3rd Circ. Must Reject EEOC's Flawed Equal Pay Theory
On Feb. 5, the U.S. Equal Employment Opportunity Commission filed an amicus brief in Cartee-Haring and Marinello v. Central Bucks School District, pending in the U.S. Court of Appeals for the Third Circuit.
The plaintiffs in the case are English teachers, and the commission supports their contention that their lower pay rate, relative to a male social studies teacher — i.e., the comparator, or person used as a standard of comparison — establishes a prima facie case of sex discrimination in pay under the Equal Pay Act.
We join them in referring to this principle as the "single comparator" rule. Under this rule, the lower pay of other males, who perform similar work and may outnumber the one or two males shown to be paid more than the plaintiffs, is deemed irrelevant to whether the plaintiffs' pay was decided on the basis of sex. Indeed, this evidence of other potential comparators was precluded at trial.
In our view, this rule allows plaintiffs to present the jury with a false picture of pay inequity. Focusing on a single pay comparison between a female plaintiff and a comparator of her choosing potentially blinds the jury to an untold number of comparisons in which men are paid less than similarly situated women. By precluding evidence regarding other male comparators, who may be paid less than the plaintiffs, as well as women who may be paid more than the chosen comparator, the single-comparator rule throws the baby out with the bathwater.
Indeed, it would be surprising if employees of each sex could not point to some employee of the opposite sex who is similarly situated but paid more. Thus, we believe prima facie evidence of an Equal Pay Act violation must demand more than anecdotal evidence regarding a single comparator, and include a wider spectrum of evidence from which a jury may decide whether a pay difference was on the basis of sex.
We illustrate the weakness in the EEOC's argument by demonstrating that, followed to its logical end, the single-comparator rule mandates pay uniformity, not pay equality. We provide a reductio ad absurdum — or reduction to absurdity — to illustrate why the EEOC's interpretation is flawed and urge the Third Circuit to abandon an interpretation of the statute that leads to illogical outcomes.[1]
Consider a thought experiment in which an employer hires one set of twins, Bill and Mary Jones, and another set of twins, Bob and Suzy Smith. These twins share all the same demographics, except gender. The employer pays each individual of each set of twins exactly the same, but there is a pay difference between the pairs of twins in the same job — i.e., Bob and Suzy Smith are paid the same, but both are paid less than Bill and Mary Jones.
By any reasonable assessment, this pay difference is attributable to Smith discrimination, not sex discrimination: Men and women are treated identically by this employer. But under a single-comparator rule, both Bob and Suzy can state a prima facie case under the Equal Pay Act, with Bob pointing to Mary, and Suzy pointing to Bill as their comparators. Thus, the same facts support a claim of sex discrimination by an employee of each sex against the same employer.
This is not the end of the story, and this no longer is a hypothetical. In any large company, there is a range of pay in every job category. No matter the sex of the highest-paid employee, an employer seeking to avoid creating prima facie evidence inevitably must pay all employees the same.
For example, if the highest-paid employee is female, then every lower-paid male can present prima facie evidence of pay inequality if the single-comparator rule provides the standard. Once their pay is adjusted, every remaining female also can provide prima facie proof of a gender pay difference. Alleged pay discrimination, or at least a prima facie case, is eliminated only when pay is uniform rather than equal.
Differences in pay among similarly situated men often exceed pay differences among similarly situated women, for reasons that researchers still don't understand. For example, the Bureau of Labor Statistics reports measures of the pay inequality separately for groups of men and in several educational groups.[2] These data consistently show greater pay differences among men than among women, all else equal.
According to December 2025 survey results, among full-time wage and salary workers who are college graduates, men at the 75th percentile of the pay distribution earned about 117% more than men at the 25th percentile. The corresponding interquartile difference for college-educated women was 105%. The larger pay differential between the 75th and 25th percentiles for men occurs for every education group, not just college graduates.[3]
The point is that pay differences are widespread and of great magnitude among workers of each sex, yet the single-comparator rule turns this innocuous fact into a pillar of the prima facie case.
