Disparity studies comparing various demographic groups based on different outcomes in education, employment, health, housing, and income have been a staple of public policy analysis for decades. Depending on the variables controlled for, it may be argued that a group is underutilized, underrepresented, or even marginalized. Nowhere, however, have these studies had such legal significance as their effect on public procurement. Now that era may be coming to an end.

　　Racial, ethnic, and gender preferences in federal contracting have existed for about fifty years. Similar programs were adopted by states, counties, cities, and special districts across the country.Sometimes these programs create overall annual goals which may be implemented on particular contracts. Elsewhere they are attached to specific expenditures for major projects where obtaining political approval requires bargaining among various constituencies.

　　Federal programs use the term Disadvantaged Business Enterprises (DBEs) to define beneficiaries, while the other programs usually use the term Minority Women Business Enterprises (MWBEs). Congressionally mandated DBE programs control federal expenditures nationwide, but the presence of a MWBE program is related to partisan politics. They are much more common in blue than in red jurisdictions.

　　The United States Department of Transportations 1983 DBE program covers federal subsidies for highways, mass transit, airports, and seaports, while the Small Business Administration’s 1988 (8a) program awards contracts to minority-owned firms in every area of federal expenditures. MWBE programs are quite varied ranging from general quotas to soft goals on specific projects. There is no comprehensive study of the hundreds of these programs. What they have in common are disparity studies which purport to show that they are remedies for discrimination, so their racial preferences diminishing or even excluding contracting opportunities firms owned by white men or stockholders are justified.

　　The persistence of these preferential programs seems hard to explain given the clear equal protection language in the Fourteenth Amendment and the Civil Rights Act’s Title VI prohibition against all racial discrimination in federal expenditures. Further, the Supreme Court in City of Richmond v. Croson (1989) and Adarand v. Peña (1995) ruled that the standard for evaluating preferential contracting programs is “strict scrutiny” requiring identification of a “compelling interest” and use of “narrow tailoring” in implementation. The answer to this conundrum can be found in the following three sections.

　　Statistical Disparities and Constitutional Violations

　　Justice O’Connor’s plurality Croson opinion became the precedent influencing all subsequent preferential contracting litigation. In overturning Richmond’s 30 percent minority business procurement quota, she articulated multiple evidentiary rules that the City did not meet. These requirements should have been the death knell for almost all race preferential contracting programs, except that Justice O’Connor went on to say, “Where there is a significant statistical disparity between the number of qualified minority contractors willing and able to perform a particular service and the number of such contractors actually engaged by the locality or the locality’s prime contractors, an inference of discriminatory exclusion could arise.”

　　Governments viewed this paragraph as an opening to use disparity studies as predicates for the creation or continuation of DBE and MWBE programs. Huge amounts of procurement money were at stake in these programs. The approximately 600 disparity studies completed probably cost taxpayers more than $300 million. A handful of for-profit consulting companies who understood what the market demanded began to dominate the disparity study mini-industry. About 70 percent of all these studies were conducted by six small firms. As a recent article concluded: “In practice, these firms are contracted to find evidence of disparity and there is institutional pressure on these firms to find it—they have failed if they do not find it or if they find the ‘wrong’ disparity (with the wrong group). These are not academically neutral studies.”

　　These studies initially created formidable defenses for racial preferences. Often several hundred pages long with multiple tables and charts, they seemed to meet the Croson test, even though they could only create a statistical “inference” of discrimination. They almost never identified any actual discrimination by the governments that commissioned them or by others in the local economy. Still, plaintiffs were confronted with the need to hire their own experts and engage in arduous litigation. Governments frequently hired private attorneys who profited from prolonged proceedings. When disparity studies were involved, small firm plaintiffs’ attempts to win by summary judgment motions were frequently unsuccessful.Although plaintiffs won some major cases after trials, that process was laborious and expensive.

　　The presumption that all persons identified somehow with a particular racial or ethnic group are socially and economically disadvantaged is surely a stereotype.

　　Despite operating in the arena of strict scrutiny, litigation about procurement preferences began to diminish. Major construction organizations became reluctant to sue governments that funded projects of interest to their members. Smaller businesses that had been hurt by preferences rarely had resources to litigate.

　　Then the tide began to turn. Early in his term, President Joe Biden instructed over 90 executive agencies to create “Equity Action Plans,” many of which aggressively used preferences to create “equity.” Examples included United States Department of Agriculture programs excluding white farmers and ranchers from loan relief, a Covid-era restaurant support program that put white owners at the back of the line where funds would run out and a Minority Business Development Administration (MBDA) initiative to provide regional centers services only to minority-owned businesses. All of these new race preferences were ruled unconstitutional by federal courts. Even the decades-old (8a) program which could set aside federal contracts for MBEs anywhere in the country was permanently enjoined. The US Department of Transportation DBE program has survived multiple challenges over its four-decade existence. On September 19, 2024, however, a federal district court ruled that its preferences for certain minorities in federally funded contracts lacked both a compelling interest and narrow tailoring.

