding three of the five sets of plaintiffs. If the unfairly prejudicial effect of the testing evidence had been substantial, the jury would have likely found against Flagstar on all the claims. We will therefore not reverse the district court's judgment or grant Flagstar a new trial on this basis.

Flagstar next contends that the district court erred in preventing it from offering into evidence "four loan files of minority applicants who were extended mortgages despite deficiencies in their applications or credit histories." In excluding this evidence, the district court stated:

There will be no comparable use — there will be no African-American comparable to an African-American file. It is not arguably relevant to any of the issues in this case.

Given[ ] the files available to [Flagstar] because of the large number of mortgages it generates, the fact that it [granted a mortgage to] an African-American wh[o] it says [is] comparable to ... one of the [ ] African-American plaintiffs[,] doesn't mean that [Flagstar] didn't discriminate against the [plaintiffs].

The plaintiffs claim that Flagstar waived this issue by failing to make an offer of proof regarding the excluded files. Selden Apartments v. HUD, 785 F.2d 152, 162 n. 13 (6th Cir.1986) (noting that Rule 103(a)(2) of the Federal Rules of Evidence requires a party to make known to the court the substance of the excluded evidence in order to preserve the issue for appeal). Flagstar has not responded to this argument, an argument that we find persuasive. In any event, even if the district court should have accepted the four files into evidence, we would have been hard-pressed to conclude that, by not doing so, it "committed a clear error of judgment." Huey v. Stine, 230 F.3d 226, 228 (6th Cir.2000).

Finally, Flagstar contends that the district court erred in excluding Flagstar's proposed Exhibits 331, 332, and 341. These exhibits consisted of Community Reinvestment Act (CRA) performance evaluations by the Office of Thrift Supervision (OTS), the federal regulatory agency charged with supervising Flagstar. The purpose of the CRA is "to require each appropriate Federal financial supervisory agency to use its authority when examining financial institutions, to encourage such institutions to help meet the credit needs of the local communities in which they are chartered consistent with the safe and sound operation of such institutions." 12 U.S.C. § 2901(b). In excluding evidence of performance evaluations performed by the OTS to determine whether Flagstar was complying with the CRA, the district court stated: "I finding nothing arguably relevant to any of the issues in this case. What [the performance evaluations] show is overall Flagstar had a relatively good compliance program. It has nothing to do with the issues in this case."

The plaintiffs have pointed out that Flagstar did not respond to the plaintiffs' motion in limine to prohibit introduction of the CRA performance evaluations. This again constitutes a waiver of this argument on appeal. Niecko v. Emro Mktg. Co., 973 F.2d 1296, 1299 (6th Cir.1992) ("It is well settled law that this court will not consider an error or issue which could have been raised below but was not."). Moreover, after reviewing the record, we would not likely have found the district court's exclusion of the CRA performance evaluations to be reversible error. Huey, 230 F.3d at 228.

Flagstar contends that the district court erred in denying its motion for judgment as a matter of law, which it filed following the jury's verdict. Specifically, Flagstar argues that the evidence at trial was insufficient to support the jury's finding of racial discrimination and, therefore, that Flagstar is entitled to judgment in its favor as a matter of law.

A motion for judgment as a matter of law "may not be granted unless reasonable minds could not differ as to the conclusions to be drawn from the evidence." McJunkin Corp. v. Mechanicals, Inc., 888 F.2d 481, 486 (6th Cir.1989). An appeals court is not to "weigh the evidence, pass on the credibility of witnesses, or substitute its judgment for that of the jury." Toth v. Yoder Co., 749 F.2d 1190, 1194 (6th Cir.1984). Instead, we must view the evidence in the light most favorable to the opposing party, drawing all reasonable inferences in that party's favor. Id.

