Office of Federal Contract Compliance Programs v. Enterprise RAC Company of Baltimore, LLC (agency decision, November 3, 2021)

Plain-English summary

OFCCP alleged that Enterprise's management-trainee hiring process discriminated against African-American applicants under Executive Order 11246. The ALJ found liability under both intentional disparate-treatment and facially neutral disparate-impact theories, relying on statistically significant hiring disparities and application records. The Board held that the ALJ wrongly required Enterprise to persuade at the rebuttal stage when the employer's disparate-treatment burden was only to produce legitimate nondiscriminatory explanations. It also held that subjective hiring criteria can be legitimate unless used as a mask for discrimination and that comparison of four applications from thousands was not enough by itself. For disparate impact, the ALJ had not identified which particular practice caused which disparity or analyzed job relatedness, business necessity, and less discriminatory alternatives. The Board vacated the recommended decision and remanded for a new analysis without deciding whether Enterprise ultimately violated the nondiscrimination requirements.

Decision snapshot

• Cited authorities: 41 C.F.R. §§ 60-1.4(a)(1) and 60-30.28; 42 U.S.C. § 2000e-2(k)
• Outcome: Disparate-treatment and disparate-impact findings vacated; case remanded for application of the proper burdens and causation standards.
• Key point: Statistical disparity establishes an inference, not final proof, and a disparate-impact claim must connect a specific practice to the measured disparity unless the practices cannot be separated.

Full text (DOL official public release)

U.S. Department of Labor Administrative Review Board
200 Constitution Ave. NW
Washington, DC 20210-0001

In the Matter of:

OFFICE OF FEDERAL                             ARB CASE NO.     2019-0072
CONTRACT COMPLIANCE
PROGRAMS, UNITED STATES                       ALJ CASE NO.     2016-OFC-00006
DEPARTMENT OF LABOR,

            PLAINTIFF,

      v.                                      DATE: November 3, 2021

ENTERPRISE RAC COMPANY
OF BALTIMORE, LLC,

            DEFENDANT.

Appearances:

For the Plaintiff:
      Kate S. O’Scannlain, Esq.; Beverly I. Dankowitz, Esq.; Radine Legum,
      Esq.; Jeffrey M. Lupardo, Esq.; Jennifer B. Frey, Esq.; Kevin J. Koll,
      Esq.; Office of the Solicitor, U.S. Department of Labor; Washington,
      District of Columbia

For the Defendant:
      John C. Fox, Esq.; Alexa L. Morgan, Esq.; Jay J. Wang, Esq.; Fox,
      Wang, & Morgan P.C.; Los Gatos, California

Before: James D. McGinley, Chief Administrative Appeals Judge; Thomas
H. Burrell and Randel K. Johnson, Administrative Appeals Judges; Judge
Johnson, concurring


                            ORDER OF REMAND

      PER CURIAM. This matter arises under the nondiscrimination requirements of
Executive Order 11246 (30 Fed. Reg. 12319), as amended, and its implementing
regulations at 41 C.F.R. Chapter 60. Enterprise RAC Company of Baltimore, LLC
                                       2

(Defendant) appeals a Department of Labor Administrative Law Judge’s (ALJ) Recommended Decision and Order (Recommended D. & O.) issued on July 17, 2019.1 Specifically, Defendant appeals the ALJ’s finding that it is liable under Executive Order (EO) 11246 for intentional and unlawful discrimination against African-American job applicants between 2007-2012 and 2014-2017.2 After fully considering the parties’ arguments and the record, we remand the Recommended D. & O. to the ALJ for further consideration consistent with this Order of Remand.

                                BACKGROUND3

   Defendant is a rental car company with a business office in Baltimore,

Maryland, and is a wholly owned subsidiary of Enterprise Holdings, Inc. Defendant, through an ongoing government contract that began on October 1, 2002, provides rental cars to the U.S. Department of Defense Military Traffic Management Program Command. Defendant is a government subcontractor within the meaning of EO 11246.

   Defendant’s management trainee position is an entry-level position that may

lead to promotion to higher paying positions within the company. The hiring process for the management trainee position is multi-stepped, consisting of a review of the applicant’s online application and resume, a phone screening, and up to three in- person interviews with varying members of Defendant’s managerial staff. If an offer of employment is considered after the interviews, a background check is performed. Each step of the hiring process aims to examine candidates further, and only those applicants who successfully pass one step move forward to the next step of the hiring process.

   Throughout the hiring steps, applicants may be rejected by use of an “S” or

“I” labeled disposition code that explains the reason for the rejection. Prior to in- person interviews, Defendant’s hiring recruiters conduct an initial application review, and on occasion, a follow-up phone screening, to determine whether an applicant meets the minimum qualifications for the management trainee position.

1 41 C.F.R. § 60-30.28.
2 EO 11246 authorizes the OFCCP to ensure that Federal contractors and subcontractors doing business with the Federal government comply with the laws and regulations requiring nondiscrimination and equal opportunity in employment, as implemented through 41 C.F.R. Part 60-30. 3 As neither party disputes the general background facts, this background follows the Recommended D. & O. and undisputed facts.
3

   Until 2008, all initial application reviews were performed by hiring

recruiters. In November of 2008, Defendant began using software services which automatically rejected applicants who did not meet the minimum qualifications. Recruiters did not review any automatically rejected applications. Rather, recruiters only reviewed applications accepted by the software to determine whether they should schedule a phone screening.

   If an applicant was rejected during the initial application review or phone

screening steps, he or she was assigned one or more “S” labeled disposition codes. The “S” labeled disposition codes examined whether an applicant possessed the Basic Qualifications required by Defendant to perform the management trainee position.4 To measure the Basic Qualifications against each applicant during the phone screening step, hiring recruiters used a prepared script designed as a questionnaire to obtain and record information from the applicant.5

   The phone screening was also used as a tool to ensure that the applicant had

an interest in a sales and customer service career, the ability to engage in polite conversation, and would be willing to accept the management trainee’s responsibilities, hours, and location of the position.6 To determine whether an applicant possessed the required sales and/or customer service experience, the hiring recruiters looked beyond the applicant’s resume in order to measure how the applicant articulated their experience. If the hiring recruiter rejected an applicant after the phone screening, the recruiter entered a disposition code and had the option to write any notes in the space provided on the questionnaire.7

   Throughout the relevant time period, Defendant offered two to three in-

person interviews to potential hires. The first interview was performed by a Recruiting Specialist or Recruiting Manager to make a preliminary assessment of the applicant’s Core Competencies, including whether the applicant had the customer service skills required for the position. When applicants were brought in for an interview, the hiring recruiter used an interview evaluation form that

4 Defendant’s Basic Qualifications include Education, Sales and/or Customer Service Experience, Job Stability, Work Eligibility, Driving Record, Criminal Convictions, and Age. Recommended D. & O. at 13, 14. 5 Id. at 16. 6 Id. 7 Id.
4

corresponded with Defendant’s core competencies, and included a space for comments at the end.8 If an applicant was rejected during one of the interview steps, Defendant would assign one or more “I” labeled disposition codes to the applicant.

