Analysis of public contracting data indicates that substantial disparities exist between minority and non-minority business enterprises. Specifically, the data show that minority business enterprises (MBEs) typically secure a lower number and dollar amount of contracts in proportion to the number of available MBEs in a relevant market. As a result, MBEs, agency officials, policy makers, and advocates have a strong incentive to understand the factors that give rise to observed contracting disparities.
In order to advance the dialogue concerning contracting disparities and inform the development of new and innovative solutions, the Minority Business Development Agency (MBDA) requested a comprehensive review of existing data and studies to address several key research questions:
What factors create barriers and cause disparities in public contracting for MBEs?
What information do existing studies provide stakeholders in assisting agencies address observed disparities?
What areas warrant further investigation and policy research with respect to contracting disparities experienced by MBEs?
This research explored existing disparity studies conducted by a variety of economic consultants that were commissioned by local and state governments nationwide. A disparity study is a comprehensive effort that analyzes a wealth of data pertaining to the legal, legislative, and contracting environment facing MBEs in a particular jurisdiction or when procuring contracts from a specific federal, state or municipal agency. The findings presented in this report are drawn from a comprehensive review of 100 disparity studies, summaries, and reports that are publicly-available and accessible via the internet. The selected set of disparity studies does not represent the full universe of studies and includes a greater focus on recent studies with information on contracting disparities affecting MBEs within the last ten years.
LEGAL PRECEDENT AND DISPARITY STUDY BASICS
The evolution and development of disparity studies arose from legal challenges to existing affirmative action or race-conscious programs1 enacted by government rules, legislation or policies intended to alleviate perceived or actual discrimination against different racial, ethnic or gender groups in public contracting. In response to the legal precedent,2 government agencies have commissioned disparity studies to examine the extent to which minority contractors are underutilized in public procurement in a particular industry and geography, such that the agency can determine whether a legally-defensible race-conscious program is justified or needed to provide remedial relief given discriminatory or exclusionary behavior.
Disparity studies typically include an overview of the legal precedent that influences the key methodologies, computations, and evidence necessary to justify or support existing or proposed contracting programs, including those that are race-conscious. City of Richmond v. J.A. Croson Co.3 (Croson) and Adarand Constructors Inc. v. Peña4 (Adarand) are two seminal legal decisions that established the evidentiary tests necessary to evaluate local, state, and federal race-conscious contracting programs. These cases introduced several key concepts and standards, including:
Ensuring that disparities in contracting are specific to the relevant geographic and product markets
Disparities are evaluated considering only firms that are ready, willing and able to bid on and perform contracts;
The importance of evidence related to marketplace discrimination to support race- conscious contracting programs; and
The importance of anecdotal evidence in supporting programs offering remedial relief of discrimination in public contracting.
There have been a number of additional challenges to existing race-conscious contracting programs. While the constitutionality of programs has been upheld, the legal decisions have often brought forth key issues related to disparity study methodologies and the evidence needed to support an inference of discrimination related to an observed disparity ratio.
In addition to the legal review, disparity studies typically include an overview of the rules, regulations, and ordinances that govern public contracting for a particular agency. This includes the existence of race-conscious programs to alleviate contracting disparities. In order to determine the extent to which disparities exist among MBEs and different racial and ethnic groups, disparity studies compute numerical disparity ratios using agency procurement data, information on winning bidders, and a comprehensive analysis of actual and potential bidders to determine which firms are ready, willing, and able to bid on contracts. Consultants use this information to determine utilization and availability, the two inputs of the disparity ratio calculation. Figure ES-1 shows a simplified illustration of the disparity ratio calculation, where the numerator represents the utilization of MBEs and the denominator shows the availability of MBEs.5
DISPARITY RATIO COMPUTATION EXAMPLE
As a general rule of thumb, a disparity ratio of less than 0.80 (or 80 if expressed on a scale that multiplies the disparity index by 100) indicates a substantial disparity.6 Utilization and availability are also specific to well defined geographic and product markets (i.e., the “relevant markets”). Market definition is an economic concept that looks to substitutability and is intended to determine who is competing for public contracts along geographic and product lines. Robust and defensible disparity studies have an explicit definition of both geographic and product markets, as these are required in order to determine who is competing for contracts and the extent to which disparities exist among these market definitions.
The review of selected disparity studies provided 2,385 distinct high-level disparity ratios presented in executive summaries, major findings, and conclusions sections. These ratios include observations for MBEs in the aggregate, as well as for the African American, Hispanic American and Asian American categories.7 In addition, studies computed disparity ratios based on industry, with the majority reporting disparity ratios for major industry groups such as construction, professional services, architecture and engineering services, and goods and supplies. However, there is no standard disparity ratio reporting method and a review of the disparity studies found wide variation in how disparity authors computed and reported disparity ratios. Some studies included a single disparity ratio covering multiple years, while others reported ratios for every calendar year or fiscal year for the time period under investigation. Furthermore, some studies only reported disparity ratios on prime contracts, while other studies distinguished between prime contracts and subcontracts.
