Opinion
Case No. 97 C 3882.
March 12, 2001.
MEMORANDUM OPINION AND ORDER
This case has had a protracted procedural history in which certain African-American employees of Abbott have sought to plead class claims regarding terms of employment. The case is now pending before the court on plaintiffs' Third Amended Complaint which was recently modified by a court ruling which dismissed all claims except a putative class claim for denial of promotions based on a desperate impact theory. Plaintiffs have now moved for class certification but have dropped the denial of promotion claim in favor of one for discriminatory pay. The motion of plaintiffs is based on the opinion of a statistician, who states that his statistical analysis, while not supporting a claim for denial of promotions, will support a desperate impact claim for discriminatory pay.
DISCUSSION
The parties do not disagree that to make a desperate impact claim, wage or promotion, the plaintiff must show a racial disparity in treatment, identify a specific employment practice, and causally connect the two. Watson v. Ft. Worth Bank Trust Co., 487 U.S. 977 (1988). In this case plaintiffs rely on a multiple linear regression analysis by their statistician, Dr. Bernard R. Siskin, Ph.D., which in his opinion shows that the average African American employee endures a statistically significant wage shortfall in five of Abbott's divisions, Abbott Diagnostic Division ("ADD"), Corporate Engineering Division ("CED"), Hospital Products Division ("HPD"), Pharmaceutical Products Division ("PPD"), and Ross Products Division ("RPD") when considered in the aggregate. The specific employment practice which plaintiffs claim is responsible for this shortfall is a company-wide system of salary guidelines that they claim permit excessive subjectivity on the part of individual decision-makers in each relevant Abbott division.
Abbott's answer to plaintiff's motion includes an analysis of Siskin's study by its own expert, Joan G. Haworth, Ph.D., an economist, who likewise performed a linear regression analysis. Dr. Haworth had four basic criticisms of Dr. Siskin's analysis: (1) Siskin lumped the five divisions together rather than considering them separately; (2) he included salary grades and years not involved in the suit; (3) he included improper factors and excluded proper factors in his model; and (4) his analysis was based on current salaries of Abbott employees and did not measure or compare salary changes on a year to year basis. If Siskin had analyzed the data properly as she did, his results would show that there is no statistically significant difference in the wages of African American and white employees.
Dr. Haworth's report precipitated a responsive report from Dr. Siskin where he breaks down his analysis, division by division, deletes the non-involved salary grades, and adds the factors deemed relevant by Dr. Haworth. While his analysis continued to show that African Americans in the aggregate earn less than similarly situated whites, when he ran his analysis division by division only two of the five divisions (HPD and RPD) showed statistically significant differences in favor of whites. The others were either not adverse to African Americans (PPD and CED) or not adverse to a statistically significant degree (ADD). Even at HPD and RPD there were no significant differences in compensation in 1998.
While Dr. Siskin does not respond to Haworth's criticism that he did not measure salary changes during the years in question to determine whether racial disparities exist, Plaintiffs contend in their reply brief that it is the total salaries paid, not salary increases, which are relevant because, as they say, ". . . the employer is deciding on the total level of compensation, and even if there had been no change at all in salaries during the relevant period, plaintiffs would still be entitled to attack the unequal compensation because every pay day brings a fresh violation of their civil rights." Citing Bazemore v. Friday, 196 S.Ct. 3000 (1986). (Reply Br. P. 7.)
However, plaintiffs miss the point of Dr. Haworth's criticism and the holding in the Bazemore decision. The uncontradicted evidence shows that the salaries of Abbott employees in the relevant grades are made up of three factors: (1) starting salary, determined at the time of hire; (2) merit increases to base salary, consideration of which occur every 12 to 16 months; and (3) promotional increases to base salary, which occur when an employee is promoted, either to a higher grade or a different job. Plaintiffs have candidly advised the court that their statistical evidence does not support a finding that Abbott discriminates in promotions and plaintiffs have neither charged nor demonstrated that Abbott discriminates in hiring, which leaves only merit increases as a possible source of illegal discrimination. However plaintiffs have not sought to prove that the cause of the salary shortfalls to the extent that any exists are caused by merit increases discriminatory granted. Consequently they have not identified the cause of the possible salary discrepancy.
