Homegrown Discrimination

Homegrown Discrimination

When foreign labor recruiters, acting on foreign soil as agents of domestic growers, intentionally prefer young, non-disabled men as temporary agricultural workers in the United States, federal antidiscrimination law traditionally has offered no recourse because of the presumption against extraterritorial application of domestic statutes.  Accordingly, prospective migrant workers face discrimination abroad by American employers’ agents without any remedy in American courts. This essay considers a new theory to address that problem.  Specifically, instead of challenging foreign labor recruiters for engaging in disparate treatment abroad, it considers the possibility of challenging domestic growers for engaging in disparate impact at home.  The essay examines the prospect of challenging a grower for deciding, within the United States, to engage and reengage a specific foreign labor recruiter to hire temporary agricultural workers when doing so results in disparate impact and lacks business necessity.  A disparate impact cause of action may be viable in some contexts depending on the statute at issue and the framing of the hiring practice, potentially giving migrant workers a new tool to combat growers’ homegrown discrimination.




From 1942 until 1964, the Bracero Program allowed United States growers to staff their farms with Mexican migrant workers.[1] Despite the facial neutrality of the laws and agreements establishing the program, “[b]raceros were almost all young men.”[2] Today, the Bracero Program has been replaced by Section H-2A of the Immigration and Nationality Act of 1952 (“INA”)[3] as the governing regime for temporary agricultural work in the United States,[4] but migrant farmers in America remain overwhelmingly young, non-disabled men.[5] Why? Foreign labor recruiters, acting as agents of United States growers, intentionally prefer young, non-disabled men for jobs requiring H-2A visas.[6] Despite the enactment of watershed federal antidiscrimination laws like Title VII of the Civil Rights Act of 1964 (“Title VII”),[7] the Age Discrimination in Employment Act of 1967 (“ADEA”),[8] and the Americans with Disabilities Act of 1990 (“ADA”),[9] the female, elderly, and disabled migrant farmworkers rejected for jobs requiring H-2A visas have found little recourse in attacking foreign labor recruiters’ sexist, ageist, and ableist actions abroad due to the extraterritorial limitations of domestic laws.[10] Worse still, the overwhelming majority of migrant farmworkers in America are Mexican and identify as Latinx/Hispanic,[11] meaning this homegrown discrimination is inherently intersectional.

This essay considers a new theory that might afford these victims of discrimination a remedy. Specifically, extant litigation challenging the discriminatory H-2A recruiting practices of foreign labor recruiters has alleged that they engaged in unlawful disparate treatment abroad (e.g., on Mexican soil) by intentionally preferring young men for jobs requiring an H-2A visa;[12] to date, no suit has alleged disability discrimination in this context. Herein, I ask the following: what if we stopped focusing on how these foreign labor recruiters act abroad (e.g., disparately treating female, elderly, and/or disabled jobs applicants on Mexican soil) and focus instead on how American growers act domestically? In other words, could a rejected female, elderly, and/or disabled job applicant establish a viable cause of action against the grower under Title VII, the ADEA, or the ADA using disparate impact theory when: (1) a grower decides, within the United States, to engage a foreign labor recruiter to staff jobs that require an H-2A visa; (2) that domestic practice causes sex-, age-, and/or disability-based disparate impact in the United States; and (3) the policy or practice resulting in preferences for young, non-disabled men is not justified by business necessity? Would plaintiffs overcome the presumption against extraterritorial application of such antidiscrimination laws given a litigation strategy that focuses on the growers’ domestic act of engaging the foreign labor recruiter, rather than any action taken by the foreign labor recruiter, as the practice that ultimately caused disparate impact? In this essay, I tackle those issues and conclude that such a cause of action might be viable under any of these statutes and that it likely would overcome the presumption against extraterritoriality, thereby affording women, older, and disabled applicants for jobs requiring an H-2A visa with a potential means of fighting back against homegrown discrimination perpetuated by U.S. growers.

Part I begins by recounting the problem. I highlight examples of discrimination against women, older, and disabled applicants for jobs needing an H-2A visa by foreign labor recruiters acting as agents of United States growers (Subpart II.A.) and consider why most extant litigation strategies attacking such practices have failed (Subpart II.B.). Specifically, I analyze some of the hallmark cases addressing the extraterritorial reach of Title VII and the ADEA in the context of job applicants requiring H-2A visas (e.g., Reyes-Gaona v. North Carolina Growers Ass’n,[13] Olvera-Morales v. Sterling Onions, Inc.[14]), as well as other landmark cases examining the reach of other labor and employment laws like the Fair Labor Standards Act of 1938 (“FLSA”).[15]

Part II concludes by proposing a solution to the problems highlighted in Part II. First, I consider the relevant history of disparate impact theory under Title VII and explain how it differs from disparate impact theory under the ADA and the ADEA. Second, I apply those theories to growers engaging foreign labor recruiters that discriminate abroad and consider three frameworks that advance workers’ litigation prospects with varying degrees of success. Third and finally, I consider some weaknesses that may thwart the viability of my thesis, chiefly the business necessity defense[16] and whether growers’ engagement of foreign labor recruiters is appropriate for disparate impact scrutiny after Wal-Mart Stores, Inc. v. Dukes.[17] Ultimately, I conclude that my theory may be viable. Thus, I argue that plaintiffs should allege that growers who acquiesce to discriminatory recruiting practices by their foreign labor recruiters violate antidiscrimination laws on a disparate impact theory. By doing so, growers would (and should) be taken to task for unjustifiably sticking their heads in the sand whilst their agents discriminate abroad.

I. Explicating the Problem

This Part lays the groundwork for the legal analysis in Part III. Namely, this Part reviews the empirical and anecdotal evidence of sexist, ageist, and ableist bias of foreign labor recruiters in recruiting for jobs requiring H-2A visas and the myriad harms that such bias wreaks. Moreover, this Part exposes the shortcomings of existing litigation strategies targeting these harms.