   Defendant categorized the “I” labeled disposition codes to reflect what it

considered the Core Competencies and qualifications for the management trainee position.9 Defendant provided a standard list of behavioral-based questions to all of the interviewers, and interviewers recorded the applicant’s answers or any relevant comments they had on the interview evaluation form. This standard evaluation form was used by each subsequent interviewer in the next interview steps.

   The second interview was performed by a local rental Area and/or Branch

Manager and included a tour with an Assistant Manager who performed a Branch Observation Checklist to provide feedback about the applicant to the Area and/or Branch Manager. Until November of 2014, applicants who successfully completed the second interview advanced to the third and final in-person interview with a senior Group Rental Manager.

   Defendant’s records included an “application packet” for applicants of the

management trainee position. Due to the nature of Defendant’s hiring process, an applicant’s packet contained the application for the position, the applicant’s resume, interview notes, and/or a disposition code(s) if the applicant was rejected for the position.

  The Office of Federal Contract Compliance Programs (OFCCP) periodically

conducts compliance reviews to determine whether covered government contractors are complying with the affirmative-action and nondiscrimination requirements of the EO laws and their implementing regulations.10 Pursuant to Section 202 of EO 11246 and Title 41, Section 60-1.4(a)(1), Defendant agreed not to discriminate against any applicant for employment because of race.

  On May 1, 2008, OFCCP’s Regional Manager notified Defendant by letter

that OFCCP had scheduled an EO legal compliance review of Defendant’s

8 Id.
9 Defendant’s Core Competencies include Customer Service, Persuasiveness/Sales Orientation, Flexibility, Results Driven, Leadership, and Communication. Id. at 27-28. 10 See 41 C.F.R. Part 60-1 (2015).
5

Linthicum, Maryland, car-leasing facility. After auditing Defendant’s hiring practices for the management trainee position, OFCCP issued a Notice of Violation on March 13, 2013. After six conciliation meetings failed to resolve the dispute, OFCCP filed a complaint alleging that Defendant engaged in discrimination violations from August 1, 2006, through July 31, 2008, and that these discrimination violations were continuing.

                   JURISDICTION AND STANDARD OF REVIEW

  The Secretary of Labor has delegated authority and assigned responsibility to

this Board to review decisions by ALJs as provided for or pursuant to Executive Order No. 11246, as amended, and 41 C.F.R. Parts 60-1 and 60-30.11

                                    DISCUSSION

   OFCCP is charged with investigating and prosecuting alleged violations of

EO 11246. A claim of employment discrimination may be established under a disparate impact or disparate treatment theory of discrimination.12 For the reasons set forth below, we conclude that the ALJ erred under both theories.

 1. The ALJ Erred in his Disparate Treatment Analysis


   To prevail in a pattern-or-practice discriminatory treatment claim, OFCCP

has the burden to produce credible evidence that there was a statistically significant racial disparity, and that intentional racial discrimination was the cause of that disparity. A pattern-or-practice discriminatory treatment claim requires that “racial discrimination was the company’s standard operating procedure” and that

11 Secretary’s Order No. 01-2020 (Delegation of Authority and Assignment of Responsibility to the Administrative Review Board), 85 Fed. Reg. 13186 (March 6, 2020). 12 While disparate treatment and disparate impact are different, it is clear that OFCCP may bring claims of liability under both as alternate theories for relief in a given case. See, e.g., Wright v. Nat’l Archives & Records Serv., 609 F.2d 702, 710-11 (4th Cir. 1979) (noting plaintiffs’ election to pursue alternate theories was permissible, but not for the purpose of establishing multiple violations on the same set of facts); Mozee v. Am. Com. Marine Serv. Co., 940 F.2d 1036, 1042 (7th Cir. 1991) (Plaintiffs brought two types of evidence, statistical evidence and evidence of discipline practices, that were “probative both of disparate impact and of a pattern of practice of disparate treatment.”).
6

racial discrimination was a “regular rather than the unusual practice.”13

   It is the OFCCP’s threshold burden to establish, by a preponderance of the

evidence, a prima facie case of discrimination.14 A prima facie case of a pattern-or- practice discrimination can be proven by both statistical disparity and anecdotal evidence of discrimination.15

  If OFCCP establishes a prima facie showing, then the burden shifts to the

employer to rebut the presumption by either offering legitimate, nondiscriminatory reasons for its actions, or by demonstrating that the statistical proof was unsound.16 The employer’s burden here is a “burden of production, of ‘going forward’ with evidence of ‘some legitimate, nondiscriminatory reason for the [action].’”17 The employer must “defeat the prima facie showing of a pattern or practice by demonstrating that the [ ] proof is either inaccurate or insignificant.”18 The burden

13 Int’l Bhd. of Teamsters v. U.S., 431 U.S. 324, 336 (1977). 14 Id. at 360. 15 See Hazelwood Sch. Dist. v. U.S., 433 U.S. 299, 307 (1977). 16 Palmer v. Schulz, 815 F.2d 84, 99 (D.C. Cir. 1987). Citing to Teamsters, OFCCP argues on appeal that “[a] framework designed to analyze singular events does not easily translate to the systemic context. That is why the Supreme Court has held that the McDonnell Douglas prima facie test is inapplicable in government broad-based pattern and practice cases.” OFCCP’s Response Brief, at 87. OFCCP’s point addresses proof of individual instances of causation, which OFCCP correctly noted are not necessary in a pattern-or-practice case. We disagree that Teamsters rejected McDonnell Douglas altogether for pattern-or-practice cases. Teamsters simply noted the need for flexibility. See Teamsters, 431 U.S. at 336 (internal citations omitted); see also id. at 360 n.46. We believe the McDonnell Douglas framework can be flexibly used by factfinders in pattern-or-practice cases. See U.S. v. City of N.Y., 717 F.3d 72, 83-91 (2d Cir. 2013) (discussing McDonnell Douglas and Teamsters in pattern-or-practice cases). 17 Wright, 609 F.2d at 713-14 (quoting Furnco Constr. Corp. v. Waters, 438 U.S. 567, 578) (1978)). 18 Teamsters, 431 U.S. at 360. As “Teamsters sets a high bar for the prima facie case the Government or a class must present in a pattern-or-practice case: evidence supporting a rebuttable presumption that an employer acted with the deliberate purpose and intent of discrimination against an entire class . . . . An employer facing that serious accusation must have a broad opportunity to present in rebuttal any relevant evidence that shows that it lacked such an
7

can be met by “provid[ing] a nondiscriminatory explanation for the apparently discriminatory result.”19 The employer’s purported, legitimate nondiscriminatory reasons must be articulated with some specificity to avoid “conceal[ing] the target” at which employees must aim pretext arguments.20

