Note from the Editor:

This article is about the use of anecdotal evidence to justify racial preferences in public contracting under City of Richmond v. Croson. As always, the Federalist Society takes no position on particular legal or public policy matters. Any expressions of opinion are those of the author. Generally, the Federalist Society refrains from publishing pieces that advocate for or against particular policies. But when we do, as here, we will offer links to other perspectives on the issue, including ones opposed to the arguments put forth in the article; these can be found at the right side of this page. We also invite responses from our readers. To join the debate, please e-mail us at


The perception is just there that if you’re Black or if you’re a woman you probably don’t know how to do X, Y and Z type of work. So they’ve already put [you] in that pigeonhole.1

This excerpt is typical of the anecdotal evidence which has appeared in hundreds of disparity studies since the United States Supreme Court’s decision in City of Richmond v. Croson, the landmark case regarding race conscious procurement programs.2 In Croson, the Supreme Court struck down Richmond’s public contracting racial preference, in part because the city’s anecdotal evidence of racial discrimination was insufficient to withstand scrutiny under the Equal Protection Clause. The city’s anecdotal evidence was testimony offered at a public hearing. Anecdotal evidence consists primarily of personal accounts of discrimination told from the perspective of the person claiming discrimination. This article examines the use and misuse of anecdotal evidence in public contracting discrimination cases since Croson.

Part I of the article examines Croson, a landmark civil rights decision about racial preferences in public contracting. Part II discusses how the Croson decision caused the proliferation of disparity studies, and the fact, often misunderstood by courts, that the government bears the burden of justifying preferences in a court challenge to the constitutionality of a public contracting program. Part III analyzes lower court decisions evaluating anecdotes and points out two common flaws of anecdotal evidence: interviewer bias and response bias. Part IV addresses the central issue with the sufficiency of anecdotal evidence: whether anecdotal claims of discrimination must be corroborated, or whether perceptions of discrimination are sufficient to justify preferences. Courts are split on this issue. Some circuits have rejected the need for verification because they do not think sufficient public policy arguments for requiring verification have been offered. This article argues in Part V that investigation and corroboration of anecdotal evidence is required, and it offers three public policy arguments for this conclusion.

I. City of Richmond v. Croson

Croson’s path to the Supreme Court began in 1983, when the city of Richmond, Virginia began to set aside 30% of its contract dollars for Minority Business Enterprises (MBEs).3 Six months later, the J.A. Croson Company was the low bidder on a project to install urinals in the city jail. The company, a white male-owned mechanical plumbing and heating contractor, was denied the construction contract because it did not meet the required MBE participation goal. Croson filed suit. The case bounced around the federal district and circuit courts for six years, eventually reaching the United States Supreme Court.4

In January 1989, the Court ruled 6-3 that the MBE set-aside requirement violated Croson’s right to equal protection under the 14th Amendment, Section 1, which mandates that “No state shall…deny to any person within its jurisdiction the equal protection of the laws.”5 The Equal Protection Clause of the 14th Amendment is an individual right, guaranteed regardless of an individual’s race, ethnicity or sex, so Croson could claim its protection even though he was not a racial minority.6

Justice Sandra Day O’Connor, writing for the plurality in Croson, noted that Richmond had offered “no evidence that qualified minority contractors have been passed over for City contracts or subcontracts.”7 Nor had the city presented any evidence about how many MBEs were in the relevant market and how many city dollars these firms had received. Discrimination against MBEs should have been carefully identified, but was not.8 Also, there was no evidence of discrimination against the various minority groups Richmond had included in its preferential program.9 Croson explicitly condemned any non-remedial purpose for preferences, saying that “Classifications based on race carry a danger of stigmatic harm. Unless they are strictly reserved for remedial settings, they may in fact promote notions of racial inferiority.”10 The Croson Court did not eliminate preferences in public contracting, but it did limit their use to the “extreme case” where patterns of deliberate exclusion are shown.11

The Croson Court established that racial preferences in public contracting are analyzed under strict scrutiny, and therefore that both a compelling governmental interest and narrow tailoring are required. As Justice O’Connor explained, “[p]roper findings in this regard are necessary to define both the scope of the injury [compelling interest] and the extent of the remedy necessary to cure its effects [narrow tailoring].”12

The Court also discussed the relevance of anecdotal evidence, stating that anecdotal evidence “of a pattern of individual discriminatory acts can, if supported by appropriate statistical proof, lend support to a government’s determination that broader remedial relief is justified.”13 However, the Court did not specifically define the methodology that would characterize proper anecdotal evidence in these cases. This article recommends methodological guidelines for obtaining reliable anecdotal evidence.

