Summary
noting that cases implicating "numerous, independent decision-makers" result "in the need for numerous individual inquiries"
Summary of this case from Case v. JuddOpinion
[Copyrighted Material Omitted] [Copyrighted Material Omitted] [Copyrighted Material Omitted]
Mary Stowell, Linda Debra Friedman, Suzanne E. Bish, Richard E. Russell, Stowell & Friedman, Ltd., Chicago, IL, Johanna J. Raimond, Law Offices Of Johanna J. Raimond Ltd., Chicago, IL, Michael M. Mulder, Thomas R. Meites, Shona B. Glink, Paul William Mollica, Meites, Mulder, Mollica & Glink, Chicago, IL, Rafael Eduardo Lazaro, Seyfarth Shaw LLP, Chicago, IL, for Plaintiffs.
Richard Cartier Godfrey, Donna M. Welch, Kirkland & Ellis LLP, Chicago, IL, Anneliese Wermuth, Brian Wegg Bulger, Eileen Elizabeth Baker, Paul R. Garry, Julie Lynn Trester, Meckler, Bulger & Tilson, Chicago, IL, Bredale Rucker, Katheleen A. Ehrhart, Khara AishaAshanti Coleman, Rachel L. Perrier, Sallie Gamble Smylie, Wendy Lynn Bloom, Kirkland & Ellis LLP (Chicago), Chicago, IL, Thomas E. Dutton, Greenberg Traurig, Chicago, IL, for Defendant.
On January 11, 2005, by consent of all parties and pursuant to 28 U.S.C. § 636(c), this case was assigned to the Honorable Michael T. Mason for all proceedings, including entry of final judgment (doc. 18, 19). Thereafter, upon Judge Mason's recusal from the case, on February 11, 2005, the case was reassigned to his Court (doc. # 23).
SIDNEY I. SCHENKIER, United States Magistrate Judge.
Katherine Puffer (" plaintiff" or " Ms. Puffer" ) filed this suit on behalf of herself and a putative class alleging that Allstate Insurance Company (" Allstate" ) has carried out a nationwide pattern or practice of sex discrimination in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. § 2000e et seq., and the Equal Pay Act, 29 U.S.C. §§ 206 and 207 (doc # 13: Am. Compl. at 11-12, ¶ 4). Plaintiff has moved to certify the following class under Federal Rule of Civil Procedure 23(b)(2) and (3): " All female managers holding the position of Division or Department manager and above who worked or continue to work at Allstate Protection at any time between May 9, 2001 and the present and who have been and continue to be injured by the alleged discriminatory employment policies and practices" (doc. # 126: Pl.'s Class Cert. Mem. at 21). The parties have conducted extensive fact and expert discovery, and have submitted equally extensive briefing, on plaintiff's motion for class certification. For the reasons that follow, plaintiff's motion for class certification (doc. # 125) is denied.
I.
We begin our discussion with a brief introduction to Allstate's organizational structure, followed by a summary of Ms. Puffer's employment history at Allstate.
A.
Prior to 2002, Allstate organized its Property and Casualty (" P & C" ) business into four operational areas: Claims, Finance, Product Operations, and Distribution (doc. # 200: Def.'s Opp'n to Class Cert. at 2). In October 2002, Allstate created a new organization called Allstate Protection (" Protection" ). Protection is the largest business unit within Allstate, employing approximately 80 percent of all Allstate personnel (Pl.'s Class Cert. Mem. at 7-8). Allstate Protection consists of different Areas of Responsibility (" AORs" ), each headed by a Vice President or Senior Vice President, i.e., " senior managers" ( id. ). The four largest AORs within Protection are Claims, Product Operations, Protection Finance, and Marketing/Distribution ( id. at 8). Allstate Protection also includes Encompass, three different marketing groups, and the Technology Shared Services Group (Def.'s Opp'n to Class Cert. at 2). During the class period, women constituted 61 percent of the employees in the four largest AORs within Protection, 36 percent of the managerial positions (Salary Grades (" Grades" ) 63 and above), and 10 percent of the senior manager positions ( id. at 2-3; Pl.'s Class Cert. Mem. at 7-8).
Employee responsibilities, supervisors, office locations, and salary grade vary widely among the four primary AORs in Protection (Def.'s Opp'n to Class Cert. at 6-7). The Claims AOR employs attorneys, human resources and IT professionals, claims investigators, claims and subrogation service managers, and project managers ( id. ). Two-thirds of the putative class worked or works in Claims ( id. at 7). The Product Operations AOR, by contrast, focuses on pricing, strategy, product and design, risk management, and research analytics, and consists of, among other employees, pricing directors and actuaries ( id. at 8). Likewise, the Finance AOR houses a different set of employees, with their own set of technical and professional skills; and in the Marketing/Distribution AOR, sales-related skills and experience are most important ( id. at 7-8).
In addition to the variety of job types in which members of the putative class are employed, plaintiff's proposed class includes a wide spectrum of managerial level positions. Eighty percent of the putative class consists of women employed in entry-level, non-officer management positions at Grades 63 and 64 or below ( id. at 6 n. 5). However, the putative class also includes women employed in senior manager positions at Grade 77, director positions at Grade 78, appointed officer positions (Assistant Vice President and Assistant Field Vice President) at Grade 80, and elected officer positions (Vice President, Field Vice President, and Senior Vice President) who are at Grade 90 and above ( id. at 5-6). There are only 200 officer positions at Allstate in a company of 38,000 employees ( id. at 3).
Supervisors are required to meet with each manager who reports directly to them at the beginning of the year to discuss major job responsibilities and individual performance goals for the manager over the next 12 months ( id. at 9). At the end of the year, the supervisor fills out a performance development summary (" PDS" ) for each manager who reports to them, based on the major job responsibilities and individual performance goals set for him or her the previous year ( id. ). The numeric ratings of the manager's major responsibilities are weighted and then tabulated to reach an overall evaluation rating ( id. at 9-10). The supervisor then submits the PDS to his or her own direct superior for input, review, and approval before it goes to the evaluated manager in a process called " one-over-one" review ( id. ). In addition, Allstate evaluates its managers based on 17 critical success factors (" CSFs" ) as well as Quality Leadership Management Surveys, which are filled out by the employees who directly report to the managers' being reviewed ( id. at 10-11). These evaluations are factored into decisions on merit or promotional increases in salary for managerial employees.
Protection is allotted a percentage amount of money each year to be used for merit or promotional salary increases (Pl.'s Class Cert. Mem. at 16). Allstate's Human Resources department (" HR" ) distributes annual salary administration guidelines to management supervisors, which set forth an overall range of salary increases in terms of a percentage amount for each performance and salary level ( id. ). Using these guidelines, together with the evaluations described above, Allstate determines merit and promotional increases for all managerial employees. This process is done through HR, with input from Protection management ( id. ). In Grades 63 to 78, pay decisions are based on the manager's performance and salary range (Def.'s Opp'n to Class Cert. at 12). Only managers in Grades 77 and above are bonus-eligible ( id. ). Allstate's 12-person senior management team is only involved in pay and promotion decisions for the 200-member officer group ( id. at 11).
Allstate generally does not post openings for managerial positions (Pl.'s Class Cert. Mem. at 11). Rather, an employee's supervisor is responsible for recommending employees for promotion when positions become vacant (Def.'s Opp'n to Class Cert. at 13). Allstate employees, however, many change jobs without going through formal promotion procedures through use of the " broadbanding system" adopted by Allstate in 2000 (Pl.'s Class Cert. Mem. at 10-11). Broadbanding groups hundreds of jobs with similar levels of responsibility and accountability into the same " job band," with each job band containing positions covering a wide salary range ( id. at 2). Broadbanding allows senior managers to change an employee's responsibilities, salary, and job, without going through formal procedures, so long as the employee stays within the same band ( id. at 10). Broadbanding gives employees cross-training and developmental opportunities in other jobs within the same band ( id. at 11.)
Job moves within a salary band, with the cross-training opportunities that go along with them, may play a role in succession planning decisions, through which senior management identifies talented and qualified candidates for more senior positions that may become available later ( id. at 12). Personality and other subjective traits, as well as performance evaluation results and an employee's prior experience, factor into succession planning decisions ( id. at 14; Def.'s Opp'n to Class Cert. at 13). Succession planning decisions are generally formalized at least once a year in succession planning and talent management meetings between senior managers in Protection and HR (Pl.'s Mot. for Class Cert. at 13). The heads of the AORs make the final decisions regarding promotional and advancement opportunities for individuals in Grades 63 and 64, and they provide recommendations to senior officers for the advancement and promotion of individuals in Grade 77 and above ( id. at 14).
B.
Ms. Puffer began her employment with Allstate in 1977 in an entry level accounting position in the Corporate Accounting department (Def.'s Opp'n to Class Cert. at 3). By 1993, after being promoted eight times, she rose to the officer position of Assistant Vice President in Finance and Planning ( id. ). Ms. Puffer worked in Finance for most of her 26-year Allstate career ( id. at 8). Ms. Puffer worked in the Policies and Forms unit within the Product Operations department from April 1998 to October 2000, but she never worked in the actuarial side of that department, the largest section in Product Operations ( id. at 8). She has not worked in a Grade 63 or 64 equivalent job since 1986 ( id. at 6 n. 5). Ms. Puffer never worked in Claims or Distribution ( id. at 7).