   If the employer satisfies its burden of production, the focus returns to the

plaintiff. The trier of fact must determine whether the plaintiff has sustained its ultimate burden of proving, by a preponderance of the evidence, that the defendant intentionally discriminated.21 In other words, if the employer offers a legitimate, nondiscriminatory reason, the plaintiff must demonstrate that the proffered reason was not its true reason, but was a pretext for unlawful discrimination.22 OFCCP may prove that the given reasons were pretextual by showing enough instances that a court could find a pattern or practice of racial discrimination.23

  The record supports the ALJ’s conclusion that OFCCP met its initial burden

because it produced statistical evidence establishing a racial disparity sufficient under the law to create a prima facie case of racial discrimination.24 The statistical evidence from both experts demonstrates that the disparity between the expected value and the observed value of offers of employment for white and African- American applicants was or exceeded two standard deviations for every year during

intent.” City of N.Y., 717 F.3d at 87 (citing Teamsters, 431 U.S. at 358); see also Reeves v. Sanderson Plumbing Prods., Inc., 530 U.S. 133, 142-43 (2000). 19 Teamsters, 431 U.S. at 360 n.46. 20 Figueroa v. Pompeo, 923 F.3d 1078, 1088 (D.C. Cir. 2019) (citing Lanphear v. Prokop, 703 F.2d 1311, 1316 (D.C. Cir. 1983)). 21 Tex. Dep’t of Cmty. Affairs v. Burdine, 450 U.S. 248, 253 (1981). 22 Id. 23 See Paxton v. Union Nat’l Bank, 688 F.2d 552, 567 (8th Cir. 1982) (“It was then incumbent on the plaintiffs and intervenors to prove that the given reasons were pretextual in at least enough instances that the court could find a pattern and practice of racial discrimination against blacks in the discharge class.”). Pretext “means a lie, specifically a phony reason for some action.” Russell v. Acme–Evans Co., 51 F.3d 64, 68 (7th Cir.1995). 24 The statistical evidence from both experts demonstrates that offers of employment by race (between white and African American applicants) had a significant two standard deviation equivalent or higher (some years exceeding three standard deviations) for every year during the relevant time period except for 2013. Plaintiff’s Exhibit (PX) 106, at 18; PX 114, at 1.
8

the relevant time period except for 2013.25 Courts have consistently found significance in disparities exceeding two standard deviations.26

   Accordingly, the burden shifted to the Defendant to rebut the presumption or

inference created by the prima facie case. Defendant, unable to establish OFCCP’s statistical proof as unsound because its expert provided similar statistical evidence, provided evidence of legitimate, nondiscriminatory hiring practices.27 Specifically, Defendant provided extensive documents consisting of individual application packets for management trainee program applicants during the time period to rebut the presumption by explaining its hiring decisions by use of nondiscriminatory disposition codes.28 Some of Defendant’s documents also included hand-written notes from interviewers explaining why they selected certain disposition codes for an applicant to demonstrate that disposition codes were applied in a race-neutral manner. Accordingly, the record shows that Defendant articulated nondiscriminatory reasons for rejecting applicants for the management trainee program.

  As noted above, Defendant’s burden in the rebuttal stage is one of production,

not persuasion. As the Supreme Court stated in Burdine, “[t]he explanation provided must be legally sufficient to justify a judgment for the defendant. If the defendant carries this burden of production, the presumption raised by the prima

25 Recommended D. & O. at 100.
26 See Hazelwood, 433 U.S. at 308 n.14; Segar v. Smith, 738 F.2d 1249, 1283 (D.C. Cir. 1984) (finding the plaintiff presented statistical analyses that met “the generally accepted
.05 level of statistical significance” and noting a two standard deviation corresponds to a statistical significant .05 level which “are certainly sufficient to support an inference of discrimination.”); H.B. Rowe Co., Inc. v. Tippett, 615 F.3d 233, 245 (4th Cir. 2010) (finding the statistical evidence demonstrated that African American subcontractors were underutilized because the level “fell outside of two standard deviations from the mean and therefore, was statistically significant at a 95 percent confidence level.”); see also Palmer, supra note 16, 815 F.2d at 92 (noting that a statistical “disparity measuring two standard deviations (to be more precise, 1.96 standard deviations) corresponds to a 5% probability of randomness”). 27 Recommended D. & O. at 108. 28 The ALJ noted that the record contained at least sixty individual application packets. Id. at 106. The ALJ identified that Defendant submitted DX 7366, via a thumb drive, and described it as “Enterprise’s application set, Excel Spreadsheet showing which applications were analyzed by Dr. Madden and Dr. White.” Id. at 2-3, 98. The record submitted to the ARB did not contain the thumb drive.
9

facie case is rebutted, and the factual inquiry proceeds to a new level of specificity. Placing this burden of production on the defendant thus serves simultaneously to meet the plaintiff's prima facie case by presenting a legitimate reason for the action and to frame the factual issue with sufficient clarity so that the plaintiff will have a full and fair opportunity to demonstrate pretext.”29

   The ALJ appeared to state the correct standard for Defendant’s rebuttal but

nonetheless conducted a persuasion analysis.30 The ALJ found that the application packet documents were not persuasive because of the subjectivity used by the recruiters to reject applicants.31 The ALJ also reviewed four applications (comparing the use of disposition codes between two white applicants with two African-American applicants) to support his conclusion that the disposition codes were not applied consistently.32 The ALJ’s analysis of Defendant’s rebuttal was legal error because at this point Defendant’s burden was one of production, not persuasion.33

 In particular, the ALJ did not explain why Defendant’s production of

documents explaining the racial disparities shown in the statistical evidence did not

