National Academies Press: OpenBook

Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (2010)

Chapter: Chapter 2 - Designing Defensible DBE Programs

« Previous: Chapter 1 - Overview of Legal Standards for Disadvantaged Business Enterprise Programs
Page 9
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 9
Page 10
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 10
Page 11
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 11
Page 12
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 12
Page 13
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 13
Page 14
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 14
Page 15
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 15
Page 16
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 16
Page 17
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 17
Page 18
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 18
Page 19
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 19
Page 20
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 20
Page 21
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 21
Page 22
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 22
Page 23
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 23
Page 24
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 24
Page 25
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 25
Page 26
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 26
Page 27
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 27
Page 28
Suggested Citation:"Chapter 2 - Designing Defensible DBE Programs." National Academies of Sciences, Engineering, and Medicine. 2010. Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program. Washington, DC: The National Academies Press. doi: 10.17226/14346.
×
Page 28

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

9Introduction The federal government apportions tens of billions of dol- lars each year from the National Highway Trust Fund to the fifty states and the District of Columbia to assist and support the construction and maintenance of the nation’s highway infrastructure.27 In the decade between FFYs28 1998 and 2007, more than $325 billion in federal highway transportation assis- tance (in 2007 dollars) was allocated to state DOTs.29 In FFY 2007 alone, the amount distributed was almost $32 billion. Table 1 shows the apportionment of federal highway funds among state DOTs in FFY 2007, as well as their 2007 rankings and their average rankings from FFYs 1998 through 2007. As a condition of receipt of these federal funds, the state DOTs must agree to abide by certain federal rules and regu- lations, including those governing the U.S.DOT’s DBE Pro- gram.30 The DBE Program, as originally enacted and as revised, was established by Congress in response to wide- spread evidence of discrimination against businesses owned by minorities and women in the construction industries, and in particular the highway construction industries.31 The principal objectives of the DBE Program are to: • Ensure “nondiscrimination in the award and administra- tion of DOT-assisted contracts in the Department’s high- way, transit, and airport financial assistance programs”; • “To create a level playing field on which DBEs can compete fairly for DOT-assisted contracts”; • “To help remove barriers to the participation of DBEs in DOT-assisted contracts”; and • “To assist the development of firms that can compete suc- cessfully in the marketplace outside the DBE program.”32 The DBE Program operates on both the supply-side and the demand-side of the market for highway construction. On the supply-side, the program encourages recipients to undertake efforts to increase the number of firms that can be certified to participate in the program and to enhance the ability of DBEs to compete effectively in the highway contracting marketplace. Such “race-neutral” activities include structuring solicitations, quantities, specifications, and delivery schedules to facilitate increased participation by DBEs; assisting DBEs to overcome barriers related to surety bonding and other types of financing; assisting start-up DBE firms to become established; imple- menting communications programs regarding contracting procedures and specific contract opportunities; implementing supportive services programs to develop business manage- ment, record keeping, and accounting skills; helping firms learn to handle increasingly significant projects and a greater diversity of project types; and assisting firms in the adoption of emerging technologies and use of electronic media.33 On the demand-side, the DBE Program requires state DOTs to encourage prime contractors and consultants to subcon- tract to DBEs by placing percentage participation goals on cer- tain contracts and requiring prime contractors and consul- tants to make good-faith efforts to meet those goals, and ensuring distribution of the DBE directory to the widest feasi- ble universe of prime contractors.34 Such demand-side activi- ties are described in Part 26 as “race-conscious” efforts.35 C H A P T E R 2 Designing Defensible DBE Programs 27 Highway Trust Fund monies are also allocated to American Samoa, Guam, the Northern Mariana Islands, Puerto Rico, and the Virgin Islands. These entities, however, are excluded from the scope of the present report. 28 The federal fiscal year runs from October 1 through September 30. 29 U.S. Department of Transportation (2008) and U.S. Department of Labor (2008). 30 49 C.F.R. Part 26. 31 See, e.g., The Compelling Interest for Affirmative Action in Federal Procurement, 61 Fed. Reg. 26050 et seq (1996). 32 49 C.F.R. 26.1. 33 49 C.F.R. 26.51(b). 34 49 C.F.R. 26.51(d). 35 Since women-owned firms can also qualify as DBEs, such efforts can also be “gender-conscious.” Following the nomenclature of Part 26, we shall refer to both types of efforts as “race-conscious.”

10 State Federal-aid Highway Fund Apportionment (in thousands) FFY 2007 Rank Federal-aid Highway Fund Apportionment (in thousands) FFY 1998–FFY 2007 Rank California $2,960,914 1 $26,397,793 1 Texas $2,624,571 2 $23,067,022 2 New York $1,512,802 4 $14,129,262 3 Florida $1,556,841 3 $14,044,320 4 Pennsylvania $1,508,597 5 $13,666,393 5 Georgia $1,148,445 7 $10,325,502 6 Ohio $1,085,685 6 $10,079,858 7 Illinois $965,678 8 $9,447,852 8 Michigan $971,738 9 $8,996,057 9 North Carolina $890,846 10 $8,150,729 10 New Jersey $831,505 12 $7,554,338 11 Virginia $819,035 11 $7,546,287 12 Indiana $751,898 13 $6,935,524 13 Missouri $739,140 14 $6,759,202 14 Tennessee $696,083 15 $6,444,907 15 Alabama $629,576 16 $5,745,715 16 Wisconsin $614,695 18 $5,658,072 17 Arizona $594,321 17 $5,184,202 18 Massachusetts $552,890 22 $5,140,683 19 Kentucky $539,127 19 $4,990,111 20 Washington $519,858 21 $4,953,246 21 South Carolina $516,771 20 $4,799,501 22 Maryland $502,039 23 $4,638,775 23 Louisiana $489,304 25 $4,526,336 24 Oklahoma $493,726 26 $4,464,640 25 Minnesota $416,012 24 $4,246,007 26 Connecticut $450,100 27 $4,217,953 27 Arkansas $396,891 29 $3,693,491 28 Colorado $391,802 28 $3,628,968 29 Mississippi $371,797 30 $3,511,338 30 Table 1. Federal-aid highway apportionments by state, FFYs 1998–2007.

As discussed in Chapter One, and in more detail in Appen- dix C, the U.S. Supreme Court’s plurality opinion in City of Richmond v. J. A. Croson Co., requires the highest level of review on the use of race-conscious classifications by public entities, even when such classifications are used as a tool for remedying the effects of past and present discrimination.36 The Court’s explanation of the type of evidence that would support a race-conscious contracting program gave rise to the “disparity study.” These studies gathered statistical and anec- dotal evidence of discrimination against M/WBEs to allow state or local policy makers to determine whether there was a “strong basis in evidence”—the standard of proof imposed by the Supreme Court—to adopt race-conscious remedies. This project seeks to provide guidance and models to state DOTs considering what evidence is necessary to meet their constitutional obligations and regulatory mandates. The guide- lines for disparity and availability studies, presented below, reflect recent judicial decisions about what is necessary to meet strict constitutional scrutiny and implement best prac- tices for DBE Programs, based upon controlling precedents, sound science, and practical experience. 11 36 488 U.S. 469. See Chapter One for an overview and Appendix C for a detailed discussion of the law governing contracting affirmative action. Oregon $355,299 31 $3,359,554 31 Iowa $334,938 32 $3,296,421 32 Kansas $340,126 34 $3,221,821 33 Alaska $281,428 37 $3,148,654 34 West Virginia $338,224 33 $3,088,318 35 Montana $301,199 35 $2,867,240 36 New Mexico $301,437 36 $2,796,547 37 Nebraska $229,493 39 $2,213,989 38 Utah $225,006 40 $2,169,214 39 Idaho $232,943 38 $2,126,670 40 Nevada $200,879 41 $2,028,138 41 Wyoming $214,290 42 $2,005,894 42 South Dakota $209,154 43 $1,995,007 43 North Dakota $194,716 44 $1,872,619 44 Rhode Island $158,583 45 $1,639,456 45 Maine $139,141 47 $1,440,198 46 New Hampshire $152,770 46 $1,411,474 47 Hawaii $140,971 49 $1,393,796 48 Vermont $133,324 48 $1,294,456 49 Delaware $124,031 51 $1,246,687 50 District of Columbia $125,354 50 $1,142,586 51 Source: U. S. Department of Transportation, Federal Highway Administration (2008). State Federal-aid Highway Fund Apportionment (in thousands) FFY 2007 Rank Federal-aid Highway Fund Apportionment (in thousands) FFY 1998–FFY 2007 Rank Table 1. (Continued).

