Quirky question # 134
My company is planning a reduction in our workforce, which largely consists of African-Americans. I understand that when a company conducts a mass layoff, it should make sure the layoff does not disproportionately affect older workers, women or men, or employees of a particular race, or else the company risks being accused of discrimination. But doesn’t the law also protect employees from discrimination based on color? If so, do we also need to worry about letting go a disproportionate number of “dark-skinned” or “light-skinned” employees? How would we even go about measuring this? I know our African-American workforce consists of a broad spectrum of skin tones.
Joel's Analysis of Quirky Question # 134, Discrimination Based on Color
You raise a good question about whether a “disparate impact” claim can successfully allege discrimination based on color. While intriguing in theory, I think for practical reasons such a claim could not succeed, so I do not believe that you need to agonize over your employees’ varying skin tones when planning your reduction in force. To reach my conclusion, it is best to begin with a brief summary of anti-discrimination law.
As you note in your question, anti-discrimination law prohibits discrimination against individuals in certain protected classes, based on characteristics such as age, sex, race, national origin, or color. Employees generally can assert discrimination claims under two legal theories: disparate treatment and disparate impact. Disparate treatment claims alleging race, sex, age, or national origin discrimination typically involve an employee in the protected class claiming her employer intentionally treated her less favorably than persons outside of the employee’s protected class. For example, a female employee might sue claiming her employer denied her a promotion, instead advancing a less-deserving male coworker. In such a scenario, the distinction between the allegedly disfavored plaintiff and the favored coworker is based on a clear and evident difference—the employees’ sex.
Disparate impact claims do not allege intentional discrimination, but rather encompass the type of scenario you describe here – a facially-neutral policy (e.g., a reduction in workforce based on financial or business reasons) that disproportionately affects individuals in a protected class. For example, a female employee might sue after a mass layoff claiming that the group of terminated employees included too many females, while too few female employees were spared termination.
Cases involving alleged color discrimination are relatively rare, and typically are brought under a disparate treatment theory. In color-based cases, the aggrieved employee can be – and often is – the same race, sex, age, and national origin as the allegedly favored coworkers and the allegedly discriminatory supervisor. That employee, however, claims she is, in the employer’s eyes, “too dark” or “too light” in color. The cases, while different than those involving other protected classes, are straightforward. To determine the merits of such cases, the court must, among other things, compare the relative skin tones of an individual plaintiff and a limited number of her coworkers.
As an example, in one older case, a light-skinned Pakistani employee sued his Pakistani employer, claiming darker-skinned Pakistani employees were favored. Ali v. Nat’l Bank of Pakistan, 508 F. Supp. 611 (S.D.N.Y. 1981). Another, more recent case involved a light-skinned Native-American employee claiming that her employer favored darker-skinned Native-American employees. Nettle v. Cent. Okla. Am. Indian Health Council, Inc., 2009 U.S. App. LEXIS 14470 (10th Cir. July 1, 2009) (unpublished). In yet another case, an employer defended itself against an employee claiming that light-skinned employees were favored, by arguing that it had in fact replaced the dark-skinned plaintiff with a darker-skinned worker. Brack v. Shoney’s, Inc., 249 F. Supp. 2d 938 (W.D. Tenn. 2003). The congressional discussions leading to the enactment of the anti-discrimination laws specifically noted the laws’ applicability to cases like these, expressly noting the law would apply when, for example, an employer refused to hire an African-American because her skin pigmentation was “too dark,” even if the employer was also an African-American. See Cong. Rec., Feb. 8, 1964, at H-2552-2555.
Unlike these intentional discrimination cases, there do not appear to be any published court decisions involving claims that an employer’s facially-neutral practice or policy had a disparate impact on persons of a particular color. In theory, such a claim could exist, like the previous example of a female worker suing her employer following a mass layoff. In your situation, a dark-skinned African-American employee could argue your company’s layoffs, while ostensibly based on neutral criteria, resulted in the termination of too many darker-skinned employees, while too few darker-skinned employees remained employed.
Perhaps the chasm between theory and reality with these color-based disparate impact cases stems from the practicality of enforcing a claim. First, there is the problem of classifying skin-tone. Even scientists have a difficult enough time categorizing the wide variety of skin pigmentation in the human race – first using the complicated 36-tone Von Luschan chromatic scale, then more recently using the six-tiered Fitzpatrick scale (primarily for grading sunburn risk). We cannot expect an employer to do much better. While it is relatively simple to determine whether an employee is male or female, of one race or another (though the idea of race does have its complexities), or to determine in the disparate treatment context whether a single plaintiff is darker or lighter than a few allegedly favored coworkers, categorizing large numbers of employees based on skin tone is not a “black or white” question.
Second, even if scientists (and employers) could perfect the practice of color-grading, doing so at the workplace or in the courts seems rather distasteful. In fact, despite Congress’s apparent approval of color-based discrimination claims, several courts, even in disparate treatment cases, have cautioned the judiciary to be wary of “the unsavory business of measuring skin color and determining whether the skin pigmentation of the parties is sufficiently different to form the basis of a lawsuit.” See Sere v. Bd. of Trs. of the Univ. of Ill., 628 F. Supp. 1543 (N.D. Ill. 1986); see also Franceschi v. Hyatt Corp., 782 F. Supp. 712 (D.P.R. 1992) (noting that courts have “shied away from grappling with cases that deal with subtle degrees of intra-racial discrimination”); Walker v. I.R.S., 713 F. Supp. 403 (N.D. Ga. 1989) (observing the “genuine and substantial” difficulties in comparing skin pigmentation).
An additional problem with color-based disparate impact claims is the lack of a legal demarcation instructing an employer how to test the lawfulness of a facially-neutral practice. For sex, the employer can easily determine whether men or women are disproportionately impacted. Likewise, determining race or national origin represented in the workforce may be accomplished without much difficulty. Even age – which, like color, exists in the workplace in a broad range – contains a bright-line cutoff under federal law, with individuals age 40 and older afforded protected from discrimination. In contrast, skin color does not have any clear cutoff.
Without such a cutoff, we cannot expect employers to ensure employees throughout the skin-tone spectrum are equally affected or disaffected by an employment practice. Courts in age-based disparate impact cases have declined to accept employees’ requests for a protection against such “subgroup” discrimination. For example, in EEOC v. McDonnell Douglas Corp., 191 F.3d 948 (8th Cir. 1999), employees claimed that a workforce reduction, while not disproportionately affecting employees age 40 and older, disproportionately affected employees age 55 and older. The Eighth Circuit Court of Appeals rejected this argument. The court explained that “if disparate-impact claims on behalf of [age] subgroups were cognizable . . . , the consequence would be to require an employer engaging in a [reduction in force] to attempt what might well be impossible: to achieve statistical parity among the virtually infinite number of age subgroups in its workforce.” Id. at 951. Given courts’ reluctance to allow a disparate impact claim based on an age subgroup, courts would likely also recognize the analogous difficulties with a color-based theory. Certainly the Eighth Circuit’s reasoning applies to color-based disparate impact lawsuits: it may well be impossible for an employer implementing a neutral practice or policy to achieve parity among various categories of skin tone (be they six or thirty-six distinct subgroups) in its workforce.
For all of these reasons, I am skeptical as to whether a legitimate color-based disparate impact case could ever be properly pled in court. In sum, then, continue to test for whether your company’s layoff has a disproportionate impact on employees with respect to age (forty and older), gender, race, and national origin. But there is little reason, and inherent practical difficulty, in also testing for color-based disparate impact.