29 Burdine, 450 U.S. at 255-56.
30 Recommended D. & O. at 98-99, 103.
31 The ALJ explained that recruiters’ testimony demonstrated the subjectivity of Defendant’s hiring process and use of disposition codes, finding that “[i]t is clear that Enterprise did not apply all of the disposition codes consistently. That is patently obvious with respect to the S3 disposition code that was used frequently and had a disparate impact on African-American applicants.” Recommended D. & O. at 105. 32 Id. at 106. 33 In a pattern-or-practice case, an employer can satisfy its burden of production by “produc[ing] any evidence that is relevant to rebutting the inference of discrimination.” City of N.Y., 717 F.3d at 85. An employer can satisfy this burden by either showing the plaintiff’s statistics are flawed, or by establishing legitimate, nondiscriminatory reasons for the observed disparities. See Teamsters, 431 U.S. at 360 n.46 (noting that the employer’s burden must “be designed to meet the prima facie case” and in cases where the government’s case consists of proof of racial disparities of a regularly followed policy, an employer may “provide a nondiscriminatory explanation for the apparently discriminatory result.”); Segar, 738 F.2d at 1267-68 (an employer can either refute the claim that a disparity exists, or “the employer can offer an explanation defense; such a defense amounts to a claim that an observed disparity has not resulted from illegal discrimination.”).
10

satisfy the burden of production.34 Accordingly, we find that the ALJ’s persuasion analysis of Defendant’s rebuttal was legal error.35

  Further, the ALJ’s burden of persuasion analysis was also flawed. The ALJ

found that Defendant’s subjective use of disposition codes did not adequately explain the racial disparities shown in the statistical evidence.36 However, in

34 Yet, the ALJ recognized OFCCP’s acknowledgement that Defendant’s race-neutral hiring practices were “legitimate standards” “on paper” for the management training position. Recommended D. & O. at 108 (“The issue here is not the process as it existed on paper, it is the process as it was applied in actual practice.”). 35 In the response brief, relying on the Board’s Bank of America decision, the Solicitor argues on appeal that the Board may accept the ALJ’s factual findings if supported by substantial evidence and limit its review to the ultimate finding of discrimination. See OFCCP v. Bank of Am., ARB No. 2013-0099, ALJ No. 1997-OFC-00016, slip op. at 12 (ARB Apr. 21, 2016) (“After a full evidentiary hearing, there is no need to engage in the burden of production analysis to determine whether the OFCCP presented a prima facie case or whether BOA presented legitimate, non-discriminatory reasons for its practices. This burden of production analysis applies to motions for summary judgment and motions for judgment as a matter of law.”). It is true that in cases arising under many statutes, we discourage the ALJ from analyzing a prima facie case for matters that have had a full evidentiary hearing. Brune v. Horizon Air Indus., Inc., ARB No. 2004-0037, ALJ No. 2002- AIR-00008, slip op. at 15 (ARB Jan. 31, 2006). However, in these EO cases, the subject matter is complex and often built upon statistical models serving as a prima facie case or inference of discrimination. Further, federal case law follows this prima facie case framework. Our problem with Bank of America’s avoiding the prima facie case is that it dispels the model in one respect, but cites cases which in turn depend on the prima facie model in another, creating a mismatch from citations operating within the inference and prima facie model but for the premise of proof of intentional discrimination. Proof of a prima facie case is not proof of intentional discrimination. “A McDonnell Douglas prima facie showing is not the equivalent of a factual finding of discrimination.” Furnco, 438 U.S. at 579-80. 36 “It is clear that Enterprise did not apply all of the disposition codes consistently. . . . The testimony of Ms. Morris, Ms. Hardesty, Ms. Lichter and Mr. Wucher showed the highly subjective and seemingly arbitrary nature of what Enterprise counted as relevant work experience; where potentially two applicants could have held the same job and one gets credit for it and the other does not.” Recommended D. & O. at 105; id. at 107. But in concluding that the Defendant was inconsistent, the ALJ did not fully analyze the employer’s explanations, seemingly disregarding subjective explanations entirely. The full quote from Licther’s testimony is as follows: “Ms. Lichter said two applicants could have held the exact same job, had the exact same job title, and worked for the exact same company and one could get credit for sales or customer service experience and the other
11

analyzing the evidence and arguments concerning Enterprise’s hiring criteria, (including its requirement of sales and/or customer service), the ALJ conflated evidence of subjectivity with evidence of discrimination without allowing an employer’s legitimate use of subjective hiring criteria.37

   Although there is a risk that a nefarious employer may use subjective

standards as cover for discrimination, subjective criteria which are facially nondiscriminatory “no matter how subjective the criteria—may constitute a legitimate reason” for rejecting applicants.38 Subjective evaluation criteria “can constitute [ ] legally sufficient, legitimate, nondiscriminatory reason[s]” for an employer’s business decisions.39 In fact, “subjective evaluations of a job candidate are often critical to the decision-making process, and if anything, are becoming more so in our increasingly service-oriented economy . . . .”40

    However, an employer’s subjective criteria is not beyond scrutiny. The

reasons given must have some substance to allow for evaluation.41 If, for example, the ALJ compared the qualifications of those rejected with those that were hired in order to show intentional discrimination, the differences must be so striking as to permit a reasonable factfinder to raise the alarm of a pattern or practice of intentional discrimination.42 Slight or even mistaken differences in qualifications fail to satisfy this burden because the ALJ does not sit as a super-personnel board

rejected for a lack of the same depending upon how they articulated their experiences during their telephone conversations with her.” Id. at 104 (emphasis added). Mr. Wucher explained “he was looking for applicants to ‘sell themselves’ at that moment and that it was helpful, but not necessary, for them to even talk about their prior work experience.” Id. at 105. 37 Id. at 104-05, 107, 110 (rejecting employer’s hiring criteria as examples of legitimate, nondiscriminatory reasons because they were subjective). 38 Figueroa, 923 F.3d at 1088. 39 Denney v. City of Albany, 247 F.3d 1172, 1185 (11th Cir. 2001) (internal quotations omitted). 40 Id. at 1185-86 (internal quotations omitted); Millbrook v. IBP, Inc., 280 F.3d 1169, 1176 (7th Cir. 2002) (similar subjective comments made of white and black candidates, negating any inference that comments were codes for race). 41 Figueroa, 923 F.3d at 1088 (internal quotations omitted) (analyzing how some subjective reasoning can be so skeletal that it can “conceal the target” in a pretext analysis). 42 Millbrook, 280 F.3d at 1180-81.
12

second-guessing the employer’s hiring practices.43

   This objective evaluation is not accomplished by reviewing and comparing a

small number of applications.44 The ALJ relied on the subjectivity of the hiring decision-making process to summarily conclude that the racial disparity shown in the statistical evidence, and the few number of applications he reviewed, was the result of racial discrimination. Thus, “absent evidence that subjective hiring criteria
[was] used as a mask for discrimination, the fact that an employer based a hiring or promotion decision on [ ] subjective criteria” does not in and of itself prove pretext for intentional discrimination.45

  Because the ALJ erred in his disparate treatment analysis we must remand

the case back to the ALJ.

