J. PAUL OETKEN, District Judge

OPINION AND ORDER

:

Plaintiffs Annette Richardson, Deborah Bowman, Debra Poe, Dino Riojas, Stephanie Thomas, Jon Watson, Brenda McKiver, and Erica Richardson, former and current employees of the Fire Department of New York ("FDNY"), bring this putative class action against Defendant City of New York, claiming that FDNY's hiring, promotion, and compensation practices for civilian employment violate 42 U.S.C. §§ 1981 and 1983 and the New York City Human Rights Law ("NYCHRL"), N.Y.C. Admin. Code § 8-101 et seq. Plaintiffs seek class-wide monetary and injunctive relief for FDNY's disparate treatment of, and policies having a disparate impact on, African Americans.

On May 26, 2020, the City filed motions for partial summary judgment and to strike Plaintiffs' disparate impact claims. That same day, Plaintiffs filed a motion to certify two classes under Federal Rule of Civil Procedure 23. For the reasons that follow, the City's motion for summary judgment is granted in part and denied in part, the City's motion to strike is denied, and Plaintiffs' motion for class certification is denied.

I. Background

The Court assumes familiarity with the allegations in the case, on the basis of the Court's prior opinions addressing the City's motion to dismiss and Plaintiffs' motion for leave to file an amended complaint. See Richardson v. City of New York, No. 17-cv-9447, 2018 WL 4682224 (S.D.N.Y. Sept. 28, 2018); Richardson v. City of New York, No. 17-cv-9447, 2019 WL 1512646 (S.D.N.Y. Apr. 8, 2019). The facts below are taken from the parties' submissions for the pending motions, and the Court resolves factual disputes as necessary to resolve the class certification issues raised. See Shahriar v. Smith & Wollensky Rest. Grp., Inc., 659 F.3d 234, 251 (2d Cir. 2011).

A. Hiring, Promotions, and Compensation for FDNY's Civilian Workforce

Plaintiffs bring their discriminatory hiring, promotion, and compensation claims with respect to FDNY's civilian workforce. The civilian workforce comprises roughly 1,750 employees distributed across six job groups: Management Specialists, Science Professionals, Health Professionals, Clerical Supervisors, Clerical, and Craft. (Dkt. No. 80-2 ("Scherbaum Rep.") at 15-16, 28.) These six job groups, in turn, comprise 92 job titles, seven of which are unique to FDNY. (Dkt. No. 80-7 at 100:14-25). Job titles range from Fire Alarm Dispatcher, the "largest title" within the Clerical group (Dkt. No. 80-3 ("Erath Rep.") at 5), to Computer Systems Manager, a title within the Science Professionals group (Scherbaum Rep. at 16).

FDNY recruits for its 92 job titles in two main ways. The first applies when the City's Department of Citywide Administrative Services ("DCAS"), a separate agency, administers an examination, uses the results to produce a civil service list of eligible candidates, and makes the list available for one of FDNY's openings. (Dkt. No. 86-1 ¶ 9.) In this circumstance, FDNY must interview and select its new hire from the three highest-scoring candidates on the list. (Dkt. No. 86-26 ¶ 9.) The second recruitment method applies when no applicable DCAS list exists; in this circumstance, FDNY proceeds with a general posting. (Dkt. No. 86-1 ¶ 21.) The responses to the general posting are reviewed by the team seeking the new hire, which chooses applicants to interview, schedules the interviews, and ultimately selects the new hire. (Dkt. No. 86-1 ¶ 22.) In the hiring period debated by the parties' experts, roughly one-third of FDNY's hiring was done through the DCAS process, and the remaining two-thirds were done through the "discretionary" process without a list. (Erath Rep. at 8.)

FDNY also promotes employees in two ways: through "in-title" promotions and through promotions to a different job title. (Dkt. No. 80-22 at 24:22-25.) In-title promotions are available only for job titles in which employees' duties vary. (Dkt. No. 80-22 at 36:12-25.) Supervisors may seek an in-title promotion on behalf of an employee when that employee is tasked with higher-level responsibilities that still fall within the scope of the job title. (Dkt. 80-22 at 26:3-11; 73:21-25.) If a job title does not have levels or has a low compensation ceiling, a supervisor may seek to promote an employee to a different job title. Promoting an employee can be "similar" to filling a vacancy through FDNY's standard hiring processes, in that the promotion position will sometimes be posted publicly or filled through a DCAS list. (Dkt. No. 80-22 at 25:4-9; Dkt. No. 80-7 at 137:13-25.) Other times, promotions are posted internally. (Dkt. No. 86-28.) Openings posted for the purpose of promoting a specific employee may be "tailor made" for the employee and have an application window of as few as three days. (Dkt. No. 80-10 at 164:1-14.) Promotions are associated with increased compensation, capped at an 8% increase or an increase to the compensation floor of the new title. (Dkt. No. 80-22 at 77:10-25; 78:8-21.)

Employees' compensation is tied to their job title and level. (Dkt. 80-22 at 24:7-8.) All titles have compensation floors and ceilings. (Dkt. 80-22 at 71:13-17; Dkt. 80-21 at 225:6-12.) For "many" titles, compensation is determined by the City's negotiations with the appropriate union, and there is "very little, if any, variation" in compensation within the title. (Dkt. 80-22 at 36:4-20; 37:9-19.) For other, eligible titles, FDNY used to allow supervisors to request discretionary increases to employees' salaries when those employees took on "additional responsibilities of a substantial nature or as a result of new mandates or programs." (Dkt. 80-22 at 33:13-19; 34:10-18; Dkt. No. 86-14 ¶ 5.) A discretionary increase had to fit within a supervisor's "allocation" for the fiscal year, and a limited number of increases could be awarded in any given month, so as to spread spending across the year. (Id.; Dkt. 80-22 at 35:16-21.) FDNY has now phased out discretionary increases for the vast majority of the civilian workforce. (Dkt. 80-22 at 30:7-25.)