II. Croson-Proofing Preference Programs Through Disparity Studies

Following the Croson decision, many states, cities, counties, and local agencies commissioned disparity studies in order to produce the strong basis in evidence needed for their preference programs to withstand strict scrutiny.14 One law journal article referred to this as “Croson-proofing” the preference programs.15

As a result of Croson, government entities began to conduct disparity studies to justify their use of preferences in awarding contracts. Disparity studies have been described as “the strange fruit of the most significant civil-rights decision of the 1980s, City of Richmond v. Croson.”16 The decision requires “proper findings” of discrimination to justify racial preferences, and such findings are offered in disparity studies. The contents and methodological approaches of disparity studies vary widely, but they all have statistical analyses of availability of MBEs, and nearly all contain anecdotal sections as well.17

The anecdotal evidence that appears in disparity studies is obtained through a variety of sources, including public hearings, interviews, focus groups, and surveys by both mail and telephone. Sometimes the anecdotal evidence is quantitatively summarized, sometimes it is paraphrased without context, and sometimes it appears in the form of verbatim excerpts. Rarely are any of the anecdotal sources named in disparity studies, forcing governments and courts to evaluate the validity of anonymous allegations of discrimination.

Disparity studies generally categorize anecdotes. The October 24, 2014 “State of Missouri Disparity Study” is typical, and includes these categories: unequal access to industry and information networks, discriminatory attitudes and negative perceptions of competence, obtaining private sector work on an equal basis, and obtaining private sector work or “no goals” work on an equal basis.18 The anecdotes often refer to a survey respondent’s or interviewee’s negative experiences in the procurement process. That negative bias may reflect reality, or it may reflect the kinds of people who choose to respond to the survey, or it may even result from editorial decisions by the study authors. Because there is rarely a third party check on the representativeness of anecdotes used in a disparity study, the study authors control the narrative.

A. Anecdotal Evidence and Strict Scrutiny

Anecdotal evidence in disparity studies has been offered to satisfy both prongs of strict scrutiny: compelling interest and narrow tailoring. For state and local governments to demonstrate a compelling interest in maintaining a racial preference, they must satisfy two conditions. First, the government must identify the discrimination it seeks to remedy, whether public or private, with enough specificity to at least approach a prima facie case. This is accomplished by showing that the government actively or passively participated in the discrimination in the local market.19 Second, there must be a strong basis in evidence to support the conclusion that remedial action is necessary.20 Remedial action addressing the present effects of identified past discrimination is the only compelling interest that can justify the use of preferences in public contracting.21

Engineering Contractors of South Florida v. Metropolitan Dade County is one of the leading lower court cases on the sufficiency of anecdotal evidence.22 The record in that case contained anecdotal complaints of discrimination by MWBEs. These anecdotes described, among other things, incidents in which suppliers quoted higher prices to MWBEs than to their non-MWBE competitors, and in which non-MWBE prime contractors unjustifiably replaced a MWBE subcontractor with a non-MWBE subcontractor.23 These kinds of anecdotes go to the compelling interest prong of strict scrutiny.

Even where a government shows a compelling interest, it is still constrained in how it may pursue that valid end. The means “chosen to accomplish the [government’s] asserted purpose must be specifically” and narrowly tailored to accomplish the compelling interest.24 The purpose of the narrow tailoring requirement is to ensure that “the means chosen ‘fit’ … the compelling goal so closely that there is little or no possibility that the motive for the classification was illegitimate racial prejudice or stereotype.”25