Guidelines for Conducting Disparity and Availability Studies The objective of the study is to develop the following tools for state DOTs: • Guidelines to assist them in determining when and if a disparity/availability study is recommended; • A Model Scope of Work to be included in future RFPs or similar solicitations for disparity/availability studies; and • Detailed recommendations on how to design and imple- ment availability and disparity studies. In Croson, the city’s central piece of statistical evidence was that “while the general population of Richmond was 50% black, only 0.67% of the city’s prime construction con- tracts had been awarded to minority businesses [between] 1978 and 1983.”37 The Court ruled, in essence, that the city’s disparity statistic compared “apples to oranges” and there- fore did not support an inference of discrimination. The correct benchmark for business discrimination cases was held to be the percentage of minority-owned firms in the business population of an area. The Court stressed that “proper” statistical evidence—evidence that revealed significant racial disparities between a government’s minority procurement and the availability of minority businesses in the surround- ing economy—would be an acceptable form of evidence in future litigation. Such evidence—possibly standing alone, or in combination with other factual showings—could be used to support an inference of discriminatory exclusion. “There is no doubt that ‘[w]here gross statistical disparities can be shown, they alone in a proper case may constitute prima facie evidence of a pattern or practice of discrimina- tion’ under Title VII.”38 In light of Croson and subsequent decisions, many state and local governments abandoned their efforts at race-conscious contracting either voluntarily or upon being enjoined by the courts.39 Others took a different approach and commissioned disparity studies to determine whether, in light of Croson, there was sufficient evidence of discrimination for their jurisdictions to justify continuation or enactment of race-conscious con- tracting programs. At the federal level, the Adarand case established that Con- gress must also meet strict scrutiny. To meet this test, the DBE Program was revised in 1999 based upon the necessary compelling interest, and narrowly tailored regulations were adopted. 49 C.F.R. Part 26 describes five methods recipients may employ to set narrowly tailored goals for their imple- mentation of the DBE Program.40 One is to conduct a dispar- ity study. Over 300 disparity studies have been conducted in the 20 years since Croson.41 Studies have varied considerably in content and in quality. At their best, they have been inde- pendent, objective, comprehensive assessments conducted in accordance with accepted practices in social science research and have been instrumental in assisting government entities, including state DOTs, to survive constitutional challenges to their M/WBE or DBE Programs. At their worst, they have been poorly conceived and executed, ultimately wasting the scarce public resources expended on them when they proved inadequate in litigation or even leading to ongoing programs being dropped without a legal challenge. Below, we describe in more detail the major and minor components that are typically found in the best disparity studies. In summary, such studies typically include the fol- lowing major elements: 1. A legal review discussing Croson and subsequent case law and legal standards; 2. An empirical assessment of the appropriate geographic market relevant to an agency’s contracting activity; 3. An empirical assessment of the appropriate product markets relevant to an agency’s contracting activity; 4. An estimate of the fraction of businesses within the agency’s geographic and product markets that are owned by DBEs (i.e., “availability”); 5. An estimate of the percentage of all prime contract and subcontract dollars earned by DBEs (i.e., “public sector utilization”); 6. A statistical comparison of public sector utilization to availability (i.e., “public sector disparity ratios”); 7. Econometric analyses of DBEs’ success, relative to non- DBEs’ (e.g., in business formation rates and in business owner earnings), and holding nondiscriminatory factors constant, in the market area surrounding the agency in question (i.e., “private sector disparity ratios”); 8. Econometric analysis of DBEs’ access to capital and credit, relative to non-DBEs’, holding balance sheet and creditworthiness information constant; 9. Qualitative evidence from DBEs and non-DBEs concern- ing experiences doing business or attempting to do busi- ness in the relevant marketplace, including experiences of institutionalized discrimination and/or individual dis- parate treatment, gathered through surveys, personal inter- views, and/or public hearings (i.e., “anecdotal evidence”); 12 37Croson, 488 U.S., at 479–80. 38 488 U.S. at 501. 39 For example, programs in New Jersey, Atlanta, Cleveland, Cook County, Columbus, the District of Columbia, Miami, and Philadelphia were enjoined. The State of Louisiana and the City of Milwaukee voluntarily dropped their race- conscious contracting programs. 40 49 CFR. § 26.45. 41 We have catalogued over 300 disparity studies performed between 1989 and 2008. Of these, 39 were performed directly for state DOTs or included state DOTs among the agencies studied and explicitly included federally assisted DOT contracts. Ten of these 39 have been superseded at their respective agencies by more recent studies. See Table 2.

10. Qualitative and/or quantitative analysis of the effective- ness of race-neutral measures to address low DBE partic- ipation in public contracting; and 11. Review of existing policies and procedures related to DBE participation, with recommendations for changes/ revisions designed to improve the effectiveness of the program and increase legal compliance. Additionally, the best studies are supervised by profession- als who can be qualified as expert witnesses under the exact- ing standards of the Federal Rules of Evidence to testify about a disparity or availability study’s data, methods, and find- ings.42 To the extent that the content of these studies largely involves economics and statistics, the courts will require expert witnesses to possess professional experience and qual- ifications, such as a doctoral degree, in these fields rather than in other tangential disciplines such as, for example, political science or anthropology. Closely related to the disparity study is the “availability study.” This has often caused confusion about the differences, if any, between the two. An availability study is a subset of a disparity study, focusing primarily on measuring the fraction of businesses in the relevant marketplace that are DBEs or potential DBEs (see Figure 1). Its main purpose is to assist federal grant recipients to meet the narrow tailoring require- ments of strict scrutiny, and it operates on the assumption that evidence of the recipient’s compelling interest in remedying discrimination is not required. Of the 11 elements that comprise a quality disparity study, a quality availability study focuses primarily on elements 2 through 4. Availability studies often include the business formation studies listed under number 7 above as part of the “Step 2” goal-setting procedure,43 and therefore provide compelling interest evidence in addition to their primary focus on narrow tailoring.44 Prior to Western States, some state DOTs chose to forgo the expense of conducting a full disparity study because recipients are not required to inde- pendently demonstrate compelling interest (since that has previously been established by Congress).45 Instead, some commissioned availability studies to assist their programs to become narrowly tailored by providing improved estimates of DBE availability from which to set annual DBE participa- tion goals, as required by 49 CFR 26. To date, this approach has been upheld by the court that has addressed the issue.46 In the remainder of this chapter, we identify best practices regarding if, when, why, and how state DOTs should conduct disparity or availability studies rather than employ one of the other four methods allowed for under 49 CFR 26.45.47 These results will contribute to improved state DOT decision mak- ing and more effective implementation of the DBE Program. When is a Study Required? It is neither possible nor prudent to provide definite or absolute answers to the question of when or if a state DOT should conduct a disparity or availability study since the law in this field is constantly evolving and each state DOT’s situ- ation is likely to be unique in several respects. With this caveat firmly in mind, we offer the following guidance. Part 26 does not require state DOTs to independently establish evidence of business discrimination within their contracting marketplace in order to satisfy the compelling interest prong of constitutional strict scrutiny. The Western States ruling, however, has created uncertainty about this principle for Ninth Circuit state DOTs.48 13 Figure 1. An availability study is a subset of a disparity study. 42 Fed. Reg. Ev. 702 (“If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or edu- cation, may testify thereto in the form of an opinion or otherwise, if (1) the tes- timony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.”) 4349 C.F.R § 26.45(d). 44This is because business formation studies use econometrics to answer the question of “how different would minority (or female) business formation rates be if they faced the same market structure as similarly situated non-minority males?” If minority or female business formation rates would be statistically sig- nificantly larger under the nonminority male market structure, this difference can be used to quantify a “Step 2” adjustment under 49 CFR 26.45. Most econ- omists would also accept it as evidence of discrimination. See, e.g. Oaxaca (1973), Fairlie and Meyer (2000), and Wainwright (2000). 45Adarand Constructors, Inc. v. Slater, 228 F.3d 1147 (10th Cir. 2000), cert. granted, 532 U.S. 941, then dismissed as improvidently granted, 534 U.S. 103 (2001) (Adarand VII); Sherbrooke Turf, Inc. v. Minnesota Department of Transportation, (8th Cir. 2003), cert. denied, 541 U.S. 1041 (2004). 46Sherbrooke Turf, Inc, 345 F.3d. at 973–74. 47I.e., (1) using DBE directories and Census data; (2) using bidders lists; (3) using the DBE goal of another U.S.DOT recipient; or (4) using other types of contrac- tor lists. 48See Chapter One, supra, and Appendix C, infra.

