The Equal Employment Opportunity Commission (EEOC) brought this action against McDonnell Douglas Corporation under the Age Discrimination in Employment Act (ADEA), see 29 U.S.C. 621-634. The EEOC alleges that during a reduction in force (RIF) between May, 1991, and February, 1993, McDonnell Douglas engaged in a pattern or practice of terminating employees 55 years and older because of their age. The EEOC's complaint makes both disparate-impact and disparate-treatment claims. The district court dismissed the disparate-impact claim and granted summary judgment in favor of McDonnell Douglas on the disparate-treatment claim. The EEOC appeals and we affirm the judgment of the district court.

We consider first the EEOC's disparate-impact claim. While several of our sister circuits have held that the ADEA cannot support claims based on disparate impact, see Mullin v. Raytheon Co., 164 F.3d 696, 701 (1st Cir. 1999) (summarizing cases), the law of this circuit is that disparate-impact claims are cognizable under the ADEA. See Smith v. City of Des Moines, 99 F.3d 1466, 1470 (8th Cir. 1996). To prevail on a disparate-impact claim, a plaintiff must prove, as a threshold matter, that the challenged employment practice, while facially neutral, has a disparate impact on certain employees "because of their membership in a protected group." Watson v. Fort Worth Bank and Trust, 487 U.S. 977, 994 (1988) (plurality opinion); see also Lewis v. Aerospace Community Credit Union, 114 F.3d 745, 750 (8th Cir. 1997), cert. denied,118 S. Ct. 1392 (1998). Under the ADEA, the protected group consists of "individuals who are at least 40 years of age." See 29 U.S.C. 631(a).

Here, however, the EEOC does not claim that the RIF at McDonnell Douglas had a disparate impact on the entire class of persons protected by the statute, that is, those employees aged 40 or older (nor would its proffered statistical evidence support such a claim). Instead, the EEOC alleges that McDonnell Douglas's practice of basing RIF decisions on considerations such as retirement eligibility, merit raises, and salary had a disparate impact on a subgroup of the protected class, namely, those employees aged 55 or older. In support of its claim, the EEOC relies on statistical evidence that shows that the company laid off 13.7 percent of its employees aged 55 or older, compared to 5.4 percent of its employees under 55. The EEOC is thus asking us to expand our recognition of disparate-impact claims under the ADEA to include claims on behalf of subgroups of the protected class. We decline to do so.

In rejecting the EEOC's disparate-impact claim, the district court relied on Lowe v. Commack Union Free School District, 886 F.2d 1364, 1373 (2d Cir. 1989), cert. denied, 494 U.S. 1026 (1990), which held that disparate-impact claims on behalf of subgroups are not actionable under the ADEA. The district court reasoned that because there are an infinite number of variations and possible subgroups within the protected class of persons aged 40 or older, any plaintiff could attempt to establish a disparate-impact claim simply by taking his or her own age as the lower limit of the subgroup, and then using statistical evidence to show a disparate impact on that subgroup. But we think that in any case it is important to note that not every plaintiff would be successful: We can certainly envision cases that would involve an age distribution in the relevant work force that would not support a claim of disparate impact on behalf of any subgroup of the protected class. Even if it were true that in many or even most cases statistically significant evidence might support the claim of a disparate impact on some subgroup, we see no reason why that should preclude disparate-impact claims on behalf of subgroups. The fact that a particular interpretation of a statute might spawn lawsuits is not a reason to reject that interpretation.

We believe that there are stronger reasons for refusing to recognize such claims. For one thing, if such claims were cognizable under the statute, a plaintiff could bring a disparate-impact claim despite the fact that the statistical evidence indicated that an employer's RIF criteria had a very favorable impact upon the entire protected group of employees aged 40 and older, compared to those employees outside the protected group. We do not believe that Congress could have intended such a result.

We agree, moreover, with the district court that if disparate-impact claims on behalf of subgroups were cognizable under the ADEA, the consequence would be to require an employer engaging in an RIF to attempt what might well be impossible: to achieve statistical parity among the virtually infinite number of age subgroups in its work force. Adoption of such a theory, moreover, might well have the anomalous result of forcing employers to take age into account in making layoff decisions, which is the very sort of age-based decision-making that the statute proscribes.

More importantly, in this case the EEOC itself maintains that McDonnell Douglas relied on criteria such as retirement eligibility, salary, and seniority in making its layoff decisions. We have held that employment decisions motivated by factors other than age (such as retirement eligibility, salary, or seniority), even when such factors correlate with age, do not constitute age discrimination. See Hanebrink v. Brown Shoe Co., 110 F.3d 644, 647 (8 th Cir. 1997); see also Hazen Paper Co. v. Biggins, 507 U.S. 604, 611 (1993). We certainly do not think that Congress intended to impose liability on employers who rely on such criteria just because their use had a disparate impact on a subgroup.

The EEOC contends that the Supreme Court's decision in O'Connor v. Consolidated Coin Caterers Corp., 517 U.S. 308 (1996), supports its argument that the ADEA recognizes disparate-impact claims on behalf of subgroups. We disagree. O'Connor addresses disparate-treatment claims under the ADEA, id. at 309, and holds merely that where there is evidence of discriminatory animus, a plaintiff can make a prima facie case of disparate treatment by showing that he or she was replaced by a younger employee, whether or not the younger employee was also within the protected class of employees aged 40 or older. Id. at 312. The Court in O'Connor did not address disparate-impact claims under the ADEA, and thus we do not think that O'Connor has any relevance to our analysis here. We therefore affirm the district court's dismissal of the EEOC's disparate-impact claim.