B. Personnel Involved in Hiring, Promotion, and Compensation Decisions

A variety of personnel, ranging from individual supervisors to the FDNY Commissioner, are involved in FDNY's hiring, promotion, and compensation decisions. In general, supervisors are responsible for initiating and driving the processes leading to these decisions. Working with supervisors, FDNY's administrative offices and top leadership often mediate hiring, promotion, and compensation processes or must approve their resulting decisions.

With respect to hiring, Human Resources assists supervisors by reviewing their proposed interview questions and ensuring that the proposed selection criteria are job-related. (Dkt. No. 80-21 at 65:2- 8.) Additionally, a Human Resources representative sits on the interview panel for any candidate who qualifies through a DCAS civil service list and for non-DCAS candidates for "hard to recruit positions, like fire alarm dispatcher and fire protection inspector." (Dkt. No. 80-21 at 65:13-15; 66:19-23.) During the hiring period at issue, five members of Human Resources were responsible for sitting on interview panels. (Dkt. No. 80-21 at 66:2-13.) Supervisors took these Human Resources representatives' post-interview feedback seriously in deciding which interviewees to select for hire. (Dkt. No. 80-18 at 126:7-25.)

After a supervisor selects a candidate, Human Resources is again involved. First, the Candidate Investigation Division in Human Resources investigates the selected candidate to ensure that the candidate actually has the qualifications required for the position. (Dkt. No. 80-21 at 77:3-5.) If the Division determines, for instance, that the selected candidate does not meet the "medical requirements" of the position, the candidate cannot be hired. (Dkt. No. 80-21 at 77:24-25; 78:2-3.) Second, and assuming a candidate passes the Division's review, the supervisor must work with Human Resources to set the candidate's compensation rate, consistent with "the parameters of the [] title." (Dkt. No. 86-13 at 261:6-9; Dkt. No. 80-21 at 224:8-11.) With the supervisor's input, and mindful of the rates paid to new hires and paid to "incumbents" currently holding the position, Human Resources confers with FDNY's Budget and Finance Office and then proposes a compensation rate to the Commissioner for approval. (Dkt. No. 80-21 at 223:3-24; 224:3-5.) To the extent a supervisor disagrees with the proposed compensation rate, the supervisor may directly petition the Commissioner. (Dkt. No. 86-13 at 264:11-13.)

Before discretionary increases were phased out, Human Resources and the Budget and Finance Office coordinated that compensation process. At the start of each fiscal year, the Budget and Finance Office would determine how much money was available for discretionary increases. (Dkt. No. 80-22 at 34:10-18.) Human Resources would then "poll" supervisors, asking them to request discretionary increases for eligible employees and to justify their requests. (Dkt. No. 86-11 at 72:7-19.) The Budget and Finance Office, Human Resources, and the Commissioner had to approve any discretionary increase. (Dkt. No. 80-22 at 35:2-6.) As a matter of course, however, the Budget and Finance Office played a limited role in reviewing requests and did not "interfere" with supervisors' requests. (Dkt. No. 86-14 ¶ 7.) While some Commissioners took a "hands-on" approach to reviewing compensation requests, others were less interested in doing so. (Dkt. No. 93-11 at 88:8-14.)

C. Unequal Results for African-American Employees

Central to this case, FDNY's hiring, promotion, and compensation decisions have not produced equal outcomes across racial groups. Most notably, FDNY's hiring processes, historically and contemporarily, have resulted in or maintained a significant "underutilization" of African Americans in the civilian workforce. For most job groups, the proportion of African Americans in the FDNY civilian workforce is lower than the proportion of African Americans that DCAS estimates are available for employment and is similarly lower than the proportion of African Americans employed by other City agencies. (Scherbaum Rep. at 12.) By contrast, whites are "generally overutilized," or overrepresented, in the job groups. (Scherbaum Rep. at 100.) To illustrate, African Americans constituted 9% of FDNY's Health Professionals in 2017, and whites constituted 72%. (Scherbaum Rep. at 67.) That same year, the City's Health Professionals, across all agencies, were 33% African American and 28% white. (Id.)

Furthermore, the African Americans in FDNY's civilian workforce tend to hold lower-level jobs and lower-compensated job titles. (Dkt. No. 80-4 ("Scherbaum Rebuttal") at 30.) In 2017, the average salary for African Americans was $63,789.34, and the average salary for whites was $81,763.32. (Scherbaum Rep. at 116.) Between 2005 and 2018, there is just one year in which African Americans did not receive the lowest average salary of any racial group employed in FDNY's civilian workforce. (Id.)

D. Proposed Classes and Representatives

Plaintiffs propose the following two classes to challenge FDNY's hiring, promotion, and compensation practices. The first is the "Rejected Applicant Class," which Plaintiffs define as

All African Americans who passed any applicable Department of Citywide Administrative Service ("DCAS") tests, possessed all other posted requirements for any posted Fire Department of New York ("FDNY") civilian vacancy, and applied and were rejected by FDNY for any such position at any time between December 1, 2014 and the date a class is certified unless they applied for a position (a) in a job title classified by the City of New York (the "City") as an administrator or manager, or (b) would have [received] a salary of at least $150,000 per year during at least one year during the period.

All African Americans who have been employed in a civilian full-time position in FDNY at any time between December 1, 2014 and the date a class is certified, unless throughout this period they (a) were in a job title classified by the City as an administrator or manager, or (b) had a salary of at least $150,000 per year during at least one year during the period.