State DOTs in the Ninth Circuit must now gather evidence of discrimination (as part of a disparity study, an availability study, or through other means), if they have not already done so, and include it as part of their annual DBE goal submis- sion. All nine state DOTs in the Ninth Circuit have opted to collect such evidence by means of a disparity study. Compliance with the federal regulations and guidance, how- ever, is only one of several considerations. First, the state DOT’s legal counsel must consider the likelihood of litigation against its implementation of the DBE Program.49 Litigation risks can be influenced by many factors, including the following: • The level of consensus in the local contracting community, both DBE and non-DBE, concerning the need for and appropriateness of race-conscious contracting goals;50 • The quality of relationships between the state DOT and the local contracting community, both DBE and non-DBE; • The extent of the race-neutral aspects of the DBE Program and the level of resources devoted to such efforts; • How vigorously the race-conscious aspects of the DBE Pro- gram are implemented and enforced; • The reasonableness of the current overall DBE goals; • The reasonableness of the current contract-level DBE goals and the method by which they are determined; • The reasonableness of the agency in granting waivers of DBE goals when legitimate good-faith efforts to achieve the goals proved insufficient; and • Whether there is a race-conscious program for state- funded transportation contracts in addition to the federal DBE Program. The greater the perceived likelihood of a legal challenge to the DBE Program, the more prudent it becomes to consider commissioning a disparity or availability study. Although we do not read the law as requiring it, if a legal challenge does occur, being able to rely on a reasonably cur- rent and high-quality disparity or availability study, and access to credible expert witnesses to testify about the data, methods, and findings of the study, will be a great asset to state DOT defendants. In Sherbrooke and Northern Contract- ing, proffering these studies along with qualified experts to testify about these studies’ data, methods, and findings, con- tributed significantly to each agency’s successful defense. In the Western States case, in contrast, WSDOT’s DBE goals at the time were not based on an availability study and no expert testimony was presented. In general, race-conscious contracting programs supported by only an outdated study, or a study with inadequate evidence, or no study at all, have not fared well when challenged.51 There is one other circumstance in which a disparity study should be considered: where all or most of a state DOT’s annual DBE goal is projected to be met through race-neutral means. Although an all race-neutral program is unlikely to draw a constitutional challenge, it must be consistent with Congress’ findings that discrimination still results in a lack of equal opportunities for DBEs. Therefore, state DOTs that propose to wholly or mostly eliminate race-conscious con- tract goal setting should be confident that such an approach to ensuring a level playing field is warranted by quantitative and qualitative evidence of a lack of discrimination in its rel- evant markets, a consistent ability in the past to meet its DBE goals without race-conscious means, and a numerical goal that fairly reflects DBE availability. Disparity Study or Availability Study? State DOTs in the Ninth Circuit would be prudent to con- duct disparity studies rather than availability studies to make 14 49It seems probable that the U.S.DOT directive and an assessment of increased likelihood of a challenge played a role in Ninth Circuit state DOTs’ decision to conduct disparity studies. 50Building and achieving consensus is important. If racial and ethnic disparities in business enterprise emerge due to unequal business opportunities caused by discrimination, then the entire country pays a price resulting from decreased competition, underdeveloped human resources, and reduced social cohesion. Marshall (1991, 10), for example, has observed that, “. . . as long as minorities and women are denied access to business opportunities, the distribution of wealth, income, and power will continue to be unfair and even to polarize, with grave consequences for the economy, polity, and society.” Similarly, Fairlie and Meyer (1994, 1–2), note that Understanding the ethnic/racial character of self-employment is impor- tant for at least three reasons. First, conflicts between ethnic and racial groups in the U.S. have often been partly caused by business ownership patterns. . . . Second, self-employment has historically been a route of economic advancement for some ethnic groups. . . . Third, small busi- ness owners have an important effect on political decisions in the U.S. The under-representation of many ethnic/racial groups in business means that these groups may possess less political power than is sug- gested by their proportion of the population. Aronson (1991, 76–77) expressed similar concerns, noting, “[h]istorically, self- employment and small business ownership have been an important path by which ethnic, racial, and religious minorities in the United States have overcome social and cultural disabilities and entered the so-called mainstream.” 51See, e.g., Thompson Building Wrecking Company, Inc., v. Augusta, Georgia, 2007 U.S. Dist. LEXIS 27127 (S.D Ga. 2007) (13-year-old study not sufficient to estab- lish ongoing compelling interest, M/WBE program enjoined); Builders Associa- tion of Greater Chicago v. County of Cook, 256 F.3d 642 (7th Cir. 2001) (no dis- parity study or analysis); Associated Utility Contractors of Maryland, Inc. v. Mayor and City Council of Baltimore et al, 83 F.Supp.2d 613 (D. Md. 2000) (10-year-old evidence to justify goals is equivalent to no evidence); Association for Fairness in Business, Inc. v. New Jersey, 82 F.Supp.2d 353 (D. N.J. 2000) (no evidence lead- ing to particularized findings of discrimination); W. H. Scott Construction Com- pany, Inc. v. City of Jackson, Mississippi, 199 F.3d 206 (5th Cir. 1999) (study addressed only prime contracts; no evidence about the availability and utiliza- tion of subcontractors); Associated General Contractors of Ohio, Inc. v. Drabik, 50 F.Supp.2d 741 (S.D. Ohio 1999) (old evidence was insufficient); Associated Gen- eral Contractors of America v. City of Columbus, 936 F.Supp. 1363 (S.D. Ohio 1996) (studies rejected as invalid and incomplete); Associated General Contrac- tor of Connecticut v. City of New Haven, 791 F.Supp. 941 (D. Conn. 1992) (no sta- tistical evidence presented).