A. The History of Foreign Labor Contractor Bias

Growers are smart enough not to, within the United States, ask foreign labor recruiters to recruit young, non-disabled men as temporary agricultural workers because such sex-, age-, and disability-based disparate treatment violates Title VII, the ADEA, and the ADA, respectively, as well as analog state and local laws. Nevertheless, the majority of temporary agricultural workers on H-2A visas are young, non-disabled men. For example, 96% of all H-2A visa holders that entered the United States between fiscal years 2009 and 2013 were men,[18] and “[t]hree-fourths of workers who entered [the United States from fiscal years 2009 to 2013 on H-2A or H-2B visas] were 40 years old or younger.”[19] Although no empirical data supports the dearth of H-2A visa holders with a disability, and the federal government does not track such data, ample anecdotal evidence demonstrates that “employers may use the H-2A program as a means to avoid hiring those who they presume to work at a slower rate, including older workers, women, and people with disabilities who would be able to do the work with accommodations.”[20]

However, the labor market demographics for farmworkers as a whole differ from these numbers significantly. According to the U.S. Bureau of Labor Statistics, in 2019, roughly 24.2% of all miscellaneous agricultural workers in America were women,[21] the median age of such workers was 39.2, and 37.0% of such workers were age 45 or older.[22] Furthermore, data from the U.S. Department of Agriculture’s Economic Research Service demonstrates that “9 percent of U.S. farmworkers (134,000 people) had a disability at some point between 2008 and 2016.”[23] Similar demographics for farmworkers in the largest home country for H-2A visa holders, Mexico, are not publicly available, although there is no logical reason why the sex, age, and disability status demographics of farmworkers in Mexico would not approximate what we see domestically.

Yet, if growers aren’t explicitly directing their foreign labor recruiters to hire younger, non-disabled men (a possibility that may exist but go undetected because such backroom, illegal discussions are not likely to be publicized), why then are H-2A visas awarded overwhelmingly to young, non-disabled men? In all likelihood, foreign labor recruiters are engaging in intentional bias based on animus or stereotypes.

This conclusion is supported by the leading caselaw in this area. For example, Marcela Olvera-Morales, a Mexican woman who worked in upstate New York on an H-2B visa in the late 1990s and early 2000s, alleged in the charge of sex discrimination that she filed with the U.S. Equal Employment Opportunity Commission (“EEOC”) and in a subsequent lawsuit raising the same claims that a foreign labor recruiter called International Labor Management Corporation “systematically placed women in H-2B jobs while placing men in H-2A jobs, which provide better pay and benefits.”[24] “In other words,” as University of North Carolina at Chapel Hill Professor of Sociology Holly Straut Eppsteiner described the phenomenon that Olvera-Morales experienced, “employers use sexist ideologies to justify women’s concentration in low-wage, low-prestige service and care work.”[25] A report from Farmworker Justice summarized the incidence of intentional bias well:

Employers can “hand-pick” a certain demographic of workers. Our government has not sought to apply U.S. anti-discrimination laws to H-2A employers’ recruitment of foreign workers that occurs abroad. Growers thus can pick their ideal workforce—mostly young men removed from daily family obligations who will work long hours for low pay.[26]

Preferring young, non-disabled men for temporary agricultural work causes incredible harm on several fronts. First, it unjustifiably discriminates between men and women, younger and older workers, and workers with and without a disability when any worker who can perform temporary agricultural work should be allowed to work regardless of their sex, age, and disability status so long as they can perform the essential functions of the position. Discrimination like this perpetuates harmful stereotypes (e.g., younger, non-disabled men are strong and more-suited for outdoor, labor-intensive work, whereas women, older people, and workers with a disability are frail and more suited for indoor work).

Second, preferences such as these subjugate classes of workers (i.e., women, the elderly, individuals with a disability) who almost certainly see higher rates of rejection for jobs than their younger, non-disabled, and male counterparts, thereby unfairly increasing the unemployment of subjugated classes in their home countries. To that end, subjugation is harmful because it further oppresses already-oppressed worker classes that deserve equality before the law, whereas discrimination is harmful on an independent basis because it perpetuates harmful stereotypes about those classes of workers.

Third, a preference for young, non-disabled men exacerbates unemployment within the United States. As Bruce Goldstein, then the Executive Director of Farmworker Justice, stated in a hearing before the U.S. House of Representatives Committee on Education and Labor in 2008, “[w]hen employers can select foreign workers based on stereotypes and other prejudices to achieve the workforce they desire, they are less likely to be willing to hire U.S. workers who fall outside those stereotypes and prejudices.”[27]

Fourth, growers allowing their workers to be selected based on these stereotypes deprive themselves of the best performing workers. In other words, the best farmworker for a particular job requiring an H-2A visa may be an elderly Mexican woman with a disability, but a foreign labor recruiter’s bias not only hurts her by denying her a job for which she was most qualified; it also hurts the grower who will be denied her superior performance. Fifth, as a corollary to that point, a hiring practice that results in suboptimal workers harms the United States economy at large by acquiescing to bias at the cost of optimal production.

In sum, foreign labor recruiters are preferencing young, non-disabled men based on their sex, age, and disability status, and such preferences cause great harm. Then, why has litigation provided no redress? The next Subpart addresses that issue.

B. The Shortcomings of Extant Litigation

This Section explains courts’ presumption against extraterritorial application of statutes, recounts the leading cases applying that presumption in the context of migrant farmworkers who challenge foreign labor recruiters’ discriminatory hiring decisions abroad, and explains why the strategy employed in those cases on behalf of applicants falls short.

