The U.S. Court of Appeals for the Third Circuit recently became the first appellate court to find that so-called “subgroup” disparate impact claims are cognizable under the Age Discrimination in Employment Act (ADEA), which prohibits age discrimination against individuals age 40 and older.¹  In Karlo v. Pittsburgh Glass Works, the Third Circuit ruled that the plaintiffs stated a disparate impact claim where a reduction-in-force (RIF) disproportionately impacted employees over 50 years old.  According to the court, plaintiffs can demonstrate a disparate impact by comparing subgroups of older employees within the protected class (such as those over 50 years old) with employees both outside of the protected class and younger employees within the protected class (such as those between 40 and 50 years old).  Karlo has created a split in the circuits, with the Second, Sixth, and Eighth Circuits previously rejecting the use of subgroup comparisons. 

The Decision

The defendant implemented a RIF that resulted in the termination of 100 salaried employees, including the plaintiffs.  In conducting the RIF, the court found that the defendant did not employ any written guidelines or policies, conduct any adverse impact analysis, review prospective terminees with counsel, or document why any particular employee was selected for inclusion.   

The plaintiffs, all of whom were over 50 years old, filed a putative ADEA collective action, claiming the RIF had a disparate impact on a subgroup of employees within the ADEA’s protected class (age 40 and over).  Specifically, the plaintiffs claimed that the RIF disproportionately impacted employees over 50 years old.    

The defendant filed a motion for summary judgment, arguing that the plaintiffs’ “50-and-older” disparate impact claim was not cognizable under the ADEA.  The defendant argued the relevant comparison on a disparate impact ADEA claim was how the RIF impacted employees over 40 versus employees under 40.  The trial court held that plaintiffs’ subgroup disparate impact claim was not cognizable and dismissed it. 

On appeal, the Third Circuit reversed.  The court found that plaintiffs can demonstrate a disparate impact ADEA claim with various forms of evidence, including 40-and-older comparisons, subgroup comparisons, or more sophisticated statistical modeling.  The court reasoned that although the ADEA limits the protected class to those who are 40 and older, that limitation did not impact the ADEA’s general prohibition against age discrimination.  In other words, if the employer selects a 50-year-old employee for termination over a 40-year-old because of his age, there still is a cognizable claim for age discrimination.  Further, the court suggested requiring 40-year-olds to be included in the disparate impact comparison group would wash out the statistical evidence of age discrimination.  Applying this reasoning to the disparate impact context, the court concluded that the plaintiffs could establish an unlawful adverse impact in a RIF, or other facially neutral policy, by proof of disparate impact on a group of employees within a subgroup of the protected class, such as employees over 50.    

Conducting Adverse Impact Analyses

Although likely to be appealed, the Karlo decision influences how employers should conservatively approach adverse impact analyses when implementing RIFs.  As most employers know, the EEOC issued regulations in 2012 addressing the employer’s “reasonable factor other than age” defense to a claim of disparate impact under the ADEA.² The EEOC specifically referred to a RIF as the example of an adverse impact claim subject to the regulations in its comments on the revised regulations.³ The EEOC stated that whether the employer assessed the adverse impact of its employment practice on older workers was a consideration relevant to use of an employment practice that was a reasonable factor other than age.⁴ The comments to the revised regulations also state that employers would be prudent to assess the adverse impact of a selection procedure but remark that the extent of the assessment depends on the circumstances.⁵ The Q & A on the final rule summarized this concept:

Where an assessment of impact is warranted, the appropriate method will depend on the circumstances, including the employer’s resources and the number of employees affected by the practice. For example, a large employer that routinely uses sophisticated software to monitor its practices for race- and sex-based disparate impact may be acting unreasonably if it does not similarly monitor for age-based impact. Other employers, lacking the resources or expertise to perform sophisticated monitoring, may show that they acted reasonably by using informal methods of assessing impact.⁶

The EEOC tends to give an expansive reading to most of its regulations, which suggests that employers should consider assessing the adverse impact in most RIF’s or other group employment decisions.  Based on the Karlo decision, it may not be sufficient for employers to conduct adverse impact analyses that compare only employees under 40 with employees over 40.  This is because such a comparison would not necessarily account for potential “subgroup” disparate impact claims the Karlo court has now recognized.   

Recommendations for Employers

• Most employers should consider conducting adverse impact analysis whenever a RIF is implemented that impacts a statistically significant number of employees. 
• The company’s legal team should be involved in the preparation of the adverse impact analysis and take steps to maintain privilege over the analysis and other documentation regarding the implementation of the RIF. 
• The adverse impact analysis should include subgroup comparisons within multiple age bands instead of focusing only on an over-40 versus under-40 comparison.  This would include, for example, an over-40 age band, an over-50 age band, and an over-60 age band.
• The appellate court made a special point in the Karlo decision that the employer did not have any policies related to RIF’s, did not assess adverse impact, did not review potential selections with counsel, or even document the reasons for selection decisions.  Regardless of whether those events actually occurred or whether it was just a matter of proof submitted to the trial court, this decision is a reminder that employers should always document termination decisions in whatever context so that evidence is available to defend the decisions if there is a subsequent challenge.  Documented policies and procedures for RIF’s also are advisable, as are working relationships with in-house or outside counsel and statistical specialists.  Even though the Karlo decision only directly applies to employers in the states comprising the Third Circuit (Delaware, New Jersey, Pennsylvania, Virgin Islands), employers should consider policies and procedures that comply with this decision in their RIF and other group decision-making processes.