228 F.R.D. 476 | S.D.N.Y. | 2005
OPINION AND ORDER
High-level female employees at Willis Group Holdings Ltd. (‘Willis”) bring this gender discrimination suit, challenging the terms and conditions of their employment. In an Opinion and Order dated September 25, 2002, this Court denied a motion to compel nationwide discovery, and this action was subsequently limited to Willis’s Northeast region. Plaintiffs have now moved for class certification pursuant to Federal Rule of Civil Procedure 23, seeking to certify a class of current and past female officers, and female employees eligible to receive officer titles. (P. Mem.2.) Defendants oppose class certification, primarily arguing that plaintiffs fail to identify an ascertainable class or to meet the commonality requirement of Rule 23(a). For the reasons stated below, the motion for class certification will be granted.
BACKGROUND
The named plaintiffs, Shelley Hnot and Heidi Scheller, allege a pattern and practice of gender discrimination by their employer, Willis, a global insurance brokerage company. Specifically, they contend that Willis delegates substantial authority regarding promotion and compensation decisions to re
Willis North America’s headquarters are located in Nashville, Tennessee. (D.Mem.2.) Willis is organized into several regional divisions, each headed by a Regional Executive Officer (REO). (D.Mem.3.) Each regional division is comprised of four to six local offices. (Id.) For example, the Northeast region is comprised of local offices in New York, New Jersey, Connecticut, Massachusetts, Pennsylvania, and New Hampshire. (D.Mem.4.) Each local office is led by a Chief Executive Officer (CEO). (D.Mem.3.) Local offices vary in structure, with some utilizing a team structure. (D.Mem.4.) In the Northeast region, a limited number of officials occupy key decisionmaking roles.
Plaintiffs challenge alleged gender discrimination related to compensation and promotion at Willis. Plaintiffs and defendants dispute how these decisions are made. Plaintiffs contend that Willis had a policy of vesting its regional and local officers with “unfettered discretion” on these matters. (D.Mem.7.) They claim that there were no written policies for awarding officer titles (P. Mem.7), and that generally, there were few or no guidelines or criteria for determining salary or bonuses. (D.Mem.8-14.) Plaintiffs’ expert, Dr. Mark R. Killingsworth, Ph. D., provided statistical analyses stating that women were statistically significantly adversely affected in compensation and promotions. (D.Mem.15.)
Defendants, in contrast, assert that compensation and promotion decisions were subject to multiple layers of review. (D.Mem.2, 5.) Defendants also provide their own expert report by Dr. Joan Gustafson Haworth, Ph.D. Dr. Haworth disputes plaintiffs’ statistical evidence, arguing that Dr. Killings-worth’s reports fail to support class certification due to faulty methods. (D.Mem.16.)
Plaintiffs are high-level female employees who work in the Northeast region. They seek to represent a class including females of officer rank, and officer-level equivalents. An officer holds an officer title, such as Assistant Vice President, Vice President, or Senior Vice President. (D.Mem.13.) Officer titles are awarded to certain senior employees, but the designation does not necessarily affect compensation. (P.Mem.6.) All employees, including these high-level employees, hold functional job titles, such as “Account Executive,” “Account Manager,” or “Assistant Account Manager,” that describe their role (and compensation) more accurately. (P.Mem.6.) After extensive discovery, plaintiffs move for certification of a class consisting of “all current and former female employees who have been employed by the defendants in positions eligible for the award of officer titles” between 1998 and 2001. (P.Mem.2.) These individuals hold positions at “officer-level or equivalent.” (P.Mem.7.)
DISCUSSION
I. Standard For Class Certification
To obtain class certification, plaintiffs must demonstrate that the class meets the express requirements of Rule 23(a), known as numerosity, commonality, typicality, and adequacy, and of one of the subsections of Rule 23(b). In re Visa Check/MasterMoney Antitrust Litig., 280 F.3d 124, 132-33 (2d Cir.2001). Plaintiffs must also show that the class meets the implicit threshold requirement of ascertainability. Dunnigan v. Metro. Life Ins. Co., 214 F.R.D. 125, 135 (S.D.N.Y.2003).