There are six requirements to show narrow tailoring in cases involving racial preferences in public contracting: (1) the program’s MWBE group classifications cannot be overly inclusive;26 (2) the program must show that race neutral alternatives have been tried, evaluated, and found insufficient;27 (3) the program’s goals must be related to the actual availability of MWBE firms;28 (4) the program and its MWBE goals must not create an undue burden on third parties; (5) the preferences must be shown to be necessary; and (6) the program must be flexible and have adequate waiver provisions.29

In Rowe v. North Carolina Department of Transportation, the Fourth Circuit found that evidence from a telephone survey, personal interviews, and focus groups was strong evidence of discriminatory treatment of African American and Native American firms.30 Based on this anecdotal evidence (coupled with statistical evidence), the court ruled that contracting preferences for those two groups were justified, but that preferences which had been in force for other MWBE groups were not justified.31 The result that some minority groups remain in the MWBE program while others are removed from the program is an application of the narrow tailoring requirement that a program’s MWBE classifications not be overinclusive.

B. Allocating the Burden

Although preferential programs supposedly supported by disparity studies have often been challenged, courts are still confused on the issue of which party bears the burden in this type of constitutional litigation. Determining who bears the burden is a three-step analytic process; courts typically get the first two steps correct and ignore the third.

Courts have properly held that, where a government has implemented a preferential program, the government bears the initial burden of showing a strong basis in evidence (typically through a disparity study) to support its program; the evidence must show that the groups who will benefit from the preferential program have suffered from patterns of discrimination against them in public contracting that should be remedied by the preference.32 After the government makes this initial evidentiary showing, the burden shifts to the plaintiff to rebut that showing.33 This is usually done by attacking the sufficiency of the statistics and anecdotes in the disparity study. Lower courts usually correctly require litigants to meet these two burdens.

In any equal protection action questioning the constitutionality of racial or gender preferences, it is the defendant-government that bears the third and ultimate burden of proving that its preferences satisfy the appropriate level of scrutiny (strict for race, intermediate for gender).34 In Johnson v. California, a case involving race-based assignments of prisoners to cells, the Supreme Court declared, “We put the burden on state actors to demonstrate that their race-based policies are justified.”35 In United States v. Virginia, a case involving differing treatment for men and women, the Court placed the “burden of justification” for the differing treatment of the sexes on the government.36 The Ninth Circuit correctly placed the burden of justification on the government in Western States Paving v. Washington State Department of Transportation, but lower courts usually err by placing the final burden on plaintiffs.37

III. The Sufficiency of Anecdotal Evidence

A. Anecdotes Alone Are Never Enough

In all the many challenges to preferential programs, no race conscious program has ever been upheld solely on the basis of anecdotal evidence.38 In Coral Construction v. King County, the Ninth Circuit noted that 57 affidavits from MWBEs alleging discrimination were not sufficient to establish the constitutionality of the preference because, “[w]hile anecdotal evidence may suffice to prove individual claims of discrimination, rarely, if ever, can such evidence show a pattern of discrimination necessary for the adoption of an affirmative action plan… the MBE program cannot stand without a proper statistical foundation.”39

However, anecdotal evidence “nevertheless is essential if a government is to defend a MBE program successfully.”40 This is because disparity ratios alone cannot identify the source of the discrimination, if any, and a preferential program in government contracts cannot remedy a problem if its source is unknown. Regression analysis, a statistical tool found in many disparity studies, suffers from this same flaw of being unable to identify the source of a problem.41

B. Croson’s Progeny on Anecdotes

Since Croson did not address anecdotes at any serious length, its lower court progeny must provide guidance.42 The two leading cases on anecdotal evidence are Engineering Contractors of South Florida v. Metropolitan Dade County43 and Associated General Contractors v. City of Columbus.44 These two cases offer the most thorough analyses of standards for anecdotal evidence. 

