Jump to Navigation | Jump to Content
American Bar Association

ABA Section of Business Law

Volume 11, Number 4 - March/April 2002

The cutback that cuts the wrong way
Older workers claim disparate impact
    By Julie Badel

If a company is out to get rid of old workers, it better be careful.

A seemingly innocuous process of ranking employees and gradually weeding out the low performers has generated a surprising amount of controversy as well as litigation against Ford Motor Co. The evaluation process instituted at Ford differed from the kind of process typically used by employers when reducing the work force only because it required a fixed percentage of employees to be ranked in the bottom group, and certain consequences followed the low ranking.

Ford instituted a new evaluation policy for its top 18,000 managers and executives in January, 2000. The performance- evaluation process called for 10 percent of the group to be graded "A," 80 percent "B" and 10 percent "C," not unlike a typical bell curve. Employees receiving a C grade were not eligible for a raise or bonus. An employee who received a C grade for two consecutive years was subject to demotion or termination. According to USA Today, this evaluation system enabled Ford to avoid layoffs in an era of declining profits and increased costs.

Most employers would agree that this is a noble goal. In addition, adopting the Ford system would help many employers avoid the problems that often occur in relying on previous performance evaluations when selecting employees for termination during a reduction in force. Because so many supervisors are loath to criticize their subordinates in a review and provide an honest evaluation of their performance, employee performance reviews often do not reflect actual performance. The grade inflation so rampant in the schools has found its way into the workplace.

As a result, many employers who rely on existing reviews in selecting the lowest performing employees for termination during a reduction in force find little or no difference among employee ratings on their reviews. The system used by Ford essentially required supervisors to distinguish among employees by ranking some of them in the bottom group.

It is difficult to conjure up a legal reason why an employer cannot rank its at-will employees on the basis of performance, withhold monetary rewards to the poorest performers, and ultimately demote or terminate them.

However, following criticism by employees and a number of lawsuits, at least one of which claims the performance process targeted older workers, Ford modified the performance system in July, 2001 to classify workers into three groups — top achiever, achiever and improvement required — rather than using letter grades. Under the modified process, no set percentage of workers had to be categorized in the lowest group, according to the California Employment Law Letter.

The challenged Ford performance plan involves two noteworthy issues, the first being the use of a somewhat unique performance ranking system that requires some employees to be at the bottom of the totem pole. This type of evaluation system has been variously referred to as a "forced distribution" or "rank and yank" system. The second involves the use of a theory of age discrimination, that of disparate impact, on which courts have issued very divergent opinions.

Some 60 managers sued Ford in Michigan state court, claiming that the performance-management process had a "disparate impact" on older workers, according to the BNA Daily Labor Report. Employment discrimination claims generally are brought under one of two theories. A "disparate treatment" claim is one of intentional discrimination and alleges that an employer treated employees in a protected group differently from employees in a nonprotected group because of their membership in that group. For example, a claim that an employer hired only male waiters and refused to hire women as waitresses states a disparate treatment claim on the basis of sex.

A "disparate impact" claim, on the other hand, focuses on a neutral rule or practice that affects a protected group more harshly than others. As an example of a disparate impact claim, hiring only workers who are at least 6 feet 2 inches tall could have a disparate impact on women, most of whom are not 6 feet 2 inches tall. An employer can defend against a disparate impact claim by demonstrating that business necessity justifies the practice. Using this same hypothetical, an airline could defend the use of a height requirement (although probably not a requirement to be 6 feet 2 inches tall) for flight attendants on the ground that they have to be tall enough to reach the overhead bins.

In an oral ruling on July 31, 2001, a Michigan judge decided that the Ford employees could pursue their age discrimination claim on a disparate impact theory, a decision that The Daily Labor Report said a Ford lawyer called an issue of "first impression in Michigan." Michigan's Elliot-Larsen Civil Rights Act, the law under which the Ford employees sued, prohibits discrimination on the basis of religion, race, color, national origin, age, sex, height, weight, familial status and marital status. MCLS §37.2101 et seq.

Age discrimination under federal law is prohibited by the Age Discrimination in Employment Act, 29 U.S.C. §623, while disparate impact claims originated in cases brought under Title VII of the Civil Rights Act of 1964, 42 U.S.C. §2000e et seq., which prohibits discrimination on the basis of race, color, religion, national origin and sex. Because the language of the two federal statutes differs, federal courts disagree whether ADEA claims can be brought on a disparate impact theory.