In the first Supreme Court opinion to consider the extraterritorial application of domestic laws, American Banana Co. v. United Fruit Co., the majority noted that “the acts causing the damage were done . . . outside the jurisdiction of the United States,” giving the court great pause in applying domestic laws that might infringe upon foreign states’ sovereignty.[28] Reflecting on this concern, in a unanimous opinion written by Justice Oliver Wendell Holmes, Jr., the court established a strong presumption that a federal statute is “confined in its operation and effect to the territorial limits over which the lawmaker has general and legitimate power.”[29]

Nearly a century later, the court revisited this principle in the context of extraterritorial application of Title VII in EEOC v. Arabian American Oil Co.[30] There, the Supreme Court invoked the presumption against extraterritorial application to conclude that Title VII does not apply to “United States citizens employed by American employers outside of the United States.”[31] Interestingly, Ali Boureslan, the plaintiff-employee, had been employed in Saudi Arabia immediately before his termination, but the opinion fails to cite evidence that the termination was the basis of his lawsuit or evidence of where the perpetrator of the allegedly-unlawful conduct was at the time of that conduct. The underlying complaint clarifies that Boureslan’s allegations involved discriminatory harassment by his supervisor and termination,[32] but neither the facts on the record before the Supreme Court, the pleadings, nor the circuit and district court opinions below provide evidence of where the alleged perpetrators were located when they undertook the allegedly unlawful actions.[33] The Court reiterated only that, in all of his claims, “the focus of Boureslan’s attack was the discriminatory treatment which he allegedly received while in Saudi Arabia from his Aramco supervisor.”[34]

This omission of the employers’ location is paramount vis-à-vis how broadly should we read Arabian American Oil Co. On the one hand, the Supreme Court appears to make a broad statement that Title VII does not apply to United States citizens employed by American employers outside the United States, suggesting that the alleged perpetrator’s location is immaterial to a Title VII extraterritoriality analysis. Yet, such a broad reading should be construed as dicta. If the record had been clear that the alleged perpetrator engaged in harassment or made the termination decision from the United States, and the court had relied on that fact in reaching its conclusion, then the case could stand for the broad proposition that domestic decisions causing effects abroad cannot run afoul of Title VII. But those are not the facts; the record is silent concerning the location of the alleged perpetrator at the time of the alleged harassment or termination decisions. Indeed, a decision to terminate an employee working in Saudi Arabia could have been made by a manager anywhere, and remote harassment of an employee working in Saudi Arabia could have been perpetrated by a manager anywhere.

Arabian American Oil Co. raises the question of whether courts ought to reject Title VII claims any time the effects of discriminatory acts are felt by employees abroad, even if those acts occurred in the United States. Existing caselaw suggests that rejecting Title VII claims in such situations would be inconsistent with the purpose of the presumption against extraterritoriality. For example, last year, in Hernandez v. Mesa, the Supreme Court limited the extraterritorial application of Bivens actions based on cross-border shootings.[35] There, the Court restated the presumption against the extraterritorial application of statutes, explaining that “[w]e presume that statutes do not apply extraterritorially to ‘ensure that the Judiciary does not erroneously adopt an interpretation of U.S. law that carries foreign policy consequences not clearly intended by the political branches.”[36]

Applying that logic, there is less concern of implicating foreign policy considerations by holding an American employer liable for the actions of someone who was located in America at the time she or he acted, even if those actions implicate concerns abroad. The downstream effect of such liability may be felt by foreign labor recruiters abroad, thereby indirectly and eventually causing foreign policy implications, but the same could be said of significant swaths of all civil liability in our increasingly-interconnected world. Therefore, I maintain that, despite the broad-sounding rule announced in Arabian American Oil Co., the court should resolve that Title VII does not apply when the alleged perpetrator was located abroad for fear of reaching into the jurisdiction of a sovereign state and interfering, but that Title VII most certainly does apply when the alleged perpetrator was located within the United States. Such a rule would clarify that the presumption against extraterritorial application of statutes is premised on avoidance of direct foreign policy implications as opposed to indirect ones (with the nuances of the resulting line-drawing reserved for subsequent, more difficult cases).

Finally, I should note for the sake of completeness that, after Arabian American Oil Co., the Civil Rights Act of 1991 (“1991 Act”), inter alia, amended Title VII and the ADA (which was not explicitly addressed in Arabian American Oil Co., but which would be implicated by that decision given the parallels between Title VII and the ADA) to cover United States citizens employed abroad by American employers or employers controlled by American employers.[37] Yet, the 1991 Act, as well as a 1984 bill that similarly amended the ADEA,[38] are irrelevant in this context given that farmworkers applying to work in jobs requiring H-2A visa are, ipso facto, not United States citizens.

Given this background on courts’ presumption against extraterritorial application of statutes, the chief shortcoming of extant litigation has been its focus on foreign labor recruiters’ actions abroad instead of domestic growers’ actions in the United States. Two primary examples are Reyes-Gaona and Olvera-Morales, both of which targeted foreign labor recruiters’ disparate treatment instead of domestic growers’ actions causing disparate impact.

