Opinion
No. 95 C 5370
February 17, 2000
MEMORANDUM OPINION AND ORDER
Plaintiffs Abraham Sonii and Rufus Jones, former consumer service technicians for General Electric, have brought claims against the company for disparate treatment and disparate impact under Title VII. Sonii and Jones allege that the system by which General Electric assigned service calls was racially discriminatory, and that it prevented them from succeeding under the company's revenue-focused performance criteria. General Electric has moved for summary judgment on all claims. For the following reasons, the court grants General Electric's motion as to Jones' disparate treatment claim, but denies it in all other respects.
Background
Sonii and Jones are former consumer service technicians for General Electric who repaired major appliances in customers' homes. They lived and worked in the area of Chicago designated by the company as the "South Zone." Sonii worked at General Electric from 1974 until he was terminated on August 19, 1994. Jones began working for General Electric in 1966, and worked as a consumer service technician from 1967 until he took a Special Early Retirement Option in October 1995.
The court rejects General Electric's argument that Jones' claim is barred by his failure to file an EEOC charge. The court previously ruled that "Jones' failure to file is not a bar to his participation in this case," given that "Sonii's [EEOC] charge sufficiently notified General Electric that the consequences of Sonii's charge could transcend his individual grievance." (6/26/97 Order at 2)
In distributing service calls, General Electric claims that it assigned zip codes to each technician based on where that technician lived, the workload in that area, and the technician's seniority. Plaintiffs contend that race also played a part in the assignments — that is, white technicians were rarely assigned to predominately black areas, and black technicians were rarely assigned to predominately white areas, even if those areas were near the technician's home. This factual dispute is at the core of this case.
The assignment system's impact on the technicians' ability to meet revenue goals is also a matter of dispute. Plaintiffs claim that their assignment to predominately black, low-income areas prevented them from bringing in the required revenue. In Sonii's case, his revenue shortfall led General Electric to place him on its Performance Improvement Program ("PIP"); when his revenue margins did not show sufficient improvement, he was terminated. Jones alleges that the revenue constraints caused General Electric to place him on PIP as well, and that he would have been terminated had he not opted for early retirement. General Electric contends that Sonii's revenue problems were not due to his assigned area, and that other technicians assigned to that area have met expectations. General Electric also claims that Jones was never placed on PIP, and that he was not in immediate danger of being terminated.
Analysis
Summary judgment is appropriate only if the evidentiary record indicates that "there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law." FED. R. CIV. P. 56(c). The court applies "this standard with added rigor in employment discrimination cases, in which intent and credibility are crucial issues." Drake v. Minnesota Mining Mfg. Co., 134 F.3d 878, 883 (7th Cir. 1998). "Because evidence directly supporting a claim of intentional discrimination is rare, affidavits and depositions must be carefully scrutinized for circumstantial proof which, if believed, would show discrimination." Id. (quoting Courtney v. Biosound, Inc., 42 F.3d 414, 418 (7th Cir. 1994)).
Disparate Treatment
In Count I of their complaint, Sonii and Jones allege disparate treatment based on General Electric's system of assigning service calls. By assigning black technicians to low-income, predominately black areas of Chicago, plaintiffs allege that General Electric limited the technicians ability to generate revenue, which is an important measure of their performance. They also claim that General Electric compounded the impact of this discriminatory policy by requiring them to meet certain revenue margin goals, despite their assignment to low-income areas, and by wrongfully terminating their employment with GE.
Plaintiffs may take two paths in satisfying their Title VII burden of proof: (1) they may offer direct evidence of discriminatory intent; or (2) they may use the indirect, burden-shifting procedure set forth in McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973). Gonzalez v. Ingersoll Milling Machine Co., 133 F.3d 1025, 1031 (7th Cir. 1998). Given the difficulty of proving discrimination directly, Sonii and Jones have opted for the burden-shifting approach.
Plaintiffs correctly point out that the McDonnell Douglas approach is not always appropriate for analyzing an employment discrimination claim. For example, the Eighth Circuit found the approach to be "of little use" where an employer maintained an essentially segregated workforce, and the plaintiffs were never even considered for the jobs in question. Lams v. General Waterworks Corp., 766 F.2d 386, 391 (8th Cir. 1985). While recognizing "the need for flexibility in application of the McDonnell-Douglas guidelines to differing factual situations," id., the court finds no reason to depart from the guidelines in this case.