While suggesting that it would decline to recognize a disparate impact claim under the ADEA, the U.S. Supreme Court has yet to rule on the matter. A case on the court's docket this term will require the court to resolve the issue. However, many of the federal courts of appeal interpret the Supreme Court's decision in Hazen Paper Co. v. Biggins, 507 U.S. 604 (1993), as a disinclination to recognize disparate impact claims under the ADEA.

Biggins claimed that his employer fired him — at age 62 — because of his age a few weeks before his tenth employment anniversary when he would have been vested in the company's pension plan. Both the jury and the federal court of appeals found in Biggins' favor. While Biggins brought his case under a disparate treatment theory and the court had no need to address the viability of a disparate impact theory, the court observed that in enacting the ADEA, Congress sought to prohibit disparate treatment.

Concerns that older workers were being deprived of employment because of inaccurate stereotypes about declining productivity and competence prompted congressional enactment of the federal ban on age discrimination, according to the court. The court noted that making an employment decision on a factor correlated with age, such as pension status, did not invoke the specter of inaccurate and stigmatizing stereotypes. It held that an employer does not violate the ADEA by interfering with an employee's pension benefits that would have vested by virtue of the employee's years of service.

Three justices concurred to underscore that the court's opinion should not be read to sanction a disparate impact theory of liability under the ADEA, noting the existence of substantial arguments for not carrying over disparate impact claims from Title VII of the Civil Rights Act of 1964 to the ADEA.

Following Hazen Paper, a number of the circuit courts of appeal, among them the First, Third, Sixth, Seventh, Tenth and Eleventh circuits, have declined to recognize a disparate impact theory under the ADEA. Adams v. Florida Power Corp., 255 F.3d 1322 (11th Cir. 2001). The lengthy set of reasons advanced by the Tenth Circuit in Ellis v. United Airlines Inc., 73 F.3d 999 (10th Cir. 1996), exemplifies the reasoning used by the other circuits that have declined to recognize a disparate impact claim.

First, the ADEA provides that it is not unlawful for an employer to take any action otherwise prohibited by the ADEA if based on "factors other than age." 29 U.S.C. §623(f). Title VII does not contain similar language. However, the Equal Pay Act mirrors this language, and the U.S. Supreme Court has interpreted that statute to preclude disparate impact claims.

Second, the court in Ellis interpreted the legislative history of the ADEA to encompass only purposeful acts of age discrimination, pointing to a secretary of Labor report recommending that Congress prohibit intentional discrimination based on stereotypes about age.

Third, the court observed that Congress amended Title VII of the Civil Rights Act of 1964 in 1991 explicitly to provide for a disparate impact cause of action. No parallel provision was added to the ADEA. As its fourth reason, the court pointed to the Supreme Court's decision in Hazen Paper. Finally, the Ellis court noted the trend among other courts to decline to recognize a disparate impact ADEA claim in the wake of Hazen Paper.

Other courts have allowed terminated employees to challenge ranking systems under a disparate treatment theory, a case in point being a Dolores Oubre, in Louisiana. Oubre's employer, Entergy Operations Inc., instituted a new employee evaluation process in which employees were ranked according to two criteria, performance and potential, in comparison with their peers. Ultimately, employees were placed into one of nine groups, and the lowest 10 percent of the employees had to be placed in the lowest group. Oubre found herself in this group. Given a choice between a severance package and an action plan for one year to upgrade her ranking, she accepted the severance package and then sued.

Oubre argued that company management deliberately implemented a ranking system that was prone to age stereotyping. The individual who devised the ranking system admitted that the system adversely affected older employees. Oubre also pointed out that prior to the new ranking system, she received positive evaluations every year. The court acknowledged that the facts lent themselves to a disparate impact theory, but concluded that the ranking process was not a neutral process. Instead, the court described the ranking as one that was impermissibly infected with age stereotyping. This exemplifies a classic case of disparate treatment, and the court denied the company's motion for summary judgment. Oubre v. Entergy Operations Inc., 1998 U.S. Dist. LEXIS 15306 (E.D. La. 1998).