In Reyes-Gaona, a Mexican man over age forty named Luis Reyes-Gaona applied in Mexico to a foreign labor recruiter looking to staff jobs requiring an H-2A visa on behalf of a grower located in the United States.[39] He alleges that the foreign labor recruiter rejected his application after telling him that it would not accept applications from anyone over forty who had not worked for the grower before; based on these facts, he accused the grower and its foreign labor recruiter of disparate treatment under the ADEA.[40] The Fourth Circuit held that the presumption against extraterritorial application of domestic statutes “prevented the ADEA from regulating events taking place in foreign countries even when they involved citizens of the United States,” and that conclusion is unaltered by the 1984 amendments to the ADEA extending its scope to cover United States citizens employed abroad by American employers.[41]

As another example, in Olvera-Morales, Marcela Olvera-Morales accused a groups of foreign labor recruiters of sex-based disparate treatment in violation of Title VII and analog state laws, arguing that, in Mexico, they steered her and similarly situated women into lower-paying jobs requiring H-2B visas and men into higher-paying jobs requiring H-2A visas.[42] She also alleged similar disparate treatment against the growers premised on their putative responsibility for their agents’ discriminatory actions in Mexico.[43] Both sets of defendants moved to dismiss the suit and moved for summary judgment on multiple grounds, including the argument that Title VII should not be applied extraterritorially to foreign labor recruiters’ actions in Mexico.[44] On that point, the court denied summary judgment, holding that Olvera-Morales’s “extensive contacts” with the United States (i.e., she applied to and was hired by a United States employer, she obtained work authorization, and she worked in the United States lawfully) made it unclear without additional evidence whether the application of Title VII here would be extraterritorial.[45]

In theory, Olvera-Morales was a win for farmworkers fighting discriminatory recruiting practices abroad because the plaintiff survived summary judgment. But, in reality, the opinion leaves much to be desired. First, there is no basis in Supreme Court or circuit precedent for the “extensive contacts” analysis when analyzing extraterritorial application of statutes; rather, that analysis is proper only in undertaking an interstate conflict of laws analysis.[46] In all likelihood, had the case been appealed, the Second Circuit would have corrected the analysis by invoking the Supreme Court precedent cited above focused on the fear of foreign policy implications, the likes of which would almost certainly be implicated if the actions of the foreign labor recruiters in Mexico had resulted in liability under Title VII (which the court conceded was a possibility in denying summary judgment). Second, even assuming arguendo that the “extensive contacts” analysis passes muster, it would remain a fact-intensive inquiry that would make certifying a class incredibly difficult, thus insulating foreign labor recruiters from collective pressure. Third, and most significantly for this essay, Olvera-Morales fails to grapple with the growers’ argument that they are insulated from any Title VII liability from actions that their agents (i.e., the foreign labor recruiters) took because they were allegedly not joint employers. Indeed, the court’s only mention of the growers in its denial of summary judgment was to conclude (correctly) that the growers may have discriminated against Olvera-Morales in the United States,[47] thus sidestepping the need for a joint employer analysis regarding the foreign labor recruiter’s actions in Mexico. If we want to target the growers’ acquiescence to discrimination, we need a different strategy.

Two reflections on these leading cases are important, especially given my argument that they do not present a promising strategy. First, on one reading, Olvera-Morales and Reyes-Gaona are incompatible as both challenge the same thing—a foreign labor recruiter in Mexico rejecting an applicant for a job in the United States—yet reach different conclusions. To that end, Olvera-Morales conceded the possibility that Title VII may apply to the challenged act, whereas Reyes-Gaona rejected such application, and the two statutes do not differ materially on this point. Rather than embrace Olvera-Morales’s proffered harmonization of these cases (i.e., Marcela Olvera-Morales received work authorization and worked in the United States, whereas Luis Reyes-Gaona did not; her extensive domestic contacts overcome the presumption against extraterritoriality), I reject the “extensive contacts” analysis, as explained above, and contend that Olvera-Morales’s reliance on that analysis rendered the opinion a relatively hollow victory for migrant workers.

Second, Reyes-Gaona only dealt with “events taking place in foreign countries” because the allegedly discriminatory act (i.e., the foreign labor recruiter’s act of rejecting Reyes-Gaona’s application) and the harm (i.e., Reyes-Gaona receiving the rejection) both occurred in Mexico, so the court was right to invoke the presumption against extraterritorial application of statutes to avoid treading into the realm of foreign affairs. However, the theory I forward herein bifurcates those two components: the allegedly-discriminatory act (i.e., a grower engaging a foreign labor recruiter) takes place in the United States, even though the harm (i.e., the applicant receiving the rejection) takes place abroad. I contend that this bifurcation exempts my theory from the fears that underly the anti-extraterritorial application presumption. Courts should avoid regulating acts outside the United States to not interfere in foreign affairs, but courts should have far less, if any, such concern regulating harms occurring outside the United States when the act causing the harm took place within the United States.

Shockingly, these are the only two cases on record in United States courts challenging the allegedly discriminatory hiring practices abroad of foreign labor recruiters tied to the H-2A visa program. Moreover, for clarity’s sake, cases like Arriaga v. Florida Pacific Farms, LLC[48] and Castellanos-Contreras v. Decatur Hotels[49] provide no additional insight here because those cases concerned, inter alia, allegations that the employers violated the FLSA by failing to reimburse temporary workers’ pre-employment expenses primarily benefitting the employers during their workers’ first workweeks.[50] Therefore, the courts were right to ignore the presumption against extraterritorial application of statutes because the allegedly unlawful act (i.e., not reimbursing expenses in the first workweek once workers were in the United States) occurred in the United States, as did the harm (i.e., workers failing to receive reimbursement during that workweek).

