Equal Treatment & College Degree Barriers: The Problem of Disparate Impact Doctrine

Introduction: Turning Civil Rights on its Head

The changes implemented by the Trump Administration to civil rights enforcement are “revolutionary,” in the original sense of the term. By targeting illegal racial preferences and hostile racial climates rooted in discriminatory equity ideologies, the administration has recommitted the civil rights state to ensuring equal treatment under law for all Americans (Exec. Order, 14190, 2025; Schorr, 2025). In nearly every instance, this approach accords well with the plain meaning and intended purposes of federal civil rights statutes. The principal exception to this rule is the disparate impact doctrine.

Disparate impact doctrine is an aberration in U.S. civil rights law. The concept of “disparate impact discrimination” inverts the traditional understanding of discrimination in ways that undermine procedural equality. Whereas “discrimination” ordinarily describes disparate treatment—and, in civil rights law, disparate treatment based on protected characteristics—disparate impact theory applies this label to group outcome differences or “disparities,” including those resulting from equal treatment (Legal Dictionary, n.d.). On this basis, courts and federal regulators (e.g., the Equal Employment Opportunity Commission [EEOC]) force employers to abandon non-discriminatory policies and encourage outcome-leveling remedies that erode equality under law.[1]

Additionally, disparate impact analysis, as applied by courts and federal regulators, harms American workers and families by distorting labor and post-secondary education markets. This is because while disparate impact law constrains employers’ use of aptitude tests and high school diploma requirements, it does not impose comparable restrictions on college degree requirements (Hess & Addison, 2019). This discrepancy is ironic given that the latter likely disadvantages black job seekers—the original intended beneficiaries of disparate impact doctrine—to a greater degree than the former.

This issue brief challenges disparate impact doctrine, addressing its rationale, entrenchment in civil rights law, and impact on American workers. It presents a range of options for how Executive Branch (“administrative”) agencies can use their discretionary authorities to mitigate the distortions caused by disparate impact doctrine. Such reforms, if implemented, would go a long way towards improving equality under law and coherence in post-secondary education.

Shaky Foundations

Disparate impact theory holds that neutral standards may violate anti-discrimination law if they produce statistical disparities in outcomes for groups, even absent discriminatory intent.[2] The doctrine entered modern civil rights law through the Supreme Court’s 1971 decision in Griggs v. Duke Power Company (hereafter, Griggs; Griggs v. Duke Power Co., 1971). There, 14 black employees and job applicants challenged Duke Power’s use of high school diplomas and standardized aptitude test requirements for job seekers. The plaintiffs argued these requirements violated Title VII of the Civil Rights Act of 1964 by disproportionately excluding black applicants and were not shown to be meaningfully related to job performance (Civil Rights Act of 1964).

In siding with the plaintiffs, the Court determined that facially neutral employment practices with disparate racial effects may constitute unlawful discrimination if employers cannot demonstrate such requirements to be both 1) job-related and 2) consistent with business necessity. The Court subsequently narrowed disparate impact liabilities for employers in Wards Cove Packing Co. v. Atonio (1989); however, Congress responded by restoring (and codifying) these liabilities in Title VII with the Civil Rights Act of 1991 (Civil Rights Act of 1991, Title VII).

Today, Title VII disparate impact claims generally proceed through a burden-shifting framework: once a plaintiff identifies a specific employment practice producing a substantial statistical disparity, the employer must demonstrate that the practice is both job-related and consistent with “business necessity.” Even then, employers may still be found liable if plaintiffs can demonstrate a less disparity-inducing alternative practice could serve the employer’s legitimate interests.

Through this and similar processes—i.e., those associated with other civil rights statutes—courts and federal agencies wield disparate impact doctrine to force employers to modify or abandon a range of policies. Examples include:

• Employment aptitude tests, such as cognitive and other written exams (Ford Motor Co. v. EEOC, 1982; U.S. Department of Justice, 2017).
• Physical fitness tests (U.S. Department of Justice, 2023).
• Criminal background checks in employment and housing (U.S. Department of Housing and Urban Development, 2016; U.S. Equal Employment Opportunity Commission, 2012).
• School discipline standards (U.S. Department of Justice & U.S. Department of Education, 2014).
• English proficiency requirements (EEOC v. Synchro-Start Products, Inc., 1999).
• Height and weight standards (Dothard v. Rawlinson, 1977).
• Teacher certification standards (Gulino v. Board of Education of the City School District of the City of New York, 2020).