 2. The ALJ Erred in his Disparate Impact Analysis


  In order to establish a disparate impact theory of discrimination, OFCCP

must establish that a particular employment practice has a disproportionately

43 See DeJarnette v. Corning Inc., 133 F.3d 293, 299 (4th Cir. 1998) (a “[c]ourt does not sit as a kind of super-personnel department weighing the prudence of employment decisions”) (internal quotation marks omitted). 44 The ALJ conceded that “[e]xamining four applications from one fiscal year out of thousands of applications submitted over an 11-year period, and observing that standards were applied in an inconsistent manner that was more lenient for the white applicants than the African-American applicants, in and of itself might not cast substantial doubt on the legitimacy of the hiring process.” Recommended D. & O. at 106. 45 Denney, 247 F.3d at 1185; see also Alvarado v. Texas Rangers, 492 F.3d 605, 616 (5th Cir. 2007) (“An employer’s subjective reason for not selecting a candidate, such as a subjective assessment of the candidate’s performance in an interview, may serve as a legitimate, nondiscriminatory reason for the candidate’s non-selection.”); Patrick v. Ridge, 394 F.3d 311, 317 (5th Cir. 2004) (the McDonnell Douglas framework of the employer’s burden of production to articulate nondiscriminatory reasons for its hiring decisions “does not mean that an employer may not rely on subjective reasons for its personnel decisions.”). It is important to note that the case before us presents a factual situation that is distinctly different from that of Alvarado, in the sense that here the employer did not exercise purely subjective judgment in a vacuum or without context. Indeed, the relevant documentation in the record contains other factors which shaped and explained the employer’s hiring decisions as a whole. See Alvarado, 492 F.3d at 617 (discussion of the absence of these types of contextual factors).
13

adverse impact on African-American applicants.46 Under a disparate impact claim, the challenged employment practice is facially neutral but has an adverse impact on a protected class.47 The principals of a disparate impact analysis were summarized by the Supreme Court in the Ricci v. DeStefano case:

   Under the disparate-impact statute, a plaintiff establishes a prima facie
   violation by showing that an employer uses “a particular employment
   practice[48] that causes a disparate impact on the basis of race, color,
   religion, sex, or national origin.” An employer may defend against
   liability by demonstrating that the practice is “job related for the
   position in question and consistent with business necessity.” Even if the
   employer meets that burden, however, a plaintiff may still succeed by
   showing that the employer refuses to adopt an available alternative
   employment practice that has less disparate impact and serves the
   employer’s legitimate needs.49

Unlike the three-part test articulated above for disparate treatment, the employer’s rebuttal to prove business necessity or job relatedness in disparate impact cases is proof by a preponderance of the evidence, not merely a burden of production.50

  We conclude that the ALJ also erred in his disparate impact analysis.

The ALJ’s review of this alleged form of discrimination is cursory (two pages), without any of the necessary analysis following the relevant statute and the case law that forms the pillars of disparate impact liability.51 The ALJ failed to identify what specific employment practice caused the disparate impact and failed to adequately discuss the issue of the respondent’s possible defenses to the prima facie

46 See Davis v. Dist. of Colum., 925 F.3d 1240, 1248-49 (D.C. Cir. 2019). 47 Griggs v. Duke Power Co., 401 U.S. 424, 430-31 (1971). 48 Plaintiff must “point to a defendant’s policy or policies causing that disparity. A robust causality requirement ensures that ‘[r]acial imbalance . . . does not, without more, establish a prima facie case of disparate impact’ and thus protects defendants from being held liable for racial disparities they did not create.” Tex. Dep’t of Hous. & Cmty. Affairs v. Inclusive Cmtys. Project, Inc., 576 U.S. 519, 542 (2015) (quoting Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 653 (1989), superseded by statute on other grounds, 42 U.S.C. § 2000e– 2(k)). 49 Ricci v. DeStefano, 557 U.S. 557, 578 (2009) (citations omitted). 50 Lewis v. City of Chicago, 560 U.S. 205, 213 (2010). 51 Recommended D. & O. at 112-13.
14

case (i.e. job relatedness or business necessity), and whether there was an alternative practice which could have served the same purposes of the practice with a lesser disparate impact. These requirements are clearly spelled out in the applicable statutes and case law.52 As a result, this, too, requires that we remand the matter to the ALJ.

                                   CONCLUSION

  The Board finds the ALJ committed legal error and that it is proper to

remand the case to the ALJ to apply the correct legal standards. Accordingly, the ALJ’s Recommended Decision and Order is VACATED, and this case is REMANDED to the ALJ for further proceedings consistent with this Order of Remand.

   SO ORDERED.

Judge Randel K. Johnson, concurring

   I concur with the majority that this decision should be remanded to the ALJ

for reconsideration of both the claims of disparate treatment and disparate impact. I write separately to elaborate on the reasons for remand regarding the ALJ’s analysis of the disparate impact claim. I also note one area where I slightly differ with the majority opinion.

   In disparate treatment cases, the ultimate issue in question is the intent of

the defendant underlying the complained of act—was there intentional discrimination based on a protected classification?53 In disparate impact cases, however, intent of the defendant is irrelevant; rather the focus of the case is on some particular practice of the defendant which adversely affects classes protected under civil rights laws to a significantly greater degree than a majority group, depending on the comparators. The ALJ, in his brief analysis, found liability under

52 Ricci, 557 U.S. at 577-78 (citations omitted); Civil Rights Act of 1991, 42 U.S.C. § 2000e–2(k). 53 On appeal, Defendant argues that the ALJ erred in finding liability under both theories of discrimination. While the theories of disparate treatment and disparate impact are very different, it is clear that OFCCP may bring claims of liability under both theories. On this point, see supra note 12.
15

disparate impact principles while discussing two cases, Griggs v. Duke Power Company, 401 U.S. 424 (1971) and Davis v. District of Columbia, 925 F.3d 1240 (2019).54

    As the ALJ implicitly recognized, the genesis of the theory of disparate

analysis of liability under Title VII of the 1964 Civil Rights Act was the seminal 1971 Griggs v. Duke Power Co. case. However, not surprisingly, that case spawned multiple subsequent Supreme Court decisions55 seeking to clarify the meaning of the decision (which had internal inconsistencies), and literally hundreds and hundreds of cases in the lower courts. The typical issues which generated litigation were: (1) what type of employment practice was even covered under disparate impact analysis; (2) what type and degree of statistical comparisons were appropriate to establish a prima facie case of a cognizable disparate impact claim; (3) to what degree did a plaintiff have to identify with specificity the employment practice complained of and prove a causal link between that employment practice and the statistically demonstrated disparate impact; (4) assuming that a prima facie case was shown, was the burden of proof on the defendant to show that (and thus justify) that the practice was justified by business necessity or job relatedness—was it a burden of production or persuasion; and (5) even upon such a justification, could the plaintiff still prevail upon showing that there was an alternative practice which the employer could have implemented with a lesser disparate impact which met the same business related reasons for that practice.56

