Taylor v. District of Columbia Water & Sewer Authority

MEMORANDUM OPINION AND ORDER

KENNEDY, District Judge.

Charles Taylor, a Black employee of the District of Columbia Water and Sewer Authority (“WASA”), alleges that WASA has unlawfully discriminated against him and other Black employees at WASA in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C.2000e et seq. (“Title VII”), and Section 1981 of the Civil Rights Act of 1871, as amended by the Civil Rights Act of 1991, 42 U.S.C. § 1981 (“Section 1981”). Alleging that WASA has subjected its Black employees, including its new hires, from October 1996 through December 2000, to a pattern and practice of discrimination with respect to compensation and promotions, Taylor seeks certification of a class of more than 800 members, comprised of all Black employees who were not appropriately compensated or advanced. On behalf of a proposed class of 800 members, Taylor seeks compensatory damages, declaratory, injunctive and other equitable relief, as well as litigation expenses and attorneys’ fees. Upon consideration of Taylor’s amended motion for class certification [# 107], the opposition thereto, and the record of the case, the court concludes that the motion must be granted in part and denied in part.

I. BACKGROUND

In 1996, WASA, formerly a part of the District of Columbia government, became an independent authority and no longer bound by D.C. personnel regulations and civil service protections.¹ Taylor alleges that when WASA became unconstrained by personnel regulations and measures that provided civil service protection, it instituted an employment system which vested supervisors with excessive discretion in making hiring, compensation, and promotion decisions. As a result, Taylor asserts that WASA’s Blacks employees encountered a “glass ceiling.” Second Am. Compl. at 3. Although they make up approximately 80 percent of the agency’s employees, they are underrepresented at higher pay grade levels and in supervisory positions. Taylor also contends that Black employees are hired at lower starting salaries than similarly situated White employees. Thus, Taylor contends that Black employees have been denied the opportunity to advance to the same level and at the same rate as similarly situated White employees.

Taylor, a civil engineering technician, asserts that he applied for several promotions from 1988 to 1996, but was passed over despite demonstrating the required qualifications. The civil engineering career ladder² provides the opportunity for advancement from grade DS-2 to grade DS-11.³ Taylor was promoted from DS-6 to DS-8 in 1996 because of a labor agreement. In 2000, his direct supervisor, Londra Watson, who is Black, recommended him for a promotion to DS-9 because he had demonstrated the skill and ability to perform at the higher level and had spent the required time in his previous grade. Nonetheless, the Director of the Department of Engineering and Technical Services, Leonard Benson, a White male, denied Taylor the career-ladder promotion. Taylor still has not been promoted since 1996, despite receiving job appraisals that are satisfactory or above.

II. ANALYSIS

A party who moves for class certification bears the burden of establishing that all requirements for proceeding as a class action have been satisfied. See Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 614, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997). A putative class representative must satisfy the four prerequisites of Rule 23(a): (1) that the class is so numerous that joinder of all members is impracticable; (2) that there are questions of law or fact common to the class; (3) that the claims or defenses of the representative parties are typical of the claims or defenses of the class; and (4) that the representative parties will fairly and adequately protect the interests of the class. Fed.R.Civ.P. 23(a). Although not an explicit requirement under Federal Rule of Civil Procedure 23, this court has recognized the “common-sense requirement” that a plaintiff also establish the existence of a class. Does I through III v. District of Columbia, 232 F.R.D. 18, 25 (D.D.C.2005). Finally, a putative class representative bears the burden of showing that the class falls within at least one of the three categories set forth in Rule 23(b).

Whether a class should be certified is a preliminary question, and disputes regarding the merits of a case or the weight of evidence are not proper considerations at the class-certification stage. See Eisen v. Car-lisle & Jacquelin, 417 U.S. 156, 177, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974) (“We find nothing in either the language or history of Rule 23 that gives a court any authority to conduct a preliminary inquiry into the merits of a suit in order to determine whether it may be maintained as a class action.”). When “determining the propriety of a class action, the question is not whether the plaintiff or plaintiffs have stated a cause of action or will prevail on the merits, but rather whether the requirements of Rule 23 are met.” Ibid. (internal quotation marks and citation omitted). A district court exercises broad discretion in deciding whether a party seeking certification has carried her burden. Hartman v. Duffey, 19 F.3d 1459, 1471 (D.C.Cir. 1994).

A. Rule 23(a) Requirements

WASA contends that Taylor cannot satisfy any of the prerequisites of Rule 28(a): numerosity, commonality, typicality, and adequacy of representation. The court will consider each requirement in turn.⁴

1. Numerosity

Rule 23(a)(1) permits maintenance of a class action if “the class is so numerous that joinder of all members is impracticable.” There is no specific threshold that must be surpassed in order to satisfy the numerosity requirement; rather, the determination “requires examination of the specific facts of each case and imposes no absolute limitations.” Gen. Tel. Co. of the Northwest, Inc. v. EEOC, 446 U.S. 318, 330, 100 S.Ct. 1698, 64 L.Ed.2d 319 (1980). That said, courts in this jurisdiction have observed that a class of at least forty members is sufficiently large to meet this requirement. See, e.g., Thomas v. Christopher, 169 F.R.D. 224, 237 (D.D.C. 1996), aff'd in part and rev’d. in part, 139 F.3d 227 (D.C.Cir.1998).

On its face, joinder of 800 plaintiffs is impracticable. Thus, the court concludes that a class containing 800 members is sufficiently numerous under Rule 23(a)(1). WASA’s only contention otherwise—that the class will shrink upon application of the commonality and typicality requirements—is unpersuasive, as explained below.