The district court in Engineering Contractors struck down Dade County, Florida’s MBE and WBE programs, declaring that the county’s statistical45 and anecdotal46 evidence was too weak to survive strict scrutiny. The court concluded that the anecdotal evidence offered could not cure the weaknesses of the statistical evidence because the anecdotal evidence showed something more akin to societal discrimination that was “not the sort of ‘identified discrimination’ contemplated by Croson.”47 Concerning anecdotes, the district court in Engineering Contractors stated:

Plaintiffs respond with several points the Court believes to be valid concerning the reliability of this anecdotal evidence. First, whether discrimination has occurred is often complex and requires a knowledge of the perspectives of both parties involved in an incident as well as knowledge about how comparably placed persons of other races, ethnicities, and genders have been tested. Persons providing anecdotes rarely have such information. Attributing an incident to discrimination when the practice is just aggressive business behavior, barriers faced by all new or small businesses, or bad communication is always a possibility . . . . individuals who have a vested interest in preserving a benefit or entitlement may be motivated to view events in a manner that justifies the entitlement. Consequently, it is important that both sides are heard and that there are other measures of the accuracy of the claims. Attempts to investigate and verify the anecdotal evidence should be made.48 

More specifically, the Engineering Contractors court focused on three issues: interviewer bias, response bias, and verification of anecdotes. After addressing the first two issues in this section, this article will go on to address the third in Part IV.

1. Interviewer Bias

According to the Engineering Contractors district court, interviewer bias could occur when the interviewer either phrases questions in a suggestive manner or implies the political purpose of the question to the respondent.49 This problem can also arise where there are only questions about discrimination and not about difficulties arising from nondiscriminatory factors.50

The other leading case on anecdotal evidence is Associated General Contractors of America v. City of Columbus.51 Like the district court in Engineering Contractors, the district court in Columbus addressed interviewer bias, insisting that “investigators should be impartial and unbiased and they should be reasonably thorough and diligent.”52 The court further warned that “investigators should consider the credibility and potential bias of witnesses and respondents.”53 After an extensive discussion of the anecdotal evidence before it, the court ruled that the Columbus MWBE program failed to meet the Equal Protection requirements of Croson. The ruling presents a thorough evaluation of the statistical and particularly the anecdotal evidence.54 The anecdotal evidence under review in the case was extensive and included a disparity study. Nevertheless, the court declared that the anecdotal evidence was “poorly executed” and “fell far short of proof of pervasive discrimination in the private sector.”55

2. Response Bias

Response bias becomes a concern when a sample of respondents is not carefully constructed and consequently is unrepresentative of all potential respondents. As the court in Engineering Contractors observed, this may occur because the people most likely to respond to surveys are those who feel most strongly about the problem under review.56 Anecdotes might be motivated by the knowledge that the continuing use of preferences is dependent on “their ability to create a record of discrimination” such that “the incentive to engage in memory contrivance, consciously or unconsciously, is substantial.”57 Investigation could well determine if memory contrivance has occurred. George LaNoue, Professor of Political Science at University of Maryland, Baltimore County, has noted that, “[w]here vested interests are so clear and constitutional rights are at stake, the researcher’s need to use careful methodologies and to report only verified information is strong. Unfortunately, disparity studies do not meet the test.”58

The Columbus opinion expressed a similar concern: “Extra care should be taken in gathering and evaluating anecdotal evidence from advocates of race- and gender-based preferences. Such informants may be prone to exaggerate or fabricate circumstances and events or omit important details.”59 One way to take the “extra care” is to ensure that those questioned “include a fair sampling of all segments of the community who have relevant knowledge and who would be impacted by such legislation.”60

IV. Verification of Anecdotes

The problems of interviewer bias and response bias can be most effectively addressed by verification of the anecdotal evidence. Anonymous, unverified anecdotes are at best hearsay evidence. Anecdotes of discrimination should be corroborated:

As the Dade County and Columbus cases make clear, it can be fundamentally important for jurisdictions to demand that [disparity study] consultants diligently seek to verify individuals’ accounts of discrimination. Adequate verification will require consultants to approach the task with skepticism. They should assess the perspectives of parties accused of discriminatory acts, and consider potential nondiscriminatory explanations.61

In Engineering Contractors, Dade County offered anecdotal evidence in two forms: testimony from program staff and the results of a survey of black-owned firms.62 The court pointed out that the anecdotal evidence offered in Engineering Contractors needed to be investigated and verified; otherwise, “[w]ithout corroboration, the Court cannot distinguish between allegations that in fact represent an objective assessment of the situation, and those that are fraught with heartfelt, but erroneous, interpretations of events and circumstances.”63 Anecdotal claims of discrimination can be unreliable, so they should “be treated cautiously” due to the inherent difficulty in verifying that they are being “remembered, perceived, or reported accurately.”64 Even if accurate, anecdotal allegations of discrimination are suspect because they may be anomalous and reflect no pattern of discrimination such as might justify a preference program. In such cases, the compelling interest prong of strict scrutiny has not been met. The possibility of response bias may make anecdotal evidence unreliable.