As a result, the disparity ratios are not an “apples to apples” comparison when examining results from one report compared to another. The studies were conducted by different authors, for different agencies, using different product and geographic market definitions and for different time periods. In addition, there are methodological differences in computing disparity ratios among consultants. Nevertheless, the comprehensive nature of the review established a distinct pattern of substantial contracting disparities for MBEs in the aggregate and for different racial and ethnic groups across different industries. 78.2 percent of all disparity ratios drawn from the set of disparity studies were less than 0.8, with a median value of 0.19. Considering that less than 0.8 is a substantial disparity, these results indicate that contracting disparities for MBEs are pervasive.
Furthermore, many studies tested whether these disparity ratios were statistically significant, where disparity study authors used statistical approaches to test whether the disparity could have arisen due to chance, or some other factor such as discrimination. For those disparity studies that explicitly indicated whether a disparity ratio was statistically significant or not, approximately 65 percent of all disparity ratio observations were classified as statistically significant by the study authors. However, this may be a conservative estimate since some disparity study consultants only reported significance at a highly aggregated level. Lastly, 99 percent of statistically significant disparities identified by study authors were less than 0.8, > lending strong support for discriminatory behavior in contracting.
Despite the detail regarding underrepresentation presented in disparity calculations, the existence of a disparity does not on its own support a conclusion of discrimination. Rather, the numerical disparity ratios necessitate additional inquiries to explain why MBEs face significant contracting disparities compared to non-MBEs. In order to determine whether disparities are the result of discrimination, disparity study consultants use both quantitative and qualitative analyses to examine the root causes of disparities in public contracting. Most studies in the research set included an analysis of marketplace discrimination, using regression analysis to investigate disparities in business formation, business earnings, and loan denials between MBEs and non-MBEs in the private sector. These analyses demonstrate the presence of discriminatory behavior in private markets by showing race as a statistically significant predictor of disparities in business owner earnings, business formation and access to capital. As a result, these analyses allow disparity studies to address whether or not public agencies were susceptible to or engaging in passive discrimination in public contracting.
USING ANECDOTAL EVIDENCE TO EXPLORE CONTRACTING BARRIERS AND CAUSES
Anecdotal evidence does not establish the predicate for race-conscious programs, but instead, aids policymakers in evaluating whether a contracting program is needed and narrowly tailored to address demonstrated discriminatory behavior. Anecdotal evidence provides first-hand accounts of barriers in public contracting and instances where discrimination is a factor in MBE underrepresentation. Critics of the validity of anecdotal evidence argue that the accounts may not be sufficiently verified and that instead of detailing actual accounts of discrimination, the evidence may only present perceptions of discrimination. Yet, legal proceedings have varied on the level of verification needed to support the importance and relevance of anecdotal evidence. In order to address these concerns, the most robust disparity studies will draw on multiple techniques to obtain anecdotal accounts from individuals that have had actual, verifiable experiences in working with a procurement agency. It is through a wide number of reliable sources that disparity studies can include instances of discrimination which are representative of the experiences of multiple minority business owners.
The disparity studies reviewed for this study provided specific, verifiable instances of discrimination which were recorded, cataloged, and analyzed using content analysis. The most robust studies identified barriers, discussed the harm that the improper conduct inflicted on the businesses in question, and examined the extent to which discriminatory exclusion and impaired contracting opportunities are systemic rather than isolated or sporadic. Figure ES-2 summarizes the most frequently cited barriers in the disparity studies.
MOST FREQUENTLY CITED CONTRACTING BARRIERS FACING MBES
Discrimination influenced multiple contracting barriers, both from the marketplace, as well as driven by either a contracting agency or non-MBE prime in the context of subcontracting. The barriers identified varied from outright prejudicial treatment and instances of exclusion based on racism, to marketplace barriers erected by systemic discrimination in both the private and public market (e.g., access to capital). Disparity studies with substantial anecdotal evidence supporting the presence of discriminatory barriers provide justification for the use of race- conscious programs in those jurisdictions. In addition, there are multiple non-discriminatory barriers, such as large project sizes, timely payment, and bid requirements that present challenges to potential bidders regardless of the race or ethnicity of the owners. However, the anecdotal evidence indicates that certain systemic discriminatory barriers can influence the perception of exclusionary practices with respect to some non-discriminatory barriers.