54 Recommended D. & O. at 112-13.
55 See Watson v. Fort Worth, 487 U.S. 977 (1988) (subjective criteria, e.g., alertness, personal appearance, ambition, leadership ability, ability to work with others; job related, manifest relationship); Conn. v. Teal, 457 U.S. 440 (1982) (written examinations; manifest relationship to the employment in question); N.Y. City Transit Auth. v. Beazer, 440 U.S. 568, 587 n.31 (1979) (rule against employing drug addicts in both safety and non-safety jobs; if goals are “significantly served by—even if they do not require—”the practice it bears a manifest relationship to the employment in question); Washington v. Davis, 426 U.S. 229 (1976) (general written test of verbal skills; job relatedness); Dothard v. Rawlinson, 433 U.S. 321 (1977) (height and weight requirements; manifest relation to the employment in question, job related); Albemarle Paper Co. v. Moody, 422 U.S. 405 (1975) (written aptitude tests; manifest relation to the employment in question, job related). 56 For an in-depth review of the case law on these issues, as of 1991 prior to the enactment of the 1991 Civil Rights Act discussed below, see House Education and Labor Committee Report accompanying the “Civil Rights and Woman's Equity in Employment Act of 1991.” H.R. Rep. No. 102-40, at 23-45, 124-137(1991).
16

   Importantly, courts put their own gloss on these issues since the Griggs

decision, with varying degrees of consistency, finally culminating in the controversial Wards Cove Packing Co. v. Antonio, 490 U.S. 642 (1989) decision. This decision triggered some three years of debate on Capitol Hill with introduction of major legislation, a veto by President Bush, and ultimately a legislative compromise embodied in the 1991 Civil Rights Act which amended Title VII. That compromise added a new subsection (k) to sec.703 of the Act, codified at 42 U.S.C sec. 2000e- 2(k), which reads, in part, as follows:

  (k)(1)(A) An unlawful employment practice based on disparate impact is
  established under this title only if –

  (i) a complaining party demonstrates that a respondent uses a particular
  employment practice that causes a disparate impact on the basis of race,
  color, religion, sex or national origin and the respondent fails to
  demonstrate that the challenged practice is job-related for the position
  in question and consistent with business necessity, or

  (ii) the complaining party makes the demonstration described in
  subparagraph (C) with respect to an alternative employment practice
  and the respondent refuses to adopt such alternative employment
  practice

  (B)(i) with respect to demonstrating that a particular employment
  practice causes a disparate impact as described in subparagraph (A) (i),
  the complaining party shall demonstrate that such particular
  challenged employment practice causes a disparate impact, except that
  if the complaining party can demonstrate to the court that the elements
  of respondent’s decision-making process are not capable of separation
  for analysis, this decision-making process may be analyzed as one
  employment practice.

  (ii) If the respondent demonstrates that a specific employment practice
  does not cause the disparate impact, the respondent shall not be
  required to demonstrate that such practice is required by business
  necessity…
                                          17

   The “exclusive” legislative history57 of the 1991 Civil Rights Act concerning

this language is as follows:

   The terms “business necessity” and “job-related” are intended to reflect
   the concepts enunciated by the Supreme Court in Griggs v. Duke Power
   Co., 401 U.S. 424 (1971), and in other Supreme Court decisions prior to
   Wards Cove Packing Co. v. Antonio, 490 U.S. 642 (1989).

   When a decision-making process includes particular, functionally-
   integrated practices which are components of the same criterion
   standard, method of administration, or test such as the height and
   weight requirements designed to measure strength in Dothard v.
   Rawlinson, 433 U.S. 321 (1977), the particular, functionally integrated
   practices may be analyzed as one employment practice.

The principals of a disparate impact analysis as delineated by the 1991 Act were recently summarized, as also noted by the majority opinion, by the Supreme Court in the Ricci v. DeStefano case:

   The Civil Rights Act of 1964 did not include an express prohibition on
   policies or practices that produce a disparate impact. But in Griggs v.
   Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971), the
   Court interpreted the Act to prohibit, in some cases, employers’ facially
   neutral practices that, in fact, are “discriminatory in operation.” The
   Griggs Court stated that the “touchstone” for disparate-impact liability
   is the lack of “business necessity”: “If an employment practice which
   operates to exclude [minorities] cannot be shown to be related to job
   performance, the practice is prohibited.” Under those precedents, if an
   employer met its burden by showing that its practice was job-related,
   the plaintiff was required to show a legitimate alternative that would
   have resulted in less discrimination.

   Twenty years after Griggs, the Civil Rights Act of 1991, 105 Stat. 1071,

57 “Section 105(b) of the 1991 Act states: ‘No statements other than the interpretive memorandum [quoted here] appearing at Vol. 137 Congressional Record S.15276 (daily ed. Oct. 25, 1991) shall be considered legislative history of, or relied upon in any way as legislative history in construing or applying, any provision of this act that relates to Wards Cove – Business necessity/cumulation/alternative business practice.’” See also Henry H. Perritt, Jr., Civil Rights in the Workplace, Vol. 1, at 4-5, 285-286 (2d ed. 1995).
18

  was enacted. The Act included a provision codifying the prohibition on
  disparate-impact discrimination. That provision is now in force along
  with the disparate treatment section already noted. Under the
  disparate-impact statute, a plaintiff establishes a prima facie violation
  by showing that an employer uses “a particular employment practice
  that causes a disparate impact on the basis of race, color, religion, sex,
  or national origin.” An employer may defend against liability by
  demonstrating that the practice is “job related for the position in
  question and consistent with business necessity.” Even if the employer
  meets that burden, however, a plaintiff may still succeed by showing
  that the employer refuses to adopt an available alternative
  employment practice that has less disparate impact and serves the
  employer’s legitimate needs.[58]

   Similarly, a leading treatise on discrimination law summarized the elements

of a disparate impact case as incorporated into the 1991 Civil Rights Act, broken out into separate paragraphs, as follows:

  Thus, the allocation of burdens now follows this scheme in Title VII
  adverse impact cases:

  (1) The Prima Facie Case: A court will consider statistical evidence
  offered by both the plaintiff and the defendant to determine whether, on
  the basis of those statistics that are most probative, the challenged
  practice or selection device has a substantial adverse impact on a
  protected group. The burdens of production and persuasion at this stage
  are on the plaintiff.

  (2) Business Necessity: If impact is established, the inquiry becomes
  whether the practice or selection device is “job-related for the position in
  question and consistent with business necessity.” The burdens of
  production and persuasion at this stage are on the defendant, but the
  precise meaning of this standard remains an open question . . .