Ms. Puffer alleges that she was sexually harassed by her immediate supervisor from late 1998 to mid-1999, and that she was passed over for promotion and advancement as a result of her rejecting the supervisor's advances and because of her gender (Am.Compl.¶ ¶ 19-23). She alleges that Allstate failed to conduct an adequate investigation into those matters ( id. ¶ 24.) Based on those alleged events, in January 2001, Ms. Puffer filed a charge of gender discrimination, sexual harassment, and retaliation with the Equal Employment Opportunity Commission (" EEOC" ). Ms. Puffer claims that after she filed that EEOC complaint, Allstate transferred her to a new department, cut her responsibilities in half, and refused to finance her business school education ( id. ¶ ¶ 27-29). Plaintiff chose not to file a lawsuit within the required time period after she received a right to sue letter from the EEOC.
In May 2003, Ms. Puffer was terminated from her position at Allstate. At that time, Ms. Puffer held the title of Assistant Vice President, which was a Grade 80 position (Pl.'s Class Cert. Mem. at 1). Plaintiff was one of fourteen officers terminated at that time: eleven were male, and three were female (Def.'s Opp'n to Class Cert. at 1). Ms. Puffer then filed a new charge with the EEOC, alleging gender discrimination and retaliation (Am.Compl.¶ 41). After receiving a right to sue letter from the EEOC, Ms. Puffer timely filed the complaint in this case. As amended, the complaint asserts three claims.
In Count 1, Ms. Puffer alleges that Allstate discriminated against her and a class of female managerial employees because of their gender, and allowed the discrimination " to exist and go unremedied for so long that it amounts to a policy or practice and constitutes Allstate's standard operating procedure" (Am.Compl.¶ 52). Plaintiff alleges this policy or practice under both " differential treatment and disparate impact theories of liability under Title VII" ( id. ¶ 53). In Count II, Ms. Puffer alleges a class claim under the Equal Pay Act, alleging that Allstate has paid (and continues to pay) putative class members " lower wages than male employees in substantially equal jobs even though plaintiffs and all others similarly situated performed similar duties requiring the same skill, effort, and responsibility of male employees" and performed " substantially equal work" ( id. ¶ ¶ 57, 59). Plaintiff claims that the difference in pay between the sexes " was not pursuant to seniority, merit, quantity or quality of production, but was due to sex" ( id. ¶ 58). In Count III, Ms. Puffer alleges an individual claim of retaliation in violation of Title VII and the Equal Pay Act.
In " disparate treatment" cases, proof of discriminatory motive is critical, although in some situations the requisite motive can be inferred from the fact of differences in treatment. Int'l Bhd. of Teamsters v. United States, 431 U.S. 324, 335, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977). In disparate impact cases, by contrast, motive is irrelevant. EEOC v. Chicago Miniature Lamp Works, 947 F.2d 292, 297 (7th Cir.1991).
Plaintiff's class certification papers do not address the appropriateness of class certification under the Equal Pay Act. That kind of undeveloped argument likely constitutes a waiver, but, in any event, a similar analysis would apply to the Equal Pay Act claim as applies to plaintiff's disparate impact claim, neither of which requires a showing of intent. Warren v. Solo Cup Co., 516 F.3d 627, 629 (7th Cir.2008).
Since the matter is before the Court solely on plaintiff's motion for class certification, we express no view on the merits of her individual retaliation claim.
II.
A plaintiff seeking class certification bears the burden of satisfying all of the criteria listed in Federal Rule of Civil Procedure 23(a)-numerosity, commonality, typicality, and adequacy of representation-as well as at least one subsection of Rule 23(b). Arreola v. Godinez, 546 F.3d 788, 797 (7th Cir.2008). The determination of class certification under Rule 23(a) turns not on the ultimate merits of the case, but rather on whether the party seeking certification meets its burden of showing all the certification requirements of Rule 23(a). Gen. Tel. Co. v. Falcon, 457 U.S. 147, 161, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982). " Failure to meet any of the Rule's requirements precludes class certification." Arreola, 546 F.3d at 794.Rule 23 gives the district courts " broad discretion to determine whether certification of a class-action lawsuit is appropriate." Id. That said, the Seventh Circuit " suggest[s] caution in class certification generally." Thorogood v. Sears, Roebuck and Co., 547 F.3d 742, 746 (7th Cir.2008).
The Seventh Circuit has directed that, " [b]efore deciding whether to allow a case to proceed as a class action, ... a judge should make whatever factual and legal inquiries are necessary under Rule 23." Szabo v. Bridgeport Machs., Inc., 249 F.3d 672, 676 (7th Cir.2001). When a plaintiff seeks certification under Rule 23(b)(3), the Court should " make a preliminary inquiry into the merits" of the claims, to decide whether " the difficulties likely to be encountered in the management of the class action" preclude certification. Id. As this Court has previously stated:
We do not believe that Szabo directs district courts to decide class certification questions based on a preliminary assessment of the ultimate merits of the plaintiffs' claims: to base class certification on a prediction of who will win the case would be at odds with Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 94 S.Ct. 2140, 40 L.Ed.2d 732 (1974). In our view, the " preliminary inquiry into the merits" discussed in Szabo is a more limited one that has as its focus not the substantive strength or weakness of the plaintiffs' claims, but rather whether the path that will need to be taken to decide the merits renders the case suitable for class treatment.
Harris v. Circuit City Stores, Inc., No. 07 C 2512, 2008 WL 400862, at *3-4 (N.D.Ill. Feb, 7, 2008). " It is in this limited sense that the Court assesses the ‘ merits' of plaintiffs' allegations in considering the class certification motion." Id. at *4.
In Parts II.A. through II.D. below, we consider whether plaintiff has satisfied the requirements of Rule 23(a). In Part III, we then address whether plaintiff has met the requirements of at least one of the subsections of Rule 23(b).
We note that in addition to meeting the requirements expressly set forth in Rule 23(a) and (b), a plaintiff seeking class certification also must show that her proposed claim is " sufficiently definite to permit ascertainment of the class members," Alliance to End Repression v. Rochford, 565 F.2d 975, 977 (7th Cir.1977). Allstate does not argue that plaintiff's proposed class definition fails to meet this standard, and so we do not discuss it further in this opinion.
A.
Rule 23(a)(1) requires that the class be " so numerous that joinder of all members is impracticable." Fed. R, Civ. P. 23(a)(1). Plaintiff asserts-and Allstate does not contest-that there are more than 1,700 women in Allstate Protection who could be covered by the proposed class definition (Pl.'s Class Cert. Mem. at 22). This number of putative class members easily meets the numerosity requirement of Rule 23(a)(1). See, e.g., Acik v. I.C. Sys., Inc., 251 F.R.D. 332, 335 (N.D.Ill.2008) (class consisted of at least 100 individuals); Cavin v. Home Loan Ctr., Inc., 236 F.R.D. 387, 392 (N.D.Ill.2006) (collecting cases).
In her amended complaint, Ms. Puffer alleges discrimination against female employees in Allstate's Property and Casualty business, rather than Allstate Protection (Am.Compl.¶ 4). Although defendant briefly notes this discrepancy between plaintiff's proposed class as alleged in the complaint and asserted in the class certification motion (Def.'s Opp'n to Class Cert. at 5 n. 4), defendant does not assert this discrepancy as a basis to deny class certification.
B.
The second requirement of Federal Rule of Civil Procedure 23(a) presents a far more difficult challenge for plaintiff-one that, in our judgment, plaintiff fails to meet. Rule 23(a)(2) requires that a plaintiff seeking class certification show that " there are questions of law or fact common to the class." Fed.R.Civ.P. 23(a)(2). " A common nucleus of operative fact is usually enough to satisfy the commonality requirement of Rule 23(a)(2)." Keele v. Wexler, 149 F.3d 589, 594 (7th Cir.1998). Plaintiff contends that the proposed class members are commonly linked by the allegedly discriminatory compensation, promotion, and training policies and practices at Allstate Protection. Plaintiff argues that these employment policies and practices are excessively discretionary, vesting all authority to choose, advance, and compensate management employees in the male-dominated, senior manager ranks, thereby allowing senior managers " to apply their own subjective views, without explanation or accountability, of whom to promote or develop" (Pl.'s Class Cert. Mem. at 1). Specifically, plaintiff argues that: Allstate's promotion decisions are excessively subjective because position openings are not posted; Allstate's decisions on merit or promotional increases are excessively subjective because senior management can distribute the money allotted for these salary increases at their discretion; and succession planning decisions are excessively subjective because the process is secretive, and the criteria used by senior management to determine who will be chosen are subjective and tend to be influenced by personalities and not by objective performance measures ( id. at 9-11, 14-16), Plaintiff further contends that Allstate's adoption of the " broadbanding system" only increased the subjectivity of pay, promotion, and development decisions, because senior managers no longer had to go through the formal promotion procedures ( id. ). Plaintiff asserts that this subjectivity leads Allstate to consciously (and unconsciously) favor men in these employment decisions, with the result that " women earned significantly less than their male counterparts, were disproportionately segregated into the lower paying jobs, were denied equal access to training and career advancement opportunities and were otherwise subjected to the same enterprise-wide employment practices" ( id. at 8-9, 23).