sure that they have examined the type of evidence of discrim- ination in their markets that establishes the need to use race- conscious measures to meet the annual DBE goal. While the court in Western States clearly held that the regulations are constitutional, and the regulations are explicit in not requir- ing a disparity study,52 the types of evidence listed by the Ninth Circuit would be included in a disparity study but not typically in an availability study.53 Outside the Ninth Circuit, an availability study may be suf- ficient to meet a state DOT’s constitutional obligations under strict scrutiny. There are risks to this approach, however, as only the Eighth Circuit has relied solely upon an availability study to uphold a recipient’s program. Therefore, if the state DOT believes there is a substantial likelihood of a challenge to the DBE Program, then serious consideration should be given to conducting a disparity study that provides quantita- tive and qualitative evidence of discrimination. An availabil- ity study, in contrast, generally does not present any qualita- tive or quantitative evidence regarding discrimination.54 If a constitutional challenge is not likely, but consensus regarding the DBE Program is still lacking in the local con- tracting community, a disparity study can still be a useful tool for building agreement about the need for a race-conscious remedy and the scope of the remedy. If consensus exists, then a periodic disparity or availability study can play a role in maintaining such consensus.55 If the state as a whole (not just the state DOT) has a race- conscious M/WBE Program for state-funded transportation contracts, then a disparity study is strongly recommended because such a program may not be able to rely upon Congres- sional findings of a compelling interest.56 In such cases, the mar- ginal cost of expanding the scope of such a study to also include a DBE availability or disparity study should be relatively mod- est. Therefore, it may be sensible for a state DOT to consider adding a DBE study onto the state-funds disparity study. Timing If a state DOT’s DBE Program that is not based on a study for its annual DBE goal setting is challenged, then it may still be possible in some cases to commission a study and employ the results at trial. The obvious advantage to this strategy is that, if no challenge occurs, it saves the resources that would otherwise be expended on a study. There are at least three dis- advantages to not collecting independent evidence prior to a legal challenge. First, the state DOT runs the risk that the trial court will either deny the motion to supplement the record or will not allow enough time to complete the work and prepare wit- nesses before trial. Such was the case in Kossman Contracting Co., Inc. v. Texas Department of Transportation.57 The chal- lenge to Texas’ Historically Underutilized Business Program for state-funded contracts was filed just before Christmas 1999, and the judge set the trial date for mid-February 2000— leaving insufficient time for the state to prepare its defense, let alone conduct a new study. As a result, TxDOT settled the case by dropping race-conscious subcontracting goals and paying the plaintiff’s legal fees and costs. Second, the costs of a study procured during litigation are likely to rise due to the time pressures involved. Nevertheless, this is preferable to going to trial with no study at all. Finally, failing to gather and evaluate evidence beyond DBE directories and bidders lists risks presenting an incom- plete picture of the barriers to full opportunities for DBEs. Adopting broad measures of availability, examining economy- wide data, and gathering anecdotal evidence gives a fuller pic- ture of the elements relevant to a narrowly tailored program as well as provides a supportable benchmark for determining the level of DBE participation in a marketplace that does not discriminate. This furthers the overall remedial purpose of the DBE Program. Quality of Disparity and Availability Studies Completing a disparity or availability study before a chal- lenge is only part of the equation. The study must also be of sufficient breadth, depth, and quality to withstand judicial scrutiny. Moreover, to the extent the study is used as a tool for consensus building, its scope and quality matter regard- less of the likelihood of a challenge. In the remaining sections of this report, we review disparity and availability studies performed for state DOTs, identifying included and excluded elements and discussing differences in how the elements have been designed and implemented. Next, we provide an overview of goal-setting methods currently in use at state DOTs along with an extended discussion of our rec- 15 52The “Department is not requiring recipients to do a disparity study.” 64 Fed. Reg. 5110. 53407 F.3d at 999–1001 (evidence of discrimination against each minority group and women in the state’s marketplace, detailed availability analysis, disparity testing, anecdotal evidence, etc.). 5449 C.F.R. §26.45 requires public input regardless of the method chosen to set DBE goals. 55The State of Maryland, for example, under the supervision of the Maryland Department of Transportation (MDOT) conducts a disparity study, by law, every 5 years. Each study is used to revise the state’s MBE Program, as well as to provide current data to support the DBE goal-setting activities at all of MDOT’s modal administrations. The study also collects feedback from DBEs and non- DBEs about the operation of the various programs. 56Therefore, a state or local government that is setting race-conscious goals on non-federally funded contracts should have a disparity study documenting that such goals are based on a compelling government interest and are narrowly tai- lored to the evidence relied upon. For contracts subject to 49 C.F.R. Part 26 or Part 23, outside the Ninth Circuit, an availability study may still suffice to demonstrate a narrowly tailored program. Since Western States, it seems clear that in the Ninth Circuit it would be prudent to supplement an availability study with evidence of discrimination. 57C.A. No. H-99-0637 (S.D. Tex. 2000).

ommended method for measuring DBE availability. Finally, we incorporate these elements into a Model Scope of Work to be used in future RFPs, along with other advice for improving the disparity/availability study RFP process. Appendix A, as well as other subsections in the report, provides detailed infor- mation on the type and quality of data necessary to support a high-quality disparity or availability study. Appendix D pro- vides a glossary of abbreviations and technical terms such as “endogenous,” and “multivariate regression,” and their mean- ing in disparity and availability studies. Appendix C is a full dis- cussion of the legal issues and judicial decisions. Review of Existing Studies State DOTs That Have Performed Disparity or Availability Studies For this report, we contacted each state DOT and the Dis- trict of Columbia and requested its annual DBE goal submis- sions for FFYs 2006, 2007, and 2008.58 We also asked whether each was currently performing, had recently performed, or had ever performed a disparity or availability study. If an agency had recently performed or was currently performing a study, we also asked for a copy of the RFP for the study. Of the 51 state DOTs we contacted, 26 have performed or are currently performing, a disparity or availability study. These 26 states account for 28 studies altogether since two states—Colorado (CDOT, ongoing) and North Carolina (NCDOT, ongoing)—are currently conducting studies to replace earlier ones. Of those 28 studies, 22 were commis- sioned directly by the state DOT while the remaining 6 included the DOT as part of a multi-agency statewide study.59 Five state DOTs have studies in progress, as of this writing— Hawaii (HDOT, ongoing) and Montana (MDT, ongoing), both in the Ninth Circuit; as well as Colorado (CDOT, ongo- ing), New York State (NY, ongoing), and North Carolina (NCDOT, ongoing). Sixteen states have completed studies within the last six years This includes the other seven Ninth Circuit state DOTs— Arizona DOT (ADOT, 2009), Caltrans (2007), Idaho Trans- portation Department (ITD, 2007), Nevada DOT (NDOT, 2007), ODOT (2007), WSDOT (2006), and Alaska Department of Transportation and Public Facilities (Alaska DOT&PF, 2008)—as well as Georgia DOT (GDOT, 2005), IDOT (2004), the State of Maryland (MD, 2006), Mn/DOT (2005), Mis- souri DOT (MoDOT, 2004), the State of New Jersey (NJ, 2005), NCDOT (2004), Tennessee DOT (TDOT, 2007), and Virginia DOT (VDOT, 2004). Seven states currently have studies that are more than 5 years old: CDOT (2001), Florida DOT (FDOT, 1999), the State of Louisiana (LA, 1990; LA, 1991);60 NDOR (2000); New Mexico DOT (NMDOT, 1995); the State of Ohio (OH, 2001a; OH, 2001b);61 and South Carolina DOT (SCDOT, 1995). Of the 14 state DOTs with current studies,62 only 863 used them, in whole or in part, as the basis for DBE goals for FFYs 2006, 2007, and/or 2008.64 CDOT used its previous study to set goals for FFY 2006 and used the City of Denver’s local study to set goals for FFYs 2007 and 2008,65 and it is currently conducting a new study (CDOT, ongoing). GDOT made explicit use of its disparity study in recommending DBE goals for FFYs 2006, 2007, and 2008 to the State Transportation Board. However, the board adopted lower goals, without explanation or modification of the evidentiary basis, in lieu of the recommended ones in all three years. ADOT’s study (ADOT, 2009) is being used as the basis for its FFY 2009 DBE goals. The remaining five DOTs66 used one of the other meth- ods from § 26.45(c) to set annual goals. Common Elements in State DOT Disparity and Availability Studies As a prelude to developing the model recommendations, we reviewed 25 of the 28 studies identified above.67 Nineteen were disparity studies; the remaining five were availability studies. Table 2 lists the 29 studies, the study type (disparity or availability), and the year they were completed. Table 3 shows which studies included which elements. 16 58Few DOTs post their annual DBE goals and supporting goal-setting methods on their departmental Web sites. Of those that did, most posted only the latest submission. In a few cases, despite providing a letter of introduction from TRB, we had to file open records/freedom of information requests to obtain the goal- setting information. In contrast, a handful of state DOTs had a great deal of infor- mation posted on their Web sites. North Dakota DOT’s [NDDOT] Web site, in particular, was excellent. See http://www.dot.nd.gov/divisions/civilrights/dbe program.htm. 59Multi-agency statewide studies in this category not only had to include the state DOT but also had to include federally assisted state DOT contracts. 60Volume 1 of the 1990 State of Louisiana Study was performed by two local uni- versity professors and completed in 1990. The state subsequently determined additional analysis was needed and commissioned another consultant, D.J. Miller & Associates, to perform Volume II, which was completed in 1991. 61Ohio’s study was conducted by two consultants. D. J. Miller & Associates con- ducted the statistical portion and Mason Tillman Associates, Ltd. conducted the historical and anecdotal portion. 62California, Georgia, Idaho, Illinois, Maryland, Minnesota, Missouri, Nevada, New Jersey, North Carolina, Oregon, Tennessee, Virginia, and Washington. 63California, Illinois, Maryland, Minnesota, Missouri, Nevada, North Carolina, and Washington. 64Alaska’s study was not completed in time to be used for FFY 2008. 65Colorado DOT used CDOT (2001) to help set goals for FFY 2006. For FFY 2007 and 2008, it used CCD (2006). 66New Jersey, Virginia, Tennessee, Oregon, and Idaho. 67We were unable to review the remaining three studies. For one that was ongo- ing, we were unable to obtain a copy of the winning proposal (NCDOT). Another study was never publicly released (TDOT), and a copy of one study could not be located (FDOT). For the other four ongoing studies (CDOT, Hawaii DOT [HDOT], Montana DOT [MDT], and New York State), we obtained copies of the winning proposals and based our review on the contents of those proposals.