For decades, economists have sought to understand why pay is so dispersed, but with limited success, causing much of this dispersion to be labeled simply an unexplained residual variance. Economists measure the residual variance in pay while accounting for multiple factors including education, experience, geographic area and occupation.
In 2024, we updated a 2006 analysis by University of British Columbia professor Thomas Lemieux, referenced below, and estimated pay dispersion among men and among women; we found that inequality in pay was greater among men. The largest gender differences in the residual variance in pay were among male college graduates and workers with more potential labor market experience.[4] Consequently, we suggest that pay differences among two randomly chosen men may exceed the pay difference between a randomly chosen man and a randomly chosen woman in similar jobs.
Yet, the Equal Pay Act prohibits only pay differences that arise on the basis of sex. Although the U.S. Supreme Court observed that "on the basis of" prescribes a causation requirement,[5] this requirement frequently has been ignored by the lower courts.
This omission is notable in the EEOC's brief, which ignores this key phrase. A better view is to require some evidence that a pay difference, even when established regarding a single comparator, is caused by sex discrimination. For example, a plaintiff may prove this by demonstrating both (1) that members of the plaintiff's sex on average are paid less for the same job, and (2) that the plaintiff in particular is paid less than an appropriate comparator.
This test responds to observations that resonated with the U.S. Court of Appeals for the Fourth Circuit in its 1993 ruling in Houck v. Virginia Polytechnic Institute: "[I]solated incidents or random comparisons demonstrating disparities in treatment may be insufficient to draw a prima facie inference of discrimination without additional evidence that the alleged phenomenon of inequality also exists with respect to the entire relevant group of employees."[6]
Accordingly, we too believe the prima facie case should require evidence supporting a reasonable inference that a pay difference is on the basis of sex. When the quantum of prima facie proof prescribed by the single-comparator rule is so low as to be within the grasp of a high fraction of employees of either sex, we should recognize that rule sweeps far too broadly.
In sum, we argue that, to avoid illogical outcomes and needless litigation, the Third Circuit should reject the EEOC's advocacy of the single-comparator rule, as articulated in its amicus brief in Cartee-Haring and Marinello v. Central Bucks School District.
Allan G. King is senior counsel at Littler Mendelson PC.
Stephen G. Bronars is a partner at Edgeworth Economics.
The opinions expressed are those of the author(s) and do not necessarily reflect the views of their
employer, its clients, or Portfolio Media Inc., or any of its or their respective affiliates. This article is
for general information purposes and is not intended to be and should not be taken as legal advice.
CITATIONS
[1] Our argument is elaborated at greater length in A.G. King and S. G. Bronars, "Pay Differences in
the Absence of Discrimination" Legislative Fallacies and Statistical Truths, 50 J. Legis. 23 (2024).
[2] See Table 5 in https://www.bls.gov/news.release/archives/wkyeng_12042025.htm.
[3] This general calculation was based on data collected over five years by the U.S. Census Bureau,
which dovetails with the Bureau of Labor Statistics reporting. See American Community Survey 2016-
2020 5-Year Data Release, U.S. CENSUS BUREAU (Mar. 17, 2022), https://www.census.gov/newsroom/press-kits/2021/acs-5-year.html.
[4] King and Bronars, supra note 1, Table 2, p. 46 and Thomas Lemieux, Increasing Residual Wage
Inequality: Composition Effects, Noisy Data, or Rising Demand for Skill?, 96 AM. ECON. REV. 461,
462 (2006).
[5] See, e.g., the discussion in Bostock v. Clayton County, 590 U.S. 644 (2020) (using "because of"
and "on the basis of" interchangeably, with each implying "but-for" causation).
[6] Houck v. Virginia Polytechnic Inst. , 10 F.3d 204, 206-07 (4th Cir. 1993).