2. Commonality

Commonality requires that the plaintiff raise claims which rest on “questions of law or fact common to the class.” Fed. R.Civ.P. 23(a)(2). This rule does not require that “every issue of law or fact be the same for each class member.” Bynum v. District of Columbia, 217 F.R.D. 43, 46 (D.D.C.2003). Rather, “[t]he commonality test is met when there is at least one issue, the resolution of which will affect all or a significant number of the putative class members.” Coleman v. Pension Benefit Guar. Corp., 196 F.R.D. 193, 198 (D.D.C.2000) (internal quotation marks omitted). Because the commonality requirement may be satisfied by a single common issue, courts have noted that it is “often easily met.” In re Vitamins Antitrust Litig., 209 F.R.D. 251, 259 (D.D.C.2002). Additionally, proposed class actions seeking injunctive and declaratory relief, such as this one, “by their very nature” present common questions of law and fact. 7A Wright, Miller & Kane, Federal Practice and Procedure § 1763 (3d. ed.2005).

Taylor’s claim of unlawful discrimination is based on two different theories. A “disparate treatment” claim alleges that the defendant intentionally based an employment decision on the race of the plaintiffs. Palmer v. Shultz, 815 F.2d 84, 90 (D.C.Cir.1987) (citing Int’l Bhd. of Teamsters v. United States, 431 U.S. 324, 335-36, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977)). Disparate treatment claims can involve an isolated incident of discrimination against a single individual, or, as in this ease, allegations of a “pattern or practice” of discrimination affecting an entire class of individuals. Id. A “disparate impact” claim, on the other hand, alleges that the defendant based an employment decision on a criterion that although “facially neutral” nevertheless impermissibly disadvantaged individuals of one race more than the other. Id.

Although discrimination claims, by their nature, often allege treatment based on membership in a class, such an allegation is not sufficient for a court to determine that such treatment was likewise suffered by a class of similarly situated persons. Gen. Tel. Co. of Sw. v. Falcon, 457 U.S. 147, 157-58, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982); Love v. Johanns, 439 F.3d 723, 728 (D.C.Cir.2006). A plaintiff must “bridge th[e] gap” between her own alleged discrimination and a “common policy” that affected the members of the putative class. Falcon, 457 U.S. at 158, 102 S.Ct. 2364.⁵ Plaintiffs alleging disparate treatment of a class must “show (i) discrimination (ii) against a particular group (iii) of which the plaintiff is a member, plus (iv) some additional factor that permits the court to infer that members of the class suffered from a common policy of discrimination.” Love, 439 F.3d at 728 (internal quotation marks and alteration omitted) (emphasis in original). Regarding a complaint of discriminatory impact, plaintiffs must make a showing “sufficient to permit the court to infer that members of the class experienced discrimination as a result of the disparate effect of a facially neutral policy.” Garcia v. Johanns, 444 F.3d 625, 632 (D.C.Cir.2006) (internal quotation marks omitted).

A common discriminatory policy may be established by “[significant proof’ that an employer’s hiring and promotion practices were accomplished by “entirely subjective decisionmaking processes.” McReynolds v. Sodexho Marriott Services, Inc., 208 F.R.D. 428, 439 (D.D.C.2002) (quoting Falcon, 457 U.S. at 159 n. 15, 102 S.Ct. 2364). It is well established that “the deliberate and routine use of excessive subjectivity is an employment practice that is susceptible to being infected by discriminatory animus.” Dukes v. Wal-Mart Stores, Inc., 222 F.R.D. 137, 149 (N.D.Cal.2004) (internal quotation marks omitted) (certifying class of 1.5 million women alleging sex discrimination), aff'd, 474 F.3d 1214 (9th Cir. 2007). In proving a policy of excessive subjectivity, plaintiffs may rely on a combination of statistical and anecdotal evidence to support their assertions that unfettered discretion pervaded the company’s decisions regarding hiring and promotions which resulted in racial discrimination against the members of the class. See Valentino v. U.S. Postal Serv., 674 F.2d 56, 68 (D.C.Cir. 1982) (“[C]lass action plaintiffs offer a combination of statistical proof and individual testimony of special instances of discrimination.”). Such proof would support both theories of disparate treatment and impact in a class action because both theories “are attacks on the systemic results of employment practices” and “proof of each claim will involve a showing of disparity between the minority and majority groups in an employer’s workforce.” Segar v. Smith, 738 F.2d 1249, 1267 (D.C.Cir.1984) (observing that there is “an important point of convergence” between disparate treatment and disparate impact claims in class actions).

As explained below, Taylor has presented sufficient evidence to support his claim that there are significant factual and legal questions common to all class members with respect to the alleged discrimination against the putative class. Each category of evidence will be discussed below.

a. Factual Evidence

i. WASA’s Racial Make-up

The majority of WASA’s workforce is Black. As of December 31, 2001, approximately 80 percent of the 1,046 employees listed on WASA’s employee roster were Black. Ex. G (Employee Roster, Dec. 31, 2001).⁶ Of the remaining employees, approximately 15 percent were White, less than 2 percent were Hispanic, and approximately 2 percent were Asian/Pacific Islander. The racial make-up of the various departments at WASA indicates that the Black employees were not evenly distributed.⁷ Several departments or sub-departments were more than 90 percent Black, including: Buildings and Grounds; Customer Services (Meter Operations); Material Management; Sewer Services (Inspection, Pumping, or Repair); and Water Service (Distribution). WASA’s White employees were concentrated in four departments, which accounted for more than 75 percent of WASA’s White employees. While Engineering and Technical Services employed only 9 percent of WASA’s workforce, it employed 25 percent of its White workers. Similarly, Wastewater Treatment employed 14 percent of the workforce, but 26 percent of WASA’s White workers; Maintenance Services (Electrical) employed 6 percent of WASA’s workforce but 13 percent of its White employees; Maintenance Services (Mechanical) employed 8 percent of WASA’s workforce but 13 percent of its White employees worked there.