Under this approach, plaintiffs "must initially establish a prima facie case of racial and/or ethnic discrimination by a preponderance of the evidence." Id. at 1032. To do so, they must show that (1) they were in a protected class; (2) they were performing their jobs satisfactorily; (3) they were the subjects of materially adverse employment actions; and (4) others outside the protected class were treated more favorably. Id. Once they establish a prima facie case, the burden shifts to General Electric to "articulate some legitimate, nondiscriminatory reason" for its action. Id. (quoting Flowers v. Crouch-Walker Corp., 552 F.2d 1277, 1281 (7th Cir. 1977)). If General Electric meets this burden, plaintiffs then must show that the articulated reason was in fact pretext for the discrimination. Id.
General Electric contends that neither Sonii nor Jones can establish a prima fade case under Title VII. As to Sonii, General Electric argues that he cannot show that he was performing his job satisfactorily, or that others outside the protected class were treated more favorably. General Electric points out that Sonii was discharged only after being placed on PIP and still having failed to achieve the performance goals set for him. Plaintiffs respond by showing that Sonii ranked ahead of several other technicians in various performance categories, including overall performance. In reply, General Electric argues that those performance categories are irrelevant because they did not underlie the termination decision; rather, Sonii's failure to meet his required margin per day led to his dismissal.
The court agrees that an employee's adequate performance in certain areas does not necessarily translate into a satisfactory overall job performance. Sonii may have had exemplary customer skills, but if he was not bringing in revenue, General Electric may have been justified in deeming his performance inadequate. The problem here, however, is that the basis on which Sonii's performance was deemed inadequate is the core of Sonii's disparate treatment allegation. Sonii argues that the company's discriminatory assignment policy hindered his ability to bring in revenue; General Electric justifies Sonii's dismissal by pointing out his failure to bring in revenue. This is circular reasoning, at best. Sonii's failure to overcome the discriminatory constraints of the assignment policy does not entitle General Electric to summary judgment. This element of the prima facie case is the core dispute in this case.
General Electric next argues that Sonii has failed to show that white service technicians were treated more favorably than he was. General Electric asserts that race was not a factor in the assignment of service areas to technicians, and that whites also made service calls to predominately minority areas. Thus, any economic impact of the assignment system would have been a function of legitimate non-racial factors, such as a technician's residence or seniority.
Plaintiffs rebut this argument with a chart showing that, even though some white technicians lived near Sonii and Jones, they were assigned to predominately white service areas in the suburbs and downtown. Plaintiffs also offer statistics showing that while blacks made up only 16% of the technicians living in the South Zone, they handled 74% of the service calls made in predominately black areas of Cook County, but only 14% of the calls in predominately white areas. Over the entire Chicago region, black technicians handled 70% of the service calls in predominately black areas, but only 7.6% of the calls in predominately white areas. Other statistics show that only 3% of the service calls made by non-black technicians were in predominately black areas.
Many of these statistics were provided in an affidavit from plaintiffs' expert Maxine Mitchell. The court previously denied General Electric's motion to strike this affidavit, but indicated that the motion's substance would be considered in the context of this summary judgment ruling. The court finds that General Electric's arguments do not warrant disregard of the affidavit.
Even more tellingly, the statistics — if found credible by the jury — show that black and non-black technicians living in the south suburbs were assigned to completely different areas. Jones and Sonii were two of six black technicians living in the south suburbs, along with twelve non-black technicians. The twelve non-black technicians handled 94.3% of the service calls in the suburbs. Two of the black technicians together handled 5.7% of the suburban calls; the other four black technicians did not handle any suburban calls.
General Electric's insistence that service areas were assigned based solely on non-racial-factors is also belied by Jones' experience. According to Jones, he made numerous requests to be assigned to the predominately white south suburbs near his home. His requests were denied. However, when the service technician assigned to the south suburbs retired, he was replaced by a white technician with less seniority than Jones. If this evidence is found credible by the jury, the jury may conclude that service assignments were influenced by the technicians' race, and that white technicians were treated more favorably than plaintiffs were.
General Electric moved to strike this portion of Jones' affidavit for three reasons. First, General Electric cites an unpublished opinion for the notion that the testimony cannot be considered because it occurred years before any claims were filed. This argument fails given the nature of the challenged practice — a covert practice of discrimination over a long period. Events outside the limitations period are admissible in this context. See Selan v. Kiley, 969 F.2d 560, 565 (7th Cir. 1992). Second, General Electric alleges that the evidence is not competent because the individuals Jones claims he spoke to did not become managers in Chicago's south zone until 1991 (and Jones alleges that he made the requests in the late 1970s through the 1980s). This argument goes to the weight and credibility of the evidence, not its admissibility. Third, General Electric argues that portions of Jones' affidavit constitute hearsay. However, the court is not relying on the substance of General Electric's response to Jones' request, but simply that his request was not granted, and the assignment ultimately went to a white technician with less seniority.