17 Table 2. State DOT disparity and availability studies reviewed. State Consultant Type of Study Year Completed AK D. Wilson Consulting Group, LLC Disparity 2008 AZ MGT of America, Inc. Disparity 2009 CA BBC Research & Consulting Disparity 2007 CO MGT of America, Inc. Disparity 2001 CO D. Wilson Consulting Group, LLC Disparity ongoing FL MGT of America, Inc. Disparity 1999 GA Boston Research Group Disparity 2005 HI NERA Economic Consulting Disparity ongoing ID BBC Research & Consulting Disparity 2007 IL NERA Economic Consulting Availability 2004 LA D.J. Miller & Associates & Lunn/Perry Disparity 1991 MD NERA Economic Consulting Disparity 2006 MN NERA Economic Consulting Availability 2005 MO NERA Economic Consulting Availability 2004 MT D. Wilson Consulting Group, LLC Disparity ongoing NC EuQuant68 Disparity ongoing NC MGT of America, Inc. Disparity 2004 NE MGT of America, Inc. Availability 2000 NJ Mason Tillman Associates, Ltd. Disparity 2005 NM BBC Research & Consulting Disparity 1995 NV BBC Research & Consulting Disparity 2007 NY NERA Economic Consulting Disparity ongoing OH D.J. Miller & Associates & Mason Tillman Associates, Ltd. Disparity 2001 OR MGT of America, Inc. Disparity 2007 SC MGT of America, Inc. Disparity 1995 TN Mason Tillman Associates, Ltd. Disparity 2007 VA MGT of America, Inc. Disparity 2004 WA NERA Economic Consulting Availability 2006 68EuQuant was formerly Boston Research Group.

The remainder of this section discusses important study elements and whether these were included in the studies we examined. For ease of reference, Table 3 is presented in three parts, 3a, 3b, and 3c: study elements (a) through (d) are ref- erenced in Table 3a, study elements (e) through (i) are refer- enced in Table 3b, and study elements (j) through (m) are referenced in Table 3c. The next section introduces our proposed Model Dispar- ity Study by examining differences in how certain elements were implemented in these 25 studies and making recom- mendations as to the most constitutionally sound approach to each study element. Executive Summary and Introduction. None of the five availability studies (0%) included an executive summary. All (100%) included an introduction. Seventeen of 18 disparity studies (94%) included an executive summary. Seventeen of 18 disparity studies (94%) included an introduction.69 Every study, though, included at least one or the other. An introductory chapter describing, at a minimum, the objectives of the study and the structure of the study report ori- ents the reader to the material to follow and is a useful contri- bution to any study. An executive summary that briefly describes the data, methods, key findings, and conclusions also provides a handy digest of what is typically a lengthy and fairly complex collection of quantitative and qualitative analyses. The longer the study, the more important an executive summary. Although an executive summary provides a handy digest, it is only that—a digest. Should a DBE Program that is sup- ported by a disparity or availability study ever be challenged, it will be helpful to be able to establish that key decision mak- ers in the organization read and carefully considered the entire study—not just the executive summary. Legal Review. Four of the five availability studies (80%) included a legal review. Fifteen of 20 disparity studies (75%) included a legal review. Since the case law in the field of race-conscious contracting is constantly evolving, it is helpful to include a section in the study that reviews and evaluates relevant case law and shows how the study methods comport with the consultant’s understanding of the law. This section of the study also pro- vides helpful information for state DOT attorneys if a pro- gram is challenged. This section allows the state DOT’s attor- neys to determine whether the consultant’s grasp of strict scrutiny and the relationship of its methods to the law is sat- isfactory, and guides the court about the legal principles as applied by the consultant. Historical Review. None of the five availability studies (0%) included any review of the history of the impact of dis- crimination on business enterprise. Only two of 20 disparity studies (10%) included such an historical review, and they were also among the oldest studies we examined (LA, 1991; SCDOT, 1995).70 Many disparity studies in the immediate post-Croson era included historical reviews, including of the civil rights move- ment, and the emergence of black-owned businesses. This approach soon fell out of favor. Nothing in Croson, however, indicates that specific historical evidence of discrimination does not have a place as part of a larger mosaic of quantita- tive and qualitative evidence produced in a disparity study, only that such studies cannot rest on such evidence in isola- tion. On the contrary, high-quality historical analyses can provide important context, for both policy makers and jurists, for understanding how current disparities arose and the link- ages between different types of past discrimination and the current disadvantaged status of many minority- and women- owned business enterprises. Such information may be solicited as part of the anecdotal interviews with local business own- ers, however, rather than as a stand alone section in a study. For example, in finding that the City of Chicago had a strong basis in evidence to continue its M/WBE program for con- struction contracts, the court reviewed the history of the “long, slow, painful and continuing process” of eradicating the continuing effects of slavery and discrimination.71 Review of Contracting Policies and Procedures. None of the five availability studies (0%) included a review of con- tracting policies and procedures. Fourteen of 20 disparity studies (70%) included such a review. A race- and gender-based contracting program must be “narrowly tailored” to any evidence of discrimination in the agency’s markets. For state DOTs, this is the crucial task for goal setting under Part 26:72 recipients must utilize race- neutral means to the maximum feasible extent to meet the overall goal73 and project the portion of the goal that it expects to meet through race-neutral measures and race-conscious subcontracting goals.74 The requirement that recipients utilize 18 69For CDOT (ongoing) and MDT (ongoing), only the proposals were available, and these did not specify whether introductory chapters or executive summaries would be provided. For FDOT (1999), neither the study nor a copy of the pro- posal could be located. For NCDOT (ongoing), the proposal was not available, and for TDOT (2007) the study was never released. 70The State of Ohio study (OH, 2001b) also included an historical section. How- ever, it was only two pages in length and did not qualify to be counted in this cat- egory. 71Builders Association of Greater Chicago v. City of Chicago, 298 F. Supp.2d 725, 727 (N.D. Ill. 2003). 7264 Fed. Reg. 5109 (Feb. 2, 1999) (section 26.45 “is critical to meeting our con- stitutional obligation to ensure that the program is narrowly tailored to remedy the effects of discrimination.”). 7349 C.F.R. § 26.51(a). 7449 C.F.R. § 26.51(c).

19 Table 3a. Elements included in state DOT disparity and availability studies. Study Type Intro-duction Execu- tive Sum- mary Legal Review and Anal- ysis Histor- ical Analy- sis Review of Con- tracting Policies and Proce- dures WSDOT (2006) A x x Mn/DOT (2005) A x x MoDOT (2004) A x x IDOT (2004) A x x NDOR (2000) A x HDOT (ongoing) D x x x x NY (ongoing) D x x x x MD (2006) D x x x x ADOT (2009) D x x x x ODOT (2007) D x x x x NCDOT (2004) D x x x VDOT (2004) D x x x x CDOT (2001) D x x FDOT (1999) D n/a n/a n/a n/a n/a SCDOT (1995) D x x x x x Caltrans (2007) D x x ITD (2007) D x x NDOT (2007) D x x x NMDOT (1995) D x x x x Notes: (1) x indicates the relevant study element was present, an empty cell indicates that the element was not present; (2) in the type column, A indicates an availability study, D indicates a disparity study; (3) n/a means the necessary information to make a determination was not available, usually because the study was ongoing and the study proposal was not available or did not address a particular element, or because the study was never released or could not be located. OH (2001) D x x LA (1991) D x x x x NJ (2005) D x x x TDOT (2007) D n/a n/a n/a n/a n/a Alaska DOT&PF (2008) D x x x x CDOT (ongoing) D n/a n/a x x MDT (ongoing) D n/a n/a x x GDOT (2005) D x x x NCDOT (ongoing) D n/a n/a n/a n/a n/a