Like Oubre, Charles Hartsell also challenged his employer's ranking system, which involved a required percentage of employees to be ranked in the bottom group in Hartsell v. The Boeing Co., 1998 U.S. Dist. LEXIS 3287 (D. Del. 1998). In preparing for a reduction in force, Hartsell's employer, Boeing, rated employees based on work competence, diligence in accomplishing assigned tasks, and work capabilities and classified them into three groups.

Hartsell was downgraded when he was viewed as the least qualified of the managers who had been placed in the lowest group. A panel that included both human resources personnel and older workers reviewed the decision and found nothing amiss in Hartsell's rating or his identification as a surplus employee.

However, the court seemed troubled that there were no performance reviews that incorporated prior corrective action notices, which Boeing pointed to in order to support its judgment of Hartsell's poor performance. As a result, the court denied Boeing summary judgment.

Other than establishing a set percentage of employees who had to be ranked in the bottom group, the evaluation process used in Oubre and Hartsell did not differ materially from the kind of ranking process typically used by employers during a workforce reduction geared toward selecting the poorest performers for termination. Provided that those processes use objective and job-related criteria, employers often prevail at the summary judgment stage.

The result when Armand Deshais and Karl Steig sued their employer, Consolidated Rail Corp., when they were laid off based on a performance rating is typical of the result an employer might expect from terminations based on a ranking system used to select the poorest performers. In Deshais v. Consolidated Rail Corp., 956 F. Supp. 230 (N.D. N.Y. 1997), the employer rated its supervisors based on performance evaluations (50 percent), a structured interview (30 percent), supervisor's recommendations (10 percent) and specific project/team activities (10 percent).

The 10 supervisors with the lowest scores, including 49-year- old Deshais and 47-year-old Steig, were selected for termination in a reduction in force. They had no evidence to support their claims of age discrimination other than an affidavit from a former employee who alleged that employees "perceived" a negative atmosphere toward older employees by management. In granting Conrail summary judgment, the court referred to the "time-consuming evaluation system" to provide a ranking of the employees.

Where does this leave employers? Performance evaluations may be more critical in today's economy than ever. Employers not bound by a union contract generally prefer to terminate their lowest performers rather than those employees with the least amount of seniority in an economic downturn. Ranking employees based on their performance evaluations is the logical step to take.

However, employers must carefully structure their evaluation systems and administer them uniformly. The following elements are critical to any evaluation system:

Employees must be evaluated on a regular basis, initially either three or six months after hire, and either annually or semiannually thereafter.

The criteria on which the employees are evaluated should relate to the job that the employee performs.

Whenever possible, the criteria evaluated should be objective (quantity of work, attendance) rather than subjective (leadership, cooperation), and when subjective criteria are used, they should be defined so all supervisors completing a review are rating the same characteristics. Subjective criteria, such as "potential" and "versatility," which could be argued as criteria to deselect older workers, should be avoided.

Evaluations should be honest. A falsely positive review is far more damaging to an employer's case than no review at all.

Reviews should be discussed with employees, and employees should be given an opportunity to reflect their comments on the review itself. Often, even a poor performer will acknowledge that a review critical of his or her performance is accurate.

All evaluations should be reviewed by human resources for consistency.

If the employee is given an overall rating, whether a descriptive word (outstanding, competent, needs improvement) or a number, the overall rating should be consistent with the individual ratings in each area. For example, a review in which an employee is rated "competent" in seven out of 10 areas but is given an overall rating of "needs improvement" may suggest a bias to provide the employee with a lower rating than his performance actually merits.

If an employee traditionally has had better reviews, the review itself should explain why it is significantly lower than the employee's reviews have been historically. Sometimes a new supervisor with higher expectations triggers a lower than normal rating. Other times, the employee has trouble learning new tasks or just seems to lose interest in the job. But whatever reason accounts for the decline, reflecting the problem in the review itself will help to stem arguments that the review is artificially critical in order to place the employee on a termination list.

Workforce reductions on the basis of performance make the most sense for employers. Employers should continue to make reduction-in-force decisions on the basis of employee performance and their resultant rank, and if performance is reviewed based on objective and job-related factors, with the safeguards discussed above, employers should be able to withstand most attacks from older workers.

Badel is a partner at Epstein Becker & Green, P.C., in Chicago. Her e-mail is jbadel@ebglaw.com.

Back to Top