Because plaintiffs have made a prima facie showing of discrimination as to Sonii, the burden shifts to General Electric to offer a legitimate nondiscriminatory reason for its service call assignment system. General Electric does so by pointing out that assigning technicians to service areas near their homes saves time and transportation costs, and that it also comports with technicians' own preferences. By concentrating technicians service calls near their homes — thereby limiting the time wasted in transit — General Electric allows technicians to maximize their productivity and performance.
Even assuming the validity of General Electric's efficiency-driven efforts to keep technicians close to their homes, the justification falls flat in light of the company's alleged reluctance to assign black technicians to nearby areas that are predominately white. Specifically, the denial of Jones' requests to be assigned to the predominately white south suburbs, along with the chart depicting the dispersion of calls assigned to black versus non-black technicians — again, if found credible by the jury — show that General Electric's assignment policy was not based solely on legitimate, nondiscriminatory objectives. For summary judgment purposes, if the evidence viewed in the light most favorable to the plaintiffs would permit the jury to infer that General Electric's justification is a lie, the plaintiffs have created a triable issue of pretext. See Anderson v. Baxter Healthcare Corp., 13 F.3d 1120, 1123-24 (7th Cir. 1994). Accordingly, General Electric's motion for summary judgment on Sonii's disparate treatment claim is denied.
General Electric next challenges Jones' disparate treatment claim, arguing that Jones cannot show that he was subject to a materially adverse employment action. Unlike Sonii, Jones voluntarily retired from his position with General Electric. He did not have his salary or benefits reduced before retiring, nor was he demoted, transferred, or subject to having his responsibilities reduced. General Electric correctly points out that a negative performance evaluation does not constitute an adverse employment action. See Sweeney v. West, 149 F.3d 550, 556 (7th Cir. 1998).
Plaintiffs concede that Jones was never terminated, but assert that, given the circumstances, his early retirement constitutes a constructive discharge. According to plaintiffs, Jones did not want to retire, but felt that he had no choice. They allege that his work was made difficult by the discriminatory assignment policy, and point out that Jones had asked his supervisor to accompany him on his service calls so that the supervisor could see the conditions in which he worked. After his union steward told him that he had been placed on PIP, Jones believed that the company was intent on terminating him.
Through the doctrine of constructive discharge, "a plaintiff who is forced out by discriminatory conduct may bring a successful Title VII claim even though the plaintiff was never officially dismissed by the defendant." Vitug v. Multistate Tax Comm'n, 88 F.3d 506, 517 (7th Cir. 1996). To establish a constructive discharge, Jones must show "that the defendant made the working conditions so intolerable as to force a reasonable employee to leave." Id. He also must show that he was constructively discharged due to his membership in a protected class. Id.
General Electric argues that Jones' 1995 retirement cannot be construed as a constructive discharge because he had made service calls in roughly the same geographical areas since 1967. Accordingly, if having to generate the required revenue in predominately black, low-income areas was intolerable to Jones, he would not have withstood those conditions for nearly thirty years before voluntarily taking early retirement. In response, plaintiffs argue that the conditions of his employment had changed since 1967 — most notably, Jones believed that he had been placed on PIP sometime in 1994. General Electric points out that, even assuming this to be an intolerable change in employment condition, Jones did not retire until October 1995.
The court agrees with General Electric. Jones' allegations reflect, at most, that General Electric's assignment policy made his job more difficult — both in terms of bringing in the required revenue, and in terms of on-the-job safety and security. The increased difficulty or decreased safety of an employment setting does not necessarily render the setting "intolerable." This is borne out by the fact that Jones did tolerate General Electric's assignment policy for many years.
The fact that Jones may have complained to his supervisor about his assignment does not help his cause; more is needed to establish a constructive discharge. See Drake, 134 F.3d at 886 ("More than ordinary discrimination is necessary to establish a constructive discharge claim; in the `ordinary' case, an employee is expected to remain employed while seeking redress."). Jones needed to show that the assignment policy — in combination with the company's performance demands — was "so onerous or demeaning that [he] was compelled to leave" the job. Harriston v. Chicago Tribune Co., 992 F.2d 697, 705 (7th Cir. 1993). Because Jones has not done so, General Electric's motion for summary judgment is granted as to Jones' disparate treatment claim.