Consider a recent dispute involving the Maryland State Police Department. In October of 2024, following a two-year investigation by the Department of Justice (DOJ), Maryland agreed to overhaul two of its job requirements for police recruits: 1) a written test covering mathematics, reading, writing, and grammar, which applicants could take up to four times for a passing score of 70% or higher; [3] and 2) a physical fitness test encompassing 18 push-ups, 27 sit-ups, a 1.5-mile run within 15 minutes and 20 seconds, and a demonstration of being able to reach 1.5 inches past one’s own toes (United States v. Maryland Department of State Police, 2024). The physical test could be taken up to three times.

Test data revealed that, on average, whites were more likely than blacks to pass the written exam (91% vs. 71%), and males were more likely than females to pass the physical exam (81% to 51%) (Crane, 2024). In line with disparate impact doctrine, this disparity triggered legal scrutiny by the DOJ. Upon concluding that Maryland had not adequately validated its (race-blind) tests as job-related and consistent with business necessity, the DOJ secured an agreement requiring Maryland to revise its standards.

In addition to raising equal treatment concerns, the Maryland police example demonstrates how disparate impact doctrine undermines basic competence in employment settings. Consider the job test validation requirement in this case: police work involves a mixture of cognitively and physically demanding tasks, as well as basic writing skills.[4] However, under disparate impact doctrine, the mere existence of statistical disparities puts the onus on employers (including public sector employers) to prove specific test designs and cutoff thresholds are sufficiently predictive of successful job performance to justify disparities to the satisfaction of courts and regulators.

At the same time, employers can also be held liable for attempting to limit their disparate impact liabilities. In Ricci v. DeStefano (2009), the Supreme Court determined the City of New Haven violated disparate treatment under Title VII when it discarded the results of a firefighter promotion exam after that test, like Maryland’s, produced a disparity favoring white test-takers. The Court concluded the city’s decision had been made expressly on the basis of race without adequate evidence that the exam itself was racially biased. In other words, by attempting to avoid disparate impact “discrimination” against black firefighters, New Haven had embraced actual discrimination (disparate treatment) against white firefighters.

For employers, such paradoxes are downstream from the central conceptual flaw in disparate impact theory: the assumed, if often unstated, causal connection between disparities and discrimination. The very phrase “disparate impact discrimination” testifies to this assumption (Congressional Research Service, 2026). Moreover, although disparate impact doctrine does not require evidence of discriminatory intent, by requiring employers to justify neutral criteria under anti-discrimination law whenever such criteria produce significant disparities, it treats disparities as presumptively suspect and as potential evidence of legally cognizable discrimination.

Is this treatment reasonable?

To be sure, statistical disparities sometimes result from disparate treatment, but they also often result from equal treatment. Consider the case of professional sports, where demographic disparities—including along protected class lines—are commonplace, despite intense merit-based selection and close monitoring of rules to ensure fairness.[5] There are many such examples, including:

• Sex-neutral military fitness standards, which reliably produce large disparities favoring males (Beynon, 2021).
• Sex-segregation in sports, which, along with safety and fairness concerns related to transgender male participation in female sports, attests to the near inevitability of male-favoring disparities in this domain (Concerned Women for America, 2025; Gaines & Schorr, 2024).
• Various non-physical elite competitions—including in mathematics, chess, and other intellectually demanding contests—which reliably produce disparities along racial, ethnic, and sex lines (Wai et al., 2010; World Chess, 2026).
• The extreme overrepresentation of Ashkenazi Jews among Nobel Prize recipients. In theory, the Nobel Committee could be biased; however, these awards are linked to verifiable advances in knowledge: a reasonably objective outcome (Anu Museum of the Jewish People, n.d.).
• Standardized academic tests scores, which reliably produce racial, ethnic, and sex disparities (CollegeBoard, 2024).
• Occupational licensing exam results, which do the same (American Bar Association, 2022; Nettles et al., 2011).
• Risk categories, as assessed by insurance companies, which differ by age and sex, among other categories (Dionne & Rothschild, 2014).
• Assorted controversial yet nonetheless well-documented group average differences in crime rates, drug use, alcohol consumption, tobacco consumption, etc. (Federal Bureau of Investigation, n.d.; Substance Abuse and Mental Health Services Administration, 2024).^[6] Even if one prefers socio-economic explanations for these disparities, it remains the case that such behavior differences exist under common conditions. For example, stealing a car is illegal and (adult) tobacco consumption is permitted, regardless of one’s race, sex, or age.

Then there are the “exceptions that prove the rule.” For example, prior to the Supreme Court’s invalidation of racial preferences in student admissions in Students for Fair Admissions, Inc. v. President and Fellows of Harvard College (2023), Asian Americans were overrepresented among undergraduate students despite being discriminated against by those same institutions. Hispanic Americans have health advantages (e.g., life expectancy) over other groups, including whites, despite lower average access to health care. And so on (Fenelon, 2013).

In short, that a given disparity exists suggests little—and arguably, nothing—about the origins of that disparity. The assumption that, absent evidence to the contrary, disparities suggest discrimination is simply unfounded. Requiring employers to justify disparities on anti-discrimination grounds is akin to requiring teachers to demonstrate that cheating is not to blame for some students outscoring others on a given test. Cheating could be the culprit, but the mere existence of a range of scores provides no evidence of it.

Under a more intellectually coherent civil rights regime, businesses would not be subjected to unnecessary, intrusive, and counterproductive compliance burdens predicated on unfounded assumptions. Protections against disparate treatment would suffice, leaving Americans free to make their own way in life under a system of fair play (see Table 1).

Table 1

How Disparate Group Outcomes Should Be Viewed Under a System Prioritizing Equal Treatment Under Law and in Public Life

Why the “New High School Diploma” May Worsen Disparities

In the meantime, disparate impact doctrine remains. Accordingly, companies seeking to establish a business necessity case for job requirements must therefore navigate the burden sharing process, understanding that they could face significant liabilities should a claimant propose a less disparate alternative employment practice. Given the plethora of potential alternative employment practices, companies are likely to either: 1) “race to the bottom” by lowering job standards to preemptively reduce disparities (at the cost of employee performance), or 2) identify other means to restrict the applicant pool to those likely to have the skills needed to do a specific job.

In weighing these options, the prospect of being investigated for racial discrimination and the stigma of being labeled a civil rights violator are substantial costs against which the benefits conferred by better hiring practices must be weighed. Such concerns may loom especially large when civil rights enforcement priorities exclude disfavored (yet protected) groups for ideological reasons (Carl, 2023).

However, as described by Frederick Hess and Grant Addison in an influential 2019 article in National Affairs, the EEOC has effectively handed employers an “out” to the legal quagmire of identifying job-relevant requirements that avoid running afoul of disparate impact doctrine (Hess & Addison, 2019). For reasons the commission has never deigned to clarify, college degree requirements are not scrutinized for disparate impacts in the same manner as high school diplomas. This double standard is unreasonable and exacerbates the inconsistencies already inherent in the application of the disparate impact doctrine.

The Court concluded in Griggs (1971) that high school diploma and intelligence test requirements disproportionately disadvantage blacks due to the “inferior education [they received] in segregated schools.” Carrying that reasoning forward, requiring additional (post-secondary) education should result in similar disparate impacts. This is because institutions of higher education—not unlike Duke Power Company in 1971—generally require applicants to have completed high school and to take a standardized aptitude test (e.g., the Scholastic Aptitude Test or the American College Testing exam). By ignoring college degree requirements, the EEOC effectively redirects the same (i.e., otherwise prohibited) job applicant screening through the college admissions process. The resulting job screens still produce racial disparities while also imposing substantial costs through tuition and time commitments for what, in many cases, amount to entry-level job requirements.