  (3) Alternatives with a Lesser Impact: To rebut the employer’s proof of
  business necessity, a plaintiff can show that the employer refused to
  implement an effective alternative practice or selection device that
  would have a lesser adverse impact.

58 Ricci, 557 U.S. at 577-78 (citations omitted).
19

   The 1991 Act confirmed the general rule—plaintiffs’ causation burden
   generally requires plaintiffs to identify the specific policy or practice
   resulting in the disparity. But, crucially, where plaintiffs can
   affirmatively prove that the individual steps in the suspect policies or
   practices cannot be separated for examination, this general rule does not
   apply. [59]

  Against this complex backdrop of judicial interpretations and statutory

requirements, the ALJ summarily concluded, over barely two pages, that the Respondent was liable for implementing practices with a disparate impact. As noted above, the ALJ briefly quoted from the Griggs case and discussed one case out of the D.C. Circuit, Davis v. District of Columbia, 925 F.3d 1240 (D.C. June 7, 2019). The Griggs case has been modified over decades of case law, and in turn shaped by the statutory language in the 1991 Act (which is equally applicable here).60 The Davis

59 1 Barbara T. Lindemann & Paul Grossman, Employment Discrimination Law 118- 19 (4th ed. 2008) (citations omitted). For a detailed review of issues leading up to passage of the 1991 Act and how the provisions of that Act both adopted some aspects of the Wards Cove decision, such as the need to specifically identify the practice in question and prove that that practice caused the disparate impact alleged, but rejected others, particularly that the employer only has the burden of production in showing that the practice in question is job-related or justified by business necessity as distinguished from the burden of persuasion, see Kent Spriggs, Representing Plaintiffs in Title VII Actions §§ 3.02 & 3.03 (2d ed. 1994); Robert E. McKnight, Jr., Representing Plaintiffs in Title VII Actions § 6.02[A] (4th ed. 2014) (“If distinct practices used in combination are ‘not capable of separation for analysis’ [citing the applicable provision in the Civil Rights Act of 1991] of consequent disparity, then the combination may be used. Practices are not incapable of separation just because they exhibit a common feature, e.g., subjectivity.”). See Henry H. Perritt, Jr., supra note 57, § 5.7, Legislative History of Disparate Impact Provisions, citing the above- referenced memorandum of interpretation noting, “his [referring to the memorandum] does not mean much, of course, without external references to Supreme Court cases applying the business necessity and job-related concepts before Wards Cove. It does however make it clear that functionally integrated practices may be analyzed as one employment practice, thus helping to interpret the discrete practice provisions in (k)(1)(A) and (b) [of the Civil Rights Act of 1991.” (emphasis added). 60 Of course the Civil Rights Act of 1991 did not directly amend Executive Order 11246. However, the proscriptions of Title VII govern the EO. See U.S. v. Trucking Mgmt., Inc., 662 F.2d 36 (D.C. 1981) (finding EO 11246 does not override Title VII’s protections of seniority). Further, OFCCP’s compliance manual recognizes the applicability of Title VII principles. See Federal Contract Compliance Manual, at 361 (“Title VII of the Civil Rights Act of 1964.”)
20

case was cited for the proposition that OFCCP need not specifically identify the individual practices which caused the alleged disparate impact.61

   The ALJ in his conclusory decision, failed to link which statistical

comparisons62 were caused by what specific employment practice, and failed to meaningfully engage on the issue of the respondent’s possible defenses to the prima facie case, (i.e. job relatedness or business necessity) as specified by the statute and case law, and regarding whether there was an alternative practice which could have met the same purposes of the practice with a lesser disparate impact and the employer failed to adopt those, again as specified in the statute.63 Echoing these

61 Recommended D. & O. at 112-13.
62 Although the 1991 Act did not define what type of statistical comparisons were necessary to prove a prima facie case, such comparisons are in fact critical and must be demonstrated to meet certain criteria as defined under the case law—which alone has many complexities. See Lindemann & Grossman, supra note 59, at 122-43 (“The Plaintiff's Prima Facie Case”); McKnight, supra note 59, at § 6.02 [A] (“The plaintiff normally proves the disparity resulting from a particular employment practice with statistical evidence that controls for other factors that might have caused or contributed to the disparity. For example, it will usually not suffice to show a disparity between the groups of people who occupy a certain type of position because such a comparison does not account for the demographics of the applicants. And if applicants are being scrutinized, it may be necessary to sort them into the categories of the qualified and the unqualified, and calculate the statistics only with reference to the qualified applicants.”). 63 “Enterprise contends that OFCCP failed to prove that any of its practices discriminated against African-Americans and that it demonstrated that it had legitimate, non-discriminatory business reasons for its hiring practices. Accepting these contentions requires taking a very myopic view of the record.” This summary paragraph alone indicates the confusion of the ALJ’s analysis in that while the concept of “legitimate, nondiscriminatory business reasons for its hiring practices” bears on the allegations of disparate treatment, the focus in a disparate impact case is on whether or not a particular practice which does have a discriminatory impact can be justified based on business related reasons. Recommended D. & O. at 112. The ALJ also noted that “[i]n this case, the evidence shows that the seemingly race neutral standards Enterprise articulated for hiring management trainees had a disproportionate adverse impact on African-American applicants. The statistical evidence presented by Dr. Madden and Dr. White show that the racial disparity in job offers to African-American applicants was well in excess of two standard deviations in every year of the charge period except for fiscal year 2013.” Id. at

113. This broad observation of overall statistics does not meaningfully address what
     disproportionate impact was caused by which practice, which the law requires unless the plaintiff can show that it is impossible to disaggregate the practices.
     21

omissions, a thorough review of the record found very little discussion of those principles at the trial hearing itself.64

   Additional discussion with regard to the requirement to identify a specific

practice is warranted. As made clear in the aforementioned case law and the 1991 Civil Rights Act, a plaintiff must establish that a particular practice caused the disparate impact. As the courts have explained, this requirement exists because blanket assertions of disparate impact arising from broad groupings of practices would effectively make it impossible for an employer to defend those practices as job-related or justified by business necessity, resulting in an employer engaging in racial balancing of workforce demographics to avoid liability.65 This requirement is not a trivial or technical one to be lightly considered by the courts. This area is one of the few addressed in the “exclusive” legislative history which indicates its importance, but also that its contours are somewhat unclear. Moreover, it is also important to make clear that these requirements do not impose a straitjacket on the plaintiff; the statute itself and the legislative history indicate that where it is not possible on the part of the plaintiff to disentangle a grouping of practices (“not capable of separation”) a plaintiff could be relieved of this requirement.66