20 Table 3b. Elements included in state DOT disparity and availability studies. Notes: (1) x indicates the relevant study element was present, an empty cell indicates that the element was not present; (2) in the type column, A indicates an availability study, D indicates a disparity study; (3) n/a means the necessary information to make a determination was not available, usually because the study was ongoing and the study proposal was not available or did not address a particular element, or because the study was never released or could not be located. Study Type Geo- graphic Market Defi- nition Product Market Defi- nition Avail- ability Anal- ysis Public Sector Utiliza- tion Public Sector Dispar- ity WSDOT (2006) A x x x Mn/DOT (2005) A x x x MoDOT (2004) A x x x IDOT (2004) A x x x NDOR (2000) A x x x x HDOT (ongoing) D x x x x x NY (ongoing) D x x x x x MD (2006) D x x x x x ADOT (2009) D x x x x ODOT (2007) D x x x x NCDOT (2004) D x x x x VDOT (2004) D x x x x CDOT (2001) D x x x x x FDOT (1999) D n/a n/a n/a n/a n/a SCDOT (1995) D x x x x Caltrans (2007) D x x x x x ITD (2007) D x x x x x NDOT (2007) D x x x x x NMDOT (1995) D x x x x OH (2001) D x x x x LA (1991) D x x x NJ (2005) D x x x x TDOT (2007) D n/a n/a n/a n/a n/a Alaska DOT&PF (2008) D x x x x CDOT (ongoing) D x n/a x x x MDT (ongoing) D x n/a x x x GDOT (2005) D x x x x x NCDOT (ongoing) D n/a n/a n/a n/a n/a

21 Table 3c. Elements included in state DOT disparity and availability studies. Study Type Private Sector Dispar- ities Credit Access Anec- dotal Evi- dence Recom- menda- tions WSDOT (2006) A x x x Mn/DOT (2005) A x x MoDOT (2004) A x x IDOT (2004) A x x NDOR (2000) A x HDOT (ongoing) D x x x x NY (ongoing) D x x x x MD (2006) D x x x x ADOT (2009) D x x x x ODOT (2007) D x x x x NCDOT (2004) D x x x VDOT (2004) D x x CDOT (2001) D x FDOT (1999) D n/a n/a n/a n/a SCDOT (1995) D x x x Caltrans (2007) D x x x x ITD (2007) D x x x x NDOT (2007) D x x x x NMDOT (1995) D x x x OH (2001) D x x x LA (1991) D x x x NJ (2005) D x x TDOT (2007) D n/a n/a n/a n/a Alaska DOT&PF (2008) D x x CDOT (ongoing) D x x x MDT (ongoing) D x x x GDOT (2005) D x x NCDOT (ongoing) D n/a n/a n/a n/a Notes: (1) x indicates the relevant study element was present, an empty cell indicates that the element was not present; (2) in the type column, A indicates an availability study, D indicates a disparity study; (3) n/a means the necessary information to make a determination was not available, usually because the study was ongoing and the study proposal was not available or did not address a particular element, or because the study was never released or could not be located.

mined based on contract and subcontract expenditure data.78 Only eight of 18 disparity studies (44%) included a product market analysis.79 Narrow tailoring also applies to product markets. The extent of discrimination may differ from industry to industry just as among geographic locations.80 Documenting the spe- cific industries that comprise a state DOT’s contracting activ- ities and the relative importance of each to contract and subcontract spending is an important study element. A care- ful product market definition allows for (1) implementation of more narrowly tailored availability estimation methods, (2) contract-level goal setting, and (3) overall DBE availabil- ity estimates and annual goals that are a weighted average of underlying industry-level availability estimates, rather than a simple average. The weights used are the proportion of dol- lars spent with each industry and allow the overall availabil- ity measure to be influenced more heavily by availability in those industries where more contracting dollars are spent, and less heavily by availability in those industries where rela- tively few contracting dollars are spent. Estimation of DBE Availability. Estimating DBE avail- ability is at the core of any disparity or availability study. Mea- sures of DBE availability are needed in order to set narrowly tailored goals and to make comparisons to utilization in order to gauge disparities. All five availability studies (100%) and all 20 disparity studies (100%) included a section describing how availability was estimated. State DOT DBE Utilization. Only one of the five avail- ability studies (20%) included a section on DBE utilization by the state DOT itself. However, all 20 disparity studies (100%) included a utilization section. Like availability, measuring the utilization of DBEs as prime contractors and subcontractors on state DOT con- tracts is a key part of any disparity study. It is not as often seen as a stand-alone section in availability studies since the focus is on assisting state DOTs with narrowly tailored goal setting rather than making compelling interest determinations of disparities (derived in part by comparing DBE utilization to DBE availability). race-neutral measures to the maximum feasible extent has been critical to the unanimous judicial conclusion that the regulations are facially narrowly tailored.75 It was also useful in defending IDOT’s DBE Program, where expert testimony was admitted to establish that the DBE Program complied with Part 26 and best practices, as well as the process for adopting the annual goal. A review of current contracting policies and procedures should therefore document whether, and to what extent, the state DOT’s implementation of the regulations meets the constitutional objective. Conducting this type of review also familiarizes the con- sultant with the agency’s policies and procedures, which allows a more effective collection and analysis of anecdotal and other qualitative information regarding the interaction of both DBEs and non-DBE with the agency’s policies and procedures. It further permits recommendations for program improvements that support the success of the state DOT’s program in leveling its playing field. Determination of Relevant Geographic Market Area. All five availability studies (100%) included a section describing how geographic markets were empirically determined based on contract and subcontract expenditure data. Nineteen of 20 disparity studies (95%) included a geographic market analysis. The importance of establishing the geographic market area stems from Croson: Finally, the city and the District Court relied on Congress’ find- ing in connection with the set-aside approved in Fullilove76 that there had been nationwide discrimination in the construction industry. The probative value of these findings for demonstrat- ing the existence of discrimination in Richmond is extremely limited. By its inclusion of a waiver procedure in the national program addressed in Fullilove, Congress explicitly recognized that the scope of the problem would vary from market area to market area.77 While the DBE Program was established by Congress and based on findings of nationwide discrimination in the con- struction industry, establishing the state DOT’s geographic market area is nevertheless an important component of nar- rowly tailoring its DBE Program. The DBE annual goal should reflect DBE availability within the specific geographic market from which that state DOT draws the vast bulk (75% or more) of its contractors and subcontractors. Determination of Relevant Product Market Area. Four of the five availability studies (80%) included a section describing how product markets were empirically deter- 22 75See, e.g., Sherbrooke Turf, Inc. v. Minnesota Department of Transportation, 345 F.3d. 964, 973 (8th Cir. 2003), cert. denied, 541 U.S. 1041 (2004). 76Fullilove v. Klutznick, 448 US 448 (1980). 77Croson, at 504 (citations omitted, emphasis added). 78The NDOR (2000) study included some statistical analysis of how the depart- ment’s contract dollars were distributed according to internal work codes (grad- ing, concrete pavement, landscaping, bridges, etc.); however, this information was not used to weight the overall DBE availability figure according to the dif- ferent levels of spending within each work code. 79CDOT (ongoing) and MDT (ongoing) were excluded from this figure since we could not determine from the proposals whether a product market analysis would be included in the study. 80See, however, Wainwright (2000), documenting that, in general, the similari- ties in the amount of discrimination present in different industries and geo- graphic locations significantly outweighs the differences.