It gets worse. Racial disparities reemerge when students enter college, both in terms of achievement and graduation rates. This means that rerouting job applicant screening through college admissions effectively subjects future workers to not one but two racial disparity-inducing screens. This assessment seems supported by the most recent available national high school graduation data (National Center for Education Statistics, 2024). As of the 2021-22 school year, 81% of blacks graduated high school in comparison to 90% of whites. Adding college, the completion gap widens further: among all four-year institutions, 42.2% of black and 65.5% of white students graduated from college within six years (National Center for Education Statistics, n.d.). Considering these differentials, the college completion achievement gap between white and black students (23.3 percentage points) is about 2.6 times greater than the achievement gap in high school graduation (9 percentage points). However, the issue of which educational diploma is associated with greater racial disparity is less important than the fact that college degree requirements layer these screens on top of one another.

This suggests that, in addition to undermining equal treatment under law, disparate impact doctrine—as applied—may worsen rather than ameliorate racial disparities in employment. If true, disparate impact doctrine would deserve a prominent place among the most counterproductive federal domestic policy interventions in U.S. history. As described in the following section, disparate impact doctrine also appears to have undermined the broader post-secondary education system—yet here, again, the damage seems to fall most disproportionately on the very populations disparate impact doctrine seeks to help.

The Disparate Impacts of College Degree Requirements

For many, college provides an invaluable experience. A college education helps students to cultivate richer internal lives and to develop valuable career-relevant skills. Regarding the latter, there are cases where occupation-relevant skills are so tightly associated with specific degrees that specific college degree requirements are defensible. However, as noted, college degree requirements are also frequently used as an entry-level applicant pool screen for generally “trainable” employees.

This practice appears to be one among several major contributors to the recent –decades' surge in college enrollment. That surge has driven a host of problems, including spiraling college costs and debt, wasteful administrative bloat, and inadequate returns on students’ educational investments (Cooper, 2024; Hanson, 2026; Robinson, 2017; Weinstein, 2023). Since 2000, the rising costs of college have outpaced inflation (i.e., 3.91% vs. 2.56% annually) (Hanson, 2025a). Looking at tuition only, this amounts to a total increase of 157.5% (or 37.5% when adjusted for inflation) over the last two and a half decades. At the end of 2025, total unpaid student debt reached a towering $1.7 trillion. Even more concerning, of the 42.8 million borrowers with current balances, 7.7 million federal student loan borrowers are in default and owe a collective debt of $180 billion (Federal Student Aid, 2026). This is partly because post-graduate earnings often lag well behind expectations: as many as three-in-ten associate, one-in-four bachelor’s, and two-in-five master’s degrees produce negative returns-on-investment (ROI) (Cooper, 2024; 2022; 2021).

The cumulative cost of unnecessary college degree seeking is staggering. On average, the direct costs of a bachelor’s degree include $38,270 per year in tuition, fees, books, supplies, and living expenses ($153,080 over four years) (Hanson, 2026b). This does not include the additional indirect and opportunity costs associated with delayed earnings, career progression, family formation and, of course, debt. While some estimates point to a lifetime earnings premium of around $1 million dollars for completing a bachelor’s degree, when one takes account all the opportunities costs associated with a four-year degree, the actual lifetime premium may be closer to $300,000—and that is if you choose a major that generally has positive ROI (Carnevale et al., 2021; Cooper 2021).

One reason so many college graduates today receive either no benefit or are financially harmed by attending college is the general misalignment between many college programs and labor market needs (Menees, 2026). Put another way: colleges generally allow student demand for low-ROI degrees to dictate course offerings rather than tailoring their course offerings to market demand for specific occupations. As a result, too many students graduate with degrees that leave them drowning in debt and lacking the skills necessary to fill high-demand occupations.