As affidavits of Abbott's managerial employees have shown, starting salaries are determined by a group of managerial employees, including at a minimum a hiring manager, a staffing manager or recruiter, and a human resources manager, and the size of the starting salary is influenced by factors such as the job title and requirements, an applicant's prior work experience, his level of education, and the then existent market conditions for the particular job skill, which can very greatly depending on demand for a particular skill at a particular time. To the extent that starting salaries differ due to non-discriminatory reasons, such as education level, prior experience, and market conditions, these differences could be expected to extend into the future since subsequent increases for merit reasons and for promotions are based on percentages of an individual's salary at the time of the raise. To the extent that discrimination exists as to starting salaries such would not be hard to prove: all one would have to do is compare starting salaries of whites and African Americans having comparable qualifications who have been hired for the same type of job. Plaintiffs, however, have not done this. Since any salary level subsequent to hire is in part dependent on the starting salary no analysis would be useful without taking the starting salary into consideration. Bazemore holds that discrimination in pay, once established, must be remedied. Here plaintiffs have failed to establish discrimination in pay.
Moreover in neither of the two relevant EEOC charges is there any contention that Abbott discriminates against African Americans in it hiring decisions including the setting of the starting salaries. A fair reading of the Third Amended Complaint likewise shows that the plaintiffs are complaining of lack of promotions and not of hiring pay. Both charges contend that Abbott discriminated against the employee by failing to promote him and all monetary losses flow from that illegal act. Dennis Walker charged in his EEOC that he was denied a promotion to a Grade Level 20 from his then Grade Level 18 and that such "continuous pattern and practice of discrimination in denying African Americans promotional opportunities, management level positions and opportunities, pay, compensation and related benefits related to promotion and management level positions . . . violates . . . Title VII." Marvin Fields complained in his EEOC complaint that he was denied a promotion to Grade Level 19 and in doing so Abbott had "engaged in a pattern and practice of discrimination against me and other African Americans . . . benefits, management opportunities, and that this pattern and practice . . . [is] in violation of Title VII." Clearly the named plaintiffs are at most claiming that they are paid less than comparable whites due to lack of promotions, not discriminatory starting pay or merit increases. If there was such evidence it would be rather easy to discern by comparing the wage rates of new African American and white hires. In fact it is relevant that neither Fields not Walker in deposition testimony made any charge that they were discriminated against with respect to pay, except to the extent that their pay was lower because they were not promoted.
Federal Rule of Civil Procedure 23(a) sets forth four prerequisites to certification of a class: numerosity, commonality, typicality, and adequacy of representation. The burden is on plaintiffs to show that certification is proper. Patterson v. General Motors Corp., 631 F.2d 476, 480 (7th Cir. 1980). While it is not appropriate to conduct an extensive inquiry into the merits of the suit in making the class certification determination, it is however appropriate to go beyond the pleadings "before coming to rest on the certification question" and consider the information that is available in the record. General Telephone Co. v. Falcon, 105 S.Ct. 2364, 2372 (1982).
In this case plaintiffs have possibly satisfied only one of the four requirements of Rule 23(a). The plaintiffs have not established that any African American employee, including themselves, has been adversely affected by Abbott's wage policy. There is a complete absence of anecdotal evidence that any African American employee, including the proposed class representatives, have suffered wage discrimination. The plaintiffs propose that all African American employees in Grades 16 through 18 at all five divisions be included in one class, even those employees who work at Abbott Divisions that their own statistical evidence shows were treated more favorably than white employees at those divisions. Consequently neither numerosity nor typicality is demonstrated. See Texas Motor Freight Systems, Inc. v. Rodriguez, 431 U.S. 395 (1977). While the issue of commonality could probably be established since there seems to be a common procedure with respect to hiring, merit pay increases and promotions at the five relevant divisions, nevertheless plaintiffs have made no demonstration that they could be successful in presenting such a claim which reflects on their adequacy as class representatives. The named plaintiffs are claiming that they were discriminated against in promotion and have testified in their depositions that they do not believe that they have suffered discrimination in pay except to the extent they have lost pay due to Abbott's failure to promote them. Clearly they have a different bone to pick and are not adequate class representatives for this type of claim. At a trial the named plaintiffs would be putting their effort into proving their individual claims for denial of promotion which would cause the class claim to take a back seat. It is clear that a class representative must possess the same interest and suffer the same injury as the class members. East Texas Motor Freight Systems v. Rodriguez, 97 S.Ct. 395, 403 (1976); Falcon, at 2370. At this point neither their testimony nor that of Dr. Siskin establishes that either they or the class they seek to represent would have a reasonable chance of success. Thus plaintiffs have failed to prove that they would adequately represent the class.
CONCLUSION
The plaintiffs have tried to change horses in midstream. While their decision to drop the class claim for discrimination in promotions, because their expert's analysis does not support such a claim, is commendably forthright, nevertheless they cannot come up with a brand-new claim that has neither been pled not vetted through the EEOC, and which their evidence does not support. Accordingly, the motion to certify a class is denied.
IT IS SO ORDERED.