Disparate Impact
In Count II of their complaint, plaintiffs allege that General Electric's assignment system had a disparate impact on black technicians by making it more difficult for them to generate the required revenue. To succeed on this claim, plaintiffs must show that a particular employment practice of General Electric has caused a disparate impact on the basis of plaintiffs' race. General Electric may rebut plaintiffs' showing by demonstrating either that the challenged practice does not cause a disparate impact, or that the practice is job-related and consistent with business necessity. 42 U.S.C. § 2000e-2 (k).
Plaintiffs portray the disparate impact by pointing to the wide disparity in median household income in areas assigned to black technicians versus areas assigned to non-black technicians. The parties agree that the median income in areas where black technicians made service calls was $27,403, compared to $47,671 in areas where whites were assigned. According to plaintiffs, household income has a significant impact on the technicians' ability to bring in revenue. For example, out-of-warranty calls are an important revenue-generator for technicians. Plaintiffs' statistics reflect that non-black technicians are assigned to 85% of the out-of-warranty calls, and that 90% of these calls are in predominately white areas. Further, even where an out-of-warranty service call is made in a predominately black, low-income area, plaintiffs allege that the customer often cannot afford to pay for the repair. Plaintiffs' statistics underlie the common-sensical core of their claim: that the revenue-starved environment to which they are assigned necessarily has a disparate impact on their ability to generate revenue.
General Electric seeks to rebut plaintiffs' showing with statistics of its own. According to General Electric, a statistical analysis based on zip code areas reveals that predominately black zip codes have lower incomplete call rates, and that there is no statistically significant difference in out-of-warranty opportunities in black versus white zip codes. General Electric also submits statistics to show that black technicians perform better than non-black technicians as measured by revenue per out-of-warranty calls and percentage of low-revenue, out-of-warranty calls.
As for the contention that customers in plaintiffs' assigned areas have less money to pay for out-of-warranty repairs, General Electric contends that the repair time saved balances out the revenue lost. In other words, according to General Electric, where the customer has no money to pay for the repair, the service call takes less time because no repair is made; thus, the technician is able to complete more calls. The flaws in this reasoning are readily apparent: if many of a technician's out-of-warranty customers are unable to pay for a repair, the technician's ability to visit more such customers may simply lead to more unrealized revenue. More fundamentally, however, General Electric's rebuttal interpretation of plaintiffs' evidence may influence the jury's weighing of that evidence, but it is not the sort of showing that warrants summary judgment.
As for the statistics offered by General Electric, plaintiffs contest their validity. For example, they impugn the validity of those statistics derived from zip codes, contending that zip codes encompass too large an area to reflect race or income levels accurately. Conversely, General Electric argues that plaintiffs' statistics derived from census blocks are irrelevant to analyzing the company's assignment system, which is based on zip codes. These are by no means the only statistical salvos fired, as each side launches both broad-brush and detailed criticisms of the opposition's submissions. The court finds it unnecessary to address each criticism raised, for the majority take aim at the credibility or probative value of the statistic offered. None justify the wholesale discard of a party's statistical submission. Absent a well-founded Daubert motion, the statistics are best left to be explored at trial.
While the court finds that the statistics are sufficient to warrant evaluation by the jury, the court makes no finding as to their probative value. See Fisher v. Transco Services-Milwaukee, Inc., 979 F.2d 1239, 1245 (7th Cir. 1992) (observing in Title VII case that "while the statistical analysis will have to be fully developed and presented at trial, we believe that the granting of summary judgment in favor of [the employer] on this [disparate impact] issue by the District Court was premature.").
For summary judgment purposes, what is important is that General Electric's rebuttal evidence does not amount to a direct or comprehensive refutation of plaintiffs' disparate impact showing. Rather, General Electric offers anecdotal statistics and evidence suggesting that black technicians' performance measures up fairly well to that of their white counterparts. The court cannot find, as a matter of law, that General Electric's evidence negates plaintiffs' disparate impact showing.
Accordingly, General Electric must show that its assignment policy is job-related and consistent with business necessity. Once again, it does so by setting forth the economic and time efficiencies served by keeping service technicians close to their homes. And once again, this justification fails where General Electric declines to assign black technicians to service calls in predominately white areas close to their homes. If the jury accepts plaintiffs' evidence on this issue, it may very well find that General Electric's assignment policy had a disparate impact on plaintiffs, and that the policy's scope was inconsistent with business necessity. General Electric's motion for summary judgment on plaintiffs' disparate impact claim is denied.
Conclusion
For the above reasons, General Electric's motion for summary judgment is granted as to Jones' disparate treatment claim, but is denied in all other respects.