The Columbus opinion shares Engineering Contractors’ concerns about the need for verification. The court maintained that “attempts should be made to verify claims of discrimination where it is reasonable to do so,” and insisted that for anecdotal evidence to be persuasive, the collection of the anecdotes must meet “minimum standards of objectivity and diligence.”65 In a somewhat exasperated tone, the court went on to say that investigators should ask the same sorts of questions “any first-year journalism student knows to ask: ‘who, what, when, where, why and how?’”66 The court was also worried that the anecdotal evidence incorrectly emphasized perceptions of discrimination, rather than actual discrimination.67

One federal court has considered the specific problem of whether perception of discrimination is sufficient to support racial preferences. The court in Phillips & Jordan v. Watts emphatically rejected perceptions as evidence of discrimination when it pointed out that:

Individuals responding to FDOT’s [Florida Department of Transportation] telephone survey have described their perceptions about barriers to FDOT’s bidding procedures. But FDOT has provided no evidence to establish who, if anyone, in fact engaged in discriminatory acts against Black and Hispanic businesses. The record at best establishes nothing more than some ill-defined wrong caused by some unidentified wrongdoers; and under City of Richmond [v. Croson] that is not enough!68  

Other courts have disagreed with the Engineering Contractors, Columbus, and Phillips & Jordan decisions, ruling instead that it is not necessary to investigate and possibly verify allegations of discrimination. For these courts, perceptions of discrimination are sufficient to justify preferences. In Concrete Works v. City and County of Denver, the Tenth Circuit held that “Denver was not required to present corroborating evidence and [the plaintiff] was free to present its own witnesses to either refute the incidents described by Denver’s witnesses or to relate their own perceptions on discrimination in the Denver construction industry.”69 The Tenth Circuit opinion assumes, at the very least, that depositions can be taken of those anecdotal witnesses claiming discrimination as part of the disparity study, or that these witnesses could be cross examined at trial. The financial burden on the plaintiff to pursue these paths to question the witnesses claiming discrimination could well prove to be overwhelming.

The Fourth Circuit in H.B. Rowe Company v. Tippett found that the State’s anecdotal evidence of discrimination against African American and Native American subcontractors, obtained through telephone surveys, sufficiently supplemented the statistical evidence to justify preferences for these two groups.70 The court ruled that the telephone survey exposed an informal, racially exclusive network which systematically disadvantaged the two groups.71 The court felt confident to rule this way since no public policy reasons were presented as to why the unverified survey answers might be untrustworthy. The court specifically rejected the requirement that anecdotal evidence be investigated or corroborated.72 The Fourth Circuit stated that:

Rowe offered no rationale as to why a fact finder could not rely on the State’s “unverified” anecdotal data. Indeed, a fact finder could very well conclude that anecdotal evidence need not–and indeed cannot–be verified because it “is nothing more than a witness’ narrative of an incident told from the witness’ perspective and including the witness’ perceptions.”73

In contrast, a disparity study company has recognized the potential problems with perceptions of discrimination. The 1993 MGT disparity study for the North Carolina Department of Transportation, the same agency whose preferences were at stake in Rowe, includes this cautionary statement: “Firms and individuals who lose contracts no doubt sometimes believe they were discriminated against even when no discrimination exists.”74

V. Public Policy Reasons for Requiring Verification of Anecdotes

Are there rationales for demanding that perceptions of discrimination be verified in order to satisfy strict scrutiny? Why should investigation and corroboration of anecdotal claims of discrimination be necessary? The court in Engineering Contractors of South Florida v. Dade County warned that there are costs of accepting unverified anecdotes:

Without corroboration, the Court cannot distinguish between allegations that in fact represent an objective assessment of the situation, and those that are fraught with heartfelt, but erroneous, interpretations of events and circumstances. The costs associated with the imposition of race, ethnicity, and gender preferences are simply too high to sustain a patently discriminatory program on such weak evidence.75