Plaintiffs propose that Annette Richardson, Deborah Bowman, Debra Poe, Dino Riojas, Stephanie Thomas serve as representatives of the Employee Class and Compensation Subclass. (Dkt. No. 55 ¶¶ 8-12.) They propose that Brenda McKiver and Joe Watson serve as additional representatives of the Employee Class. (Dkt. No. 55 ¶¶ 13-14.) Finally, they offer Erica Richardson as the sole representative of the Rejected Applicant Class. (Dkt. No. 55 ¶ 15.)

E. Procedural History

On February 5, 2018, the City filed a motion to dismiss Plaintiffs' original complaint. (Dkt. No. 11.) Based on Plaintiffs' allegations regarding African Americans' underutilization in FDNY's civilian workforce, in combination with Plaintiffs' allegations about their experiences with FDNY's hiring and promotion processes, the Court denied the City's motion with respect to Plaintiffs' discriminatory hiring and promotion claims. (Dkt. No. 26 at 14-18.) The Court, however, dismissed Plaintiffs' discriminatory compensation claims based on the original pleadings. (Dkt. No. 18-20.) On November 13, 2018, Plaintiffs sought leave to file an amended complaint, in which they attempted to replead their compensation claims. (Dkt. No. 33.) The Court granted leave to file in light of Plaintiffs' added allegations regarding African Americans' relatively low compensation rates in four job titles. (Dkt. No. 49 at 5.) The City then answered Plaintiffs' amended complaint. (Dkt. No. 58.)

Since November 27, 2018, the parties have engaged in discovery for the purpose of filing and contesting the present motion for class certification. (Dkt. No. 37 at 3; Dkt. No. 60 at 3 ("In this case provisions concerning summary judgment motions . . . will be more appropriate for the case management order to be entered after a class certification ruling.").) On May 26, 2020, the day on which Plaintiffs' motion for class certification was due, the City filed two unexpected motions, one for partial summary judgment and another to strike the disparate impact claims in Plaintiffs' amended complaint. Plaintiffs filed their motion for class certification. The Court now addresses these motions.

II. Discussion

In its self-styled motion for summary judgment and motion to strike, the City argues that (1) Annette Richardson, Debra Poe, and Stephanie Thomas may not seek injunctive or declaratory relief because their claims were mooted when they retired from FDNY; (2) Erica Richardson and Dino Riojas lack class standing; and (3) Plaintiffs' disparate impact claims, as pleaded in the amended complaint, are legally deficient. In their motion for class certification, Plaintiffs argue, as they must, that they satisfy the demands of Rule 23(a), including its requirement that the proposed classes share common questions of law and fact. These arguments are addressed in turn.

A. Motion for Summary Judgment

"To qualify as a case fit for federal-court adjudication, an actual controversy must be extant at all stages of review, not merely at the time the complaint is filed." Arizonans for Official English v. Arizona, 520 U.S. 43, 67 (1997) (internal quotation marks and citation omitted). A case is moot, and no longer regards a live controversy, when a plaintiff's situation changes such that the complained-of injury cannot be "redressed by a favorable judicial decision." Spokeo, Inc. v. Robins, 136 S. Ct. 1540, 1547 (2016); United States Parole Comm'n v. Geraghty, 445 U.S. 388, 397 (1980). The mootness of a case deprives the Court of its subject matter jurisdiction, and "[i]t is axiomatic that a lack of subject matter jurisdiction may be raised at any time" — typically through a motion to dismiss pursuant to Federal Rule of Civil Procedure 12(b)(1) or 12(h)(3). Wight v. BankAmerica Corp., 219 F.3d 79, 90 (2d Cir. 2000); U.S. ex rel. Smith v. New York Presbyterian Hosp., No. 06-cv-4056, 2007 WL 2142312, at *4 & n.29 (S.D.N.Y. July 18, 2007).

In their motion for summary judgment, the City presents evidence indicating that Annette Richardson, Debra Poe, and Stephanie Thomas retired during the pendency of this case. (Dkt. No. 77 ¶¶ 2-4, 19-20; Dkt. No. 98.) The City argues that these retirements render moot the claims of Annette Richardson, Debra Poe, and Stephanie Thomas for equitable relief and preclude them from serving as representatives of the Employee Class and Compensation Subclass. The Court agrees. See Arizonans for Official English, 520 U.S. at 72 (holding that a plaintiff's "resignation from public sector employment" necessitated the dismissal of her case because "it became plain that she lacked a still vital claim for prospective relief" against her public sector employer). Were the Court to order FDNY to change its hiring, promotion, or compensation practices, the ensuing changes would have no cognizable effect on FDNY's former employees. This form of relief would not redress any of the alleged injuries of Annette Richardson, Debra Poe, and Stephanie Thomas. That Plaintiffs seek to bring this case on behalf of a class has no bearing on the mootness analysis here, where the claims at issue were mooted before any class was certified. See Sosna v. Iowa, 419 U.S. 393, 399 (1975) (holding that a class "acquire[s] a legal status separate from the interest asserted by [the plaintiff]" at the time of certification). Annette Richardson's, Debra Poe's, and Stephanie Thomas's claims are dismissed.

Less persuasively, the City argues that Erica Richardson and Dino Riojas lack class standing. Class standing, unlike its Article III cousin, does not implicate subject matter jurisdiction. See NECA-IBEW Health & Welfare Fund v. Goldman Sachs & Co., 693 F.3d 145, 159 (2d Cir. 2012); Policeman's Annuity & Ben. Fund of the City of Chicago v. Bank of America, NA, No. 12-cv-2865, 2013 WL 5328181, at *4 (S.D.N.Y. Sept. 23, 2013). Although the notion of class standing is "derive[d] from constitutional standing principles," Retirement Bd. Of the Policeman's Annuity & Ben. Fund of the City of Chicago v. Bank of New York Mellon, 775 F.3d 154, 161 (2d Cir. 2014), the presence of class standing "is assessed based on allegations rather than evidence," In re LIBOR-Based Fin. Instruments Antitrust Litig., 299 F. Supp. 3d 430, 459 (S.D.N.Y. 2018). Under the test established in NECA-IBEW Health & Welfare Fund v. Goldman Sachs & Co., a plaintiff has class standing if:

he plausibly alleges (1) that he personally has suffered some actual injury as a result of the putatively illegal conduct of the defendant, and (2) that such conduct implications the same set of concerns as the conduct alleged to have caused injury to other members of the putative class by the same defendants.