State DOT Disparity Analysis. Only one of the five availability studies (20%) included a section on disparities in the state DOT’s own utilization. However, all 20 disparity studies (100%) included such a section. Assessing disparities between DBE participation on DOT contracts and DBE availability in the DOT’s relevant markets is a central element of any disparity study. To determine whether DBEs have or have not been used by the state DOT in proportion to their availability in the marketplace, the con- sultant should ideally examine contracting expenditures that were not subject to race-conscious goals. However, since a state DOT’s federally assisted contracting expenditures are subject to the DBE Program, its data may not show evidence of a lack of participation even if it exists in the private sector of the state’s relevant market area. Instead, the state DOT’s own data are typically most useful for examining the effec- tiveness of its DBE policies during the study period. On the other hand, of course, if actual participation in state DOT contracts still turns out to be significantly less than availabil- ity in any industry or procurement category, then the state DOT’s data will still provide evidence of adverse impact. Economy-Wide Disparity Analyses. Four of the five availability studies (80%) included a section on economy-wide disparity analyses. Fifteen of 20 disparity studies (75%) included such a section. Assessing the presence or absence of disparities in contract- ing and other business activities that are not already subject to race-conscious goals is at the core of the compelling interest inquiry. As the Tenth Circuit noted in Adarand VII: [T]he evidence presented by the government in the present case demonstrates the existence of two kinds of discriminatory barri- ers to minority . . . enterprises, both of which show a strong link between racial disparities in the federal government’s disburse- ments of public funds for construction contracts and the chan- neling of those funds due to private discrimination. The first dis- criminatory barriers are to the formation of qualified minority . . . enterprises due to private discrimination, precluding from the out- set competition for public construction contracts by minority enter- prises. The second discriminatory barriers are to fair competition between minority and non-minority . . . enterprises, again due to private discrimination, precluding existing minority firms from effectively competing for public construction contracts. The govern- ment also presents further evidence in the form of . . . studies of local subcontracting markets after the removal of affirmative action programs.81 In disparity studies, econometric studies of minority and female business formation rates address the first type of bar- rier, and econometric studies of minority and female business owner earnings address the second type of barrier. These analyses focus on the private sector as opposed to the public sector because race-conscious contracting in the private sec- tor is far less common and there are a significant number of construction firms that work in both sectors. The business formation disparity analysis can also be impor- tant for assessing “Step 2” adjustments under §26.45(d)(3), as a way to quantify how much higher “Step 1” DBE availability would be “but-for” the “present effects of past discrimination.” This type of disparity analysis can additionally be con- ducted by comparing DBE participation on public sector contracts with goals to participation on contracts without goals. There are several possible sources of data on contracts without goals. Within the public sector, they can come from non-federally assisted state DOT contracts without M/WBE requirements (such as, for example, in California or Wash- ington State);82 representative state DOT “control” or “zero goals” contracts, such as IDOT presented at the Northern Contracting trial;83 and from comparing different time peri- ods within the same geographic and product markets when goals were and were not in force. Moreover, comparisons can also be made between the state DOT and other public agen- cies in the relevant market without race-conscious pro- grams.84 In the private sector, F.W. Dodge Reports or Reed Construction Data can be used, in some circumstances,85 to compare M/WBE participation on private contracts in a given geographic market to M/WBE participation on public contracts. Some studies also used building permit data for this purpose. This type of economy-wide evidence is relevant not only to the presence or absence of discrimination in the market- place but also to whether any discrimination would be suf- ficiently ameliorated through solely race-neutral methods. Surprisingly, some disparity studies omit these critical economy-wide analyses, making it impossible to determine whether an agency is a passive participant in a discrimina- tory marketplace. 23 81Adarand Constructors, Inc. v. Slater (Adarand VII), 228 F.3d 1147, 1167–68 (10th Cir. 2000), cert. granted, 532 U.S. 941, then dismissed as improvidently granted, 534 U.S. 103 (2001) (emphasis added). 82Both states have passed citizen-initiated propositions banning the use of race- conscious affirmative action in public contracting. Cal. Const. art. 1, sec. 31(a); RCW 49.60.400. 83See fn. 136 and the accompanying discussion. 84Comparisons between DBE participation at IDOT and the Illinois State Toll Highway Authority were mentioned previously. In a recent study performed for the Commonwealth of Massachusetts, similar comparisons were made between two agencies operating race-conscious programs, the Massachusetts Housing Finance Agency and the Division of Capital Asset Management, and two agen- cies whose programs had been either legally enjoined (Massachusetts Bay Trans- portation Authority) or voluntarily suspended pursuant to a disparity study (City of Boston) (MA, 2006; MA, 2008). 85The completeness of these data sources varies significantly by region. Some parts of the country have very complete data (e.g., NCDOT, 2004, Ch. 7), others much less so (e.g., ODOT, 2007, Ch. 7).

Credit Market Access. None of the five availability stud- ies (0%) included a section on lending discrimination. Nine of 20 disparity studies (45%) included this section. Discrimination in access to capital is one of the most com- mon and long-standing problems voiced by minority and women entrepreneurs. As long ago as 1944, Gunnar Myrdal’s seminal study of race in America found: The Negro businessman . . . encounters greater difficulties than whites in securing credit. This is partly due to the marginal posi- tion of Negro business. It is also partly due to prejudicial opinions among whites concerning business ability and personal reliability of Negroes. In either case a vicious circle is in operation keeping Negro business down.86 Evidence of this type of discrimination has proven repeat- edly persuasive to courts in upholding race-conscious con- tracting programs. In determining that the DBE Program met strict scrutiny, the Tenth Circuit wrote: The government’s evidence is particularly striking in the area of the race-based denial of access to capital, without which the for- mation of minority subcontracting enterprises is stymied. . . . [One study,87 for example] . . . surveyed 407 business owners in the Denver area. It found that African Americans were 3 times more likely to be rejected for business loans than whites. The denial rate for Hispanic owners was 1.5 times as high as white owners. Disparities in the denial rate remained significant even after controlling for other factors that may affect the lending rate, such as the size and net worth of the business. The study con- cluded that “despite the fact that loan applicants of three differ- ent racial/ethnic backgrounds in this sample (Black, Hispanic and Anglo) were not appreciably different as businesspeople, they were ultimately treated differently by the lenders on the crucial issue of loan approval or denial.” Lending discrimination alone of course does not justify action in the construction market. . . . However, the persis- tence of such discrimination, which is already unlawful under federal law, supports the assertion that the formation, as well as utilization, of minority-owned construction enterprises has been impeded.88 The first formal econometric analysis of credit market dis- crimination was introduced during the Concrete Works v. Denver trial. Using data compiled by the Federal Reserve Board and the U.S. Small Business Administration, Denver’s expert economists demonstrated that large and statistically significant differences in commercial loan denial rates between minority and nonminority firms were evident even when balance sheet and creditworthiness measures were held constant.89 Denver also introduced the Colorado Center for Community Development (CCD, 1996) study, as well as tes- timony from the former top state banking official in Col- orado who examined the financial records of several DBEs who had been recently denied credit and concluded that the denials could not be justified by their finances. Additionally, there was testimony from MBEs and WBEs concerning their difficulties in obtaining capital for their businesses and require- ments that were placed on them by lenders that were not placed on majority firms. Upon review, the Tenth Circuit noted: The City presented evidence of lending discrimination to sup- port its position that M/WBEs in the Denver MSA90 construc- tion industry face discriminatory barriers to business forma- tion. . . . [Plaintiffs] did not present any evidence that undermines the reliability of the lending discrimination evi- dence but simply repeats the argument, foreclosed by circuit precedent, that it is irrelevant. . . . [However,] in Adarand VII we took “judicial notice of the obvious causal connection between access to capital and ability to implement public works construction projects.” Outside of litigation, analysis of lending discrimination problems has been relatively rare in disparity studies. Statis- tical tabulations comparing M/WBE and non-M/WBE answers to a number of survey questions relating to credit and bonding were included in Louisiana (LA, 1991). Formal econometric analysis of race and gender differences in lend- ing discrimination did not appear, however, until Maryland (MD, 2001). Recently, Caltrans (2007), ITD (2007), NDOT (2007), ODOT (2007), and ADOT (2009) have included sim- ilar analyses.91 Anecdotal Analyses. One of the five availability studies (20%) included anecdotal and/or other types of qualitative evidence. Nineteen of 20 disparity studies (95%) included anecdotal evidence. Qualitative or anecdotal evidence of the direct experiences of minority and female business owners with discrimination is a crucial complement to the statistical evidence in any dis- 24 86Myrdal (1944, 308). 87Colorado Center for Community Development (1996). See also, U.S. Depart- ment of Justice (1996, at 26057–58). 88Adarand VII at 1169–70, fn 13, citations omitted. 89In the interest of full disclosure, Denver’s trial experts were NERA economists and Colette Holt served as outside counsel. 90“Metropolitan Statistical Area.” See also Appendix D—Glossary. 91The two proposals from D. Wilson Consulting, LLC (CDOT, ongoing; MDT, ongoing) mention the main data set that has been used to produce econometric studies of lending discrimination (the Survey of Small Business Finances) but give no indication how it will be used: “The Wilson Group will use data from the National Survey of Small Business Finances (NSSBF) conducted by the Federal Reserve Board and the U.S. Small Business Administration, the Current Popu- lation Survey (CPS) and the Five Percent Public Use Microdata Samples (PUMS) from the 2000 decennial census. These data are used to examine the incidence of minority and female business ownership and [their] earnings. . . .” However, the NSSBF cannot be used for analyses of M/WBE ownership or earnings and the CPS and PUMS cannot be used to analyze disparities in credit access.