Although Whites made up only 15 percent of WASA’s workforce, they were also disproportionately represented in management. For example, six of the eleven employees (or 54 percent) who report directly to Walter Bailey, Director of Wastewater Treatment, were White. Of the five remaining positions, one manager is a native of India; two managers were Black; and two other positions filled by Blacks (administrative manager and executive assistant) were primarily administrative and not managerial. Similarly, all of the branch chiefs who report to Taylor’s supervisor, Leonard Benson, the Director of Engineering and Technical Services, were White.

ii. WASA’s Promotion Policies

An employee at WASA may be promoted in one of two ways: career-ladder promotions (which are only available to union members) and competitive promotions. A career-ladder position is a union position in which an employee is eligible to be promoted up to a certain specified grade without having to compete for the promotions provided she receives satisfactory performance appraisals, spends the required time in-grade, and receives the recommendation and approval of her superiors. To obtain a career-ladder promotion, an employee must demonstrate to the satisfaction of management, starting with her supervisor, that she has mastered certain particular job kills, beyond those required in her current grade. According to WASA, “ability” is determined by the employee’s supervisors, and includes dedication to the job, getting along well with coworkers and clients, and technical aptitude. An employee may not receive a career-ladder promotion without the recommendation of her supervisor and the concurrence of the head of her department. An employee may not nominate herself for such a promotion.

With regard to competitive promotions, it was WASA’s policy that vacant positions were to be posted and an employee must submit an application. Applications were reviewed by the Department of Human Resources (“HR”), and all employees who met the position’s minimum qualifications were put on a list of those to be interviewed for the position. Prior to 2000, however, WASA did not routinely interview internal applicants. Ex. B (Walter Bailey Dep.) at 37:2-38:2. For those applicants who were interviewed, the hiring department drafted a list of questions which were reviewed by HR to ensure the questions were appropriate. The interviews were conducted by a panel of three to five people, usually consisting of employees at the next level above the vacant position, who were chosen by HR. At the interviews, the applicants were asked the same questions, and panelists scored them on a scale of one to five, as well as their answers to each question, after discussing the applicants. HR collected the score sheets and turned the scores into a ranking. The panel then recommended one candidate to be hired, and forwarded that recommendation to HR. In preparation for this process, WASA briefed the panelists about what questions would be inappropriate or illegal but provided no other training on how to conduct the interviews or score the candidates. WASA had never conducted a study to determine whether or not its hiring, promotion or compensation processes were equitable, nor whether the career-ladder promotion system functioned equitably.

iii. Compensation Policies

With regard to compensation policies, WASA had wide discretion to set the salaries of both its union and non-union workers. For union workers, management had the discretion to hire a union employee at a grade above the minimum grade in that career ladder, which would result in a higher starting salary. Pay increases for union workers were dependent on their advancement within their career-ladder or to another position. For non-union workers, compensation was based on a competitive “open-range,” meaning that WASA could hire an employee at any salary between the minimum and maximum range for the position, and, in some circumstances, at a salary higher than that range. Ex. C. (Barbara Grier Dep.) at 38:13-22; Ex. I (WASA Personnel Policy and Procedure) at 215-16. Salary increases or lump-sum bonuses for non-union workers were based on “performance” as defined by the employee’s supervisor. Ex. I at 216-18. Like union workers, non-union workers could also receive a raise through a promotion to a higher grade or new position.

In her deposition, Barbara Grier, head of human resources, explained that “there is general acknowledgment” that the grade compensation system for union employees was “broken.” Ex. C (Grier Dep.) at 66:11— 12. For example, one employee might be in a position that was designated as having no advancement potential, and another employee doing the same job would be in a position that had the potential for career-ladder promotions to four grades higher. She indicated that the problems were being addressed through collective bargaining in late 2000 and 2001.⁸ Despite these acknowledgments, WASA has never undertaken an independent study of its compensation system to determine whether it results in discrimination.

b. Anecdotal Evidence

Taylor points to the following anecdotal evidence of racially discriminatory patterns and practices at WASA:

• In addition to his personal experiences, Taylor observed White employees in the Department of Engineering and Technical Services (“DETS”) who were frequently promoted over Blacks who were more experienced and qualified. Ex. J (Taylor Decl.) at 3. He also observed that White employees in career-ladder positions were promoted without having to serve the required time in their present grade, while Black employees were required to spend five years or more in grade before receiving a career-ladder promotion. He noted that White workers were generally hired at higher grades than Black employees, often after working at the Washington Suburban Sanitary Commission.

• George Ginwright, a Black surveying aide with a bachelor’s degree in civil engineering, asserts that he was “stuck” at DS-9 for thirteen years before promotion. Ex. L (Ginwright Decl.) at 1. He contends that in 1996, he applied for several DS-12 positions and he was recommended by his supervisor and determined to be qualified by HR, but two White supervisors rejected his applications. He also asserts that he subsequently applied for two supervisory openings in DETS which were given to two White employees without college degrees. Ginwright also alleges he observed other discriminatory hiring practices: that White employees from the Washington Suburban Sanitation Commission were hired as managers to replace Blacks even though they had fewer qualifications; that DETS would not promote Blacks to full DS-11, even though they were on a career ladder to that grade; and that Whites would receive greater advancement or pay even when doing the same job as similarly situated Blacks.

• Clarence Stith, a Black civil engineering technician, contends that he waited seven years to be promoted to a full GS-4 in 1996, which only occurred as a result of a union grievance, and then waited another four years to be promoted to GS-5, despite the fact that he was in a career ladder position which progresses to GS-11 and he always had satisfactory performance appraisals. Ex. 0 (Stith Deel.) at 1-2.

• Philip Browne, a Black supervisory environmental engineer and division chief in Solids Processing, explained that he was “bumped” from his position in 1991 due to a reduction in force and was never notified of his right to reclaim the job once it became vacant again. Ex. Q (Browne Aff.) at l.⁹ In 1998, when he applied for the same position as supervisory environmental engineer which he had already held, he was rejected by Bailey in favor of a White male.