In their amended complaint, Plaintiffs propose that Erica Richardson could represent the Rejected Applicant Class. In support of this proposal, Plaintiffs allege that "Erica Richardson . . . applied and was rejected for one or more Civilian positions at FDNY for which she was qualified" and that her rejection was racially motivated. (Dkt. No. 55 ¶ 221.) With respect to Dino Riojas, whom Plaintiffs propose to represent the Employee Class and Compensation Subclass, Plaintiffs allege that he held a job title in the Science Professional group, that he had "not received a discretionary increase in his 35 years as a Computer Specialist," and that his manifest talents, qualifications, and achievements went unappreciated at FDNY. (Dkt. No. 55 ¶¶ 157-60.) Plaintiffs chalk up FDNY's failure to promote or increase the compensation of Dino Riojas to racial animus. (Dkt. No. 55 ¶¶ 199, 205.) There is no doubt that these pleadings describe injuries felt and conduct faced by the members of Plaintiffs' envisioned classes. Erica Richardson and Dino Riojas have class standing.

To the extent the City challenges the factual underpinnings of Plaintiffs' allegations or otherwise attempts to use evidence from discovery to distinguish Erica Richardson and Dino Riojas from Plaintiffs' putative class members, such issues "are best resolved on a motion for class certification," not on a separate motion. Fernandez v. UBS AG, 222 F. Supp. 3d 358, 373 (S.D.N.Y. 2016); In re LIBOR-Based Fin. Instruments Antitrust Litig., 299 F. Supp. 3d at 537 n. 103.

B. Motion to Strike

The City returns to the amended complaint in moving to strike Plaintiffs' disparate impact claims. After a year and a half of wide-ranging discovery, and more than two years after its motion to dismiss, the City argues, for the first time, that Plaintiffs' disparate impact claims are insufficiently pleaded or are foreclosed by Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011). The City contends that, contrary to the demands of our case law, the amended complaint fails to allege that "FDNY relies upon a 'common mode of exercising discretion'" over its hiring, promotion, and compensation processes. (Dkt. No. 76 at 16.) The City faults the amended complaint's inability to "pinpoint facially neutral policies that are applicable to the putative class." (Dkt. No. 76 at 15.)

Irrespective of the merits of these arguments, the Court declines to consider them in the context of this motion to strike. "[T]he sufficiency of class allegations should be determined in the context of a motion for class certification and [] this process may not be circumvented by utilizing a motion to strike." Rahman v. Smith & Wollensky Restaurant Grp., Inc., No. , 2008 WL 161230, at *3 (S.D.N.Y. Jan. 16, 2008). The analysis the City presents in its motion to strike overlaps substantially with the commonality analysis the Court must perform to resolve Plaintiffs' motion for class certification; the sufficiency of Plaintiffs' disparate impact theory under Dukes cannot be construed as "separate and apart from the issues that will be decided on [the] class certification motion." Id. Had the City wanted to challenge the sufficiency of the pleadings with respect to Dukes, it should have done so in its motion to dismiss or in opposing Plaintiffs' motion for leave to file an amended complaint. The motion to strike is denied.

C. Motion for Class Certification

Plaintiffs' proposed classes and subclass bring disparate treatment and disparate impact claims under §§ 1981 and 1983, as well as the NYCHRL. In debating the appropriateness of class certification, the parties chiefly dispute whether these claims present questions of law and fact that are common to the classes and subclass. This is time well spent because, "[a]s is true in many [employment discrimination] class actions, the crux of this case is commonality." Kassman v. KPMG LLP, 416 F. Supp. 3d 252, 274 (S.D.N.Y. 2018) (internal quotation marks and citation omitted). And, here, the Court concludes that Plaintiffs have not carried their burden of showing commonality, as is required by Federal Rule of Civil Procedure 23.

1. Legal Standard

Class certification is governed by Rule 23. Section (a) of Rule 23 requires the party seeking certification to establish four prerequisites:

(1) the class is so numerous that joinder of all members is impracticable;
(2) there are questions of law or fact common to the class;
(3) the claims or defenses of the representative parties are typical of the claims or defenses of the class; and
(4) the representative parties will fairly and adequately protect the interests of the class.

The Rule 23 requirements are more than a "mere pleading standard." Dukes, 564 U.S. at 350. The party seeking class certification must establish Rule 23's requirements by a "preponderance of the evidence." Teamsters Local 445 Freight Div. Pension Fund v. Bombardier Inc., 546 F.3d 196, 202 (2d Cir. 2008). Courts must "conduct a rigorous analysis to determine whether a class action is appropriate, considering materials outside of the pleadings and weighing conflicting evidence as necessary." Jackson, 298 F.R.D. at 159.

2. Commonality

Pursuant to Rule 23(a)(2), a class can be certified only if "there are questions of law or fact common to the class" that give "cause to believe that [the class members'] claims can productively be litigated at once." Dukes, 564 U.S. at 349-50 (internal quotation marks and citations omitted). This requires more than a showing that the proposed class members "have all suffered a violation of the same provision of law." Id. at 350. The claims instead "must depend on a common contention," the "truth or falsity [of which] will resolve an issue that is central to the validity of each one of the claims in one stroke." Id. Where, for instance, "the same conduct or practice by the same defendant gives rise to the same kind of claims from all class members, there is a common question" within the meaning of Rule 23(a)(2). Johnson v. Nextel Commc'ns Inc., 780 F.3d 128, 137 (2d Cir. 2015) (citation omitted).