Plaintiffs have the burden of establishing that the class satisfies all requirements. Gen. Tel. Co. v. Falcon, 457 U.S. 147, 156, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982). District courts must conduct a “rigorous analysis” of the Rule 23 requirements, which
II. Rule 23(a) Requirements
A. Ascertamability
The requirement of aseertainability, though not expressly mentioned in Rule 23, is fundamental. The class that plaintiffs seek to certify must be readily identifiable so that the court can determine who is in the class, and thus, who is bound by the ruling. See Dunnigan, 214 F.R.D. at 135 (“Plaintiffs must ... demonstrate that the aggrieved class can be readily identified.”); People United for Children, Inc. v. City of New York, 214 F.R.D. 252, 256 (S.D.N.Y.2003) (noting that courts should ensure that the class definition is precise, objective, and presently ascertainable). A class is ascertainable when defined by objective criteria that are administratively feasible, without a subjective determination. Dunnigan, 214 F.R.D. at 135; Fears v. Wilhelmina Model Agency, 02 Civ. 4911 HB, 2003 WL 21659373, at *2 (S.D.N.Y. July 15, 2003).
Initially plaintiffs simply sought to certify a class consisting of women officers at Willis. After learning in discovery that officer status was not actually linked to compensation, plaintiffs attempted to recharacterize the class to include women who were “officer-level and equivalents” in order for the class to more accurately capture individuals affected by Willis’ alleged discriminatory practices. (P. Mem.6.) See also Hnot v. Willis Group Holding Ltd., No. 01 Civ. 6558(GEL), 2004 WL 1794493, at *2-4 (S.D.N.Y. Aug.10, 2004) (describing efforts to amend complaint to include employees at levels “equivalent” to officers).
Plaintiffs’ proposed class includes women of officer rank, as well as women who are eligible for officer rank. Employees of officer rank include individuals with titles of Assistant Vice President, Vice President, or Senior Vice President. (P. Reply 8.) An individual eligible for officer rank holds a functional job title that is also held by a person who is an officer. That is, every officer has a job title that describes that his or her job function, and these same job titles are also held by employees who do not have officer titles. For example, an individual with an officer rank of Assistant Vice President might hold a job title of Account Executive. Those individuals who are also Account Executives are then considered officer-equivalents, even if they are not of officer rank. (Killingsworth Decl. ¶ 9.)
Defendants argue that plaintiffs’ class is ill-defined, contending that plaintiffs have failed to define or identify the non-officers that they seek to include in the class, and that there is no ready means to identify who is and who is not in the class. (D.Mem.2829.) Indeed, at first glance, plaintiffs’ proposed class is not intuitively identifiable. The concept of “officer-equivalents,” on its face, appears imprecise and lacking in clear standards of definition.
In the course of arguing that plaintiffs wrongfully included certain individuals in the class, defendants suggest that plaintiffs’ class is overbroad. For example, defendants cite to a job code that was included in plaintiffs’ analyses that should not be considered an officer-equivalent. (D.Mem.17, 29.) Essentially, they argue that some in the class do not share all the relevant class characteristics. This, however, is properly a question of commonality or typicality, which is addressed below. The true definitional problems appear to have been resolved after the dialogue between the experts through their statistical reports. See Killingsworth Decl. ¶ 11 (describing revised database to address defendants’ criticisms).
B. Commonality
Rule 23(a) requires a showing that common issues of fact or law affect all class members. Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 155 (2d Cir. 2001). Plaintiffs contend that Willis had a common policy of vesting regional and local officers with unfettered discretion in making promotion and compensation decisions, resulting in discrimination against women in high level positions. (P. Mem.7, 8.) This allegation appears to be patterned on that of the plaintiffs in Caridad, where African-American employees sought to certify a class, claiming that company-wide policies delegating authority to department supervisors on employee discipline and promotion were racially discriminatory. Caridad, 191 F.3d at 286. The Caridad defendants argued that an organization-wide pattern of discrimination was implausible due to the decentralization of these decisions. While Judge Walker, in dissent, maintained that a practice of delegating personnel decisions could not constitute a policy or practice to satisfy the commonality requirement, Caridad, 191 F.3d at 296-97 (Walker, J., dissenting), the panel majority rejected this view, holding that a “challenge [to] the subjective components of company-wide employment practices does not bar a finding of commonality under either the disparate treatment or disparate impact model.” Id. at 292. This is so, even when it is likely to be extremely difficult to prove that a grant of discretionary authority has actually resulted in discriminatory practices. Id. at 291. Therefore, plaintiffs’ argument and evidence demonstrating unfettered discretion in regional and local officers may support a finding of commonality.
Moreover, the class at issue here demonstrates greater commonality than plaintiffs in Caridad. The class is limited in geographic scope to the Northeast region. Individuals in the class are not purportedly discriminated against by different managers across the country, but rather, the delegation policy they challenge has subjected them all to discrimination at the hands of the same regional officers. For example, John Kelly was the regional officer of the Northeast region and the local officer responsible for the New York and New Jersey offices for much of the class period. (P. Mem. 4-5; D. Mem. 8 n.