1.
At the threshold, Allstate argues that its employment practices cannot constitute a company-wide policy of discrimination unless they are " entirely subjective" (Def.'s Opp'n to Class Cert. at 24-25). For that proposition, Allstate relies on a footnote from the Falcon decision, in which the Supreme Court stated that " [s]ignificant proof that an employer operated under a general policy of discrimination conceivably could justify a class of both applicants and employees if the discrimination manifested itself in hiring and promotion practices in the same general fashion, such as through entirely subjective decisionmaking processes," Falcon, 457 U.S. at 159 n. 15, 102 S.Ct. 2364.
Courts of appeals that have considered the issue are divided on the significance of the Falcon footnote. The Sixth Circuit has construed this footnote to create a bright-line rule: class treatment is appropriate only if an employment practice challenged because of its subjectivity is entirely subjective. Bacon v. Honda of Am. Mfg., Inc., 370 F.3d 565, 571 (6th Cir.2004). On the other hand, the Ninth Circuit has read Falcon as identifying an " entirely subjective" decision-making process as an example of one that could warrant class certification, without excluding the possibility that decision-making processes that blend subjective and objective elements also might qualify for class certification. Staton v. Boeing, Co., 327 F.3d 938, 955 (9th Cir.2003). This distinction is significant here, as there is no question that Allstate's employment process is not entirely subjective, but blends subjective and objective elements.
The Seventh Circuit has not held forth on the meaning of the Falcon footnote. In the absence of direction from the Seventh Circuit, we find that the Staton analysis comports with the plain meaning of the language the Supreme Court selected. The words " such as" commonly introduce a non-exclusive list, and we see no reason to conclude that the Supreme Court in Falcon intended that those words have a different meaning. Accordingly, we reject Allstate's contention that we may not certify a class merely because its employment decision-making process is not wholly subjective.
That said, we hasten to add that nothing in Title VII prohibits employers from using subjective criteria in their employment practices, Sattar v. Motorola, Inc., 138 F.3d 1164, 1170 (7th Cir.1998). Indeed, it is hard to envision a system that could thoroughly evaluate the suitability of persons for management jobs, or their performance in those positions, without taking into account subjective assessments of qualities that objective criteria may be unable to capture. As the Seventh Circuit has stated:
Subjective evaluations of a job candidate are often critical to the decisionmaking process, and if anything, are becoming more so in our increasingly service-oriented economy.... Personal qualities ... factor heavily into employment decisions concerning supervisory or professional positions. Traits such as common sense, good judgment, originality, ambition, loyalty, and tact often must be assessed primarily in a subjective fashion, yet they are essential to an individual's success in a supervisory or professional position.
Blise v. Antaramian, 409 F.3d 861, 868 (7th Cir.2005) (quoting Chapman v. A.I. Transport, 229 F.3d 1012, 1033-34 (11th Cir.2000)); see also Watson v. Fort Worth Bank & Trust, 487 U.S. 977, 999, 108 S.Ct. 2777, 101 L.Ed.2d 827 (1988) (" It is self-evident that many jobs, for example those involving managerial responsibilities, require personal qualities that have never been considered amenable to standardized testing" ).
As Falcon stated, a plaintiff seeking class treatment for an alleged general policy of discrimination must offer " significant proof" that the alleged policy manifested itself in the " same general fashion" as to all putative class members. Falcon, 457 U.S. at 159 n. 15, 102 S.Ct. 2364. Subjective employment processes may provide a common basis for a classwide claim of discrimination if the subjective employment criteria were uniformly " used as a mask for discrimination," or " to evade statutory anti-discrimination rules." Millbrook v. IBP, Inc., 280 F.3d 1169, 1176 (7th Cir.2002) (internal citations and quotations omitted). See, e.g., Robinson v. Metro-North Commuter R.R. Co., 267 F.3d 147, 156 (2d Cir.2001) (holding that the delegation of discretionary authority to supervisors for discipline and promotion constituted a policy or practice sufficient to satisfy the commonality requirement). After careful consideration, we conclude that plaintiff has failed to make that showing. Plaintiffs attempt to establish commonality is undermined by: (a) the diversity of both the proposed class members in terms of their job duties, and the people who made the relevant employment decisions; (b) the limitations of plaintiff's statistical evidence; (c) the conflicting anecdotal evidence found in the declarations offered by each side; and (d) the unconvincing nature of plaintiff's social science evidence concerning a " culture of paternalism" at Allstate, as that evidence bears on class certification. We address each of these in turn.
a.
To satisfy the commonality requirement, Allstate's employment policies or practices must have affected the putative class members in the same or similar ways. Thorogood, 547 F.3d at 747-48 (reversing district court's certification of class where plaintiff's claim depended on class members' uniform understanding of the significance of Sears's labeling or advertising on clothes dryers). See also Donaldson v. Microsoft Corp., 205 F.R.D. 558, 567 (W.D.Wash.2001) (holding that the proposed class included such a wide range of employees, subject to such a wide range of criteria for individual performance evaluations, that " [a]bsent some evidence that Microsoft's benefits system effected [ sic ] these various employees in the same or similar ways, the enormous diversity of this putative class defeats a finding of commonality" ), Allstate argues that plaintiff cannot show commonality here, because " [t]he staggering heterogeneity of plaintiff's proposed class requires answering questions about decisions made by hundreds of managers, across varied business divisions and segments, with employees who perform entirely different functions, who hold hundreds of different jobs, and who are scattered across a multitude of regions and office locations" (Def.'s Opp'n to Class Cert. at 25-26). We agree.
As explained above, the putative class members were supervised and reviewed by many different people; had a wide variety of salary levels, jobs, and responsibilities; and had annual goals individually tailored to them. Several decisions in this district have denied class certification in discrimination cases where a plaintiff's class definition implicated numerous, independent decision-makers, resulting in the need for numerous individual inquiries. See, e.g., Ellis v. Elgin Riverboat Resort, 217 F.R.D. 415, 424 (N.D.Ill.2003); Williams-Ellis v. Mario Tricoci Hair Salons and Day Spas, No. 05 C 5030, 2008 WL 294731, at *4 (N.D.Ill. Jan.31, 2008); Sandoval v. City of Chicago, No. 07 C 2835, 2007 WL 3087136, at *6 (N.D.Ill. Oct.18, 2007); but see Smith v. Nike Retail Servs., Inc., 234 F.R.D. 648, 661 (N.D.Ill.2006) (finding commonality in question of whether Nike discriminated in assigning positions to African-American employees by means of a corporate policy that left hiring decisions to the unfettered discretion of local level managers).
Ms. Puffer nonetheless argues that Allstate's " salary administration, performance evaluation, job assignment, and advancement policies and practices are uniform throughout Allstate Protection" (doc. # 247: Pl.'s Class Cert. Reply at 7). In aid of that argument, Ms. Puffer offers the opinion of Dr. Cristina Banks, an industrial organization psychology expert, that " Allstate has developed uniform policies and practices related to performance management, leadership assessment, succession planning, job assignment, placement and compensation that have been applied broadly across Allstate Protection" (doc. # 239: Ex. 5, Banks Report at 2), She reasons that Allstate's policies open up the possibility of bias in performance appraisals because senior managers assign employees major responsibilities, performance standards, and CSFs at their discretion ( id. at 15-17, 19).
However, Dr. Banks concedes that evaluations and CSFs are not applied consistently between regions or between supervisors in the same regions, and that their assessments of employees are not comparable; " essentially their evaluations do not permit ‘ apples to apples' comparisons" ( id. at 17-18, 27). This incomparability undermines plaintiff's claim of commonality: even if one supervisor used Allstate's subjective employment practices " as a mask for discrimination," that does not inexorably lead to the conclusion that any other supervisors did so. See, e.g., Morrison v. Panduit Corp., No. 03-3081, 2004 WL 3049816, at **2, 5 (7th Cir. Dec.21, 2004) (holding that evidence did not show that vice president discriminated against employee because vice president was not involved in many of the employment decisions at issue). Dr. Banks's opinion evidence does not overcome the heterogeneity of the class members and individual decision makers that weighs heavily against a finding of commonality.
b.
In seeking to establish commonality, plaintiff places great weight on statistical evidence. While we are mindful of the importance of statistical evidence, we likewise must acknowledge that " [i]t is said that you can find a statistic to prove anything." Barricks v. Eli Lilly and Co., 481 F.3d 556, 559 (7th Cir.2007). As Mark Twain once said, " [t]here are three kinds of lies-lies, damned lies and statistics." Id. Thus, the Court recognizes both the potential power of statistics, and their potential frailty.