First, it is a matter of basic fairness that any preferential program which disadvantages some people according to their race, ethnicity, and gender while advantaging others should only be implemented where discrimination has been shown to be real, and not simply perceived. Unless they are investigated, these anecdotal accounts remain mere perceptions of discrimination. Political scientist Mitchell Rice has observed that “[a]necdotal evidence, interviews and affidavits must be from reliable and trustworthy sources and should include counter explanations and rebuttals from sources accused of bias. In other words, the gathering of evidence utilizing these approaches must be fair and deliberative.”76

Second, unless the claims of discrimination can be verified, the right remedy to that discrimination cannot be fashioned. To find and implement the most effective remedy, perception must be distinguished from reality. For example, consider a situation in which a telephone survey includes a MBE owner’s claim that he was denied a bank loan due to discrimination, but investigation of the claim would have revealed that he had a faulty business plan and prior bankruptcy? Simply accepting this perception of discrimination as accurate would mean the most effective remedy–one which addresses the business plan and bankruptcy—would never happen. The district court in Associated General Contractors of Connecticut v. New Haven addressed this need to connect evidence of discrimination to its proper remedy.77 The court discussed a number of anecdotal complaints of discrimination, such as having tools stolen, being harassed, having difficulty in obtaining loans, enduring animus on the part of trade unions, and problems in training, bonding, or insurance, but concluded that these anecdotes did “not rise to the level of showing a systematic pattern of discrimination to the exclusion of any other explanation.”78

Third, strict scrutiny demands investigation of anecdotal claims of discrimination. Consider the following hypothetical. Two cars collide at an intersection. The police arrive and take statements from both drivers. Unsurprisingly, each claims the other is at fault. The police would investigate. Are there skid marks to support Mr. Smith’s claim that he braked suddenly to avoid the other car running a stop sign? Ms. Jones claims she saw the other driver on a cell phone; is a cell phone visible? For that matter, is there a stop sign? Has either party been drinking alcohol? What were the weather and road conditions? The police and involved insurance companies would undoubtedly investigate in order to determine fault.

Now, vary the hypothetical to include this single claim: Mr. Smith alleges Ms. Jones deliberately hit him because she is a racist. Should Ms. Jones be convicted of a hate crime on thebasis of his accusation alone? Shouldn’t that claim of discriminatory action be investigated if at all possible? If a fender bender is thoroughly investigated, doesn’t strict scrutiny require investigation of claimed discrimination when the constitutional rights of others to not be disadvantaged because of race are at stake? There is at present no consistently applied judicial approach to the verification of anecdotes on the discrimination issue. Consistency is certainly needed. Once courts have been presented with the public policy reasons for verification offered in this article, as the Fourth Circuit in Rowe requested, that inconsistency may change.

VI. Conclusion

The 1989 Supreme Court decision City of Richmond v. Croson is a landmark civil rights ruling. Croson launched the disparity study industry, which arose as governments attempted to meet Croson’s requirement that discrimination be identified with enough specificity so that effective remedies could be fashioned. When disparity studies are challenged in court, the burden of justification should be on the government. Virtually all disparity studies contain anecdotal accounts of discrimination, which can be more or less reliable depending on the extent to which anecdotes are investigated. Verification of anecdotal evidence supporting racial preferences is a matter of fairness to those who are disadvantaged by those preferences. Without verification, the right remedy for discrimination cannot be fashioned.



1  Colette Holt & Associates, State of Missouri Office of Administration Disparity Study (2014), 110, available at

2  488 U.S. 469 (1989).

3  Related acronyms which appear in this article are MWBE (minority and/or women-owned business enterprise); WBE (women-owned business enterprise); and DBE (disadvantaged business enterprise).

4  Croson, 488 U.S. at 477.

5  Similarly, the Due Process Clause of the Fifth Amendment of the United States Constitution, into which Equal Protection principles have been incorporated, protects an individual’s right to due process of the law: “No person shall be . . . deprived of life, liberty, or property without due process of law.” The obligations imposed by the Fifth and Fourteenth Amendments are indistinguishable. Adarand v. Peña, 515 U.S. 200 (1995). 