Plaintiffs’ statistical reports also support a finding of commonality. Dr. Killingsworth’s reports show significant disparities in promotion and compensation between high-level women and men at Willis. For example, in 1998, women officers and individuals in “officer-eligible” positions made $30,212 less in salary plus bonus than their male counterparts, with a standard deviation of 3.39.
The main thrust of defendants’ argument is that plaintiffs’ statistical evidence is flawed. This argument, however, has minimal force at the class certification stage. Class certification is a procedural device addressed to the nature of claims and the way in which they will be adjudicated, rather than a substantive issue going to the merits of plaintiffs’ claims. Largely for this reason, Caridad instructs that “[c]lass [plaintiffs need not demonstrate at this stage that they will prevail on the merits,” and that “ ‘statistical dueling’ is not relevant to the certification determination.” Candad, 191 F.3d at 292. Defendants present their own expert report in an effort to demonstrate the unpersuasiveness of plaintiffs’ expert’s findings of discrimination, but this is exactly what the Second Circuit cautions against. Whether or not plaintiffs’ statistical evidence is ultimately sufficient to establish that plaintiffs and the members of their proposed class have actually suffered discrimination, it is certainly adequate to establish that whether or not Willis’s promotion and compensation policies subject class members to discrimination is an issue common to all class members.
Under Caridad, to establish commonality, plaintiffs must show that “the challenged practice is causally related to a pattern of disparate treatment or has a disparate impact.” Id. Plaintiffs’ expert must “control[ ] for various factors that one would expect to be relevant” to promotion and compensation. Id. at 293. See also Watson v. Fort Worth Bank and Trust, 487 U.S. 977, 994-95, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988) (“[C]ausation must be proved; that is, the plaintiff must offer statistical evidence of a kind and degree sufficient to show that the practice in question has caused the exclusion of applicants for jobs or promotions because of their membership in a protected group.”) In Caridad, where the Second Circuit found that plaintiffs alleging racial employment discriminatory practices met the commonality requirement, the plaintiffs’ expert conducted regression analyses controlling for variables such as years of tenure, department and union status, and age, when examining the effect of race on promotion and discipline. Here, plaintiffs’ expert likewise has conducted several regression analyses, controlling for ethnicity, education, age, years of service and military status. (Fourth Report of Dr. Killingsworth (“Fourth Report”) Ex. 15 at 5.) See also Warren v. Xerox Corp., No. 01-CV-2909 (JG), 2004 WL 1562884, at *3 (E.D.N.Y. Jan.26, 2004) (recommending that class certification be granted where plaintiffs’ expert controlled for salary grade, sales experience and seniority in racial employment discrimi
Defendants boldly assert that plaintiffs’ expert witness’s report fails the admissibility test outlined in Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). Defendants cite to cases where the court has refused to admit expert opinion. These citations are unconvincing, however, as many of the cases cited do not involve class certification, and most involve studies that are blatantly problematic. See, e.g., Amorgianos v. Nat’l R.R. Passenger Corp., 303 F.3d 256, 268-69 (2d Cir. 2002) (finding that the expert’s report was inadmissible because he did not follow the methodology that he himself claimed was necessary); Hollander v. Am. Cyanimid Co., 172 F.3d 192, 202-04 (2d Cir.1999) (noting the use of biased statistical adjustments to skew the figures more favorably towards the plaintiff). Reasonable statisticians can disagree about what tests and what controls should be utilized, but at the class certification stage, only a plausible position needs to be set forth. Although the factfinder may ultimately determine that defendants’ expert analysis is more accurate, a commonality finding at this stage is still appropriate. Even if Dr. Killingsworth has failed to include all necessary variables in his regression analyses, this affects only the probativeness of the report and not its admissibility. Bazemore v. Friday, 478 U.S. 385, 400, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986). Plaintiffs’ expert report is not junk science. Cf. In re Visa Check/Mastermoney Antitrust Litig., 192 F.R.D. 68, 78 (E.D.N.Y.2000) (rejecting defendants’ contention that plaintiffs’ expert opinion was “junk science” at the class certification stage, where report was based on economic theory rather than economic or quantitative analysis). Plaintiffs’ expert is a reputable expert in statistics and has applied scientific methodology sufficient for the class certification stage.
Defendants argue that class certification should be denied because plaintiffs failed to present anecdotal evidence. The Second Circuit has made clear that statistical evidence is sufficient to establish a prima facie case in Title VII cases. Robinson, 267 F.3d at 158-59; Rossini v. Ogilvy & Mather, Inc., 798 F.2d 590, 604 (2d Cir.1986). The standard for establishing a prima facie ease is higher than the standard at the class certification stage. Cf. Caridad, 191 F.3d at 292-93 (finding commonality where statistical evidence “tends to establish” that being black affected promotion). Class certification can be based on statistical evidence alone.