" Gross statistical disparities" alone may constitute prima facie proof of a pattern or practice of discrimination in a proper case, but statistical evidence alone is rarely sufficient. Hazelwood Sch. Dist. v. United States, 433 U.S. 299, 307-08, 97 S.Ct. 2736, 53 L.Ed.2d 768 (1977). For example, " [s]tatistical evidence which fails to properly take into account nondiscriminatory explanations does not permit an inference of discrimination." Radue v. Kimberly-Clark Corp., 219 F.3d 612, 616-17 (7th Cir.2000). Although an expert need not include all measurable variables in conducting a statistical analysis, see Bazemore v. Friday, 478 U.S. 385, 400, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986), certain factors are crucial to statistical evidence in Title VII cases, such as identifying not only those individuals who are qualified for a position, but also those who are potentially interested in it. Bennett, 295 F.3d at 697. The Seventh Circuit has explained that " [i]n order to be considered, the statistics must look at the same part of the company where the plaintiff worked; include only other employees who were similarly situated with respect to performance, qualifications, and conduct; [and] the plaintiff and the other similarly situated employees must have shared a common supervisor...." Balderston v. Fairbanks Morse Engine Div. of Coltec Indus., 328 F.3d 309, 319-20 (7th Cir.2003). " [W]hen the statistical evidence does not adequately account for the diverse and specialized qualifications necessary for the positions in question, strong evidence of individual instances of discrimination becomes vital to the plaintiff's case." E.E.O.C. v. Sears, Roebuck & Co., 839 F.2d 302, 311 (7th Cir.1988) (internal citations and quotations omitted).
Dr. Janice Madden, plaintiff's statistical expert, performed a statistical analysis of Allstate's personnel data, which included compensation and employment records for any person holding an exempt salary band or officer level position at any time between 1999 (at least two-and-a-half years prior to the class period as plaintiff defines it) and 2006, to study gender differentials in compensation, job assignment, and advancement (Pl.'s Class Cert. Mem, at 2-3), Plaintiff submitted Dr. Madden's original expert report on November 6, 2007; a " Revised Report" and a " Rebuttal Report" on July 22, 2008; a " Revised Rebuttal Report" on August 18, 2008; and another " Revised Rebuttal Report" on August 21, 2008. The revised reports purport to address certain criticisms of Dr. Madden's original analysis that were pointed out by Defendant's statistical expert, Dr. Robert Topel, while the original and rebuttal reports contain substantive research, analyses, and conclusions.
On October 24, 2008, plaintiff filed a motion for leave to file a " Declaration of Dr. Janice Madden to Correct the Record; " i.e., to correct certain errors Dr. Madden made on a table attached to her rebuttal report (doc. # 253). The new declaration, however, appeared to contain new research and analysis in addition to the corrections. On November 12, 2008, this Court denied plaintiff's motion (doc. # 257).
Dr. Madden performed a multiple regression analysis using five different models, controlling for various combinations of age, tenure at Allstate, education, salary grade, job code, location, and time in grade (Pl.'s Class Cert. Mem. at 5). Based on her analyses, Dr, Madden concluded that: " women earn significantly less than men within (the same) Salary Bands in each year between 2001 and 2005; women are steered by Allstate into jobs that pay significantly less, on average, than the other jobs within the band; and the percentage of female representation in these lower paying jobs is significantly greater than the female representation in the band as a whole. Men, on the other hand, are assigned to male-dominated jobs where they earn significantly more" ( id. at 3).
Multiple regression analysis uses statistics to compare characteristics, or factors, within a particular set of data to determine how one set of characteristics ( e.g., qualifications) is related to another, single characteristic ( e.g., wages). Thomas J. Campbell, Regression Analysis in Title VII Cases, 36 Stan. L.Rev. 1299, 1300 (1984). In employment discrimination cases, regression analysis is used to factor out the effect of influences other than alleged discrimination and to test whether the remaining influence of alleged discrimination is statistically significant. Id. at 1302.
Defendant has moved to strike Dr. Madden's report on the grounds that it is irrelevant to this case because it fails to identify whether any of the pay differences she found were the result of decisions made within the class period, and it fails to link any pay differential that she found to any Allstate policy or practice (doc. # 206: Def.'s Mot. to Strike Madden at 1-2). The Court disagrees that these deficiencies in Dr. Madden's report make it irrelevant or inadmissible; however, the Court finds limited probative value in Dr. Madden's report on the issue of class certification.
First, Dr. Madden originally calculated what the Court will refer to as the " total difference" between male and female employees' salary each year, rather than the " incremental difference" in compensation each year. For example, if in 2001, men in a certain salary grade earned $100 and women of the same grade earned $90, the " total difference" in compensation is $10, If in 2002, the men and women both received a 10% increase in salary (such that the men now earned $110 and the women now earned $99), the " total difference" in compensation would increase to $11. The " incremental difference" in compensation from 2001 to 2002, however, is only $1. Dr. Madden's original method of regression analysis thus included the effect of wage differentials that existed between male and female employees as a result of employment decisions that predated the class period sought by plaintiff.
That method of analysis runs headlong into Ledbetter v. Goodyear Tire & Rubber Co., Inc., 550 U.S. 618, 127 S.Ct. 2162, 167 L.Ed.2d 982 (U.S.2007). In Ledbetter, the Supreme Court held that each paycheck issued after a prior discriminatory pay decision does not constitute an actionable act of discrimination unless the individual paycheck was accompanied by new discriminatory intent. Id. at 2174-75. See also Lewis v. City of Chicago, 528 F.3d 488, 492 (7th Cir.2008) (distinguishing between injury resulting from a " fresh act" of discrimination and injury resulting from the " automatic consequence" of an earlier act of alleged discrimination). By basing her analyses on " total difference" between salaries of male and female, Dr. Madden improperly incorporated pay-setting decisions that took place before the class period, and thus used an analysis that is contrary to Ledbetter. An " incremental difference" analysis, on the other hand, avoids this problem by focusing on pay decisions made within the asserted class period.
We recognize that plaintiff filed her complaint and asserted the rejected " paycheck" theory-prior to Ledbetter. In the wake of Ledbetter, plaintiff and Dr. Madden seek to salvage their method of analysis by recasting the paycheck theory. They contend that an employee's current annual compensation is the result of a current decision by her employer regarding total salary, not merely percentage annual increase, and thus that Dr. Madden's method of analysis comports with Ledbetter (doc. # 223: Ex. 2, Madden Rebuttal at 16-18). Dr. Madden contends that Allstate attempted each year to set the " right current level of earnings" and to correct for " inappropriate" differentials in earnings with a gender-neutral policy of giving higher earnings increases to employees whose earnings levels were low relative to comparable employees and lower earnings increases to employees whose earnings levels were high relative to comparable employees ( id. at 16-18, 29). Dr. Madden concluded that because women are systematically more likely than men to be below their expected pay levels, women were discriminated against when they did not receive a larger salary increase than men to correct for prior pay differentials ( id. at 26-27). The Court is not persuaded that this approach avoids the Ledbetter limitation. Under Ledbetter, discrimination is not measured by a defendant's failure to correct for a discrimination that may have occurred in the past, but by discriminatory decisions made during the relevant time period. Under Ledbetter, a decision by Allstate not to give women higher salary increases to correct for prior pay differentials would not constitute actionable discrimination.
The Supreme Court's opinion in Bazemore v. Friday, 478 U.S. 385, 106 S.Ct. 3000, 92 L.Ed.2d 315 (1986), does not change this analysis. In Bazemore, the Supreme Court held that the fact that the government employer discriminated with respect to salaries prior to the time the employer was covered by Title VII " does not excuse perpetuating that discrimination after the [employer] became covered by Title VII." Id. at 395, 106 S.Ct. 3000. Further, the Court held that " [e]ach 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." Id. at 395-96, 106 S.Ct. 3000. Ms. Puffer interprets Bazemore to mean that despite Ledbetter 's clear repudiation of a " paycheck accrual rule," whereby each paycheck would trigger a new EEOC charging period, every paycheck Allstate issued to female managers was a separate act of discrimination (Pl.'s Class Cert. Mem. at 22; doc. # 233: Pl.'s Resp. to Mot. to Strike Madden at 7).
Plaintiff suggests that Ledbetter does not apply here because, unlike Bazemore , Ledbetter was not a pattern or practice case. Ledbetter, however, rejected the paycheck accrual rule in the context of all disparate treatment cases, holding that allowing a plaintiff to recover for an earlier discriminatory pay decision would " effectively eliminate the defining element of [a] disparate treatment claim" -intent-which is key to pattern and practice cases as well. Ledbetter, 127 S.Ct. at 2170.
We disagree, because Bazemore involved a unique situation not present here. Unlike the case here, the employer's compensation policy at issue in Bazemore had facially discriminated between white and black employees. The Supreme Court explained that allowing the effects of such a facially discriminatory practice to continue simply because it had been instituted before Title VII became effective " would have the effect of exempting from liability those employers who were historically the greatest offenders of the rights of blacks." Bazemore, 478 U.S. at 395, 106 S.Ct. 3000. Unlike the plaintiffs in Bazemore, who could not have brought their Title VII claims earlier because Title VII had not yet been made applicable to public employers, Ms. Puffer (and many of the putative class members whom she seeks to represent) could have filed charges with the EEOC long before the class period asserted here.