• Betty Williams, a Black operator in Thickening and Digestion from 1987 to 2000, observed two White employees being given training opportunities that resulted in promotions. When she inquired of Bailey about how she could volunteer for such training opportunities, she was told they “did not exist.” Ex. R (Williams Aff.) at 2. In 1998 or 1999, Williams requested Bailey assign her to light duty while she recovered from foot surgery, but he rejected her request. A White male recovering from a hernia, however, was assigned to lighter duty.

• Eugene Scott, a Black foreman who retired in 2001, stated that in 1993, Bailey installed a White supervisor over all “dewatering operations” without permitting others to compete for the position. Ex. R (Scott Aff.) at 2. In 1997, Bailey assigned the same White supervisor to the position of “acting chief’ of the Dewatering Branch, as well as appointing another White employee, without a competitive hiring process.¹⁰

c. Statistical Evidence

Commonality may be established “by raising an inference of class-wide discrimination through the use of statistical analysis.” Dukes, 474 F.3d at 1228; see also Caridad v. Metro-North Commuter R.R., 191 F.3d 283, 292 (2d Cir.1999).

i. Plaintiffs Report

In support of his motion, Taylor presents the expert report of Alexander Vekker, Ph. D., who concludes that there is statistical evidence of class-wide discrepancies likely attributable to racial discrimination at WASA. Vekker performed a regression analyses that measured the compensation differences for Blacks with the same job tenure and same job title as White counterparts at WASA. Ex. A (Vekker Report, Jan. 21, 2005) at l.¹¹ Overall, Vekker determined that in 1999 and 2000, Black employees received significantly lower levels of compensation than White employees with the same tenure and job title. He was unable to reach any conclusions regarding WASA’s selection of employees for promotions because he was not provided sufficient data.

In his report, Vekker included the following summary of his findings:

[[Image here]]

Vekker concluded that, after controlling for job title and tenure, Black employees earned on average 4.6 percent less than Whites with the same tenure and same job in 1999. The disparity grew in 2000 to 5.6 percent.¹² The standard deviation for 1999, for example, was 4.82, which is well in excess of what scientists and courts consider to be statistically significant. Vekker Rep. at 5. A standard deviation of 4.82 indicates that the possibility that these racially differentiated results are a result of something other than race is less than seven in 100,000. Ibid. Thus, Vekker concluded, the wage disparity at WASA is caused by race. Ibid.

Vekker also concluded that there were racial disparities in initial, or base, salaries due to race, but that they were statistically insignificant. After adjusting for differences in qualifications at the time of hiring (i.e., experience and job title), the results showed that Black employees’ initial salaries were 2.8 percent lower than Whites hired in the same year into the same job title. The standard deviation, however, was only 0.99, which is not statistically significant. Id. at 6.

Vekker based his analysis on data provided by WASA, which included (1) active employees’ job and compensation history; (2) employee rosters by department at the end of each year from 1999 to 2002; and (3) inactive employees’ job and compensation history. In his analysis, Vekker did not include formal education of employees, the pay grade of particular employees, nor any indication of which employees were on career ladders, as such information was not provided to him by WASA even though it was requested. See Vekker Rep. at 6; Pis. Reply at 24-25. He indicated that his conclusions were preliminary and might be affected by receipt of additional data. Vekker Rep. at 7.

ii. Defendant’s Report

In response to the Vekker report, WASA submits the expert report of Paul F. White, Ph.D., in which he separately conducted a regression analysis of salary disparities on each department rather than on WASA’s workforce as a whole, as Vekker had done. White concludes that two of ten categories of Black employees had disparities which were statistically significant: (1) Regular Wage (“RW”) in 1999 earned on average $327 less annually, a standard deviation of 2.22; and (2) Open Grade (non-union) employees in 2001 earned on average $3,138 less annually, a standard deviation of 2.46. Def.’s Opp’n, Ex. D (White Report of Sept. 15, 2005) at 6.¹³ Otherwise, White asserts that a department-level analysis reveals that there is no widespread pay disparity at WASA.

iii. Statistical Analysis

WASA does not dispute the accuracy of the Vekker Report but rather contends it should be discounted because of alleged methodological flaws: The absence of key variables, improper aggregation of the data, and the failure to conduct a so-called “Chow” test. WASA’s contentions regarding Taylor’s statistical analysis are not persuasive.

While the omission of variables from a regression analysis “may render the analysis less probative than it otherwise might be, it can hardly be said, absent some other infirmity, that an analysis which accounts for the major factors must be considered unacceptable as evidence of discrimination.” Bazemore, 478 U.S. at 400, 106 S.Ct. 3000 (Brennan, J., concurring, joined by all Justices) (citations and quotation marks omitted). This principle does not require acceptance of regressions “from which clearly major variables have been omitted—such as education and prior work experience.” Roger v. Reno, 98 F.3d 631, 637 (D.C.Cir.1996). But a defendant “cannot undermine a regression analysis simply by pointing to variables not taken into account that might conceivably” have affected the analysis. Id. “[I]n most cases a defendant cannot rebut statistical evidence by mere conjectures or assertions, without introducing evidence to support the contention that the missing factor can explain the disparities as a product of a legitimate, nondiscriminatory selection criterion.” Palmer, 815 F.2d at 101.

WASA first contends that Vekker’s analysis is flawed because he fails to include prior work experience. As prior work experience was not provided to either expert, White estimates prior experience by using a “proxy”: an employee’s tenure at WASA minus 18 years. White Rep. at 4. WASA has failed to show, however, how White’s estimate of prior experience reveals a nondiscriminatory reason why Vekker found pay disparities. White indicates that he included the estimate of prior experience because he found “multiple instances of employees at higher levels of the pay grades with relatively fewer years at WASA.” Id. The allegations in this case, however, indicate that White employees with less education or experience were promoted over more qualified Black employees. White’s finding that employees with high pay had little experience at WASA could just as easily be explained by discrimination in hiring and promotion. Thus, WASA has failed to show how a failure to include this estimated factor requires that the court reject the conclusions of the Vekker Report.