　　The US Department of Justice has not appealed any of these decisions. Since 2020, the Biden administration has been zero for nine in defending programs to redistribute economic benefits by race.Federal courts consistently ruled the constitutional standard was equal protection for individuals not equity for groups.

　　Disparity Studies Without a Compelling Interest

　　Disparity studies must document both a compelling interest and narrowly tailoring to provide a constitutional predicate for race-preferential contracting programs.

　　In the first three decades of such litigation, plaintiffs and judges took the studys text at face value. That strategy began to change in 2016 when plaintiffs found that the disparity study’s regression analysis used to support Montana’s DBE program had disappeared and could not be reviewed or defended. In a 2020 case against Shelby County, Tennessee, the government admitted that it had never seen their study’s underlying data. So, the plaintiffs issued a subpoena for the data to Mason Tillman Associates, one of the largest disparity study producers in the country. When they would not produce that data because they no longer had it or had any confidence in it, the Ninth Circuit found MTA in contempt. Consequently, both the Montana and Shelby County cases were settled on very favorable terms for the plaintiffs.

　　Governments relying on disparity studies to support preferential procurement programs now can expect to have to produce underlying data. Disparity studies often inflate the number of MWBEs or DBEs in relevant markets. The studies usually include many firms that never bid or had the capacity to perform major government procurements or were not even certified as MWBEs or DBEs and thus could not fulfill the preferential goals the disparity ratios suggested. Plaintiffs also seek interview transcripts to show that the studys anecdotal snippets were taken out of context, were not accurate, or could not be verified.

　　In deposition, plaintiffs can ask whether relevant government bureaucrats and political officials had ever seen the underlying data. Almost always the answer will be no.There was little incentive for bureaucrats to question the data supporting an MWBE program for which their political superiors had paid hundreds of thousands of dollars. Local legislators had neither the time, inclination nor skills to wade into the underlying data. That creates an awkward problem. If that data had not been examined by the government being sued, how could any defendant confirm its disparity study results accurately provided a compelling interest?

　　Even if a study created a valid statistical disparity, that would be only an “inference” of discrimination according to Croson.In Vitolo v. Guzman, the Sixth Circuit ruled that strict scrutiny required more:

　　The government has a compelling interest in remedying past discrimination only when three criteria are met.

　　First, the policy must target a specific episode of past discrimination.

　　Second, there must be evidence of intentional discrimination in the past. Statistical disparities don’t cut it, although they may be used as evidence to establish intentional discrimination.

　　Third, the government must have had a hand in the past discrimination, it now seeks to remedy.

　　In 2023, Students for Fair Admissions, Inc. v. President and Fellows of Harvard College and Students for Fair Admissions, Inc. v. University of North Carolina (hereafter SFFA), racial preferences in college admissions were found to violate the Fourteenth Amendment and Title VI. The Supreme Court’s clear rule was that the compelling interest for race-based governmental actions was “remediating specific, identified instances of past discrimination that violated the Constitution or a statute.” It said that “the twin commands of the Equal Protection Clause [are] that race may never be used as a negative and that it may not operate as a stereotype.”

　　Those are very high evidentiary bars. In defending preferential programs under attack, the DOJ relied on several new disparity studies produced by federal agencies and veterans of the disparity study industry to show compelling interest. Federal district trial courts did not believe any of these documents provided the necessary compelling interest and rarely discussed them in any detail.

　　Disparity Studies Without Narrow Tailoring

　　Defining who is a minority eligible for preferences may be even a higher bar. DBE and most MWBE programs use a three-decade-old list created by the Small Business Administration that had almost never been altered and now includes persons who identify as:

　　Black Americans; Hispanic Americans; Native Americans (Alaska Natives, Native Hawaiians, or enrolled members of a Federally or State recognized Indian Tribe); Asian Pacific Americans (persons with origins from Burma, Thailand, Malaysia, Indonesia, Singapore, Brunei, Japan, China (including Hong Kong), Taiwan, Laos, Cambodia (Kampuchea), Vietnam, Korea, The Philippines, U.S. Trust Territory of the Pacific Islands (Republic of Palau), Republic of the Marshall Islands, Federated States of Micronesia, the Commonwealth of the Northern Mariana Islands, Guam, Samoa, Macao, Fiji, Tonga, Kiribati, Tuvalu, or Nauru); Subcontinent Asian Americans (persons with origins from India, Pakistan, Bangladesh, Sri Lanka, Bhutan, the Maldives Islands or Nepal.

　　This list encompasses persons from dozens of countries spanning four continents. It makes no distinction between recent immigrants and persons whose families have lived in the United States for decades.