The district court concluded that "[t]here is no basis for a judgment as a matter of law in favor of Flagstar," because viewing "[t]he evidence in a light most favorable to the Paschals and Edwards[es] supports the jury's verdict that both parties proved by a preponderance of the evidence that race was a factor in the way in which Flagstar dealt with each of them." Edwards, 109 F.Supp.2d at 704. According to the district court, the following evidence was particularly central to its decision: (1) "[Calvin] Bradford's conclusions that the only variable in [the plaintiffs'] treatment by Flagstar compared to white applicants was that they were treated less favorably because of their race," (2) "the [Home Mortgage Disclosure Data (HMDA)] data which generally proved that Flagstar acted with a discriminatory animus," and (3) "the testing data which displayed a discriminatory animus." Id.

The testing data has been discussed at length in Part II.C.1. above. We now will describe briefly the testimony of two of the plaintiffs' expert witnesses, Calvin Bradford, Ph.D., and Martin Wing, Ph.D. Bradford testified for the plaintiffs as a mortgage-lending discrimination expert. After examining the documents pertaining to the plaintiffs' applications for mortgages from Flagstar, Bradford also checked the files of various Caucasian applicants whose mortgage applications were approved by the bank. Although Bradford acknowledged that no two loan applications are ever exact matches, he compared the plaintiffs' applications with those of Caucasian applicants who had similar financial qualifications. Bradford first compared the Paschals' file with those of six Caucasian applicants. He concluded that the differences in how Flagstar treated the Paschals could not be explained by any valid underwriting criterion. Bradford then compared the Edwardses' file with those of two Caucasian applicants. Again, Bradford concluded that the differences in how Flagstar treated the Edwardses could not be explained by any valid underwriting criterion.

Wing testified for the plaintiffs as an expert in econometrics. The plaintiffs asked Wing to determine "if there was any evidence of differences in treatment between African-American applicants and White applicants for home mortgage loans at Flagstar Bank." Wing provided a detailed statistical analysis of Flagstar's lending practices based on Home Mortgage Disclosure Data (HMDA) for 1988 through 1996 that Flagstar had provided to the Federal Financial Institutions Examination Council. According to Wing, HMDA data "consists of a set of variables that financial institutions must record for all loan applications, for all home mortgage loan applications." Wing testified that between 1988 and 1996, Flagstar denied African-American applicants three times more often than it denied Caucasian applicants. Based on his statistical analysis of the HMDA data, Wing concluded that "the denial rates depend on race. The denial rate and race are dependent, not independent." In other words, Wing concluded that the disparity between the rate at which Flagstar denied applications of African-Americans and the rate at which it denied applications of Caucasians could not be fully explained by other factors such as income or credit history.

Flagstar claims that the evidence offered by the plaintiffs was insufficient to support the jury's finding of racial discrimination. First, Flagstar points out that the testimony of the Paschals and Edwardses about alleged discrimination was completely circumstantial. It notes that even the district court characterized this evidence as being "rather weak." Second, Flagstar claims that the testing evidence lacked probative value because, as Schrupp admitted, the tests were inconclusive as to whether Flagstar was less likely to grant a mortgage to an African-American, than it was to grant a mortgage to a Caucasian. Flagstar also emphasizes that the testing evidence involved Flagstar's treatment of individuals during the preapplication process as opposed to during the underwriting process. Third, Flagstar contends that the statistical evidence presented by Wing was of limited reliability because he omitted several key variables and he aggregated data across different loan programs with different underwriting standards. Fourth, Flagstar claims that the files on which Bradford based his comparisons involved applications that were governed by different underwriting standards. As a result, Flagstar claims that Bradford's testimony "could not have convinced a reasonable juror that Flagstar discriminated against [the plaintiffs]."