WASA also contends that it was inappropriate to aggregate employees in different pay grades, and that Vekker should have performed a so-called “Chow test” prior to aggregating the data across various pay plans. A “Chow test” (so named after the statistician who created it) is utilized to analyze whether it is appropriate to aggregate two or more sets of data into a single sample in a statistical model. See Dukes, 222 F.R.D. at 157. The result of a Chow test might indicate that the data should be disaggregated—that is, that separate regression analyses be run for distinct years, regions or pay plans, for example. WASA’s contention regarding Vekker’s aggregation of the data is neither supported by authority nor the circumstances presented here. Id. (concluding that Chow test is not required at class certification stage where plaintiff class presents data aggregated over regional level). Moreover, in cases alleging discrimination in promotion, the inclusion of grade variables would be “inappropriate” because an employee’s grade may itself reflect previous discrimination. Valentino, 674 F.2d at 72 n. 30 (observing that, in cases alleging of inconsistent and biased promotion criteria, “there is no assurance that level or rank is an appropriate explanatory variable, untainted by discrimination”). And in light of the fact that the majority of Black workers are concentrated in the RW pay plan which, as shown by White’s own analysis, indicate a pay disparity between Blacks and Whites, the aggregation did not improperly create an inference that would not otherwise exist.

More significantly, such “statistical dueling” is irrelevant to the certification determination. See Caridad, 191 F.3d at 292 (observing that the defendant’s critique of the plaintiffs’ evidence “may prove fatal at the merits stage, the Class Plaintiffs need not demonstrate at this stage that they will prevail on the merits.”). The only relevant question at this stage is whether Taylor has put forth evidence that is sufficient to raise an inference of discrimination to create a common question regarding discrimination in pay at WASA. See id. The court concludes that Taylor has met this burden. Even setting aside Vekker’s conclusions, WASA’s own analysis supports an inference of disparities likely attributable to race. White concludes that employees in the RW salary schedule in 1999 earned on average $327 less annually, with a probability that it was caused by something other than race of 3 percent (or a standard deviation of 2.22). A standard deviation of 1.96 or higher indicates a “level of statistical significance [that] is sufficient to establish a prima facie case of both disparate treatment and disparate impact.” Anderson v. Zubieta, 180 F.3d 329, 340 (D.C.Cir.1999). The RW salary schedule accounts for 539 Black employees at WASA in 1999, or approximately 51 percent of the workforce at WASA included in White’s analysis, and 60 percent of the Black workforce. White Rep. at 6. Given that so many Black employees are concentrated in the area that White concludes had a wage disparity strongly correlating to race, these statistics raise the inference of a common policy that results in racially inequitable pay.

Furthermore, although neither expert reached any conclusions regarding discrimination in promotions, the results of their analyses also support such an inference of discrimination. An employee’s compensation is largely dependent on (1) receipt of a career-ladder or competitive promotion; and/or (2) approval of salary increases, and compensation is an indirect result of such decisions. Thus, a racial disparity in compensation raises the inference of a discrimination in the promotion decisionmaking process.

d. Conclusion

Overall, Taylor has put forth evidence supporting the inference that WASA’s policies and procedures are sufficiently subjective and susceptible to racial discrimination to satisfy the commonality requirement. WASA’s promotion policies and practices are governed by minimal objective criteria or central guidelines, and as such vest excessive discretion in the hands of mostly White supervisors. A career-ladder employee’s chance at promotion rests entirely in the hands of her immediate supervisor or that person’s supervisor, and those supervisors are given little guidance and oversight when it comes to making such determinations. An employee’s chances at competitive promotion also rest entirely in the hands of her supervisors and the interviewing panel who are given no guidance or training as to how to interview, rank, and select appropriate candidates. Indeed, WASA even acknowledged that prior to 2000, internal candidates were not even routinely interviewed for competitive positions. Moreover, there is evidence that certain jobs were not posted nor filled through the regular or competitive hiring process, increasing the chance for subjectivity to influence such decisions. The discretion and subjectivity in the promotional process is compounded by the fact that there is no monitoring nor systematic review by upper-level management to ensure that ground-level decisions are not being based on impermissible criteria.

Taylor has also brought forth evidence raising the inference that such policies have resulted in a disproportionate concentration of Black employees in lower-ranking and lower-paying jobs, and that Blacks wait longer to obtain promotions. Taken together, this evidence sufficiently raises the inference that WASA’s allegedly discriminatory policies and practices affect all plaintiffs in a common manner, which satisfies the commonality requirement of Rule 23(a)(2).¹⁴

3. Typicality

Typicality requires that the claims of the representative be typical of those of the class. Fed.R.Civ.P. 23(a)(3). The typicality requirement and the commonality requirement “tend to merge” because both serve as “guideposts” as to whether a particular class action is practical and whether the claims of the plaintiff and class are sufficiently interrelated to protect the class members in their absence. Falcon, 457 U.S. at 157 n. 13, 102 S.Ct. 2364.

WASA’s contention that Taylor’s claim is not sufficiently typical rests primarily on its argument regarding commonality which, as discussed above, this court rejects. For example, WASA contends that Taylor’s claim is not typical because “his own expert ... could not demonstrate any statistical significance in the promotion rates” for Black and White employees. Def.’s Opp’n at 21. In fact, it was not that Yekker failed to find promotion disparities but that he was “not able to conduct a meaningful analysis of promotions since [he] did not have any information” on pay grade or career-ladders in the data that WASA provided him. Vekker Report at 6. WASA also contends that Taylor cannot represent the class because he has not served in all of the positions included in the class. There is no requirement, however, that a class representative have served in each job category that he seeks to represent; such a requirement would always defeat class certification. See Hartman, 19 F.3d at 1471 (concluding that unsuccessful applicant may represent other applicants for different jobs where the class is challenging similar discriminatory practices). While the court may later determine to separate plaintiffs into different sub-classes based on the distinctions between career-ladder and competitive positions, for example, “it would not be necessary to the validity of the class certification to do so.” Id.