Courts do have discretion to require or rely on anecdotal evidence, but they commonly do so where the statistical evidence is weak. See, e.g., Cokely v. N.Y. Convention Ctr. Operating Corp., No. 00 Civ. 4637(CBM), 2003 WL 1751738, at *3 (S.D.N.Y. Apr.2, 2003) (requiring plaintiffs to produce either statistical information or sworn statements when plaintiffs sought certification after only producing one complaint and one affidavit); Ross v. Nikko Sec. Co. Int’l Inc., 133 F.R.D. 96, 98 (S.D.N.Y.1990) (finding both statistical and anecdotal evidence insufficient); Carter v. Newsday, Inc., 528 F.Supp. 1187, 1197 (E.D.N.Y.1981) (“[With] nonexistent statistical discrimination, anecdotal testimony must by itself support an inference of sex-based hiring decisions.”). Anecdotal information can be important in bringing “the cold numbers convincingly to life,” but is not a necessity. Rossini, 798 F.2d at 604, quoting Int’l Brotherhood of Teamsters v. United States, 431 U.S. 324, 339, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977). Moreover, plaintiffs do provide some anecdotal evidence of discrimination, both from the named plaintiffs and from other individuals. (P. Mem.23-24.) In sum, plaintiffs meet the commonality requirement.
Rule 23(a)’s typicality requirement is satisfied when “each class member’s claim arises from the same course of events, and each class member makes similar legal arguments to prove the defendant’s liability.” Robinson, 267 F.3d at 155 (citation omitted). The typicality requirement can be met if plaintiffs’ evidence shows that an employer discriminated in the same general fashion against employees and other class members through the agency of the same key actors. Rossini, 798 F.2d at 598. Hnot and Scheller allege suffering from similar discriminatory practices as the other class members, at the hands of the same group of individuals, and in the same manner, by being paid less than similarly situated men, and by being denied or not being considered for promotions. They have presented evidence that their experience of alleged gender discrimination is typical of other high level female employees in the Northeast region of Willis.
Defendants argue that plaintiffs are not typical of the class because they do not occupy all officer positions in the class, and that neither of the named plaintiffs works in the New York office. (D.Mem.45-49.) However, plaintiffs can satisfy typicality even if there are factual differences between the named representatives and some of the class members, so long as the disputed issue of law or fact “occupy essentially the same degree of centrality” between the named representatives and the class. Caridad, 191 F.3d at 293, quoting Krueger v. N.Y. Tel. Co., 163 F.R.D. 433, 442 (S.D.N.Y.1995). That the named plaintiffs do not hold every single officer or officer-equivalent title or work in every office location does not mean that their legal claims are so different from the class members to bar typicality. See Meyer v. Macmillan Publ’g Co., 95 F.R.D. 411, 414 (S.D.N.Y.1982) (“[T]he fact that the jobs performed by the named plaintiffs are ... unique ... is not a bar to their being class representatives. If it were, no class of professional employees could ever be certified.”).
Defendants also point out that the named representatives are higher level officers that some class members. Even if one female officer supervised another, it is still possible, as plaintiffs allege, that they all suffered from gender discrimination by the key decisionmakers. Defendants’ reliance on Sheehan v. Purolator, Inc., 103 F.R.D. 641 (E.D.N.Y.1984), is misguided. In Sheehan, high-level executives were held to be inappropriate representatives for a class that included lower level staff such as secretaries due to major factual differences with the class claims. Id. at 650-51. In contrast, here both the named plaintiffs and potential class members here are all officers or eligible for officer titles. As discussed above, class members all hold functional job positions that would permit them to be awarded an officer title. Therefore, plaintiffs meet the typicality requirement.
D. Numerosity
Defendants acknowledge that plaintiffs meet the numerosity requirement. If a proposed class is forty members or more, the numerosity requirement is met. Consol. Rail Corp. v. Town of Hyde Park, 47 F.3d 473, 483 (2d Cir.1995). In this case, even counting only women who held officer titles between October 1998 to December 2001, excluding the officer-equivalents, the class would include 106 women. (White Deck ¶ 11.)
E. Adequacy of Representation
Plaintiffs also meet the requirement of adequacy of representation. To determine whether a named plaintiff will be an adequate class representative, courts inquire whether: “(1) plaintiffs interests are antagonistic to the interest of other members of the class and (2) plaintiffs attorneys are qualified, experienced, and able to conduct the litigation.” Baffa v. Donaldson, Lufkin & Jenrette Sec. Corp., 222 F.3d 52, 60 (2d Cir. 2000). Both factors are satisfied here.