6  Adarand, 515 U.S. at 224.

7  Croson, 488 U.S. at 510.

8  Id. at 497.

9  Id. at 506. The Croson court declared that the city had no evidence of past discrimination against “Spanish-speaking, Oriental, Indian, Eskimo, or Aleut persons in any aspect of the Richmond construction industry.” The Court elaborated: “The random inclusion of racial groups that, as a practical matter, may never have suffered discrimination in the construction industry in Richmond, suggests that perhaps the city’s purpose was not in fact to remedy past discrimination. If a 30% set-aside was ‘narrowly tailored’ to compensate black contractors for past discrimination, one may legitimately ask why they are forced to share this ‘remedial relief’ with an Aleut citizen who moves to Richmond tomorrow?” Id. 

10  Id. at 493.

11  Id. at 509.

12  Id. at 506, 510.

13  Id. at 509.

14  As of December 2014, more than 350 disparity studies had been released, according to a list compiled by the Project on Civil Rights and Public Contracting. The estimated total cost of these studies likely exceeds $150 million.

15  Docia Rudley & Donna Hubbard, What a Difference a Decade Makes: Judicial Response in State and Local Minority Business Set-asides Ten Years After City of Richmond v. J.A. Croson, 25 S. Ill. U. L.J. 39, 42 n.17 (2000).

16  George R. LaNoue, Social Science and Minority “Set-Asides,” 110 The Public Interest 49, 49 (Winter 1993).

17  One circuit court has referred to this variety as an “evidentiary mosaic.” Concrete Works of Colorado, Inc. v. City of Denver, 36 F.3d 1513, 1520 (10th Cir. 1994). In the early years after Croson, disparity studies often had historical sections, though such sections are rare today. George LaNoue, Selective Perception: The Role of History in the Disparity Study Industry, 17 The Public Historian 2 (Spring 1995).

18  See supra note 1 at 109-16 (Qualitative Evidence of Race and Gender Barriers in the Missouri Economy).

19  Croson, 488 U.S. at 492, 499-500.

20  Id. at 500, quoting Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 277 (1968).

21  Id. at 493. In public education, by way of contrast, preferences may also be justified by a desire for diversity.

22  943 F. Supp. 1546 (S.D. Fla 1996).

23  Id. at 1579.

24  Shaw v. Hunt, 517 U.S. 899, 908 (1996).

25  Croson, 488 U.S. at 493.

26  A preferential program is narrowly tailored only if its application is limited to those groups which have actually suffered discrimination. Otherwise, the program provides MWBE groups who have not encountered discriminatory barriers with an unconstitutional competitive advantage at the expense of both non-MWBEs and women or minority groups that have been actual victims of discrimination.

27  On the narrow tailoring facet of race neutral alternatives, Supreme Court jurisprudence has advanced substantially in the 25 years since Croson. Grutter v. Bollinger, decided in 2003, declared that racial preferences will only be permitted for another 25 years, at most. 539 U.S. 306, 342 (2003). By the end of 2028, preferences will be unconstitutional, and only race neutral alternatives will be allowed.

28   In Croson, Justice O’Connor supplied the statistical test which a jurisdiction should use as its beginning point for the kind of proper findings which would determine whether patterns of discriminatory exclusion have been identified: “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.” Croson, 488 U.S. at 509-10.

29  United States v. Paradise, 480 U.S. 149, 171 (1987).

30  615 F.3d 233 (4th Cir 2010).

31  Id. at 251.

32  See, e.g., Rothe Development Corporation v. Department of Defense, 545 F.3d 1023, 1036 (Fed. Cir. 2008); Monterey Mechanical Company v. Wilson, 125 F.3d 702, 713 (9th Cir. 1997); Hershell Gill Consulting Engineers, Inc. v.  Miami Dade County, 333 F.Supp. 2d 1305, 1316 (S.D. Fla. 2004).

33  Engineering Contractors, 122 F.3d at 916.

34  Gratz v. Bollinger, 539 U.S. 244, 270 (2003). In Gratz the Court held that the admissions program used by the University of Michigan’s College of Literature, Science and the Arts violated the Equal Protection Clause, Title VI of the Civil Rights Act of 1964, and 42 U.S.C. § 1981.