　　A major challenge to whether that list was narrowly tailored was issued by a Sixth Circuit judge in Vitolo. Amul Thapar came from an immigrant Indian family whose father owned an air conditioning supply business and his mother owned a restaurant. He was the first South Asian-American appointed to a federal circuit court. Judge Thapar questioned the SBA list. He wanted to know why there were “preferences for Pakistanis, but not for Afghans; Japanese but not Iraqis; Hispanics but not Middle Easterners—[which] is not supported by any record evidence at all?” He added there was no clear definition of SBA’s concept of having “origins from” a specific listed country. Immigrant or fifth generation?

　　Although the Supreme Court has not discussed the specific SBA federal list, perhaps the fatal blow to its validity is reflected in SFFA. Chief Justice Roberts’ majority opinion declared that the Constitution forbids governmental use of race to benefit one group and harm another group or to use race as a stereotype in judging a person’s social condition or attitudes. The presumption that all persons identified somehow with a particular racial or ethnic group are socially and economically disadvantaged, which underpinned so many federal programs, is surely a stereotype. Roberts pointed out that the racial categories the universities used were “imprecise,” “overbroad,” and “arbitrary, The Asian and Hispanic general categories did not distinguish between persons from many different national origins, while the treatment of applicants from various Middle Eastern backgrounds was uncertain. 

　　Justice Gorsuch, in a concurring opinion, questioned the checkboxes used for racial and ethnic identifications.He noted they were created by bureaucrats without the help of “anthropologists, sociologists, ethnologists, and other experts.” Moreover, federal regulators had cautioned that these racial categories should not be interpreted as “scientific” nor “be viewed as determinants of eligibility for participation in any Federal programs.” Justice Gorsuch’s analysis of “incoherent” racial stereotypes went into detail about problems with the various racial classifications now in use by so many institutions.

　　Shortly thereafter, using the federal list for contracting awards was specifically undermined because these programs rely on the listed national origins to determine presumed “socially and economic disadvantaged” awardees.Citing SFFA, decided 21 days earlier, a District Court ruled the 8(a) program lacked both a compelling interest and narrowly tailoring because it did not purport to remedy any specific past discrimination against the beneficiary firms.

　　Based on discovery, however, the Court determined:

　　The Defendant SBA has not added a group to the list of those entitled to the rebuttal presumption since 1999. Further, the Defendant SBA has never removed a group from that list for being no longer adversely affected by the present effects of discrimination and Defendant SBA does not have criteria to evaluate whether a group should be removed from the list. Defendant SBA has not considered any race-neutral alternative to the use of the rebuttable presumption since 1986.

　　In 2023, a District Court found MBDA’s race-based presumption of social and economic disadvantage was both under-and-over inclusive, because of the many irrationalities in the national origins of the included and excluded groups. MBDA was unable to offer any rubric for these determinations. Further, according to MBDA rules, “if a business owner belongs to an enumerated group, he or she is entitled to services without regard to their life circumstances, financial performance, or any social or economic metric of “disadvantage.”Plaintiffs were able to document that policy based on the text of MBDA documents without hiring an expert or even taking a deposition even though the DOJ had a $700 per hour expert and entered 2200 pages of studies to support its position.

　　In considering the same racial and ethnic presumptions in the DBE program, a District Court said in a 2024 decision there must be evidence that “the Department of Transportation has previously discriminated against those groups. It cannot group all minority-owned businesses into one gumbo pot, but then try to scoop out only the sausage and not the okra” (Mid-America Milling Company v. US DOT).

　　No race-preferential contracting program anywhere in the United States has ever had evidence of discrimination against all the groups on the federal list. Finding specific public contracting discrimination against the thousands of firms that were beneficiaries of the social and economic presumption will be a formidable evidentiary and bureaucratic problem. The invalidation of the federal minority list also creates a substantial issue for existing and future state and local disparity studies. Out of convenience and because there was no better alternative, almost all of them adopted that list in making their statistical calculations. Now, their programs are overinclusive just as the federal programs are. Moreover, if they are to meet the Croson test that their disparity results must be “statistically significant” it will be very difficult to fine-tune the data to reflect actual differences among the racial and ethnic groups on the list and still have statically significant results.

　　Disparity studies face a bleak future as predicates for race-preferential procurement programs. Older disparity studies almost never identify any constitutional or statutory violations and governments will be reluctant to fund future studies that find them guilty of committing or tolerating such violations. To reach statistical significance, studies need to use racial and ethnic group clusters courts are likely to find overinclusive. Pretrial required admissions and depositions can easily establish those facts and cases may be resolved without more expensive litigation. Still, plaintiff lawyers may wish to calibrate how much evidence about the deficiency of a disparity study a specific judge and possibly an appellate court will require. The new legal vulnerabilities facing disparity studies create a real opportunity to overturn discriminatory policies. As Chief Justice Roberts stated bluntly for the SFFA majority, “Eliminating racial discrimination means eliminating all of it.”