Contrary to Flagstar's arguments, we are of the opinion that there was sufficient evidence to support the jury's finding of racial discrimination. First, the Edwardses and the Paschals both experienced numerous, seemingly unnecessary delays in applying for loans. According to Bradford, none of the Caucasian applicants whose files he compared with the plaintiffs' experienced such a multitude of "tactics of delay" and "hindrance." Second, although the testing focused on preapplication contacts with Flagstar, it is not unreasonable to draw the inference that the disparate treatment encountered by African-American testers during their initial interactions with Flagstar might be indicative of what applicants for mortgages would experience. Schrupp, who has been the executive director of the FHC for 22 years, testified that it has been his experience that this sort of testing can provide "quite a bit of information about how people are treated and may be treated differently by loan officers." Third, the methodological concerns that Flagstar raises regarding Wing's testimony are generally insubstantial. As the Supreme Court has held, "it is clear that a regression analysis that includes less than all measurable variables may serve to prove a plaintiff's case." Bazemore v. Friday, 478 U.S. 385, 400, 106 S. Ct. 3000, 92 L. Ed. 2d 315 (1986) (Brennan, J., concurring) (internal quotation marks omitted). Wing's finding that, between 1988 and 1996, Flagstar denied African-American loan applicants three times more often than Caucasian applicants provides support for the jury's verdict. Finally, the district court found that the expert testimony provided by Bradford was compelling. The district court stated that "Mr. [Bradford] was one of the best-prepared experts that I have heard in 20 years." Later, the district court noted that, "[y]ou know, without hearing the cross-examination, Bradford painted a pretty bleak picture for [Flagstar]."

Flagstar is correct that, when broken down into discrete parts, the plaintiffs' evidence might be subject to attack. For example, Steven Laidlaw, who manages Flagstar's disclosures under the Home Mortgage Disclosure Act, testified that Flagstar approved mortgage applications at the following rates: 89.1% for African-Americans, 96% for Caucasians in 1994; 92.8% for African-Americans, 97% for Caucasians in 1995; and 91.3% for African-Americans, 96.5% for Caucasians in 1996. These approval rates do not, however, tell us anything about whether the mortgage applications that were approved discriminated in their terms and conditions on the basis of race.

Given the combined effect of the testimony offered by the plaintiffs, we are of the opinion that there was sufficient evidence for reasonable minds to "differ as to the conclusions to be drawn from the evidence." McJunkin Corp., 888 F.2d at 486. Accordingly, we conclude that the district court did not err in holding that the evidence "supports the jury's verdict that both [plaintiffs] proved by a preponderance of the evidence that race was a factor in the way in which Flagstar dealt with each of them."

In the alternative, Flagstar contends that the district court erred in denying its motion for a new trial. Flagstar maintains that it is entitled to a new trial because the evidence was insufficient as to both liability and damages. It raises the same arguments in support of this claim as it did in support of its claim that the district court erred in denying its motion for judgment as a matter of law.

A district court's decision to deny a motion for a new trial will not be set aside unless the court abuses its discretion. Logan v. Dayton Hudson Corp., 865 F.2d 789, 790 (6th Cir.1989). Such an abuse will be found only if we have "a definite and firm conviction that the trial court committed a clear error of judgment." Id.

In evaluating Flagstar's motion for a new trial, the district court concluded that, "viewing the evidence in a light most favorable to the Paschals and Edwards[es], the Court has no firm conviction that there is an injustice in the liability verdict or that it is a mistake to allow the liability verdict to stand." Edwards, 109 F.Supp.2d at 704. The district court noted that "[t]he sequence of events, the length of time, the comparable data, the HMDA data, the testing data, all support the conclusion that the verdict was certainly one which a jury could reasonably have come to as to liability." Id. Based on an analysis of a number of comparable cases in which damages were awarded for discrimination, the district court also concluded that "the compensatory damages awarded by the jury are not excessive or out of line with awards in similar cases." Id. at 705.

As explained above, the plaintiffs' evidence, when considered as a whole, provides sufficient support for the jury's verdict. We therefore conclude that the district court did not abuse its discretion when it denied Flagstar's motion for a new trial.

Flagstar's final contention is that the evidence does not support the jury's award of $325,000 in punitive damages to the Paschals. Because we have concluded that the Paschals' claim is barred by the statute of limitations (see Part II.A. above), this issue is moot.

For all of the reasons set forth above, we AFFIRM the district court's judgment in favor of the Edwardses, REVERSE its judgment in favor of the Paschals, and REMAND the case with instructions to dismiss the Paschals' complaint.