In the context of an employment discrimination class action, commonality demands that there be "some glue holding the alleged reasons for all [the challenged employment] decisions together . . . [that] will produce a common answer to the crucial question why was I disfavored." Dukes, 564 U.S. at 352 (emphasis in original). That glue may be supplied by a "companywide evaluation method that can be charged with bias" or "significant proof that [the employer] operated under a general policy of discrimination." Id. at 353 (internal quotation marks omitted). Typically, an employer's policy of "allowing discretion by local supervisors over employment matters" will not present questions and answers common to a large class. Id. at 355 (emphasis omitted). Although some supervisors, left unchecked, will "be guilty of intentional discrimination that produces a [class]-based disparity," others will make employment decisions by reference to "[]neutral, performance-based criteria . . . that produce no actionable disparity." Id. Demonstrating that prejudice animates one supervisor's unfettered decision-making "will do nothing to demonstrate the invalidity of another's." Id. at 355-56.

It follows that, in assessing Plaintiffs' disparate impact and disparate treatment claims, the Court considers whether Plaintiffs have shown that African Americans at FDNY are subjected to "a common mode" or "general policy" of discrimination or whether, instead, Plaintiffs have done no more than show that FDNY has a "policy against having uniform employment practices." Dukes, 564 U.S. at 355-56 (emphasis in original).

a. Disparate Impact Claims

Plaintiffs bring their disparate impact claims under the NYCHRL, which requires Plaintiffs to establish an "unlawful discriminatory practice," N.Y.C. Admin. Code § 8-107(17), or a facially neutral policy that "disproportionately affected [members of their class] in comparison to [others]," Bennett v. TimeWarner Cable, Inc., 138 A.D.3d 598, 598 (1st Dep't 2016). Plaintiffs "need not prove discriminatory intent to establish [their] case of disparate impact," but they must show that the disproportionate effects of their challenged practice "cannot be justified by any explanation other than . . . discrimination." Mete v. New York State Off. Of Mental Retardation & Dev. Disabilities, 21 A.D.3d 288, 297 (1st Dep't 2005). Of course, in the absence of a "specific employment practice" pertinent across a putative class, merely pointing to "an overall [class]-based disparity does not suffice." Dukes, 564 U.S. at 357.

Plaintiffs identify three aspects of FDNY's hiring, promotion, and compensation processes that they contend rise to the level of an "unlawful discriminatory practice" and affect all FDNY employees and job-seekers. First, they point to the involvement of Human Resources, the Budget and Finance Office, and the Commissioner in FDNY's hiring, promotion, and compensation processes, arguing that these offices' "centralized decision-making" is biased and common to all African-American applicants and employees. (Dkt. No. 80 at 32.) Second, they argue that FDNY imposes a "framework for discretion" on all supervisors, insofar as supervisors can be required to hire through the DCAS process and there are rules in place that govern when and by how much an employee's salary may be increased. (Dkt. No. 80 at 33-34.) Third, they problematize FDNY's failure to involve its Equal Employment Opportunity ("EEO") Office in Human Resources decisions. (Id.) But the evidence in the record undermines Plaintiffs' suggestion that these supposed practices were uniformly applied or had common effects on the putative class members.

Starting with the suggestion that FDNY employed centralized decision-making, Plaintiffs are correct that "management's involvement" across employment processes can support a disparate impact class action. Ellis v. Costco Wholesale Corp., 285 F.R.D. 492, 511-14 (N.D. Cal. 2012). If a "close-knit, centralized management team" "oversees and directs" employment decision-making for an organization, id., or a small group of individuals is itself responsible for selecting candidates for employment, promotion, or a raise, see, e.g., In re Johnson, 760 F.3d 66, 69-70 (D.C. Cir. 2014), that involvement will define the employment outcomes of all class members. To the extent those individuals' decision-making is infected with bias, all class members are liable to have suffered unequal treatment. The same cannot be said when management's involvement is limited to approving "the recommendations of lower level direct supervisors and managers," and there is "no evidence showing that any member of [] management changed . . . recommendations . . . with any frequency." Jones v. Nat'l Council of Young Men's Christian Ass'ns of the United States of America, 34 F. Supp. 3d 896, 908 (N.D. Ill. 2014). A toothless "final approval" process affords ample "freedom and independence to local supervisors" and dictates the conclusion that these varied supervisors, rather than a centralized management team, are the causal force behind any unequal outcomes. Kassman, 416 F. Supp. 3d at 280-81.

Plaintiffs have failed to show that the involvement of Human Resources, the Budget and Finance Office, and the Commissioner is sufficiently "consistent," "pervasive," or "classwide" to produce common questions and answers across the putative classes. Ellis, 285 F.R.D. at 511-12. The roles of these offices, and thus their ability to produce a disparate impact, vary substantially with respect to FDNY's employment processes. Some but not all members of the Rejected Applicant Class would have interviewed with one of five members of Human Resources. Some but not all members of the Employee Class would have applied for a promotion through a DCAS list and had a similar interview experience. Some but not all members of the Compensation Subclass would have had a job title in which the level of compensation could be influenced by Human Resources, the Budget and Finance Office, or the Commissioner, instead of union negotiations. Furthermore, despite the Candidate Investigation Division's authority to reject job-seekers whose credentials do not meet the requirements of a job title, there is no indication that the Division did or does so with any frequency. Similarly, the record suggests that the Budget and Finance Office and the current Commissioner, despite their authority to independently assess and reject compensation recommendations, "rely on [others'] judgment" in approving compensation increases. (Dkt. No. 86-14 ¶ 7; Dkt. No. 9-11 at 89:4-10.) Altogether, even if Human Resources, the Budget and Finance Office, or the Commissioner discriminated against certain FDNY employees or applicants, Plaintiffs have not shown that all of their proposed class members would have been subjected to the same mode of discrimination.