Taylor claims that he was denied a promotion, and thus an increase in compensation, based on the discriminatory practices and unchecked discretion that permeate such decisions at WASA. His claim is typical of the claims he seeks to advance on behalf of the class, and thus meets the requirement of Rule 23(a)(3).

4. Adequacy

The fourth and final requirement of Rule 23(a) requires that the court determine whether the proposed representatives can adequately represent the interests of the class. In making this determination, the court must be assured that (1) the proposed representative does not have any conflicts of interest with other class members; and (2) that the representative will vigorously prosecute the interests of the class through qualified counsel. McReynolds, 208 F.R.D. at 446. WASA contends that Taylor is not an adequate representative because (1) non-supervisors cannot represent a class containing supervisors; (2) Taylor is not a suitable representative because he lost an arbitration on his claim; and (3) that counsel is not qualified.

First, WASA contends that the D.C. Circuit’s holding in Wagner v. Taylor forecloses the possibility that a non-supervisor can represent a class containing supervisors. See 836 F.2d 578, 595 (D.C.Cir.1987). The Wagner court, in actuality, held the opposite—that supervisors could not adequately serve as representatives of non-supervisors in the circumstances of that case. Id. The court reasoned that supervisors could not be adequate representatives where they were responsible for evaluating the performances of other members of the class, and where supervisors were accused of discrimination. Id. Here, in contrast, Taylor is not a supervisor who reviews any other members’ work and there is no evidence that he nor any members of the putative class are themselves accused of discrimination. See McReynolds, 208 F.R.D. at 447 (concluding Wagner did not prevent certifying mixed class where representatives were not supervisors and complaint did not allege discriminatory conduct against other potential class members). Moreover, the potential that an eventual award of injunctive relief might apply “against a few supervisory members of the class—who most likely did not exert significant influence over departmental policy-making—is fairly characterized as de minimis relative to the value of such an injunction in protecting those same supervisors from epidemic discrimination.” Id. at 448 (internal quotation marks omitted).

WASA also contends that Taylor is inadequate because his promotion claim was resolved against him in a union arbitration in 2002. It is well-settled, however, that “the named plaintiff need not demonstrate a probability of success on the merits” in order to serve as the class representative. 2 Newberg & Conte on Class Actions § 3:29 (4th ed.2002); see also Eisen, 417 U.S. at 178, 94 S.Ct. 2140 (“[T]he question is not whether the plaintiff or plaintiffs have stated a cause of action or will prevail on the merits.” (internal quotation marks and citation omitted)). A plaintiff who may ultimately lose on the merits may nonetheless serve as an adequate representative if she has an interest in the outcome of the case. A class representative has two legally cognizable interests in the outcome of a class action: “One is the claim on the merits; the other is the claim that he is entitled to represent a class.” Richards v. Delta Air Lines, Inc., 453 F.3d 525, 529 (D.C.Cir.2006) (internal quotation marks and citation omitted). When the first of these interests expires, the class representative still has a “personal stake” in the class certification claim. Id.; see also Wilson v. Heckler, 580 F.Supp. 1387, 1389 (D.D.C.1984) (concluding that “although plaintiff’s] claim concerning the method by which her benefits were calculated no longer presents a live controversy for Art. Ill purposes, her claim that she is entitled to represent a class is still viable and supplies her ‘personal stake’ in the litigation” (quoting U.S. Parole Comm’n v. Geraghty, 445 U.S. 388, 402, 100 S.Ct. 1202, 63 L.Ed.2d 479 (1980))). Regardless of the merits of his individual claim for damages, Taylor has an ongoing interest in the injunctive relief sought by this claim against the alleged discrimination by his employer. Moreover, Taylor has demonstrated his adequacy by his diligent prosecution of this case on behalf of the class. Jarvaise v. Rand Corp., 212 F.R.D. 1, 3-4 (D.D.C.2002) (observing that “the interest and persistence of the named plaintiffs cannot be questioned in light of their prolonged prosecution of this case.”).¹⁵

WASA also contends that plaintiffs counsel is not adequate to represent the class because of its performance in Pigford v. Veneman, in which the D.C. Circuit criticized class counsel’s inability to represent claimants adequately due to repeated failures to meet deadlines. See 292 F.3d 918, 925-27 (D.C.Cir.2002). The record in that case reveals, however, that Taylor’s counsel was not the “class counsel” criticized in Pigford, but rather was one of many other attorneys representing the class and who was added to the case after class counsel missed most of the crucial deadlines. Thus, WASA has raised no genuine concern about counsel’s adequacy.

Accordingly, the court concludes that Taylor has satisfied all requirements of Rule 23(a).

B. Rule 23(b) Requirements

Compliance with one sub-part of Rule 23(b) is also required for certification of a class. Taylor contends that he has met the requirements of Rule 23(b)(2) and (3). WASA asserts that he cannot satisfy either.

Rule 23(b)(2) permits certification of a class where “the party opposing the class has acted or refused to act on grounds generally applicable to the class, thereby making appropriate final injunctive relief or corresponding declaratory relief with respect to the class as a whole.”¹⁶ Certification is permissible under Rule 23(b)(3) where questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and a class action is superior to other available methods for the fair and efficient adjudication of the controversy.

In challenging Taylor’s ability to meet the requirements of either Rule 23(b)(2) or (3), WASA merely repeats its contentions under Rule 23(a)—that WASA has not acted in a class-wide manner and that Taylor has not shown that common questions even exist. As the court explains above, it rejects these contentions. The inquiry, however, does not end there, as the court must be satisfied that the class should be certified under one or more sub-parts of Rule 23(b) before proceeding.