This distinction is an important one. While the plaintiffs in Bazemore could not have filed a suit based on events that predated the statute of limitations period (because that would have predated the applicability of Title VII), that is not true here. Plaintiff in this case (and the putative class members) labored under no similar disability. Plaintiff could have filed-and, indeed, did file-an earlier EEOC charge. But plaintiff failed to file an EEOC charge that would extend the class period back to 1999, the period included in Dr. Madden's analysis. The statute of limitations for filing EEOC charges-and then filing suits based on these charges-is part and parcel of " Title VII's integrated multistep enforcement procedure." Ledbetter, 127 S.Ct. at 2170. As in other types of cases, a defendant in a Title VII case is entitled to rely on the statute of limitations and to treat claims beyond the statute of limitations period as beyond the reach of a lawsuit,
That would not be the case if Allstate actively concealed the alleged discriminatory actions. But we see no evidence of that here. To the contrary, plaintiff has offered declarations of 49 persons who say they were aware of alleged disparate treatment as long as 20 years ago, and plaintiff herself alleges that she was subjected to unlawful gender discrimination as early as 1998 (Am.Compl.¶ ¶ 18-19). No reason has been offered for their failure to file an earlier EEOC charge.
Second, when Dr. Madden analyzed incremental salary differentials in her rebuttal report (which she referred to as " percentage change" in salary by year), that analysis yielded much smaller differentials-measured in standard deviations-than those presented in her initial report analyzing the total difference in earnings by year (Madden Rebuttal at 34-37, Tables 3R through 6R). The following table demonstrates the different results Dr. Madden found when she used incremental salary difference rather than total salary difference (when earnings are defined as W-2 earnings minus stock and settlement income). The negative figures indicate where women in fact received greater salary increases than men:
Dr. Madden did not analyze incremental difference in salaries by grade, Thus, we do not know whether, within a given year, there were any significant variations in the level of standard deviations among the various grades. That is important in considering whether Dr. Madden's analysis shows commonality across all grades (and across putative class members). We do note that Dr, Madden's original total difference analysis did show significant variations, within each year, in the level of standard deviations by grade ( see page 23, infra ).
Control Factors
Education, Age, Allstate Tenure, Salary Grade
Education, Age, Allstate Tenure, Salary Grade, Job Code
Education, Age, Allstate Tenure, Salary Grade, JobCode, Location
Education, Age, Allstate Tenure, Salary Grade, JobCode, Time in Grade
Education, Age, Allstate Tenure, Salary Grade, Time inGrade
[Note: The following table/form is to wide to beprinted on a single page. For meaningful review of itscontents the table must be assembled with part numbersin ascending order from left to right. Row numbers,which are not part of the original data, have beenadded in the margins and can be used to align rowsacross the parts.]************************************************************************************* This is piece: 1***********************************************************************
1
Salary
2
Measure
Incr
Total
Incr
Total
Incr
Total
Incr
Total
Incr
3
2001-02
2.11
9.52
0.27
3.12
0.11
3.38
0.22
2.87
-2.19
4
2002-03
3.99
10.63
3.68
6.06
2.26
5.36
3.65
5.66
4.28
5
2003-04
2.89
10.74
2.72
6.31
3.30
6.53
2.64
6.00
3.04
6
2004-05
1.59
10.25
0.52
4.85
0.15
4.99
0.45
4.50
1.80
7
2005-06
-.91
4.43
-0.56
1.64
-1.11
0.97
-0.53
1.64
-1.04
************************************************************************************* This is piece: 2***********************************************************************
1
2
Total
3
9.49
4
10.42
5
10.63
6
10.08
( See Madden Report, Table 9; Madden Rebuttal, Table 3R.) The more that the standard deviations depart from zero, the more likely the factor in question played a role in the employer's decisionmaking process. Adams v. Ameritech Servs., Inc., 231 F.3d 414, 424 (7th Cir.2000). In general, " [t]wo standard deviations is normally enough to show that it is extremely unlikely ... that [a] disparity is due to chance." Id. However, the Seventh Circuit rejects a bright-line rule that would find statistical evidence of less than two standard deviations inadmissible. See Kadas v. MCI Systemhouse Corp., 255 F.3d 359, 362 (7th Cir.2001) (holding that statistical evidence of less than two standard deviations may be relevant when corroborated by other evidence).
7
4.46
Dr. Madden's revised analysis demonstrates that when calculating the incremental salary changes, the statistical evidence does not provide strong evidence that salary differentials arose because of gender across the board. In three of the five analyses, the standard deviation was less than two in three of the five time periods; in no analysis did the standard deviation exceed two in more than three of the five time periods. And, in each analysis, the differential favored women once (and in one analysis, twice). That is so even though Dr. Madden's analysis never controlled for all considered variables-education, age, tenure, salary grade, job code, location, and time in grade-in any one analysis.
Allstate's expert, Dr. Robert Topel, found no statistically significant difference in salary growth, promotion rates, or performance evaluation scores for putative class members and comparable males (Def.'s Opp'n to Class Cert. at 14). Plaintiff filed a motion to strike the parts of Dr. Topel's original report that make legal conclusions (doc. # 222), The Court denies the motion; but, insofar as Dr. Topel's report expresses opinions or conclusions of law, the Court has disregarded those portions of the report.
Third, Dr. Madden's regression analysis does not demonstrate uniformity in salary differentials by grade. In her original report, Dr. Madden's results showed that differences in salary based on gender varied widely between grades. Her results are measured in standard deviations as follows:
Grade
2001
2002
2003
2004
2005
2006
63
7.81
8.11
9.43
9.06
7.04
2.84
64
1.89
3.44
2.62
2.86
1.78
0.71
77
3.19
3.73
3.76
4.17
3.49
2.27
78
2.09
2.27
2.90
2.61
2.46
0.63
80
2.04
2.34
1.70
2.15
0.96
2.27
90
0.62
0.62
1.15
1.00
1.69
1.28
91
0
0
0
0
0
0
(Madden Report, Tables 1-6 (" 0" indicates that the sample size was too small to determine standard deviations).) While most of Dr. Madden's original results exceed the level of two standard deviations (and in some instances by quite a lot), they do not show a uniform pattern or practice of discrimination because of the wide discrepancy in standard deviations between grades. And, as we have explained, these results use a method of analysis that does not comport with Ledbetter.
Fourth, in both her original and revised analyses, Dr. Madden failed to consider certain important non-gender related variables that may have contributed to the observed salary differentials. For example, Dr. Madden's studies do not identify whether putative class members were interested in the management jobs that Dr. Madden found were underrepresented by women, an important consideration. See Bennett, 295 F.3d at 697. In addition, she did not identify whether the putative class members were qualified for the positions in which women were underrepresented, because she did not factor into her analyses the employees' performance evaluations, or the quantity and quality of their work.
Plaintiff concedes the possibility that evidence may show " that the measured gender differences can be explained in terms of legitimate measures of skills or productivity" (Pl.'s Class Cert. Mem. at 7 (citing Madden Report at 21)). Dr. Madden reiterates in her rebuttal report that gender differences in salary are not evidence of discrimination unless productivity and other characteristics associated with productivity have been " appropriately controlled" (Madden Rebuttal at 6-8, 13). While Dr. Madden states that she " investigated all of the data that are available on productivity," that does not include employees' major job responsibilities, individual performance goals, Performance Development Summaries, Quality Leadership Management Surveys, Critical Success Factors, or any other actual measure of the quantity or quality of an employee's performance ( id. at 8).
Although plaintiff is not required to include all measurable variables in her statistical analysis, Dr. Madden's analysis is only minimally probative of commonality without these important variables. See Sears, 839 F.2d at 334 (upholding district court's finding that the plaintiffs statistics were flawed because they were based on the false assumption that all plaintiffs were equally qualified and equally interested in the promotions at issue); see also Radue, 219 F.3d at 616-17 (" Statistical evidence which fails to properly take into account nondiscriminatory explanations does not permit an inference of discrimination" ). For this reason, together with the ones we have already discussed, Dr. Madden's reports do not show commonality in plaintiff's putative class.
In her rebuttal report, for the first time, Dr. Madden notes in passing that she considered " base earnings level and performance evaluations" to determine the annual percentage, or incremental, increases in earnings (Madden Rebuttal at 19). However, she gives no further details about which or how many performance evaluations she considered. Thus, this unelaborated statement does not enhance the strength of her analysis on the issue of commonality.
c.
In addition to Dr. Madden's statistical analysis, plaintiff has submitted 49 declarations by current and former Allstate female management employees purporting to show that discrimination occurred uniformly across Allstate Protection (Pl.'s Class Cert. Mem. at 2). Allstate filed a motion to strike the declarations on the grounds that some declarants are not potential class members, and other declarations are conclusory, speculative, or vague; contain hearsay, improper opinion, or irrelevant information; or were made without personal knowledge (doc. # 168: Mot. to Strike Decls.). In addition, Allstate filed 36 declarations from 35 different employees purporting to rebut plaintiff's declarations (doc. # 188).
Although plaintiff submitted 50 declarations, one declarant does not allege gender discrimination at Allstate. Her declaration merely provides the organizational structure of All state's property/compliance division (Pl.'s Decl. 37). This declarant also has filed an affidavit in support of All state's position, stating that Allstate has not discriminated against her on the basis of gender (Def.'s Decl. of Donna Rosemeyer).