Plaintiffs' argument that FDNY imposed a framework for discretion on supervisors is no more persuasive. Plaintiffs liken FDNY's constraints on hiring, promotions, and compensation to those at issue in Chen-Oster v. Goldman, Sachs & Co., where the court held that Rule 23(a)'s commonality requirement was satisfied. 325 F.R.D. 55, 73-75 (S.D.N.Y. 2018). The analogy does not hold. In Chen-Oster, the employer required managers to use a "360 review process" that assigned numerical values to employees' various skills, as well as a "quartiling" process that required managers to rank employees based on seven predetermined factors. Id. at 64-66. The 360 review and quartiling processes determined which employees could be promoted. Id. Although these processes did "not strip managers of all flexibility in compensation and promotion decisions," they did force managers to focus on certain qualities, such as an employee's communication skills and leadership abilities, and they constrained managers' discretion in balancing employees' virtues and vices. Id.; see also In re Johnson, 760 F.3d at 73 (holding that commonality is satisfied when "every class member was evaluated upon the same criteria and scored using the same numerical system"). Plaintiffs identify no comparable constraints on FDNY supervisors' authority to select candidates from a DCAS list or to determine which employees deserve a promotion or raise. Instead, the record shows that supervisors are free to choose among DCAS candidates and that supervisors, before the discontinuation of the discretionary-increase program, were free to offer a discretionary increase to any eligible employee who had taken on additional responsibilities or mandates. Likewise, that pay increases upon promotion are typically capped at 8% has no bearing on a supervisor's ability to select an employee for promotion or to recommend whether the employee deserves, say, a 2% or an 8% increase. See Kassman, 416 F. Supp. 3d at 277-78 (holding that a relevant framework for discretion is one that "dictates . . . how [supervisors] will exercise their discretion," not one that "sets salary ranges and recommended pay increases and bonuses").

Finally, Plaintiff's argument that FDNY "walled off" its EEO Office is foreclosed by Dukes. Plaintiffs fault the EEO Office for not "review[ing] individual civilian selection decisions" and failing to impose a "system for evaluating and rewarding managers based on compliance with EEO policies." (Dkt. No. 80 at 17-18.) In essence, Plaintiffs complain that supervisors have been allowed too much discretion to make employment decisions. But too much discretion, as Dukes explains, is "the opposite of a uniform employment practice that would provide the commonality needed for a class action." 564 U.S. at 355.

In addition to the shortcomings of each of Plaintiffs' proposed practices, Plaintiffs' own statistical analysis undermines the idea that a uniformly applied, common practice is driving the unequal outcomes at issue. If Human Resources or the 8% cap on promotion-related raises were the root cause of the inequality, their impact would presumably be felt and ascertainable across FDNY's six job groups. Yet Plaintiffs' statistics expert found that African Americans were not underutilized as Science Professionals between 2005 and 2019, and between 2016 and 2019, white Science Professionals were underutilized. (Scherbaum Rep. at 41-42.) Plaintiffs' statistics expert also found that any race-based disparity in the compensation of Clerical Supervisors "was statistically insignificant in each year." (Scherbaum Rep. at 123.) It seems unlikely that Human Resources would exempt Science Professionals from its predilection to favor whites and discriminate against African Americans. Similarly, it stretches credulity to suggest that the Clerical Supervisors job group, though beholden to the 8% cap, is uniquely situated to escape its discriminatory effects. A simpler explanation for these statistical findings is that different supervisors populate the Science Professionals and Clerical Supervisors job groups, and these different supervisors are less inclined to discriminate against African Americans in their hiring or compensation practices.

Plaintiffs have not identified any practice that would have disparately impacted all members of their proposed classes. The proposed classes "cover[] a myriad of job [titles]," subject to the oversight of a myriad of supervisors. Kassman, 416 F. Supp. 3d at 277. The record shows that these supervisors exercise substantial control over FDNY's employment decisions. Plaintiffs' motion for class certification is denied with respect to their disparate impact claims.

b. Disparate Treatment Claims

Unlike Plaintiffs' disparate impact claims, their disparate treatment claims "do not require [them] to identify a specific company-wide employment practice responsible for the discrimination." Id. at 281. Instead, they can show that FDNY engaged in "widespread acts of intentional discrimination against individuals," or that "intentional discrimination was [FDNY's] standard operating procedure." Reynolds v. Barrett, 685 F.3d 193, 203 (2d Cir. 2012) (citation and internal quotation marks omitted). "[S]tatistics alone may be sufficient" to establish a pattern or practice of discrimination, though "the statistics must not only be statistically significant in the mathematical sense." Burgis v. New York City Dep't of Sanitation, 798 F.3d 63, 69 (2d Cir. 2015). They "must also be of a level that makes other plausible non-discriminatory explanations very unlikely." Id.