In addition to these domestic litigations, in 2016, a group of Mexican women who worked in the United States on guest worker visas led by two Petitioners, Elisa Tovar Martínez and Adareli Ponce Hernández, filed a petition with Mexico’s National Administrative Office under the North American Agreement on Labor Cooperation (“NAALC”) accusing the United States government of, inter alia, failing to enforce its antidiscrimination laws

by allowing the recruiters and employers of the Petitioners, as well as the majority of employers and recruiters of H-2A and H-2B workers, to perpetrate sex discrimination in the H-2 programs[, including] . . . steering women workers to the H-2B program which provides fewer benefits and protections [than the H-2A program].[51]

Essentially, Martínez and Hernández reiterated the alleged unlawful actions that formed the basis of Olvera-Morales’s lawsuit. As of January 2018, NAALC had taken no action on the petition,[52] and given the recent abolishment of NAALC by the U.S. Mexico-Canada-United States Agreement (“USMCA”)[53] and the failure of the USMCA to replace NAALC in any meaningful fashion aside from an as-yet ill-defined Labor Council,[54] it is unlikely that their case will ever see resolution. Moreover, even if NAALC or some comparable body ever does render a decision, it would lack binding legal effect.[55]

In sum, this Subpart concludes that litigation challenging growers’ reliance on foreign labor recruiters’ discriminatory acts abroad has fallen short because of the litigants’ focus on the foreign acts of rejecting the applicants instead of the domestic acts of engaging the recruiters. In the next Part, I contend with a possible solution to such shortcomings.

II. Proposing a Solution

So, what can a rejected applicant do? This Part tackles that question, arguing that Title VII likely provides recourse, whereas the ADA and ADEA may provide recourse. To reach that conclusion, I briefly recount the relevant history of disparate impact theory before applying that theory on these facts.

Section 703(a) of Title VII prohibits workplace discrimination without specifying what constitutes workplace discrimination. Clearly, refusing to hire an applicant because of the applicant’s race violates the law, but what about refusing to hire someone for race-neutral reasons that cause disparate impact on the basis of race? To that end, in Griggs v. Duke Power Co.,[56] the Supreme Court considered an employer’s promotion policies that caused disparate impact against black applicants on the basis of race and held that such disparate impact violated Title VII because the employer could not justify the policies as being job related and consistent with business necessity.

In the years since Griggs, the business necessity defense has been broadened (e.g., when the Supreme Court in Wards Cove Packing Co. v. Atonio concluded that there “is no requirement that the challenged practice be ‘essential’ or ‘indispensable’ to the employer’s business for it to pass muster” under the business necessity defense[57]) and narrowed again (e.g., when Congress reacted to Wards Cove and other cases by passing the 1991 Act, inter alia, to resurrect Griggs-era law regarding business necessity[58]). Today, there remains a debate regarding whether the business necessity defense requires literal necessity or a “manifest relationship [to] legitimate employment goals.”[59] However, that debate is irrelevant for our purposes.

What matters is this basic disparate impact framework that courts follow presently. The standard is as follows: under Title VII, an applicant can establish a prima facie case of disparate impact liability by identifying an employer’s policy or practice that caused disparate impact, after which the only relevant defense would be the business necessity defense, and that defense requires the employer to prove that the policy or practice is, at a minimum, a legitimate employment goal (and possibly that the policy or practice is literally necessary).

In Raytheon v. Hernandez, the Supreme Court endorsed disparate impact theory under the ADA.[60] Given the material similarities between Title VII and the ADA here, it is likely that the Supreme Court would apply the same disparate impact framework for cases under the ADA as it applied under Title VII. Yet, the logic of Wards Cove applies equally to the ADA as it does to Title VII, yet the narrowing of the business necessity defense by the 1991 Act had no effect on the ADA because the 1991 Act amended only Title VII.[61] Accordingly, Wards Cove most likely governs ADA disparate impact claims, thereby giving employers a defense if they can prove any legitimate business reason for the policy or practice at issue.

Finally, though the Supreme Court in Smith v. City of Jackson endorsed disparate impact theory under the ADEA,[62] it also clarified that “textual differences between the ADEA and Title VII make it clear that even though both statutes authorize recovery on a disparate-impact theory, the scope of disparate-impact liability under ADEA is narrower than under Title VII.”[63] First, the ADEA offers an additional defense not available under Title VII or the ADA: exempting employers from liability if they prove that any decision was made “based on reasonable factors other than age.”[64] Second, as with the ADA, the 1991 Act did not touch the ADEA. As such, not only could employers defend against a prima facie case of disparate impact liability under the ADEA by proving that the policy or practice causing the disparate impact was justified by at least a legitimate business reason, but employers could likewise point to that legitimate business goal as a reasonable factor other than age (essentially qualifying for two defenses at once).

Given this background, how would growers fare when confronted with a disparate impact challenge? Foremost, let us use the facts underlying Reyes-Gaona or Olvera-Morales as our starting point. A grower acts in the United States to engage a foreign labor recruiter (e.g., by signing, within the United States, a contract for services). Then, the foreign labor recruiter acts abroad to reject applicants based on their sex, disability, or age. Finally, more young, non-disabled men end up working in the United States for the grower. At the outset, it is clear that the employer has maintained a policy or practice that caused disparate impact (i.e., engaging the foreign labor recruiter). Furthermore, the employer’s engagement of the foreign labor recruiter occurred in the United States, overcoming the main concern animating the presumption against extraterritorial application of domestic statutes. Accordingly, I maintain that a rejected female, disabled, or elderly applicant challenging this grower under Title VII, the ADA, or the ADEA, respectively, will have established a prima facie case of disparate impact. However, whether the claim survives the business necessity defense is entirely a question of framing the policy or practice at issue.

First, consider what I contend would be a losing frame for the applicants: the growers maintained a policy or practice of engaging a foreign labor recruiter. The grower would rightly invoke the business necessity defense here, arguing that the U.S. Department of Labor has certified, inter alia, that “there are not sufficient U.S. workers qualified and available to perform the labor involved in the petition,”[65] so the recruitment of temporary agricultural workers is a literal necessity. Under any iteration of the business necessity defense, the grower would be exempted from liability.