Many of these same miseducated graduates could have benefited from alternative pathways to good-paying jobs. Today, for example, the U.S. economy faces job shortages in critical industries—including in industries vital to national defense—many of which are attainable through apprenticeships and short-term industry credentials (Bureau of Labor Statistics, 2025).

College degree requirements are not the sole cause for the myriad of economic distortions associated with the “college-for-all” paradigm;[7] however, research links degree requirements to excessive college enrollment. Given its role in propping up such requirements, disparate impact law would seem to bear at least some of the blame for these distortions as well (Cooper, 2023; Fuller & Raman, 2017).

Interestingly, it also appears that, through its impact on college degree requirements, disparate impact doctrine disparately impacts (harms) black students and workers. Beginning with students, black students complete four-year degrees at lower rates than their white peers and are more likely to take on federal student loan debt (81.4% vs. 59.1%)—and at higher amounts ($32,210 vs. $24,530) (National Center for Education Statistics, 2023).[8] This includes the arguably predatory Parent PLUS loans, which are used twice as often, on average, by black vs. white families (25.9% vs. 12.6%).[9] Given these borrowing disparities, it is unsurprising that black borrowers are also twice as likely as white borrowers to fall behind on paying their student loans (26% vs. 13%; Board of Governors of the Federal Reserve System, 2025).

Turning to workers, research suggests college degree requirements disproportionately exclude low-income Americans and racial minorities—both of whom attend and complete college at lower rates than other groups (Fuller & Raman, 2017). According to the U.S. Census, among Americans over the age of 25, 30% of blacks, and 22% of Hispanics possessed at least a 4-year degree, in comparison to 61% of Asians and 43% of non-Hispanic whites (United States Census Bureau, 2024). In effect, college degree requirements summarily exclude perhaps two-thirds of Americans from consideration for job opportunities (Burning Glass Technologies, 2014). This share includes nearly 70% of black workers, nearly 80% of Latino workers, and more than 70% of rural Americans from all backgrounds (Auguste, 2021).

Interestingly, businesses acknowledge their own role in promoting unnecessary degree requirements—in a recent study, 61% of employers admitted to rejecting applicants who had the required skills and experience for a job but lacked a college degree (Fuller & Raman, 2017). It follows that, today, businesses are among those negatively impacted by college degree requirements (they lose access to qualified applicants), even if they bear part of the blame for this situation. Ironically, it seems that disparate impact doctrine—a body of law that presupposed a conflict between certain racial minority group interests and business interests—has developed in such a way as to undermine both.

Minimizing the Damage Caused by Disparate Impact Doctrine

Disparate impact doctrine presents policymakers with a dilemma: it creates significant problems, yet remains firmly embedded in statutory and case law.[10] Even so, administrative agencies retain substantial discretion in how they implement the law. For example, the Trump Administration recently withdrew support for disparate impact claims under Title VI (Exec. Order No. 14281, 2025). Those claims were unusual because the statute offers no clear textual basis for them. Still, the Court has allowed agencies to create and enforce disparate impact requirements for recipients of federal funding (Alexander v. Sandoval, 2001). And even where disparate impact law is more firmly established, agencies still have room to act.

Consider the Title VII burden-sharing framework.[11] Much of disparate impact doctrine’s practical force arises at the initial stage, where plaintiffs must show a prima facie statistical disparity. That demonstration is governed by the Uniform Guidelines on Employee Selection Procedures (UGESP, 1978), a joint guidance document issued by the EEOC, DOJ, Department of Labor (DOL), and Civil Service Commission. The guidelines include the “4/5th Rule,” under which a policy presumptively triggers disparate impact review if the selection rate for a disadvantaged group is less than 80% of the rate for the highest-selected group (UGESP, 1978; U.S. EEOC, 1979).