As courts have recognized, meeting the burden to show that an employer abided by a pattern or practice of discrimination "overlaps" with satisfying Rule 23(a) commonality. Kassman, 416 F. Supp. 3d at 281. Although the commonality requirement can be satisfied "via statistical and anecdotal evidence," the evidence must offer "significant proof" of a general policy of discrimination. Chen-Oster, 325 F.R.D. at 76. In other words, the evidence should expose "the kind of gross disparities that on their face would suggest discriminatory intent." Kassman, 416 F. Supp. 3d at 282; Dukes v. Wal-Mart Stores, Inc. ("Dukes II"), 964 F. Supp. 2d 1115, 1121 (N.D. Cal. 2013) ("In a rare and extreme case, the statistics might speak for themselves."). This could be satisfied, for instance, if a class-based difference is "statistically significant by several standard deviations," and a substantial portion of the difference is traceable to the employer's choices. Chen-Oster, 325 F.R.D. at 76 (finding that 72% of the statistically significant pay gap could be explained by the scores given to men and women in 360 reviews and the quartiling process). It would not be satisfied by an equivocal statistical analysis or an analysis that "obfuscate[s] the principal explanatory variable" to create the appearance of difference. Kassman, 416 F. Supp. 3d at 283; Dukes II, 964 F. Supp. 2d at 1120-21 (characterizing statistical evidence as "underwhelming" when there were no "statistically significant disparities in even a majority of the relevant decision units"). Here, Plaintiffs' statistical evidence cannot buoy the disparate treatment claims.

With respect to hiring, Plaintiffs' statistics expert, Dr. Charles Scherbaum, focused on whether African Americans are underutilized relative to their presence in the labor market and in other City agencies. Scherbaum identified that the underutilization with respect to the overall labor market was statistically significant "in at least 14 of the 17 quarters reviewed for the management specialist, health professional, clerical supervisor, clerical, and craft job categories." (Scherbaum Rep. at 40.) He determined that the underutilization with respect to other City agencies was statistically significant "for each year" between 2005 and 2019 "in five of the six job categories." (Scherbaum Rep. at 69.) Scherbaum estimated that, between 2005 and 2018, FDNY had hired 184 fewer African Americans than would have been hired through an even-handed hiring process. (Scherbaum Rep. at 74.)

At first glance, Scherbaum's findings are troubling and imply persistent disparity across FDNY's job categories. Upon further consideration, the implications are less clear. As Defendants' statistics expert, Dr. Charles Erath, explains, assessing underutilization by job group provides an incomplete picture of hiring at FDNY. To start, 47% of FDNY's civilian employees between 2005 and 2019 were hired in 2004 or earlier. (Erath Rep. at 3.) That nearly half of FDNY's civilian employees were hired outside the period of inquiry renders any analysis predicated on the composition of the workforce, such as Scherbaum's underutilization analysis, of limited probative value in determining whether African Americans were under-hired between 2005 and 2019. Furthermore, the composition of the labor market may differ from that of the pool of candidates available to FDNY supervisors hiring through the DCAS process. Cross-referencing the DCAS civil service lists and FDNY's hiring records from 2004 to 2019, Erath found that FDNY had hired 29.53% of African-American applicants from the lists and 29.46% of white applicants. (Erath Rep. at 10.) Analyzing the same lists and records in response, Scherbaum was unable to find a statistically significant difference in FDNY's hiring rates without recoding certain instances in which an applicant was not hired as a "good" outcome for the applicant and removing a large swath of supposedly "ambiguous" data. (Dkt. No. 80-5 at 4, 8-9; Dkt. No. 80-7 at 122:3-19.) Even with these alterations, Scherbaum acknowledged that the "shortfall of African American[s] is either 22 or 15," or roughly one candidate per year. (Dkt. No. 80-5 at 9.)

The most important limitation of Scherbaum's underutilization analysis, however, is that it does not "conform[] to the level of decision for the challenged practices." Ellis, 285 F.R.D. at 523. In the wake of Dukes, courts have been skeptical of "aggregated statistical evidence . . . 'derived from hundreds of employment decisions made by myriad decision makers, at different times, under mutable procedures and guidelines, in different departments, . . . [and] concerning employees at varying levels of experience, responsibilities, and education.'" Kassman, 416 F. Supp. 3d at 282 (citing Jones, 34 F. Supp. 3d at 909). In assessing underutilization by job group, Scherbaum ignored the twin realities that FDNY hires based on job title and that FDNY's needs within a job group may differ from the needs of other City agencies. To illustrate this point, Erath offered that "the position of Fire Alarm Dispatcher does not exist in any agency apart from FDNY but is by far the largest title within the Clerical group. . . . If African-American availability differs for this job relative to other Clerical jobs, [the labor market availability] figure will not apply." (Erath Rep. at 5-6.) Indeed, African Americans represent just 27.5% of identifiable applicants for the Fire Alarm Dispatcher title, despite the 56% availability figure for African Americans in the overall Clerical job group. (Id.) In the same vein, Erath showed that FDNY's largest job titles in the Health Professionals and Craft groups — Case Management Nurse and Auto Mechanic, respectively — have substantially lower availability figures for African Americans than have the aggregated Health Professionals and Craft groups. (Id.) Erath's concerns about the level of analysis are bolstered by the analysis of applicant flow data that Scherbaum performed for his initial report. In that analysis, Scherbaum found that African Americans were under-hired, to a statistically significant degree, in only seven of the 32 titles that had attracted at least 15 African-American and white job-seekers. (Scherbaum Rep. at 102.) Such a disparity may suggest that African-American applicants to seven job titles faced invidious discrimination, but it does not suggest the kind of pattern or practice of discrimination that would create common questions and answers across the far-reaching Rejected Applicant Class. See Dukes II, 964 F. Supp. 2d at 1121 ("[I]f Plaintiffs had been more circumspect in the scope of their proposed class, their statistics may have had a bigger impact. But under the class actually proposed, . . . the statistics [] do not reflect 'significant proof' of a 'general policy of discrimination' in each [division] across the challenged decisions.").