Second, consider what I argue would be a better frame: the growers maintained a policy or practice of engaging this foreign labor recruiter in seasons two and beyond. To that end, one might forgive a grower for engaging the foreign labor recruiter; seeing their fields populated with young, male, and non-disabled workers; and deciding thereafter that nothing reasonable could be done about it this season. After all, the growers may have a literal business necessity to maintain the workforce that the foreign labor recruiter secured lest they be forced to shutter their business for the season. However, there is no excuse for having such demographics play out year after year. There is no literal business necessity for growers to reuse the same foreign labor recruiter without intervening to bar that recruiter from discriminating in its recruiting practices, especially when failing to intervene will disparately impact women, the elderly, and individuals with a disability. Yet, if growers can qualify for the business necessity defense by simply showing a legitimate business goal, they may be able to do so by showing a volume discount or ease in administrative burdens in reusing the same foreign labor recruiter year after year. Hence, in this framing, the result depends upon the legal standard for business necessity.

Given the weaknesses of these two framings, I conclude with what I consider the best framework for success: the growers maintained a policy or practice of engaging a foreign labor recruiter, year after year, without prohibiting that recruiter from discriminating in its recruiting practices (e.g., barring discriminatory practices in the master services agreement between the grower and recruiter, threatening to rescind the arrangement upon evidence of discrimination). When framed in this manner, not only does the policy or practice establish a prima facie case of disparate impact, but there is simply no business justification for growers to stick their heads in the sand as their workforces fill up time and again with nothing but young, non-disabled men.

However, even this optimal framing suffers from several potential problems. First, there have been steady headwinds against disparate impact theory recently, with some arguing that the theory itself may be unconstitutional because it forces employers to act based on impermissible factors like race or sex to avoid disparate impact liability (e.g., Justice Scalia’s concurrence in Ricci v. DeStefano[66]). Thus, it would be calling on the judiciary to fight against those winds to stake out a novel interpretation of disparate impact theory.

Second, after Wal-Mart Stores, Inc. v. Dukes, courts should not impose disparate impact liability on growers based on a negative policy or practice (i.e., not doing something), as opposed to a positive policy or practice.[67] Applied here, engaging and reengaging this particular foreign labor recruiter is certainly a positive policy, but failing to prohibit discrimination in its foreign labor recruiter contracts sounds decidedly like not doing something. It remains unclear whether a court would view this framing as one barred by Dukes, but I contend that the employer’s action (e.g., signing and renewing a services contract) is precisely the sort of positive action for which disparate impact scrutiny is appropriate, Dukes notwithstanding.

Third, the reality of human supply chains suggests many growers take a hands-off approach to the human side of the operation, meaning those growers that contract out management of their workforce may be able to escape liability by arguing that they did not know and should not have known that their policy or practice caused disparate impact because they never actually saw their fields filled by young, non-disabled men. To date, courts have yet to grapple with the extension of disparate impact liability to employers based on unknown effects, and although the text of the applicable statutes offer no defenses to employers on this basis, courts may espouse hesitation to apply disparate impact theory to an employer who did not know, and arguably should not have known, any better.

Fourth, courts may extend the presumption against extraterritorial application of domestic statutes to situations like this where the ostensibly unlawful actions occurred within the United States, but application of those statutes even to such domestic actions causes downstream effects abroad (i.e., by changing the behavior of foreign labor recruiters in other countries). This suggests the need for courts to clarify the presumption against statutes’ extraterritorial application as I outlined earlier in this essay.

All that being said, disparate impact theory carries with it the unique potential to sidestep the presumption against extraterritorial application of statutes that has precluded applicants for migrant farmworker jobs from challenging in domestic courts the hiring discrimination they feel abroad. For that reason, litigators representing applicants rejected for migrant farmworker jobs because of their sex, age, or disability status should embrace disparate impact theory despite its potential drawbacks.


Using the paradigm in this essay, disparate impact claims targeting growers-employers for acquiescing to the discriminatory practices of their foreign labor recruiters might withstand scrutiny. In any event, it would certainly increase the risk profile of such growers in litigation, thus incentivizing growers to pressure their recruiters to discriminate less and allowing plaintiffs’ counsel greater leverage to negotiate more favorable settlements for their migrant worker clients. Furthermore, this theory could be similarly used outside the agricultural context if foreign labor recruiters engaged in similar discriminatory recruitment practices when staffing jobs requiring other types of visas. Hopefully, this theory can at least provide leverage to the majority-Latinx women, elderly, and individuals with a disability who have been repeatedly discriminated against by growers and their foreign labor recruiters who, but for a series of unfortunate legal loopholes, would have been held liable under domestic antidiscrimination laws a long time ago.


Ryan H. Nelson is a Research Associate at the Harvard Law School Project on Disability and a member of the adjunct faculty at Boston University School of Law and New England Law | Boston. He received his LL.M. from Harvard Law School; his J.D., cum laude, from the Benjamin N. Cardozo School of Law, Yeshiva University; and his B.S.B.A. from the University of Florida. He would like to thank Jennifer J. Rosenbaum and Sabrineh Ardalan for inspiring this essay.

           [1].     Bracero History Archive, About, http://braceroarchive.org/about (last visited Jan. 31, 2021).

           [2].     David Bacon, ‘Close to Slavery’ or Legalization? The Farmworkers’ Hard Choice, Am. Prospect (Nov. 25, 2019), https://prospect.org/labor/close-to-slavery-legalization-undocumented-farmworkers/.

           [3].     8 U.S.C.A. § 1101(a)(15)(H)(ii)(a) (West 2021); see also U.S. Dep’t of Labor, Fact Sheet #26: Section H-2A of the Immigration and Nationality Act (INA), https://www.dol.gov/sites/dolgov/files/WHD/legacy/files/whdfs26.pdf (last visited Jan. 31, 2021) [hereinafter “DOL H-2A Fact Sheet”].