The EEOC's disparate-treatment claim alleges that during the RIF McDonnell Douglas engaged in a pattern or practice of discriminatory treatment of employees aged 55 or older. To establish a prima facie case of pattern-or-practice discrimination, a plaintiff must prove that the employer "regularly and purposefully," International Brotherhood of Teamsters v. United States, 431 U.S. 324, 335 (1977), treated members of the protected group less favorably and that unlawful discrimination was the employer's "regular procedure or policy," id. at 360. "Proving isolated or sporadic discriminatory acts by the employer is insufficient ... rather it must be established by a preponderance of the evidence that ... 'discrimination was the company's standard operating procedure -- the regular rather than the unusual practice.' " Cooper v. Federal Reserve Bank of Richmond, 467 U.S. 867, 875-76 (1984), quoting Teamsters, 431 U.S. at 336.

The EEOC directs our attention to certain evidence that, it asserts, shows that McDonnell Douglas engaged in a pattern or practice of discrimination against employees aged 55 or older. First, it points to statistical evidence indicating that 13.7 percent of employees aged 55 or older were laid off during the RIF, compared to 5.4 percent of employees under 55. Employees 55 or older were thus more than twice as likely as younger employees to be laid off.

We have previously recognized, however, that an important statistic to consider in the RIF context is the difference in the percentage of older employees in the work force before and after the RIF. See Holley v. Sanyo Manufacturing, Inc., 771 F.2d 1161, 1167 (8 th Cir. 1985) (no statistical basis for age discrimination claim where percentage of employees in protected class was 25.8 percent before layoffs and 26.0 percent after layoffs); see also Hanebrink, 110 F.3d at 646 (no statistical basis for age discrimination claim where average age of employees declined by only one-half year after layoffs). Here, the evidence establishes that employees aged 55 or older comprised 14.7 percent of the total work force at McDonnell Douglas prior to the RIF, and 13.6 percent of the work force after the RIF. We agree with the district court that the difference in these "before and after" numbers is insignificant, and does not support an inference of age discrimination. We conclude, therefore, that the EEOC's statistical evidence, standing alone, is insufficient to establish a prima facie case of pattern-or-practice age discrimination.

The EEOC also relies, however, on anecdotal evidence that some managers at McDonnell Douglas selected employees aged 55 or older for layoff because of their age, higher salaries, and retirement eligibility (under McDonnell Douglas's pension plan, an employee must be at least 55 years old and have at least ten years of service to be eligible for retirement). Specifically, several managers testified that retirement eligibility was a factor in layoff decisions, and some managers' files contained rosters listing employees' ages, projected retirement dates, years of service, and other retirement-related data. In addition, some former employees testified that they were told by supervisors that they had been selected for layoff because of their age, years of seniority, high salary, and retirement eligibility. Finally, witnesses testified that McDonnell Douglas "manipulated" its performance evaluation system by rating employees aged 55 or older lower than comparable younger employees, so that these evaluations could be used to justify laying off the older employees during the RIF.

After carefully reviewing the EEOC's proffered evidence, we believe that a reasonable jury could not conclude that to the extent retirement eligibility and salary were factors in layoff decisions, managers chose retirement-eligible employees for the RIF because of their age. As we have already said, employment decisions motivated by factors other than age (such as salary, seniority, or retirement eligibility), even when such factors correlate with age, do not constitute age discrimination. Id.; see also Hanebrink, 110 F.3d at 647. As for the EEOC's anecdotal evidence, it demonstrates, at most, isolated discriminatory acts on the part of certain managers, rather than McDonnell Douglas's "standard operating procedure," Teamsters, 431 U.S. at 336. (We note that the various "RIF Guidelines" developed and distributed by McDonnell Douglas divisions participating in the RIF advised decision-makers to consider factors such as skills and experience, disciplinary record, and length of service in identifying employees for layoff. None of the "RIF Guidelines" mentioned age as an appropriate factor in layoff decisions.). We come to the same conclusion regarding the alleged "manipulation" of older employees' performance evaluations. In both of these last instances, there is insufficient evidence from which a reasonable jury could conclude that McDonnell Douglas engaged in a pattern or practice of age discrimination.

Finally, the EEOC contends that in the years leading up to and during the RIF, senior management officials at McDonnell Douglas expressed a strong regard for younger employees and a corresponding disregard for older employees, and that this "cultural focus of youth" at the company created an environment of pervasive age bias that tainted RIF decisions. In support of this contention, the EEOC points to a 1986 memorandum authored by John McDonnell, who was president of McDonnell Douglas at that time and CEO of the company during the RIF. In the memorandum, Mr. McDonnell stated that in order "to stay competitive" the company needed "to attract and retain capable young people." The EEOC also points to the suggestion contained in a 1987 human resources memorandum that the company organize forums for "young change agents." In addition, a 1989 company document solicits nominations for "high potential/high achiever candidates" and specifies candidates who "have NOT reached their 40 th birthday." Other documents express concern about the company's "aging workforce" and "aging management team." Finally, the EEOC points to a 1991 videotaped presentation in which CEO McDonnell stated that employees over 50 have trouble changing.

We agree with the district court that most, if not all, of this evidence is too remote in time from the events at issue in this case to be of any probative value. In any event, we do not believe that a reasonable jury could conclude that this evidence indicates a pervasive age bias that affected layoff decisions during the RIF at a huge company like McDonnell Douglas.

We conclude that based upon the statistical and anecdotal evidence in the record no reasonable jury could find that McDonnell Douglas engaged in a pattern or practice of discrimination against employees aged 55 or older during the RIF. We therefore affirm the district court's grant of summary judgment in favor of McDonnell Douglas on the EEOC's disparate-treatment claim.

For the foregoing reasons, we affirm the judgment of the district court.