Plaintiffs' statistical evidence with respect to promotions is similarly flawed. In this context, too, Scherbaum performed an underutilization analysis with respect to the six job groups. (Scherbaum Rep. at 105.) Although an underutilization analysis would, in principle, be of assistance if it compared the demographics of FDNY's civilian workforce against the demographics of its promoted employees, that is not the comparison that Scherbaum performed. Instead, Scherbaum determined that using "the current composition of [FDNY] would not provide an unbiased estimate of the promotion pool" "given the findings of the disparities in the hiring of African Americans," and he compared the promoted employees against the labor market availability of African Americans in each job group. (Id.) But this approach muddles the effects of FDNY's hiring decisions and its promotion decisions, rather than isolating any evidence of discriminatory promotion practices. It also relies on the unfounded assumption that job group availability figures are a meaningful metric for assessing FDNY's employment decisions. Erath, for his part, reviewed the job title with "the largest number of promotions," Fire Protection Inspector. (Erath Rep. at 15.) He found that 33.6% of all Fire Protection Inspectors were African American and that 39.5% of Fire Protection Inspectors promoted to the job title Associate Fire Protection Inspector were African American. (Id.) It is improbable that this analysis of a single job title would sufficiently rebut robust statistical evidence showing an organization-wide disparity in promotions, but Plaintiffs offer no such evidence. Plaintiffs therefore fall short of showing that common questions and answers would arise with respect to the Employee Class.

The Compensation Subclass, though narrower than the Employee Class, fares no better. Once again, Scherbaum's approach presumes that the job group is the appropriate level of analysis. (Scherbaum Rep. at 121.) The error of this presumption is evident from Scherbaum's analysis of the Craft job group: Scherbaum's model identifies a statistically significant pay gap between white and African-American Craft employees, even though employees in this job group "work[] under collective bargaining agreements that [do] not differentiate in pay between persons with the same job title[]." (Scherbaum Rep. at 118, 128.) Scherbaum's approach renders it "impossible to tell whether his results are due to differing racial distributions by job," as must be the case for the Craft job group, or pervasive discrimination. (Erath Rep. at 17.) Tellingly, when Erath performed a competing compensation analysis that included "indicators for job title/level," he found no statistically significant disparity in the Management Specialist or Science Professional job groups that make up the Compensation Subclass. (Erath Rep. at 18.) And Scherbaum, in his rebuttal to Erath's report, conceded that "[m]aking job titles an independent variable leads race to no longer be[] statistically significant." (Scherbaum Rebuttal at 36.) This dooms Plaintiffs' subclass. When the challenged "disparities largely disappear[]" after taking job title into consideration, the Court can conclude that the "decision to aggregate data across function . . . obfuscated the principal explanatory variable," job title, and that the "statistical evidence is insufficient to show any common issue that would warrant [the proposed] class." Kassman, 416 F. Supp. 3d at 283.

It bears mention that Plaintiffs' anecdotal evidence cannot salvage their claims. "[M]ore specific testimony about discrimination" will rarely "in itself support an [organization]-wide remedy," and "[a]necdotal evidence is most useful as a supplement to strong statistical evidence." O'Donnell Constr. Co. v. Dist. of Columbia, 963 F.2d 420, 427 (D.C. Cir. 1992); accord Kassman, 416 F. Supp. 3d at 284. Moreover, the anecdotal evidence Plaintiffs offer is generally thin or purely speculative with respect to whether race animated a challenged employment decision. In several of the accounts challenging hiring or promotion decisions, the complainants inferred discrimination from nothing more than the fact that they were not selected. They had no knowledge of the credentials, and often the race, of the individual actually selected for the job or promotion, and they were unable to speak to whether the supervisor had a practice of favoring, or disfavoring, African Americans. (Dkt. No. 80-8 at 58:7-59:10; Dkt. No. 80-11 at 59:9-71:18; Dkt. No. 80-14 at 81:13-15.) For several of the accounts challenging compensation decisions, the complainants had a disciplinary history at FDNY or compared themselves to higher-compensated employees with greater experience and responsibilities. (Dkt. No. 86-14 ¶¶ 11-19.) Furthermore, the charges of discrimination against Donay Queenan, FDNY's former Head of Human Resources, are driven in substantial part by the complainants' belief that "she is ashamed of or otherwise wants to hide her African American lineage" because she, a biracial woman, "identifies as bi-racial in the City's databases." (Dkt. No. 80-29 ¶ 29; Dkt. No. 80-11 at 145:19-25 (expressing the belief that Queenan discriminated against African Americans because "[Queenan] thinks she's white.").) To be sure, a handful of the anecdotes, such as Dino Riojas's account of his compensation vis-à-vis that of more junior employees in the same job title, at the same level (Dkt. No. 80-13 at 75:4-77:17), and Stephanie Thomas's recollection of a supervisor's comment that "Asians were more hard-working than black people, smarter than black people" (Dkt. No. 80-14 at 57:22-25), show or suggest discrimination. These anecdotes may support individual claims, but see Thomas v. City of New York, 953 F. Supp. 2d 444 (E.D.N.Y. 2013), but they cannot carry a class action in the absence of strong statistical evidence.

Having reviewed the parties' competing statistical analyses and the pertinent anecdotal evidence, the Court concludes that Plaintiffs have not offered significant proof of a general policy of discrimination that would have affected the whole of the Rejected Applicant Class, Employee Class, or Compensation Subclass. Accordingly, Plaintiffs have not satisfied Rule 23(a) commonality, and their proposed classes cannot be certified.

III. Conclusion

For the foregoing reasons, the City's motion for summary judgment is GRANTED in part and DENIED in part, and the City's motion to strike is DENIED. Plaintiffs' motion for class certification is also DENIED.

The Clerk of Court is directed to close the motions at Docket Number 75 and 79.

SO ORDERED. Dated: May 12, 2021

New York, New York

/s/_________

J. PAUL OETKEN

United States District Judge