           [4].     U.S. Citizenship & Immigration Servs., H-2A Temporary Agricultural Workers, https://www.uscis.gov/working-united-states/temporary-workers/h-2a-temporary-agricultural-workers (last visited Jan. 31, 2021).

           [5].     Infra Subpart II.A.

           [6].     Id.

           [7].     Civil Rights Act of 1964, Title VII, Pub. L. No. 88-352, 78 Stat. 255 (1964).

           [8].     Age Discrimination in Employment Act of 1967, Pub. L. No. 90-202, 81 Stat. 602 (1967).

           [9].     Americans With Disabilities Act of 1990, Pub. L. No. 101-336, 104 Stat. 327 (1990).

         [10].     Infra Subpart II.B.

         [11].     Farmworker Justice, Selected Statistics on Farmworkers (2015-16 Data), http://www.farmworkerjustice.org/wp-content/uploads/2019/05/NAWS-Data-FactSheet-05-13-2019-final.pdf (last visited Jan. 31, 2021).

         [12].     Infra Subpart II.B.

         [13].     Reyes-Gaona v. N.C. Growers Ass’n, 250 F.3d 861 (4th Cir. 2001) (holding that foreign labor recruiters’ disparate treatment abroad is not prohibited by the ADEA given the presumption against extraterritorial application).

         [14].     Olvera-Morales v. Sterling Onions, Inc., 322 F. Supp. 2d 211 (N.D.N.Y. 2004) (extending reach of Title VII to certain disparate treatment abroad despite presumption against extraterritorial application).

         [15].     Fair Labor Standards Act of 1938, Pub. L. No. 75-718, 52 Stat. 1060 (1938).

         [16].     42 U.S.C. § 2000e-2(k)(1)(A) (2021).

         [17].     Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338 (2011).

         [18].     Centro de los Derechos del Migrante, Inc., Engendering Exploitation: Gender Inequality in U.S. Labor Migration Programs 12 (2018) (citing U.S. Gov. Accountability Office, GAO-15-154, H-2A and H-2B Visa Programs: Increased Protections Needed for Foreign Workers 18 (Mar. 2015), https://www.gao.gov/assets/690/684985.pdf [hereinafter “2015 GAO Report”]).

         [19].     2015 GAO Report, supra note 18, at 19. There is no age demographics for H-2A visa holders as a distinct group.

         [20].     Farmworker Justice et al., Comments on NPRM re: Temporary Agricultural Employment of H-2A Nonimmigrants in the United States, RIN 1205-AB89, at 8, http://www.farmworkerjustice.org/wp-content/uploads/2019/09/2019-H-2A-Advocate-Comments-DOL-Final.pdf [hereinafter “Farmworker Justice Comments on NPRM”].

         [21].     U.S. Bureau of Labor Statistics, Labor Force Statistics from the Current Population Survey, Household Data Annual Averages, Table 11: Employed Persons by Detailed Occupation, Sex, Race, and Hispanic or Latino Ethnicity, https://www.bls.gov/cps/aa2019/cpsaat11.pdf (last visited Jan. 31, 2021).

         [22].     U.S. Bureau of Labor Statistics, Household Data Annual Averages, Table 11b. Employed Persons by Detailed Occupation and Age 9, https://www.bls.gov/cps/aa2019/cpsaat11b.pdf (last visited Jan. 31, 2021).

         [23].     United States Department of Agriculture, Economic Research Service, Disabilities in the U.S. Farm Population 1 (2019).

         [24].     Mary Bauer & Meredith Stewart, Close to Slavery: Guestworker Programs in the United States, S. Poverty L. Ctr. (Feb. 19, 2013), https://www.splcenter.org/20130218/close-slavery-guestworker-programs-united-states; see also Olvera-Morales, 322 F. Supp. 2d at 213–15.

         [25].     Holly Straut Eppsteiner, La Vida Jaibera: The Gendered Work and Migration Experiences of Female Guestworkers in the Rural Southeast 13 (2013), https://cdr.lib.unc.edu/downloads/41687j31f.pdf.

         [26].     Farmworker Justice, No Way to Treat a Guest: Why the H-2A Agricultural Visa Program Fails U.S. and Foreign Workers 17 (2011), http://www.farmworkerjustice.org/wp-content/uploads/2012/05/7.2.a.6-No-Way-To-Treat-A-Guest-H-2A-Report.pdf.

         [27].     Do Federal Programs Ensure U.S. Workers Are Recruited First Before Employees Hire from Abroad?: Hearing Before the H. Comm. on Educ. & Labor, 110th Cong. 90, at 17 (2008), https://www.govinfo.gov/content/pkg/CHRG-110hhrg41982/html/CHRG-110hhrg41982.htm.

         [28].     Am. Banana Co. v. United Fruit Co., 213 U.S. 347, 355 (1909) (emphasis added).

         [29].     Id. at 357.

         [30].     EEOC v. Arabian Am. Oil Co., 499 U.S. 244 (1991).

         [31].     Id. at 248.

         [32].     Complaint at ¶¶ 10, 14, EEOC v. Arabian Am. Oil Co., 499 U.S. 244 (1991) (No. 89-1838), 1990 WL 10022972.

         [33].     See Boureslan v. Aramco, 857 F.2d 1014, 1016 (5th Cir. 1988), aff’d sub nom. EEOC v. Arabian Am. Oil Co., 499 U.S. 244 (1991).

         [34].     Id.

         [35].     Hernandez v. Mesa, 140 S. Ct. 735 (2020).

         [36].     Id. at 747 (quoting Kiobel v. Royal Dutch Petroleum Co., 569 U. S. 108, 116 (2013) and citing Arabian Am. Oil Co., 499 U.S. at 248).

         [37].     Civil Rights Act of 1991, Pub. L. No. 102-166, § 109, 105 Stat. 1071 (1991)

         [38].     S. 2603, 98th Cong § 802(a) (1984).