Federal agencies could issue revised guidance to make clearing this initial hurdle more challenging. For example, they could:

• Raise the threshold for review beyond 4/5ths. Under a 1/5th Rule, the selection rate for an underrepresented group would need to fall to 20% rather than 80% of the rate of the highest selected group to meet the presumptive threshold for review.
• Address omitted variables. Currently, challenged disparities are commonly evaluated by comparing selection rates. That approach does not allow for consideration of relevant omitted variables that could help to establish non-discriminatory causes for disparities.[12] Options for strengthening this standard include requiring plaintiffs to demonstrate disparities using statistical methods, such as multivariate regression analysis with controls for relevant variables. Plaintiffs could also be required to match similarly qualified applicants when assessing disparities under alternative approaches.
• Address sample sizes. Under current guidance, in cases where sample sizes are too small to allow for meaningful group mean comparisons, plaintiffs are permitted to incorporate data from other businesses to bolster their case for the presence of a reviewable disparity. The process seems to treat the absence of evidence as a problem to be overcome rather than a reason for not pursuing a claim. Given that a system of law generally takes the opposite approach, it may be prudent to instead establish sample size thresholds for group comparisons to screen out poorly evidenced disparity claims.
• Address effect sizes. Along similar lines, agencies could set minimum thresholds—e.g., numbers of individuals impacted—below which disparities would not be considered reviewable. Given the difficulty of establishing whether disparities are caused by disparate treatment, it makes sense to exempt minimally impactful disparities from consideration.
• Require plaintiffs to challenge specific practices. The 1991 updates to Title VII, require that a plaintiff asserting a disparate impact must establish a link to “particular employment practice that causes a disparate impact on the basis of race, color, religion, sex, or national origin and the respondent fails to demonstrate that the challenged practice is job-related for the position in question and consistent with business necessity” (Cornell Law School, n.d.). Under current guidance, this provision is interpreted very broadly—in essence, plaintiffs are permitted to challenge entire (bundled) hiring policies. A more rigorous approach would instead require plaintiffs to establish reviewable disparities associated with specific components of overall hiring policies.

Moving beyond the initial showing of disparate impact, the burden shifts to employers to prove that a challenged policy is job-related and supported by business necessity. The UGESP (1978) also governs this stage. These agencies could revise that guidance to make it easier for employers to satisfy this requirement. For example, they could formally recognize certain facially neutral, professionally developed practices as legitimate business necessities:

• Aptitude tests, including written and physical fitness exams where the aptitudes tested directly pertain to job requirements.
• Criminal background checks, to safeguard employees, customers, and property.

Agencies could also clarify that merit-based policies qualify as business necessities, including selection policies that recruit and promote top achievers based on performance.

Regarding validation procedures for job tests, agencies could revise the UGESP (1978) to clarify that Title VII job-related and business necessity requirements do not necessitate the use of formal psychometric validation methods such as criterion, content, or construct validation. In their place, employers could instead be required to provide less burdensome evidence, perhaps including job task analyses, industry practices, professional standards, operational requirements, and safety-based performance considerations, provided such evidence reasonably demonstrates a meaningful relationship between the selection criterion and actual job performance or operational needs. In practice, a police department’s minimum fitness thresholds for running speed or strength might be justified by job task analysis and safety requirements demonstrating the need for, and measures related to, police officers successfully pursuing suspects on foot and physically subduing them.

In the final hurdle in the Title VII disparate impact claims process, plaintiffs may propose alternative, so-called “less discriminatory” (i.e., less disparity-inducing) policies to address employers’ legitimate business necessities. Here, establishing a successful alternative establishes a disparate impact on employer liability. Here, agencies could limit the likelihood of such liabilities being established by clarifying the terms by which a policy qualifies as a viable “alternative.” They could, for example, require plaintiffs to demonstrate that a proposed alternative is truly comparable to the existing policy along various business necessity lines. For example, an alternative might be required to maintain (not disrupt) existing productivity/profitability, safety, administrative efficiency, and employer autonomy. It might also be required to avoid imposing additional compliance costs or legal risks.