First, there is no conflict of interest between the named plaintiffs and class interests. The named plaintiffs and proposed class members are all relatively high level women employees. Defendants point out that one of the named plaintiffs, Hnot, conducted supervisory reviews of other female officer-level employees. (D.Mem.47.)
Second, counsel for plaintiffs are qualified and competent. They have ably and zealously represented the interests of the class so far, and are experienced in employment class action suits. (Firm Biography of Cohen Milstein, Hausfeld & Toll, P.L.L.C. Ex. 23; Firm Biography of Warshaw, Burstein, Schlesinger, & Kuh, L.L.P.Ex. 24) Therefore, the plaintiffs adequately represent the class.
III. Rule 23(b) Requirements
Finally, to qualify for class certification, plaintiffs must meet the requirements of one of the subsections of Rule 23(b). Plaintiffs seek to certify a class here under either Rule 23(b)(2) or Rule 23(b)(3). 'The Court finds that plaintiffs meet the requirements under Rule 23(b)(2) on the issues of liability and injunctive relief, and defers consideration of whether Rule 23(b)(3) requirements are met as to individual damages.
Rule 23(b)(2) requires that the defendant “has acted or refused to act on grounds generally applicable to the class,” thus making final injunctive or declaratory relief appropriate for the class. Fed.R.Civ.P. 23(b)(2). Defendants here have purportedly acted against the whole class that has resulted in discriminatory pay and promotions. To prevail at the liability stage of the case, plaintiffs will be required to prove that defendants have engaged in a pattern or practice of disparate treatment, or that their actions result in a disparate effect on the group. If they succeed, then plaintiffs will be entitled to the injunctive and declaratory relief that they seek. Latino Officers, 209 F.R.D. at 92.
Class treatment under Rule 23(b)(2) is not appropriate if plaintiffs mainly seek monetary relief. In Robinson, the Second Circuit rejected the “bright-line” approach, finding that the absence of all claims of compensatory and punitive damages is not necessary to certify a class. 267 F.3d at 164 (emphasis omitted). Instead, courts should assess the relative importance of the remedies sought, given the context of the case. Id. Plaintiffs here do ask for monetary damages, but declaratory, injunctive, and equitable relief predominate. Plaintiffs seek to reform defendants’ practices to provide for equitable employment opportunities and compensation for women. At any rate, the primary monetary relief sought, damages for back pay awarded under Title VII, is considered equitable relief. Robinson, 267 F.3d at 157. Consequently, plaintiffs meet the requirement under Rule 23(b)(2).
Because this Court finds that plaintiffs meet the requirements under Rule 23(b)(2) for class certification, it is unnecessary at this point to determine whether plaintiffs’ class is also appropriate for (b)(3) class certification. See id. at 167 n.12 (nothing that the court was not required to respond to plaintiffs’ request for (b)(3) certification after finding the class appropriate for (b)(2) certification). Additionally, Title VII civil rights cases may be divided into liability and remedial phases. Id. at 158. Therefore, the issue of whether plaintiffs meet the requirements of Rule 23(b)(3) is deferred until after the liability phase of this action.
CONCLUSION
Plaintiffs have demonstrated that their proposed class and its representatives satisfy the requirements of Federal Rule of Civil Procedure 23(a) and 23(b). Accordingly, plaintiffs’ motion for class certification is
SO ORDERED:
. For example, the same individual, John Kelly, was the CEO for Willis New York for nine years, and the REO for the Northeast region for three years. For a number of years, Willis New Jersey did not have its own CEO, and was run by the New York CEO until 2000. Willis Massachusetts had its own CEO. During the class period, there was little turnover in these positions. (D.Mem.8 n. 12.)
. Defendants also assert that plaintiffs’ expert report incorrectly included individuals who are not part the proposed class. (D.Mem.16-19.) Plaintiffs counter by claiming that defendants provided the wrong database from which the reports were run. (P. Mem. 11.) However, these arguments are irrelevant in determining the ascertainability of a potential class. The key question is whether plaintiffs can demonstrate that members of their proposed class are identifiable when utilizing objective standards. See Fears, 2003 WL 21659373, at *3 (finding an ascertainable class because whether models had an oral
. Dr. Killingsworth reports that a standard deviation figure of 1.96 or greater is statistically significant. (Third Report Ex. 14 at 3.)
. Dr. Killingsworth also references several academic articles, including articles cited by defendants' expert, in the field of statistics and economics that use the variables of education, age, and race in analyzing earnings. (Fourth Report at 5 n. 3.)