   The ALJ appears to be aware of this requirement by noting it was an issue in

64 In quoting from the case law, the ALJ makes passing reference to the concept of job relatedness and business necessity as a defense to a practice that has a disparate impact, but makes no reference at all to the subsequent prong of impact analysis which provides that a plaintiff can show that an alternative practice exists and the employer refuses to adopt such an alternative. Both are set out in the applicable language under the 1991 Civil Rights Act and thoroughly discussed in case law, albeit often with a lack of clarity, but are critical to a disparate impact analysis. See The Civil Rights Act of 1991: The Business Necessity Standard, 106 Harv. L. Rev. 896 (1993); Lindemann & Grossman, supra note 59, at 148-56; McKnight, supra note 59, at § 6.02[C] (“Since the Supreme Court’s opinion in Albermarle Paper Co., and in the codification of disparate impact case law, plaintiffs have a last chance in the pattern of proof: if the defendant produces evidence that the challenged practice was job-related and consistent with business necessity, the plaintiff ‘may still succeed by showing that the employer refuses to adopt an available alternative employment practice that has less disparate impact and serves the employer’s legitimate needs.’”) (internal citations to the applicable provision of the 1991 Civil Rights Act omitted). 65 See, e.g., Watson, 487 U.S. at 993-996. 66 See Spriggs, supra note 59, § 3.03 [3] on how the statutory language and the interpretive memorandum appears to soften a strict application of the Wards Cove admonition on the need to identify a specific practice, when the practices are intertwined to some (unspecified) degree.
22

the Davis v. District of Columbia case67 (which involved a reduction-in-force (RIF)), but provides little to no insight as to how he applies the requirement to the very different facts (which involve hiring), before him. The ALJ broadly notes the “race neutral standards” used by the Respondent and later discusses the “disproportionate adverse impact on African-American applicants” arising from three categories (customer service and sales experience, communication ability, and compatible career and direction), referencing Dr. White’s analysis. Dr. White’s analysis does provide information on disparities, but the ALJ provides little analysis as to how he believes they are relevant here, in terms of which criteria caused a disparate impact, much less moving to the next logical step of discussing whether or not they were job-related, or required by business necessity.

   Further, the Davis case cited by the ALJ reasserts the importance, with a

lengthy discussion in both the majority and the dissent, of identifying the particular practice in question and the rationale for this requirement. The court also noted that this was a case of first impression68 and its limited findings, stating “We need not generally decide whether a RIF as such might ever be a ‘particular employment practice’ under section 2000e – 2(k)(1)(A)(i). Terminating a large group of employees in a compressed time frame is clearly an adverse employment action within the meaning of Title VII, and an employer’s assertion that the firings were a required by budget cuts does not somehow immunize them from Title VII scrutiny.”69 The fact that the Davis case and its reasoning, focused on a one-time, immediate reduction-in-force, brings into question its applicability to the case before us where the facts involved allegations concerning hiring practices spread out over several years. The case certainly does not stand for the proposition that several hiring selection criteria can be simply grouped together, in the absence of a showing that such practices are not capable of separation. Such a proposition would be clearly contrary to the case law and the 1991 Act.

67 “The district court granted summary judgement for the agency saying that the plaintiffs failed to identify a specific employment practice that had a racially disparate impact. The D.C. Circuit disagreed. The D.C. Circuit held that a disparate impact claim could include ‘the process by which the Agency identified plaintiffs’ jobs for elimination as a particular employment practice.’” Recommended D. & O. at 112 (citing Davis, 925 F.3d at 1251) (internal citations omitted). 68 “This is the first time this court has been asked whether a RIF, or more precisely, the practices through which an employer implements a RIF are subject to disparate impact review under Title VII, but we see no basis to exempt such practices from otherwise applicable law.” Davis, 925 F.3d at 1250. 69 Id. at 1251-52.
23

   Of course, what quantum of proof will be “particular” or specific enough to

meet the criteria under the law will clearly be a question of fact—with gray areas being unavoidable. But here, the ALJ has provided little to no insight with regard to his reasoning as to how this critical aspect of disparate impact case law was applied to the facts before him and provided no reasoning in discussing how a plaintiff could be relieved of this burden if the practices are incapable of separation. This important aspect of the 1991 Act cannot be given short shrift. In this case, as a consequence of the ALJ’s failure to analyze and apply other principles of disparate impact in reaching his conclusions, a remand is clearly necessary.

   In closing, I note some concern over the analysis contained in Footnote 35 of

the majority opinion. In my view, the case law with regard to the application of the so-called McDonnell Douglas–Burdine tripartite order of proof (beyond orders for summary judgment), is unclear. What has become clear, with the advent of jury trials under Title VII following the 1991 Civil Rights Act, is that juries should not be instructed in the intricacies of this tripartite order of proof. To avoid confusion, the rules as to when it must be applied in bench trials and when it should fall to the wayside (either as a matter of discretion or a matter of law) to allow the decision- maker to look at the evidence as a whole to determine the ultimate question of intentional discrimination is unsettled.70

   Given this lack of clarity, I cannot say that the approach taken in the Board’s

Bank of America decision was wrong. There are, simply, a variety of strands in the case law. However, I do agree with the included sentence in the footnote that “[p]roof of a prima facie case is not proof of intentional discrimination” under McDonnell Douglas, but this appears to state a truism. Indeed, the prima facie case raises an inference of discrimination but that inference would only lead to conclusive “proof of discrimination” in an exceedingly rare situation where the

70 As one jurist put it, “I write separately to call attention to the snarls and knots that the current methodologies used in discrimination cases of all kinds have inflicted on courts and litigants alike. The original McDonnell Douglas decision was designed to clarify and simplify the plaintiff's task in presenting such a case. Over the years, unfortunately, both of those goals have gone by the wayside.” See Coleman v. Donahoe, 667 F.3d. 835, 862 (7th Cir. 2012) (concurring opinion); see also Brady v. Sargeant at Arms, 520 F.3d. 490, 494 (D.C. Cir. 2008) (“Much ink has been spilled regarding the proper contours of the prima- facie-case aspect of McDonnell Douglas . . . . It has not benefited employees or employers; nor has it simplified or expedited core proceedings. In fact, it has done exactly the opposite, spawning enormous confusion and wasting litigant and judicial resources.”).
24

employer absolutely offered no evidence in rebuttal.71

   Accordingly, the ALJ committed legal errors in his findings that OFCCP

satisfied its burdens of establishing both a pattern or practice of intentional discrimination and a disparate impact in this matter, and I join in the majority’s order of remand.

71 See McKnight, supra note 59, § 9.03 (“Once the employer has not remained silent – i.e., once the employer has articulated at trial its legitimate nondiscriminatory reason – then whether the plaintiff proved a prima facie case is generally deemed to be irrelevant, and decisions on motions for judgment as a matter of law should dispense with analysis of the prima facie.”) (emphasis in original).