One especially impactful change to agency guidance in this final stage of Title VII review would be to carve out a “safe harbor”—consistent with Title VII recognition of business necessity—for employers who implement skills-based hiring from alternative policy comparisons. The term “skills-based hiring” refers to a “method for finding talent [by focusing] on individuals’ abilities and competencies rather than their education and direct experience” (Agovino, 2024). Such assessments have been shown to be more predictive of job performance than educational attainment, for example (Hunter & Hunter, 1984; Schmidt et al., 2016; Schmidt & Hunter, 1998). Perhaps unsurprisingly, given the noted disparate impacts of college degree requirements, skills-based hiring reforms have also been shown to reduce racial disparities and boost black employment.[13]

Skills-based hiring thus presents a unique opportunity to address business necessities while also minimizing disparities. This would seem to present a powerful argument against additional review under disparate impact doctrine. Skills-based hiring is also consistent with other suggested adjustments, such as meritocratic selection and reasonable, evidence-based validation procedures.

As a complement to the skills-based hiring safe harbor, agencies could clarify that college degree requirements are also subject to disparate impact analysis. In the absence of disparate impact law, this change would not be needed—the utility of a given job requirement would be left to employers and the labor market to sort out for themselves. However, as noted, the bizarre application of disparate impact doctrine to education requirements contributes to a range of postsecondary and labor market distortions as well as disparate racial impacts on students and workers. So, while creating a truly level playing field will require federal legislative change, the current field can be made more level by correcting the distortion that favors college degree requirements.

Implementing the suggested changes to the Title VII review process would greatly improve fairness in federal civil rights law, help students, workers, and families by reducing harmful distortions to post-secondary education and labor markets; and reduce government intrusion and business compliance burdens. While implementing these changes, agencies could also substantially reduce the damage caused by disparate impact doctrine by revoking existing guidance documents and deprioritizing enforcement of disparate impact claims, where possible.

In considering proposals from the proposed menu of options, among others, agencies should carefully consider which of these—alone or in combination—are most likely to survive judicial scrutiny. Each proposal appears to fall within agencies’ discretionary authorities; however, caution is still warranted, particularly given the recent surge in judicial activism (Trump v. CASA, Inc., 2025). Administrative agencies should therefore prioritize guidance revisions with an eye towards survivability as well as impact. They should also consider whether similar reforms may be appropriate to address disparate impact doctrine as it operates in other (i.e., non-Title VII) contexts.

What follows is a streamlined list of recommendations for Executive Branch officials and policymakers:

Recommendation 1. Administrative agencies should issue revised guidance to UGESP (1978), clarifying standards associated with each stage of the Title VII disparate impact claim review process.

Recommendation 2. Administrative agencies should establish a skills-based hiring safe harbor from alternative practice challenges arising under disparate impact claims.

Recommendation 3. Administrative agencies should clarify that college degree requirements are subject to the same disparate impact claims review process as high school diploma requirements.

Recommendation 4. Congress should prohibit the use of disparate impact doctrine at all levels of government.

Conclusion

Disparate impact doctrine distorts civil rights law—and with it, post-secondary education and labor markets. At its base, the doctrine offers a vision of civil rights that conflicts with our nation’s fundamental commitment to procedural equality. It breaks with this tradition by attempting to remedy what it regards as an unfortunate byproduct of treating people equally—namely, that some people invariably do better than others. Yet, as this issue brief has shown, disparate impact law creates its own disparate impacts, including those that affect members of the very groups on whose behalf it ostensibly works.

The distortions introduced by disparate impact doctrine to postsecondary education are substantial. Employers require skilled workers; however, disparate impact law renders their efforts to find such workers risky. Yet, due to an inexplicable omission in EEOC guidance, employers can use college degree requirements as a substitute for otherwise prohibited job screens. This appears to be among the key drivers of excessive demand for college degrees, which results in a host of downstream problems, including excessive cost and debt growth, inadequate ROI, and a general misalignment of post-secondary education with labor market needs.

Fortunately, Executive Branch agencies possess the necessary authorities to curb the most detrimental applications of disparate impact doctrine within the bounds of statutory and judicial law. Should they wield their authorities in this way, these agencies would help to return civil rights enforcement to its original, proper emphasis on ensuring commitment to equal treatment under law and help to curb the unintended harms wrought by disparate impact doctrine.