parity study. As we have already noted in Chapter One, anec- dotal evidence of experiences with discrimination in con- tracting opportunities is highly relevant since it addresses whether observed statistical disparities are due to discrimi- nation and not to some other nondiscriminatory cause or causes. As stated in Croson, “. . . [e]vidence of a pattern of individual discriminatory acts can, if supported by appropri- ate statistical proof, lend support to a local government’s determination that broader remedial relief is justified.”92 Tes- timony about personal experiences can bring “cold numbers convincingly to life.” As the Supreme Court has put it, in the context of an employment discrimination case: The company’s principal response to this evidence is that statis- tics can never in and of themselves prove the existence of a pat- tern or practice of discrimination, or even establish a prima facie case shifting to the employer the burden of rebutting the infer- ence raised by the figures. But, as even our brief summary of the evidence shows, this was not a case in which the Government relied on “statistics alone.” The individuals who testified about their personal experiences with the company brought the cold numbers convincingly to life. . . . In any event, our cases make it unmistakably clear that “[s]tatistical analyses have served and will continue to serve an important role” in cases in which the existence of discrimination is a disputed issue. We have repeat- edly approved the use of statistical proof, where it reached pro- portions comparable to those in this case, to establish a prima facie case of racial discrimination in jury selection case. Statistics are equally competent in proving employment discrimination. We caution only that statistics are not irrefutable; they come in infinite variety and, like any other kind of evidence, they may be rebutted. In short, their usefulness depends on all of the sur- rounding facts and circumstances.93 Anecdotal evidence is not often included in availability stud- ies, since they are addressed primarily toward narrow tailoring, with the understanding that the existence of discrimination has already been established. The Ninth Circuit’s Western States decision, as previously noted, has confused and conflated the distinction between the two prongs of strict scrutiny. Reflect- ing this, WSDOT (2006) supplemented its availability study with anecdotal evidence to support resuming the use of DBE subcontracting goals to meet its annual goal. Findings and Recommendations. All availability studies and disparity studies included a final chapter containing find- ings and recommendations. Recommendations allow the consultant to opine whether the evidence developed in its study could or could not sup- port the adoption or continuation of race-conscious mea- sures. Some consultants further advise the client on what DBE goals to adopt; other researchers, including ourselves, do not because it usurps the agency’s role as policy maker. Translat- ing availability measures into DBE goals is the responsibility of agency policy makers, not the consultant, because it reflects the application of judgment to the findings of the study in relationship to overall agency objectives. Current State DOT Goal-Setting Methods Of the 51 state DOTs examined, only 10 set DBE goals dur- ing any portion of FFYs 2006–2008 using disparity or avail- ability studies. Table 4 and the accompanying Figures 2–4 show the goal-setting method used by each DOT in each of the three fiscal years studied. Clearly, the most commonly employed method for establish- ing annual DBE goals is the use of a bidders list.94 Just over half the state DOTs opted for this approach. For FFY 2006, 27 state DOTs used bidders lists, and two employed a combination of a bidders list approach and a disparity study. For FFY 2007, 27 state DOTs employed the bidders list approach, one employed a combination of a bidders list approach and a dis- parity study, and one used a combination of a bidders list approach and two other approaches.95 For FFY 2008, 25 state DOTs employed a bidders list approach, one used a combina- tion of a bidders list approach and a disparity study, and one used a combination of a bidders list approach and two other approaches. After the bidders list approach, the next most frequently used is an “alternate method.”96 In all cases, the alternate method has been essentially the bidders list approach—but with a different list. In the vast majority of these cases, the alternate list chosen for use was a list of prequalified contrac- tors. State DOTs using prequalified contractor lists to set DBE goals were Indiana, Kentucky, Michigan, Ohio, Pennsylvania, Tennessee, Virginia, and West Virginia.97 By comparison to the use of bidders lists and alternates such as prequalified contractor lists, the two remaining goal- setting options were employed with much less frequency. Other than Maine DOT in FFYs 2007 and 2008, no state DOT has used the goals of other DOT recipients to set its own goals, and only five states have employed the DBE directory and Census Bureau data approach. 25 92Croson, at 509. 93International Brotherhood of Teamsters v. United States, et al., 431 U.S. 324 (1977), at 339; see also Concrete Works II, 36 F.3d at 1521 (“Croson impliedly endorsed the inclusion of personal accounts of discrimination”). 9449 C.F.R. § 26.45(c)(2). 95Maine DOT combined of a bidders list approach and two other approaches. For 2007 and 2008, Maine calculated Step 1 availability using three different approaches: bidders list; DBE directory and Census Bureau data; and the goals of other DOT recipients. It then used the average of all three to arrive at the Step 1 figure. 9649 C.F.R. § 26.45(c)(5). 97Other types of lists employed included lists of licensed contractors (Nevada), plan holders (Iowa), and contractors with approved EEO policies (Kansas).

26 Table 4. Goal-setting methods employed by state DOTs, FFYs 2006–2008. State FHWA DBE Overall Goal 2006 Method FHWA DBE Overall Goal 2007 Method FHWA DBE Overall Goal 2008 Method California 1 1 3 Colorado 2 & 3 3 3 Georgia 3 3 3 Illinois 3 3 3 Maryland 1 & 3 3 1 & 3 Minnesota 3 2 3 Missouri 3 3 3 Nevada 5 5 3 North Carolina 2 & 3 2 & 3 2 & 3 Washington 3 3 3 Hawaii 1 1 1 New Jersey 1 1 1 New York 1 1 1 Alabama 2 2 98 Alaska 2 99 100 Arizona 2 2 2 Arkansas 2 2 2 Connecticut 2 2 2 D. C. 2 2 2 Delaware 2 2 2 Florida 2 2 2 Idaho 2 2 2 Louisiana 2 2 2 Massachusetts 2 2 2 Mississippi 2 2 2 Montana 101 2 2 98Alabama DOT (ALDOT) was awaiting approval of its 2008 goal documents by FHWA. 99Alaska DOT&PF indicated that data problems prevented it from setting a DBE goal in FFY 2007 or FFY 2008 using any of the methods available in § 26.45(c). Instead, Alaska set a 4% all race-neutral goal in each year. 100See fn 99. 101According to our correspondence with Montana DOT (MDT), it did not sub- mit an annual DBE goal to FHWA for FFY 2006.

27 Table 4. (Continued). Note: Method 1 corresponds to 49 C. F. R. 26.45(c)(1) (used DBE directories and Census Bureau data); method 2 or 5 corresponds to 49 C. F. R. 26.45(c)(2) or (c)(5) (used bidders list or other types of contractor lists); method 3 corresponds to 49 C. F. R. 26.45(c)(3) (used disparity or availability study) method 4 corresponds to 49 C. F. R. 26.45(c)(4) (used the goal of another DOT recipient, 26.45(c)(5) “Use an alternative method.” 2 2 North Dakota 2 2 Oklahoma 2 2 2 Oregon 2 2 2 Rhode Island 2 2 2 South Carolina 2 2 2 South Dakota 2 2 2 Texas 2 2 2 Utah 2 2 2 Vermont 2 2 2 Wisconsin 2 2 2 Wyoming 2 2 2 Indiana 5 5 5 Iowa 5 5 5 Kansas 5 5 5 Kentucky 5 5 5 Michigan 5 5 5 Ohio 5 5 5 Pennsylvania 5 5 5 Tennessee 5 5 5 Virginia 5 5 5 West Virginia 5 5 5 Maine 2 1, 2 & 4 1, 2 & 4 State FHWA DBE Overall Goal 2006 Method FHWA DBE Overall Goal 2007 Method FHWA DBE Overall Goal 2008 Method Nebraska 2 2 New Hampshire 2 2 2 New Mexico 2 2 2

Figure 2. Goal-setting methods used by state DOTs, FFY 2006. Figure 3. Goal-setting methods used by state DOTs, FFY 2007. Figure 4. Goal-setting methods used by state DOTs, FFY 2008. 8% 76% 10% 6% DBE Directories & Census Data Bidders List or Similar Disparity or Availability Study Other Methods Source: Table 15 8% 76% 12% 4% DBE Directories & Census Data Bidders List or Similar Disparity or Availability Study Other Methods Source: Table 15 6.1% 71% 16% 6.1% DBE Directories & Census Data Bidders List or Similar Disparity or Availability Study Other Methods Source: Table 15

Next: Chapter 3 - Model Disparity Study »
Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Report 644: Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program explores guidelines for state departments of transportation (DOTs) on how to conduct effective and legally defensible disparity and availability studies to meet the requirements of the Disadvantaged Business Enterprise (DBE) program for federally funded projects. The report includes guidance designed to assist DOTs in determining when and if a disparity or availability study is recommended, a model scope of work that may be used in a request for proposals, and detailed recommendations on how to design and implement disparity and availability studies.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!