We agree that many of plaintiff's declarants shed little light on commonality during the proposed class period, because they speak to pre-class period events. Contrary to plaintiff's arguments, the class period begins not on May 9, 2001, but on July 13, 2002, which is 300 days before she and Allstate signed a " Standstill Agreement" (Def.'s Opp'n to Class Cert. at 5 n. 4). See Chaudhry v. Nucor Steel-Indiana, 546 F.3d 832, 836-37 (7th Cir.2008) (before challenging an unlawful employment practice under Title VII, an employee must first file an EEOC charge within 300 days after the alleged unlawful employment practice occurred or else the employee may not challenge the practice in court). Although plaintiff alleges a " pattern or practice" case, plaintiff offers no facts demonstrating a " continuing violation" which would extend the class period back beyond the 300-day period (doc. # 235: Resp, to Mot. to Strike Decls. at 8-9). The doctrine of continuing violation allows a plaintiff to delay suing until a series of acts by a prospective defendant turns into a wrongful injury, such as in a sexual harassment case. Lewis, 528 F.3d at 492-93. The acts plaintiff complains of here, however, such as discrimination in compensation, promotions, and training, would be actionable in and of themselves as of the time they occurred. See Maclin v. SBC Ameritech, 520 F.3d 781, 787-88 (7th Cir.2008). Moreover, pay discrimination in employment is no harder to detect than other forms of employment discrimination and does not warrant giving a plaintiff additional time to file a charge with the EEOC. Ledbetter, 127 S.Ct. at 2177.
Indeed, as we noted earlier ( see note 9, supra ), the fact that plaintiff offers 49 declarations describing alleged discrimination occurring as far back as the 1980s runs contrary to the notion that plaintiff and her putative class could not have detected earlier the discrimination they now allege.
Moreover, even using plaintiff's proposed date of May 9, 2001, for the start of the class period, nine declarants stopped working at Allstate before the class period began (Pl.'s Decls. 1, 2, 3, 4, 6, 11, 19, 30, 46), and five declarants stopped working at Allstate in 2001 (Pl.'s Decls. 13, 22, 24, 34, 50). In addition, five declarants were not employed in either Allstate P & C or Allstate Protection during the class period, and thus are not potential class members (Pl.'s Decls. 26, 31, 36, 42, 43). Although the observations of employees outside the class period may be relevant to provide background information about Allstate's employment policies and practices, see Nat'l R.R. Passenger Corp. v. Morgan, 536 U.S. 101, 113, 122 S.Ct. 2061, 153 L.Ed.2d 106 (2002), these non-class member declarations do not support plaintiff's argument for commonality among class members.
Other declarations are of little probative value on the issue of commonality because the declarants offers either no approximate date as to alleged discriminatory occurrences or no allegations at all of specific instances or observations of discrimination (Pl.'s Decls. 14, 17, 18, 20, 21, 23, 25, 27, 32, 38, 39, 40, 41, 48, 49). While some of these declarants allege generally that Allstate has a " male-dominated" or " boys' club" culture, these allegations, unaccompanied by a time frame or specific discriminatory act, are too vague to be helpful. See Mattenson v. Baxter Healthcare Corp., 438 F.3d 763, 770 (7th Cir.2006). Likewise, conclusory legal statements, such as that something happened to them " because of" their gender, are not probative of commonality without first-hand factual support.
The remaining 15 declarations may be probative of individual instances of alleged discrimination, but they do not show commonality among the class (Pl.'s Decls. 5, 7, 8, 9, 10, 12, 15, 16, 28, 29, 33, 35, 44, 45, 47). In fact, these declarations highlight the differences and uniqueness in the individual plaintiffs' claims. Seven of these declarants worked at some point during their career in Allstate's home office in Northbrook, Illinois, but it is unclear whether their time at the home office or their supervisors overlapped. The remaining declarants worked at different Allstate offices in Ohio, New Mexico, Texas, California, Colorado, Mississippi, Florida, Michigan, and Texas. These individuals worked in different departments within Allstate Protection, had different supervisors, and allegedly experienced different forms of discrimination.
Finally, the declarations submitted by Allstate further undermine plaintiff's commonality argument. Those 35 putative class members describe promotions, advancement, and pay increases they received in accordance with their personal choices and qualifications, leading them to believe that they were not discriminated against on the basis of gender. During the class period, nine of Allstate's declarants worked in the home office, and the rest worked in Allstate offices in New York, Ohio, California, Nevada, Michigan, Washington, Texas, Florida, Colorado, Indiana, Alabama, Tennessee, Pennsylvania, and Virginia. Collectively, the parties' 85 declarations represent about five percent of the putative class. The Court can only presume that the remaining possible 1,615 class members would have their own unique stories as well. Taken together, these declarations underscore the lack of commonality in plaintiff's proposed class.
d.
The fourth piece of evidence of commonality that plaintiff submitted was an expert report by Dr. Barbara Reskin, an organizational sociologist with expertise in gender discrimination and employment organizational structures. In preparing her report, Dr. Reskin relied on documents provided by plaintiff, including plaintiff's 50 declarations, Dr. Madden's initial report, Allstate's Human Resources Policy Guide, and the depositions of various witnesses (Pl.'s Class Cert. Mem., Ex. 10, Reskin Report at 2). Based on this evidence, together with her knowledge and research in her field, Dr. Reskin opined that Allstate Protection has a " uniform culture of paternalism across the company," and that " Allstate's paternalistic culture and the discretion it requires and permits decision makers have contributed to the systematic disadvantage of female managers at [Allstate Protection], including their lower compensation, bonuses, and stock options and their underrepresentation in top-level managerial jobs" (Reskin Report ¶ ¶ 2.1, 2.5). In addition, Dr. Reskin concluded that Allstate's employment practices fail to check the biases that discretion permits ( id. ¶ 2.4).
Allstate has filed a motion to strike Dr. Reskin's report under Daubert (doc. # 207). We need not decide that question because, even assuming that Dr. Reskin's report passes muster under Daubert, we do not find it persuasive on the issue of commonality. Most of Dr. Reskin's opinion relies on laboratory studies, " experimental research," and various historical trends and ideas which lead her to the conclusion that subjective decisionmaking is inherently flawed: a conclusion that flies in the face of the governing case law. See Blise, 409 F.3d at 868 (holding that subjective decisionmaking plays an important and legitimate role in employment decisions). Based on this improper assumption, Dr. Reskin concludes that Allstate's employment policies " invite [ ] distorted appraisals both through automatic biases such as ingroup favoritism and sex stereotyping and through open favoritism" (Reskin Report ¶ 7.1). Beyond the admitted existence of discretion in Allstate's decisionmaking structure, Dr. Reskin's analysis does not show that senior managers applied their discretion in a uniform way that caused, or was intended to cause, gender disparities in the management ranks of Allstate Protection.
The Ninth Circuit's decision in Dukes, 509 F.3d at 1183, is distinguishable. In that case, the appellate court upheld the district court's grant of the motion for class certification because the appellate court found that in that case, unlike in the instant case, the plaintiffs had produced considerable, reliable statistical evidence of differential pay and promotions for female employees, as well as substantial evidence of Wal-Mart's centralized company culture and policies, thus showing commonality.
For all these reasons, plaintiff has not met her burden of showing commonality.
C.
Plaintiff also has not met her burden of demonstrating typicality. Rule 23(a)(3) requires plaintiff to prove that " the claims or defenses of the representative parties are typical of the claims or defenses of the class." Fed.R.Civ.P. 23(a)(3). Whether plaintiff's claims are typical of those of the class members she seeks to represent " is closely related to" the commonality inquiry. Keele, 149 F.3d at 595. " A claim is typical if it arises from the same event or practice or course of conduct that gives rise to the claims of other class members and her claims are based on the same legal theory." Arreola, 546 F.3d at 798 (internal citations and quotations omitted). Some factual variations may not defeat typicality, but the named representative's claims must have " the same essential characteristics" as the claims of the class. Id.
Ms. Puffer alleges that the following demonstrates that Allstate discriminated against her because she is a woman: Allstate asked her to take over as head of Policy Management but did not give her a Vice President title as it had given men in that position; Allstate denied her request to be transferred to a regional field manager position; Allstate refused to consider her for a products manager position; Allstate never promoted her to a vice president position; and Allstate denied her request to transfer to the IT department in 2003 (Pl.'s Class Cert. Mem. at 20). These claims, however, are not typical of the claims of other putative class members, especially the vast majority who never held an officer position and thus were not subject to the same considerations and decisionmakers as Ms. Puffer. The variations among the putative class members that we have described above would necessitate individual inquiries for each putative class member in order to determine whether she suffered gender discrimination. In addition, if certain class members are able to establish a prima facie case of gender discrimination, Allstate's defenses would vary between class members. For example, if Allstate sought to present evidence that a class member's job performance was the legitimate, nondiscriminatory reason for her salary or job position, such evidence would differ for each individual class member, Allstate also may be able to raise a statute of limitations defense against some class members but not others, because as explained above, some proposed class members may have been aware of the existence of possible gender discrimination more than 300 days before Ms. Puffer filed her EEOC charge. See note 9, supra; see also Beamon v. Marshall & Ilsley Trust Co., 411 F.3d 854, 860-61 (7th Cir.2005) (holding that statute of limitations barred Title VII claim where the plaintiff would have been aware of discrimination more than 300 days before EEOC charge filed). Where, as here, a court would have to examine numerous individualized factors to determine the parameters of individual claims, the typicality requirement is not met. Payton v. County of Carroll, 473 F.3d 845, 854 (7th Cir.2007); see also Rahman v. Chertoff, 530 F.3d 622, 627 (7th Cir.2008) (reversing grant of class certification where proposed class was so broad that claims did not satisfy the typicality requirement).