There are two basic factors that must be present in order to certify a class under Rule 23(b)(2): (1) the defendant’s action or refusal to act must be generally applicable to the class;¹⁷ and (2) plaintiffs must seek final injunctive relief or corresponding declaratory relief on behalf of the class. Bynum, 217 F.R.D. at 48. The Advisory Committee Notes on Rule 23 explain that “cases in the civil rights field” are “illustrative” of the types of class actions suitable for certification under (b)(2). Fed.R.Civ.P. 23(b)(2) advisory committee notes; see also Eubanks v. Billington, 110 F.3d 87, 92 (D.C.Cir.1997). Certification under Rule 23(b)(2) is not always appropriate, however, when a class seeks monetary damages in addition to injunctive relief; indeed, it is not permitted if such monetary claims predominate. See Taylor v. D.C. Water & Sewer Authority, 205 F.R.D. 43, 47 (D.D.C.2002). Thus, when a Title VII class action seeks compensatory damages in addition to injunctive relief, the court may exercise its discretion to certify a(b)(2) class as to the claims for declaratory or injunctive relief, and a(b)(3) class as to the claims for monetary relief, otherwise known as “hybrid” certification. See Eubanks, 110 F.3d at 96 (race discrimination claim seeking $8.5 million in compensatory damages; certification under (b)(2) affirmed).¹⁸ Another approach, known as “partial certification,” is to bifurcate the action into liability and damages phases and certify a(b)(2) class on liability only, postponing consideration of whether class certification is appropriate for the damages phase until plaintiffs have made it that far. See Does I through III, 232 F.R.D. at 29-30 (certifying class as to injunctive relief and postponing determination regarding monetary damages); see also Fed.R.Civ.P. 23(e)(4)(A) (permitting the partial certification of a class when “an action may be brought or maintained as a class action with respect to particular issues.”). If the plaintiffs succeed in establishing liability in the first phase, the court may order class-wide injunctive and declaratory relief and may then hold so-called “Teamsters hearings” to make individualized determinations as to whether a particular employee suffered an adverse employment action as a result of the defendant’s discriminatory police, and whether she is entitled to individualized equitable relief such as back pay and front pay, or compensatory damages. See Teamsters, 431 U.S. at 336, 97 S.Ct. 1843; Hartman, 19 F.3d at 1462 n. 2.

Here, the court has no difficulty concluding that, as to liability, WASA is alleged to have acted or refused to act on grounds generally applicable to all class members and that injunctive relief would be applicable to the entire class. Moreover, litigating the pattern-or-practice liability phase for the class as a whole would both reduce the range of issues in dispute and promote judicial economy. See Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 168 (2d Cir.2001) (observing that (b)(2) certification is appropriate in such cases because “the liability phase is largely preoccupied with class-wide statistical evidence directed at establishing an overall pattern or practice of intentional discrimination”). Accordingly, the court partially certifies the class as to liability under Rule 23(b)(2) for injunctive and declaratory relief, and postpones the decision as to whether to certify the class as to damages, or whether to hold Teamsters hearings, until it is determined whether the class can prevail in the first phase.¹⁹

The court also concludes that the members of the putative class should be notified of this partial certification, pursuant to Rule 23(c)(2)(A), and of their right to opt-out of this class, see Fed.R.Civ.P. 23(d)(5). Although opt-outs are not usually required under (b)(2), the court may require opt-outs in a certification of a class under (b)(2) if the court finds that “the assumption of cohesiveness” underlying certification of a(b)(2) class is inapplicable to the individual class members’ claims for monetary damages. See Thomas v. Albright, 139 F.3d 227, 236 (D.C.Cir.1998) (observing that permitting opt-outs in a(b)(2) certification requires that the court have “some reason to believe that the assumption of cohesiveness underlying a subsection (b)(2) class action does not apply to the individual claims for monetary damages”). The possibility that the class may later be certified under (b)(3) to resolve individual damage determinations is sufficient to require opt-outs in these circumstances to protect the members of the class. Fed. R.Civ.P. 23(d)(5) (allowing the court to “make appropriate orders ... requiring for the protection of the members of the class or otherwise for the fair conduct of the action”); see also Eubanks, 110 F.3d at 96 (concluding that this rule is “broad enough to permit the court to allow ... [ (b)(2) opt-outs] ... when necessary to facilitate the fair and efficient conduct of the litigation”).

III. CONCLUSION

For the foregoing reasons, it is this 13th day of March, 2007 hereby

ORDERED that the Second Amended Motion to Certify the Class [# 107] pursuant to Rule 23(b)(2) is GRANTED in part and DENIED in part; and it is further

ORDERED that the putative class of Black employees at WASA who sought and were denied positions, career ladder promotions, or other advancement, or whose advancement was delayed, or whose compensation was otherwise affected by WASA’s alleged unlawful discrimination, from October 1996 through December 2000, is hereby CERTIFIED; and it is further

ORDERED that on or before April 10, 2007, the parties shall file a proposed case management plan including deadlines that shall govern the future proceedings in this action. If the parties are unable to agree, each shall file its own proposed case management plan and deadlines.

1 . WASA’s service area covers approximately 725 square miles, including the District, Montgomery and Prince George’s County in Maryland, Fairfax and Loudon Counties in Virginia, and Vienna, Virginia.

2 . A career-ladder position is one in which a union employee is eligible for promotion to a certain specified grade provided she receives satisfactory performance appraisals, spends the required time in-grade, and receives the recommendation of her superior.

3 . Jobs within WASA are classified on a DS scale (similar to the GS scale used by the federal government) depending upon an employees's level of expertise and experience. A higher DS correlates to a higher salary.

4 . WASA does not challenge the existence of a class. While not designed to be a particularly stringent test, plaintiffs must at least establish that the "general outlines of the membership of the class are determinable at the outset of the litigation.” 7A Wright, Miller & Kane, Federal Practice and Procedure § 1760 (3d. ed.2005). The court is satisfied that the contours of Taylor’s proposed class are sufficiently well defined as to render the proposed class sufficiently discrete and identifiable and to make management of the class administratively feasible.