Arguably the most difficult barrier to address with respect to discrimination is the exclusionary networks that MBEs encountered in public contracting. On one hand, network exclusion can arise due to normal business operating procedures, often dictated by the desire to work with companies that have prior experience, demonstrated work product, and a solid reputation. Yet, in other instances, discriminatory attitudes of agency personnel and non-MBE primes facilitated excluding MBEs from informal networks that influence learning about and obtaining public contracting opportunities.
The review of existing disparity studies yielded several common themes and insights beyond the characterization of contracting barriers and evidence of discrimination. These included:
The “needle has not moved” with respect to overcoming disparities. Every study identified significant contracting disparities and many supported these findings with additional quantitative and anecdotal evidence that supported the need for both race- neutral and race-conscious remedial efforts. Yet, over time disparities were prevalent even within the same jurisdiction.
Disparity studies often reported the same race-neutral remedies (e.g., unbundling large contracts, improving payment processes, improving data collection) and race-conscious remedies (e.g., improved goal setting and monitoring) to address contracting disparities, yet what is missing is the extent to which agencies have actually implemented and measured the success or failure of these recommendations.
Race-conscious programs typically helped MBEs when enacted; however the legal history has illustrated that these programs need to comply with the strict scrutiny standard and be narrowly tailored.
In addition to common observations, the disparity studies and anecdotal evidence highlighted common problems and issues with contracting disparities experienced by MBEs. These include:
Enforcement and accountability of race-conscious programs by contracting agencies. There is a perception that prime contractors do not engage in good faith efforts to comply with race-conscious programs and agencies do not monitor or enforce these efforts.8
Resource constraints are a major issue facing contracting agencies. Many suggestions for program improvements, both race-neutral and race-conscious, require a substantial monetary investment (both human capital and infrastructure) at the public agency level. Based on the political and economic environment, some of these recommendations are prohibitive given lack of resources.
There is often insufficient analysis and evidence of subcontracting activity at the agency level. Given that subcontracting is an important and critical component of increasing MBE participation in public contracting, greater oversight and accountability of subcontracting behavior coupled with better and more reliable data collection should be a priority.
The disparity study review indicated that both discriminatory and non-discriminatory actions lead to contracting disparities for MBEs. Additional research is needed to understand what steps public agencies have taken to address these disparities. Specifically, whether agencies have been effective at implementing the common policy prescriptions most disparity studies include and to what extent these policies have either succeeded or failed. Beyond this, there are a number of areas to explore and research with respect to lessening barriers faced by MBEs in public contracting. MBEs, advocacy groups and policy makers should explore new and innovative ways to increase engagement, oversight, enforceability and accountability within the public contracting process. This requires leveraging data sharing and transparency, exploring race-neutral means and the efficacy of these means, and also evaluating what race-conscious methods have been not only defensible, but successful, in alleviating the effects of discrimination.
1 This report uses the terms “affirmative action programs,” “race-based programs,” and “race-conscious programs” interchangeably, where the terms imply a government initiated program that specifically includes racial or ethnic preferences in alleviating discriminatory behavior.
2 City of Richmond v. J.A. Croson Co. (488 US 469 (1989)) and Adarand Constructors Inc. v. Peña (515 US 200 (1995)) are two seminal legal decisions that established the evidentiary tests necessary to evaluate local, state, and federal race-conscious contracting programs.
5 This simplified example assumes uniform contract and firm sizes, such that the disparity ratio would be equivalent whether one considers utilization based on the number of contracts or dollars awarded per contract.
6 Given the lack of standardization in evaluating the levels of underutilization, many studies employ the Equal Employment Opportunity Commission’s (EEOC) “80 percent rule” in Uniform Guidelines on Employee Selection Procedures. In the context of employment discrimination, an employment disparity ratio below 80 indicates a “substantial disparity” in employment.
7 This does not represent the totality of disparity ratios reported in the 100 studies. In certain cases, disparity study consultants also included Native Americans and Subcontinent Asian Americans, but these instances were relatively few or often contained inadequate data to compute a disparity ratio. In addition, most studies reported disparity ratios for women-owned businesses, although differences existed with respect to approaches separating out Caucasian-owned WBEs versus non-Caucasian owned WBEs. Furthermore, some studies reported an aggregate M/WBE disparity ratio, as opposed to just an MBE disparity ratio. The results presented in this report include the combined M/WBE ratios, but do not include WBE-only disparity ratios. Lastly, many studies provided hundreds of different additional disparity ratios based on smaller geographic regions, combining across industries, looking at different funding sources, or looking at different time periods. In order to minimize double counting, the research findings in this study do not include the subset of disparity ratios based on the multiple iterations that some disparity study consultants performed. The primary purpose of the disparity ratio review was to demonstrate that these studies identified contracting disparities, sufficient to assess causal factors.