35  543 U.S. 499, n.1 (2005).

36  518 U.S. 515, 553 (1996).

37  407 F.3d 983 (9th Cir. 2005).

38  “Anecdotal evidence” includes trial testimony, evidence at trial, and anecdotal material in disparity studies.

39  941 F.2d 910, 919 (9th Cir. 1991).

40  Jeffrey Hanson, Hanging by Yarns: Deficiencies in Anecdotal Evidence Threaten the Survival of Race-Based Preference Programs for Public Contracting, 88 Cornell L. Rev. 1433, 1447-48 (2002).

41  See, e.g., NERA Economic Consulting, The State of Minority- and Women-Owned Business Enterprise: Evidence from New York (April 2010), 139-86, available at

42  There seems to be a conflict among the circuits on the issue of whether appellate review of a strong basis in evidence, which includes reviewing anecdotes, is treated as a factual determination not to be overturned by the circuit court unless clearly erroneous (Engineering Contractors Association of South Florida v. Metropolitan Dade County, 122 F.3d 895, 903 (11th Cir. 1997)) or whether the strong basis in evidence is a legal determination to be reviewed de novo (Concrete Works of Colorado v. City of Denver, 321 F.3d 950, 958 (10th Cir. 1996)).

43  943 F. Supp. 1546. 

44  936 F. Supp. 1363 (S.D. Ohio 1996), vacated on jurisdictional grounds by 172 F.3d 411 (6th Cir. 1999) (unconstitutionality of the Columbus MWBE preference on construction contracts undisturbed).

45  943 F. Supp. at 1560-77.

46  Id. at 1577-80.

47  Id. at 1580. On appeal, the Eleventh Circuit affirmed the district court’s ruling without commenting directly on the district court’s extensive criticisms of the anecdotal evidence. Engineering Contractors, 122 F.3d 895. The appeals court included a brief discussion of the anecdotal evidence because the County’s statistical evidence was weak, and the court noted that “only in the rare case will anecdotal evidence suffice standing alone.” Id. at 925.

48  943 F. Supp. at 1579.

49  Id.

50  Susan A. MacManus, Minority Business Contracting with Local Government, 25 Urb. Aff. Q. 455 (1990).

51  936 F. Supp. 1363.

52  Columbus, 936 F. Supp. at 1426.

53  Id. at 1428.

54  The court issued a 71 page opinion and 20 page appendix. The discussion of the anecdotal material runs more than 40 pages.

55   936 F. Supp. at 1373. The appeals court praised the extensiveness of the lower court’s opinion: “The record in this case is voluminous and the district court’s effort in reviewing that record and issuing its ruling was thorough and exhaustive.”

56   943 F. Supp. at 1579.

57  George R. LaNoue, Standards for the Second Generation of Croson-Inspired Disparity Studies, 3 The Urban Lawyer 485, 524 (Summer 1994).

58  Id. at 525.

59  936 F. Supp. at 1426.

60  Id.

61  Hanson, supra at 1468-69.

62  943 F. Supp. at 1546, 1577-1580.

63  Id. at 1579.

64  LaNoue, Standards for the Second Generation of Croson-Inspired Disparity Studies, supra at 525.

65  936 F. Supp. at 1426.

66  Id.

67  Id. at 1373.

68  13 F. Supp. 2d 1308 (N.D. Fla. 1998) (emphasis in original).

69  321 F.3d 950, 989 (10th Cir. 2003).

70  Rowe, 615 F.3d at 250.

71  Id. at 251.

72  Id., quoting Concrete Works, 321 F.3d at 989.

73  Id. at 249.

74  MGT of America, Inc., Study of Minority and Women Business Participation in Highway Construction, 7.

75  943 F. Supp. at 1584 (emphasis added).

76  Mitchell F. Rice, Government Set-Asides, Minority Business Enterprises, and the Supreme Court, 51 Public Adm. Rev. 2, 485 (March/April 1991).

77  791 F. Supp. 941 (D. Conn. 1992).

78  Id. at 948.