5 . The requirement of a common policy applies to intentional discrimination under both Title VII and Section 1981. See Jett v. Dallas Independent School Dist., 491 U.S. 701, 735-736, 109 S.Ct. 2702, 105 L.Ed.2d 598 (1989) (holding that plaintiff must show “that the violation of his 'right to make contracts' protected by § 1981 was caused by a custom or policy” of the municipal employer); Alexander v. City of Milwaukee, 474 F.3d 437, 448 (7th Cir.2007) (observing that § 1981 requires a plaintiff to establish "an offidal policy or custom in order to allow for municipal liability”) (citing Jett).

6 . All exhibits referred to herein are those submitted by Taylor in support of his Amended Motion for Class Certification, unless noted otherwise.

7 . WASA is divided into departments, which include (1) Engineering and Technical Services, (2) Wastewater Treatment, (3) Water Services, (4) Sewer Services, (5) General Counsel, (6) Human Resources, (7) Fleet Management, (8) Facilities and Security, (9) Occupational Health and Safety, (10) Procurement and Material Management, (11) Finance and Budget, (12) Information Technology, (13) Customer Service, (14) Risk Management, (15) Public Affairs and (16) Internal Audit. See Ex. E (WASA Web Site, General Information); Ex. F (Departmental Summaries) at 4222.

8 . WASA notes that in late 2000 and 2001, it began to eliminate automatic step increases and to implement performance based increases. To the extent that such a policy change affected the class period which Taylor seeks to certify, such a limitation on the class period, or on damages, should be addressed later upon greater factual development of the impact of this change.

9 . This and other affidavits submitted by Taylor were originally obtained in the matter of El-Amin v. WASA, a 2002 case in which the court found that a Black chemist at WASA had raised a sufficient showing of race discrimination against El-Amin to deny WASA summary judgment. No. 00-2956 (D.D.C. Aug. 7, 2002) (granting in part and denying in part WASA’s motion for summary judgment). The parties in that matter eventually settled.

10 . Some of the anecdotal evidence presented by Taylor occurred before 1996, the first year covered by the putative class, which the court may consider under the continuing violation theory. A plaintiff may present evidence of a discriminatory policy that began prior to the alleged acts for which she seeks to recover, as long as the unlawful acts themselves fall within the appropriate limitations period. See Bazemore v. Friday, 478 U.S. 385, 395-96, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986) (Brennan, J., concurring, joined by all Justices) ("Each week’s paycheck that delivers less to a black than to a similarly situated white is a wrong actionable under Title VII, regardless of the fact that this pattern was begun prior to the effective date of Title VII."); Shea v. Rice, 409 F.3d 448, 452-53 (D.C.Cir. 2005) (applying Bazemore to conclude that evidence of discrimination prior to the limitations period may be presented).

The court will not consider, however, evidence of this pattern and practice that post-dates the years of his putative class. See Ex. Q (Waller Decl.) at 1-2 (asserting that in 2004, Sharon Waller was required to take ten exams to qualify for an office position while the White employee, who had less experience, received the job after taking four tests;) Ex. O (Stith Deck) at 2 (asserting that in 2004, when Stith's supervisor requested that he be promoted, Benson "laughed at her” and denied the request; the supervisor resigned the following day). Such allegations do not illuminate the allegations of unlawful discrimination during the putative class period.

11 . Generally speaking, a regression analysis is designed to isolate the extent to which an independent variable (such as race) has influenced dependent variables (such as compensation and promotion). McReynolds v. Sodexho Marriott Services, Inc., 349 F.Supp.2d 1, 22, n. 23 (D.D.C. 2004). If such an analysis includes enough nondiscriminatory independent variables (such as experience, performance, etc.) then it will indicate whether the chances that the results are random or whether they are attributable to race. Id.

12 . Vekker also included results for 2001, which the court has not consider as they fell outside the time frame relevant for which class certification is sought.

13 . White only provided a dollar amount, but not a percentage, for the disparities in pay revealed by his analysis.

14 . Because Vekker finds no significant racial disparity in starting salaries, however, the court concludes that claims based on discrimination in initial pay must be excluded.

15 . There are several other members of the class who have been identified by way of their declarations or affidavits submitted in this case who could be substituted for Taylor if at some further point in the litigation it is clear that he is no longer an adequate representative. See 1 New-berg & Conte on Class Actions § 2:26 (4th ed.2002) (observing that “the court may send Rule 23(d)(2) notice to potential class members inviting intervention for the purpose of assuming the responsibility of the class litigation”); id. (explaining that "intervention by absentee members is freely allowed in order to substitute them as class representatives”).

16 . The rule further provides that the "matters pertinent to the findings include: (A) the interest of members of the class in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already commenced by or against members of the class; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; (D) the difficulties likely to be encountered in the management of a class action.” Fed.R.Civ.P. 23(b)(2)

17 . An inquiry into whether the defendant acted on grounds generally applicable to the 23(b)(2) class is often considered to be encompassed by the commonality requirement of Rule 23(a), which the court has already found to have been met. See Does I through III, 232 F.R.D. at 28 n. 12.

18 . Before it can certify a hybrid class, the court is required to find that the "assumption of cohesiveness’’ underlying its Rule 23(b)(2) certification for purposes of equitable relief does not apply to the class members’ individual claims for monetary damages. See Eubanks, 110 F.3d at 96.

19 . The liability phase will consist of a juiy trial to determine liability on the disparate treatment claim and the Section 1981 claim, and a bench trial to determine liability on the disparate impact claim. If liability is found on either claim, the court will then order injunctive relief as necessary, and will determine at that time whether to certify the class under (b)(3) for monetary damages or whether to hold Teamsters hearings.