D.
To satisfy Rule 23(a)(4), Ms. Puffer must show that she " will fairly and adequately protect the interests of the class." Fed.R.Civ.P. 23(a)(4). " To determine if the plaintiff has met the adequacy requirement of Rule 23(a)(4), the Court must ask whether the named plaintiff: (1) has antagonistic or conflicting claims with other members of the class; (2) has a sufficient interest in the outcome of the case to ensure vigorous advocacy; and (3) has counsel that is competent, qualified, experienced and able to vigorously conduct the litigation." Quiroz v. Revenue Prod. Mgmt., Inc., 252 F.R.D. 438, 442 (N.D.Ill.2008) (internal quotations omitted); see also Rosario v. Livaditis, 963 F.2d 1013, 1018 (7th Cir.1992).
There is no dispute that plaintiff's counsel is qualified, having represented plaintiffs in numerous Title VII cases, and there is no dispute that Ms. Puffer has diligently pursued this action. Allstate argues, however, that Ms. Puffer is not an adequate class representative because she worked in a different division and was at a higher salary grade than most of the putative class members. In addition, Allstate contends that the putative class contains numerous individuals who reported to, and were evaluated by, other putative class members, leading to antagonistic or conflicting claims (Def.'s Opp'n to Class Cert. at 17). Some courts have held that a conflict of interest may arise where a class contains both supervisory and non-supervisory employees, and the supervisors implemented the employment system which the class litigation challenged. See, e.g., Hines v. Widnall, 334 F.3d 1253, 1258 (11th Cir., 2003) (upholding district court's finding that the interests of the African-American supervisors and the non-supervisory plaintiffs were not co-extensive); Wagner v. Taylor, 836 F.2d 578, 595 (D.C.Cir.1987) (upholding district court's determination that plaintiff supervisor was not adequate class representative because his interests clashed with those of other members of the class).
Allstate also argues that Ms. Puffer's individual claim of retaliation makes her inadequate to represent the class. This argument is a nonstarter. " If a retaliation claim were a bar to certification, any defendant could insulate itself from class certification by retaliating against anyone who brought or threatened a lawsuit." Meiresonne v. Marriott Corp., 124 F.R.D. 619, 624 (N.D.Ill.1989).
By contrast, Ms. Puffer contends that although she participated in some succession planning meetings and supervised some of the putative class members, she did not possess decision-making authority as to employee job moves, succession planning, compensation, or promotions, and thus she did not actually implement any of Allstate's allegedly discriminatory employment policies (Pl.'s Class Cert. Mem. at 20-21; Pl.'s Class Cert. Reply at 17-18). Rather, Ms. Puffer claims that she witnessed the subjective nature of Allstate's employment processes and the resulting disadvantages to women in salary, promotion, and career advancement opportunities ( id. ). See Wagner v. NutraSweet Co., 170 F.R.D. 448, 451-52 (N.D.Ill.1997) (holding that supervisor was adequate representative because her interests were not antagonistic to those of other class members where her involvement in implementing allegedly discriminatory salary system was minimal and she had no salary-setting authority with regard to any class members). Thus, we agree that Ms. Puffer's claims and interests do not conflict with those of the other putative class members.
In addition, although Ms. Puffer worked in a different division and was at a higher salary grade than most of the putative class members, we do not see that this places her at odds with putative class members at lower salary grades. We conclude that Ms. Puffer has a sufficient interest in and incentive to press the claims for monetary relief she brings on behalf of the class, and would be an adequate class representative had she satisfied all other requirements for class certification.
III.
Not only does plaintiff's proposed class fail to satisfy Rule 23(a)'s requirements, her proposed class would fail under Rule 23(b) as well. Plaintiff seeks to have her class certified as a hybrid Rule 23(b)(2) and 23(b)(3) class. Under Rule 23(b)(2), plaintiff seeks an " appropriate injunction compelling defendant Allstate to cease and desist from the wrongful practices here alleged and hereafter established" (Am. Compl. at 13). Under Rule 23(b)(3), plaintiff seeks compensatory, punitive, and liquidated damages for the individual class members ( id. at 13-14),
Although the Seventh Circuit has left open the possibility of a " hybrid certification" under Rules 23(b)(2) and 23(b)(3), see Arreola, 546 F.3d at 797, such hybrid certification is discouraged where class relief falls much more heavily into one category. Here, we find that plaintiff has failed to show certification is proper under either 23(b)(2) or (b)(3).
A.
A Rule 23(b)(2) class may be maintained when " the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole." Fed.R.Civ.P. 23(b)(2), Rule 23(b)(2) certification is generally limited to class actions " [w]hen the main relief sought is injunctive or declaratory, and the damages are only ‘ incidental.’ " In re Allstate Ins. Co., 400 F.3d 505, 507 (7th Cir.2005). Rule 23(b)(2) certification " does not extend to cases in which the appropriate final relief relates exclusively or predominantly to money damages." Fed.R.Civ.P. 23 Advisory Committee's note.
In this gender discrimination case, final relief appears to relate predominantly to money damages, and, indeed, a substantial amount of court time would have to be devoted to litigating each individual class member's liability and damages claims. For these reasons, Rule 23(b)(2) certification would be inappropriate even had plaintiff met all the requirements of Rule 23(a).
B.
To meet the prerequisites for class certification under Rule 23(b)(3), a plaintiff must demonstrate " that the questions of law or fact common to the members of the class predominate over any questions affecting only individual members, and that a class action is superior to other available methods for the fair and efficient adjudication of the controversy." Szabo, 249 F.3d at 676 (citing Fed.R.Civ.P. 23(b)(3)). When making the determination of predominance and superiority, a court must consider, among other things: " (A) the class members' interests in individually controlling the prosecution or defense of separate actions; (B) the extent and nature of any litigation concerning the controversy already begun by or against class members; (C) the desirability or undesirability of concentrating the litigation of the claims in the particular forum; and (D) the likely difficulties in managing a class action." Fed.R.Civ.P. 23(b)(3). In this case, plaintiff fails both the predominance and superiority requirements. 1.
" The Rule 23(b)(3) predominance inquiry tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation." Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997). This inquiry " trains on the legal or factual questions that qualify each class member's case as a genuine controversy ..." Id. The predominance requirement is similar to the typicality of claims and defenses requirement of Rule 23(a)(3), but is " far more demanding" than Rule 23(a)'s commonality requirement. Id. at 623-24, 117 S.Ct. 2231.
Here, the Court has already determined that plaintiff's proposed class fails the commonality and typicality requirements. Plaintiff admits that the amount of damages will vary significantly among putative class members, as evidenced by her suggestion that the Court bifurcate the liability from the damages stage (Pl.'s Class Cert. Mem. at 26). Individual issues, however, would predominate even before the damages stage. Because plaintiff will not be able to prove class-wide, company-wide discrimination with statistical evidence alone, a jury would need to consider each putative class member's claim individually to determine liability. To prove that an individual plaintiff suffered gender discrimination under disparate impact or disparate treatment theories of Title VII or the Equal Pay Act, an individualized inquiry into each putative plaintiff's circumstances would be necessary to determine whether employment actions taken toward a particular employee were based on gender or instead on legitimate, nondiscriminatory reasons. See Warren v. Solo Cup Co., 516 F.3d 627, 629-30 (7th Cir., 2008) (holding that if a plaintiff establishes a prima facie case of wage discrimination under the Equal Pay Act, the employer may raise one of four statutory defenses, attributing the difference in pay to: " (i) a seniority system; (ii) a merit system; (iii) a system which measures earnings by quantity or quality of production; or (iv) a differential based on any other factor other than sex" ); Atanus v. Perry, 520 F.3d 662, 672-73 (7th Cir.2008) (holding that in Title VII disparate treatment case, in order for a plaintiff to establish a prima facie case, the plaintiff must proffer evidence, among other things, that she performed her job according to her employer's legitimate expectations; if the plaintiff establishes the prima facie case, the presumption shifts the burden to the employer to produce a legitimate, nondiscriminatory reason for its actions); Farrell v. Butler Univ., 421 F.3d 609, 616 (7th Cir.2005) (holding that in Title VII disparate impact case, an employer may rebut prima facie case of disparate impact by demonstrating that the employment policy at issue is related to the employees' job performance and justified by business necessity).
Such an inquiry would involve different witnesses and proofs for various putative class members to determine, among other things, the motivation of each individual supervisor and the quality and quantity of production of each employee, " If the class certification only serves to give rise to hundreds or thousands of individual proceedings requiring individually tailored remedies, it is hard to see how common issues predominate ..." Andrews v. Chevy Chase Bank, 545 F.3d 570, 577 (7th Cir.2008). We agree, and conclude that common issues do not predominate here.