         [39].     Reyes-Gaona, 250 F.3d at 863.

         [40].     Id.

         [41].     Id. at 864–65; see also supra note 38 (1984 ADEA amendments).

         [42].     Olvera-Morales, 322 F. Supp. 2d at 213–14.

         [43].     Id. at 214–15.

         [44].     Id. at 214.

         [45].     Id. at 221.

         [46].     See id. (citing Torrico v. Int’l Bus. Mach. Corp., 213 F.Supp.2d 390, 399–405 (S.D.N.Y.2002) (citing Second Circuit conflict-of-laws cases for support in conducting an analysis of extraterritorial application of Title VII)).

         [47].     See id.

         [48].     Arriaga v. Fla. Pac. Farms, LLC, 305 F.3d 1228 (11th Cir. 2002).

         [49].     Castellanos-Contreras v. Decatur Hotels, 622 F.3d 393 (5th Cir. 2010).

         [50].     Arriaga, 305 F.3d at 1231–32; Castellanos-Contreras, 622 F.3d at 396–97.

         [51].     Petition on Labor Law Matters Arising in the United States, Submitted to the National Administrative Office of Mexico under the North American Agreement on Labor Cooperation Regarding the Failure of the U.S. Government to Effectively Enforce its Domestic Labor Laws, Promote the Elimination of Employment Discimination [sic], and Promote Equal Pay for Men and Women 4–5, 12 (July 15, 2016), https://cdmigrante.org/wp-content/uploads/2018/01/NAALC-Petition-2016-English.pdf.

         [52].     Centro de los Derechos del Migrante, Inc., Re: Supplement to Petition /Public Communication Mex 2016-1 (Jan. 22, 2018), https://cdmigrante.org/wp-content/uploads/2018/03/NAALC-Supplement-MEX-2016-1-english.pdf.

         [53].     Kathleen Claussen, RIP NAALC: North American Agreement on Labor Cooperation, Int’l Econ. L. & Pol. Blog (Dec. 4, 2018), https://worldtradelaw.typepad.com/ielpblog/2018/12/guest-post-rip-naalc-north-american-agreement-on-labor-cooperation.html.

         [54].     See Chapter 23: Labor, Agreement between the United States of America, the United Mexican States, and Canada (Dec. 13, 2019), https://ustr.gov/sites/default/files/files/agreements/FTA/USMCA/Text/23-Labor.pdf (see Article 23.14).

         [55].     Int’l Labor Rights Forum, NAFTA, Labor, and Immigration: A Package Deal, https://laborrights.org/blog/201104/nafta-labor-and-immigration-package-deal (last visited Mar. 4, 2020) (“The NAALC petition procedure is a long, bureaucratic process that is ultimately unable to sanction companies directly.”).

         [56].     Griggs v. Duke Power Co., 401 U.S. 424, 431 (1971).

         [57].     Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 659 (1989).

         [58].     Supra note 33 at §§ 2(2), 3(3).

         [59].     Linda Lye, Note, Title VII’s Tangled Tale: The Erosion and Confusion of Disparate Impact and the Business Necessity Defense, 19 Berkeley J. Emp. & Lab. L. 315, 348–54 (1998).

         [60].     Raytheon v. Hernandez, 540 U.S. 44, 53 (2003).

         [61].     Supra, note 37, at § 105.

         [62].     Smith v. City of Jackson, 544 U.S. 228, 232 (2005).

         [63].     Id. at 240.

         [64].     29 U.S.C. § 623(f)(1).

         [65].     See DOL H-2A Fact Sheet, supra note 3.

         [66].     Ricci v. DeStefano, 557 U.S. 557, 594 (2009) (Scalia, J., concurring).

         [67].     Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 356 (2011).

Ryan H. Nelson, Homegrown Discrimination, 12 Calif. L. Rev. Online 1 (May 2021), https://www.californialawreview.org/homegrown-discrimination.

More From California Law Review Online

A Brief Reflection on the Doctrinal Entrenchment of Inequality: Brach v. Newsom

In the spring of 2020, many parents of children in California schools closed during the global pandemic had had enough. They filed a lawsuit challenging the executive orders requiring compliance with state public health directives that in turn mandated the shuttering of schools. A majority of the three-judge panel recognized compelling constitutional interests in education for the parents likely to be more privileged—because of their ability to pay for private education—and denied such protection to parents whose children attended public schools. It is perhaps no wonder that as of this writing, the case is headed for en banc review. This brief essay explores the path taken in the appellate opinion to provide a critique of the different doctrinal treatment of the parents whose children attend public school relative to those whose children attend private schools.

Prosecutors’ Ethical and Constitutional Duties to Criminal Defendants: It’s Time to Reel in the Zeal

When the police violate a person’s Fourth Amendment right, the Exclusionary Rule bars evidence resulting from that violation from being used against the person in trial. The rule is designed to deter police from engaging in illegal searches and seizures. Nevertheless, prosecutors often bring charges even when the initial police report and body-worn camera footage […]

Hindsight’s 2020

Throughout the coronavirus pandemic, strong executives emerged in the President and the country’s many Governors. While the situation required extraordinary measures, expanding executive powers evidence a worrying trend of unchecked, unilateral power. This piece calls for new legislation to automatically impose checks on emergency executive power after a brief interlude, as ‘emergencies’ must be viewed […]

Wrongful Imprisonment and Coerced Moral Degradation

Despite the ever-growing number of exonerations in the U.S.—and the corresponding surge in scholarly interest in wrongful convictions in recent years—research on the carceral experiences of wrongfully-convicted persons remains strikingly limited. In this essay, we draw on in-depth interviews with 15 exonerated men to explore the moral dimensions of the experience of wrongful imprisonment. We […]