[1] In Ricci v. DeStefano (2009), the Supreme Court recognized mitigating racial disparities by discarding test results, or by post hoc score banding to obscure disparities, violates Title VII’s prohibition against disparate treatment . These “remedies” deny equal treatment by subjecting neutral, individual performance-based scores to race-based interventions. As the Court explained, if banding is used “to make lower black scores seem higher,” it is forbidden.

[2] The limits of these statistical assessments are addressed on page 10.

[3] While applicants were tested on mathematics, there was no minimum score requirement for the math component.

[4] Consider the “basic duties” listed by Mayland for State Police Troopers: "investigating and completing criminal investigations," "interviewing witnesses, victims and suspects," and "preparing cases for court and testifying in court" are all cognitively demanding tasks; "arresting and processing suspects" and "providing basic medical attention to ill and injured persons" are potentially very physically demanding tasks; and "issuing traffic citations and warnings," "writing field reports," and (again) "preparing cases for court" require writing skills (Maryland State Police, n.d.).

[5] Open discussion of such differences—much less of their causes—is generally taboo; yet these differences are readily apparent. Across professional and Olympic sports, certain racial and ethnic groups are regularly over- and underrepresented in specific athletic disciplines (National Collegiate Athletic Association, 2018; Saltin et al., 2003).

[6] See: Discrimination and Disparities (Sowell, 2018).

[7] The “college-for-all” paradigm assumes, and advocates for, every American to obtain a college degree after high school. College-for-all began in earnest after WWII when veterans returned home to a rapidly advancing job market that required new skills. When the federal government passed the “GI Bill” to help fund veteran postsecondary education, it radically expanding access to college education. Through the Higher Education Act of 1965 and the Education Amendments of 1972, the federal government again poured more funding into higher education by creating the federal student loan and Pell Grant programs. Given this easy access to tuition dollars provided by federal loans and backed by personal student debt, colleges became active participants in attempting to enroll as many high school graduates as possible in their programs—whether doing so would financially benefit the student or not.

[8] Black student loan borrowers fare even worse when analyses account for total educational debt, including graduate and private student loans (Hanson, 2025b).

[9] Parent PLUS loans are nondischargeable student loans taken by parents. Interest rates, origination fees, and default rates are typically much higher for these loans, in comparison to other student loans (Schorr and Shires, 2025).

[10] As noted, the Civil Rights Act of 1991 explicitly codified disparate impact doctrine in Title VII of the Civil Rights Act of 1964 (Civil Rights Act of 1991). Before that codification, the Supreme Court in Griggs v. Duke Power Co. (1971) effectively read disparate impact into Title VII. The Court later read disparate impact into the Age Discrimination in Employment Act in Smith v. City of Jackson (2005) and into the Fair Housing Act in Texas Department of Housing and Community Affairs v. Inclusive Communities Project, Inc. (2015)—though, in the latter case, the Court reasoned that Congress had effectively ratified disparate impact liability through the FHA’s 1988 amendments.

[11] Disparate impact claims arising under other civil rights statutes governing housing, education, disability, and age discrimination likewise use burden-sharing frameworks. These frameworks differ in strength (i.e., how onerous on employers) and derive from a mixture of statutory and judicially and administratively constructed law (see: Congressional Research Service (2026).

[12] The “4/5^th Rule” allows plaintiffs to bring forth lawsuits using descriptive comparisons of selection rates across groups. These comparisons do not account for other variables that could be related to selection outcomes that are unrelated to protected class status. For example, a basic literacy test may produce differences across groups, however, other variables such as parental education, number of parents in the household, average weekly screentime, hours parents read to applicant in childhood, etc. could explain differences – and are entirely unrelated to protected class status. By allowing cases to be brought forth on the grounds of basic descriptive statistics, these comparisons run the risk of conflating differences with causation. More rigorous analyses, such as multivariate regression, can isolate the relationships between certain variables, including protected class status, and group outcomes.

[13] A 2019 study found skills-based hiring reforms eliminated white/black hiring disparities and boosted county-level black employment by 40% following one year of implementation (Towards Employment, 2019).