2.
Likewise, " it is hard to see ... how a class action would be the superior means to adjudicate the claims." Id. Plaintiff dismisses this point by saying that " there is nothing terribly intimidating about this prospect" of over a thousand individual proceedings, as " [i]t is not unprecedented to manage such volume" (Pl.'s Class Cert. Reply at 24). Whether or not it is " unprecedented," the prospect of some 1,700 individual proceedings, especially where Title VII gives each individual the right to a jury trial, gives us pause about the superiority of the class action vehicle here for several reasons.
First, obvious difficulties in managing plaintiff's proposed nationwide class abound. Plaintiff argues that individual trials would be manageable if the Court utilized " imaginative solutions" (Pl.'s Class Cert. Mem. at 29-30). The Seventh Circuit has held that: In a case where the defendants' liability is a singular, uniform question, Rule 23 allows district courts to devise imaginative solutions to problems created by the presence in a class action litigation of individual damages issues including: bifurcating liability and damage trials; appointing a magistrate judge or special master to preside over individual damages proceedings; decertifying the class after the liability trial and providing notice to class members concerning how they may proceed to prove damages; or creating subclasses; among other things.
Carnegie v. Household Int'l Inc., 376 F.3d 656, 661 (7th Cir.2004). Plaintiff suggests bifurcating this case such that the first jury determines the issue of whether Allstate's employment policies and practices are discriminatory, and multiple follow-on juries determine whether Allstate's employment policies and practices discriminated against individual female managerial employees. However, a class proceeding that would decide one issue, but would not eliminate the need for 1,700 jury trials to determine whether each putative class member suffered discrimination and, if so, their individual damages, offers little in the way of efficiency. See Clark v. Experian Info. Solutions, Inc., No. 06-3330, 2007 WL 4219471, at *3 (7th Cir. Nov.30, 2007) (affirming district court decision not to partially certify class because with the need for individualized findings in such a large class, little efficiency would be gained by certifying a class for only particular issues).
Second, plaintiff's proposal raises constitutional concerns. " The right to a jury trial in federal civil cases, conferred by the Seventh Amendment, is a right to have juriable issues determined by the first jury impaneled to hear them ... and not reexamined by another finder of fact." Matter of Rhone-Poulenc Rorer, Inc., 51 F.3d 1293, 1303 (7th Cir.1995). In Rhone-Poulenc, the Seventh Circuit held that the district judge exceeded his authority by proposing to divide the trial between the issues he had certified for class treatment and the thousands of other individual issues, because the court may not divide issues between separate trials in such a way that the same issue is reexamined by different juries. Id. at 1302-03, In Rhone-Poulenc, the first jury would not have been able to completely determine liability, but merely whether one or more of the defendants was negligent, while juries in individual follow-on trials would have to decide such issues as comparative negligence and proximate cause, necessitating a reexamination of the same issues determined by the first jury. Id. Similarly, in this case-in light of all of the individualized inquiries that would be needed before liability was determined as to each plaintiff-any finding by an initial jury that Allstate's policies and practices generally were discriminatory (assuming such a finding were made) would likely be unconstitutionally reexamined by follow-on juries in determining whether any particular plaintiff suffered discrimination.
We are mindful that Federal Rule of Civil Procedure 23(c)(4) recognizes the possibility that " [w]hen appropriate, an action may be brought or maintained as a class action with respect to particular issues." Such partial certification of class issues, however, is not appropriate where it would not offer significant efficiencies, and instead would create constitutional concerns.
Although plaintiff points to Romasanta v. United Air Lines, Inc., 717 F.2d 1140 (7th Cir.1983), as a successful example of conducting a large number of individual hearings in the context of a class action, Romasanta is distinguishable. In Romasanta, the plaintiff class challenged United Air Lines's rule prohibiting the continued employment of any female flight attendant who got married. Id. at 1146-47. This was a facially discriminatory policy; there was no room for individual discretion. As a result, in the hundreds of individual hearings that were held in that case to identify class members, the court did not have to examine the motivation of each individual supervisor in terminating the employee, but only the motivation of each individual class member in leaving her job ( i.e., did she leave her flight attendant position due to the " no-marriage" rule or for some other reason). By contrast, in the instant case, a jury would have to look into each supervisor's motivations in making employment decisions for each individual class member before making a liability determination. Moreover, the individual hearings in Romasanta were conducted before a judge, not a jury, while in this case each class member-and Allstate-would have a right to request a jury trial on each putative class member's liability claim.
Third, class certification is not needed here to afford putative class members an opportunity to litigate claims of discrimination. An important policy behind certifying a class action is " to overcome the problem that small recoveries do not provide the incentive for any individual to bring a solo action prosecuting his or her rights. A class action solves this problem by aggregating the relatively paltry potential recoveries into something worth someone's (usually an attorney's) labor." Amchem Prods., 521 U.S. at 617, 117 S.Ct. 2231 (internal quotations and citations omitted); see also Thorogood, 547 F.3d at 744 (" The class action is an ingenious device for economizing on the expense of litigation and enabling small claims to be litigated" ).
When a class member has a sufficiently large stake in litigation to be able to afford to litigate on his own, this consideration weighs against allowing a suit to proceed as a class action. Nagel v. ADM Investor Servs., Inc., 217 F.3d 436, 443 (7th Cir.2000); see also Glazer v. Abercrombie & Kent, Inc., No. 07 C 2284, 2008 WL 4853641, at *4 (N.D.Ill. Nov. 10, 2008) (denying motion for class certification because large individual damage claims increase plaintiff's interest in individually controlling the prosecution and defense of separate actions). Indeed, the Seventh Circuit has held that damages as low as $50,000 provide adequate incentive to litigate a claim individually. See, e.g., Andrews, 545 F.3d at 577-78 (holding that class action was not superior to individual TILA rescission action because under Truth In Lending Act, a prevailing debtor with a typical loan can expect to receive over $50,000 plus attorney's fees and costs in a rescission action, and many debtors do in fact bring rescission claims).
Plaintiff argues that individual proceedings are impracticable because of the " prohibitive costs widely out-of-scale to the possible individual stakes" (Pl.'s Class Cert. Mem, at29). The Court disagrees. Countless gender discrimination claims are brought individually. And here, the putative class members would have sufficient financial incentive to pursue claims individually. The proposed class consists of managerial level employees. The average base salaries for Allstate P & C managers from 2001 to 2006 ranged from $82,389 in Grade 63 to $391,833 in Grade 91, and the average W2 wages ranged from $95,399 to $807,450 (doc. # 223: Ex. 1, Dr. Topel's Report, Ex. F1). Claims of economic loss for employees in these salary brackets could be substantial, giving many putative class members an interest in individually controlling the prosecution of their lawsuits.
Fourth, plaintiff posits that " cases such as this one" do settle, " which would obviate [these] problem[s] altogether" (Pl.'s Class Cert. Reply at 22 n. 29). The parties, however, have already spent nine months with a mediator trying unsuccessfully to settle this case. At this point, converting this case into a class action could place undue pressure on Allstate to settle regardless of the actual merit of the suit. Isaacs v. Sprint Corp., 261 F.3d 679, 681 (7th Cir.2001); see also Szabo, 249 F.3d at 675; Rhone-Poulenc, 51 F.3d at 1298. The Court declines to certify a class that fails the superiority test and other requirements of Rule 23(a) and (b), in the hope that a settlement will allow us to avoid the serious problems we otherwise would confront.
CONCLUSION
In determining whether a class action is manageable, " judges must resist" any temptations " to alter doctrine in order to facilitate class treatment" because the Court must ensure that " all parties' legal rights may be respected." In re Bridgestone/Firestone, Inc., 288 F.3d 1012, 1020 (7th Cir., 2002) (reversing district court order certifying two nationwide classes). In this case, plaintiff's proposed class does not meet several of the requirements of Rule 23: commonality, typicality, predominance, and superiority. While a subjective company policy may, in some circumstances, provide the basis for a Title VII or Equal Pay Act class action suit, we conclude that this is not such a case. Accordingly, plaintiff's motion for class certification (doc. # 125) is denied.
In light of our ruling on class certification, we deny as moot Allstate's motions to strike the Madden and Reskin reports (docs. 164, 166), Allstate's motion to strike the declarations of current and former Allstate employees (doc. # 168), and plaintiff's motion to strike, in part, Dr. Topel's expert report (doc. # 222).
This opinion does not prevent individuals who are not otherwise barred by the statute of limitations from filing an individual discrimination suit against Allstate. The filing of a class action lawsuit tolls the statute of limitations for all the proposed members of the class, and the statute resumes running for class members when class certification is denied. Culver v. City of Milwaukee, 277 F.3d 908, 914 (7th Cir.2002). By January 30, 2009, the parties may submit their views, in memoranda not to exceed five pages, as to what notice-if any-should be given to the putative class members of the denial of class certification. See id. at 913-915 (citing Fed.R.Civ.P. 23(e)).
We express no view on the merits of the individual claims brought by Ms. Puffer, or of any putative class member who may wish to file suit.