Preliminary analysis prepared independently by House Strategies Group LLC from public data. Not affiliated with or endorsed by the City of Boston. Figures illustrate analytical approach and are subject to revision with primary data. Not a procurement-disparity finding.

Module · The barriers framework

The barriers, and how the study investigates each

A disparity study earns its conclusions by naming the specific barriers that keep ready, willing, and able firms out of public work, then measuring each one. This module sets out an exhaustive, cited framework of those barriers and, for every one, the exact quantitative and qualitative method the full Boston study would use to investigate it.

A statistical gap is the starting point of a disparity study, not the conclusion. Under City of Richmond v. Croson, which controls in Massachusetts as it does everywhere, a defensible study has to connect each measured disparity to an identified barrier and corroborate it with evidence. The First Circuit has no modern disparity-study precedent as fully developed as the Fourth Circuit's H.B. Rowe v. Tippett, so a Boston record has to satisfy Croson directly, with H.B. Rowe as persuasive guidance on what a rigorous record looks like. This framework is built for that standard.

Each barrier below carries its own legal and academic authorities, a designated bench lead, and a paired method: what the study counts, and what the study asks. The approach mirrors how recent, court-tested studies are built, including the City of Houston's 2024 study, which paired a custom-census availability survey, utilization and marketplace regressions, and a structured anecdotal record to move from a gap to a finding.

Barriers
16
cited and method-paired
Authorities
120
cases, regs, studies
Study methods
210
quant + qualitative

These barriers are not hypothetical here

An independent, public-data read already finds a substantial disparity

Before bidding, House Strategies Group built a preliminary analysis of Boston's own market from Census and City data, the analysis this portal presents. It finds minority- and women-owned firms taking a far smaller share of the dollars than their share of the available, qualified market, well below the 0.80 line courts use to flag substantial underutilization. Because the read is independent and built to report what the data shows either way, that signal is credible. It is also not a verdict. A gap this size is a serious question, and the barriers below are the candidate causes a full study would test, one at a time, to separate a neutral explanation from a genuine barrier.

The map

Five families of barrier

The framework groups into five families. Jump to any barrier, or read straight through.

Screening at the gate

Eligibility and scoring rules that rank firms on accumulated history before price or technical merit is reached.

How the buy is structured

Contract size, bundling, and lower-tier rules that decide whether smaller firms ever see a piece they can win.

Information and networks

Who learns of opportunities, who is introduced, and who is mentored, the informal channel Loury calls discrimination in contact.

Upstream and the wider market

Business formation, private-sector treatment, and labor-market access that shape the supplier pool before the City buys anything.

Barrier family

Screening at the gate

Eligibility and scoring rules that rank firms on accumulated history before price or technical merit is reached.

01

Firm Experience / Past-Performance / Track-Record Requirements

Bench lead

Access-to-capital / labor economist (econometrician) leads the quantitative marketplace and utilization regressions, partnered with the disparity-study methodologist who owns the NCHRP 644 disparity-index and availability framework; a survey methodologist / qualitative-research lead directs the owner survey, interviews, and public-hearing record; legal counsel (Croson/H.B. Rowe compliance) sets the evidentiary verification standard. HSG's role: City of Boston solicitation/award data extraction, requirement-coding, threshold-simulation modeling, and the interactive disparity-visualization layer feeding all of the above.

Solicitation provisions that condition eligibility or scoring on a firm's accumulated history: minimum years-in-business, prior-similar-contract ("must have completed X projects of similar scope/dollar value"), and supplier/client reference requirements. These are facially neutral capacity screens, but they ration access to whatever firms already have a track record. Because minority-, women-, and other diverse-owned firms entered the market later and in smaller numbers as a result of past exclusion, an experience floor mechanically converts that historical disadvantage into a present-day disqualification: a firm cannot accumulate the references and prior awards a solicitation demands without first winning the awards the requirement blocks. The barrier is the contracting analog of Croson's "qualified, willing and able" inquiry: the question is whether otherwise-ready firms are screened out before bid evaluation even reaches price or technical merit.

Why it is a barrier

The exclusion operates as a feedback loop the disparity-study literature treats as a primary mechanism of underutilization. (1) Later market entry as a product of past discrimination: Fairlie & Robb (MIT Press, 2008) document that Black-owned firms start smaller, with far less startup capital (nearly half of Black families hold under $6,000 in total wealth) and less pre-business managerial experience acquired through family firms; the result is fewer, younger, smaller MBEs at any point in time, so any experience floor disproportionately catches them. (2) Capacity-constrained, not absent: BBC's 2020 City of Boston disparity study found minority- and women-owned firms available for 16.9 percent of the City's contract dollars but paid only 11.0 percent, with Black American-owned firms substantially under-used across every work type studied, a pattern consistent with capacity and track-record screens (not raw availability) binding first. (3) Anecdotal corroboration courts credit: in H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010), majorities of African American and Native American subcontractors reported that prime contractors "have higher standards for minority subcontractors and view them as less competent", the experiential face of a track-record screen, and the Fourth Circuit treated that anecdotal evidence as corroborating the statistical disparity. (4) The remedy structure confirms the diagnosis: 49 C.F.R. 26.51(b)(7) directs recipients to "assist new, start-up firms, particularly in fields in which DBE participation has historically been low," and (b)(6) to help firms "handle increasingly significant projects", the regulation presumes that experience/scale thresholds are a barrier to be actively neutralized, not a neutral fact.

How the full study investigates it
Quantitative
  • Disparity index/ratio analysis (NCHRP 644 framework): compute utilization-to-availability ratios for M/WBE prime and subcontract dollars, then disaggregate by contract size band and by NAICS/industry to show whether disparity widens precisely where experience/prior-similar thresholds are highest (e.g., large construction vs. small services).
  • Availability estimation that holds capacity constant: build the availability base from firms that are 'ready, willing, and able' for the specific work, then test whether disparities persist even after restricting to firms meeting capacity proxies, isolating track-record screens from true readiness.
  • Marketplace (business-formation) regressions per NCHRP 644: logit/probit models of self-employment and business ownership on race, sex, and controls (human capital, wealth, industry) to estimate that but for discrimination there would be more M/WBE firms, quantifying the missing firms an experience floor presumes do not exist.
  • Business-owner earnings/firm-size regressions: model firm revenue/receipts and owner earnings on ownership demographics plus controls to show diverse firms are systematically smaller/younger, demonstrating that years-in-business and prior-contract-value thresholds proxy for race/sex.
  • Firm-age and firm-survival analysis: distribution of M/WBE vs. non-M/WBE firm age and survival; quantify the share of available M/WBEs that fail a stated 'minimum X years in business' threshold to size the mechanical exclusion.
  • Conditional logit / win-probability models on bid-level data: estimate the marginal effect of meeting (vs. narrowly missing) experience, prior-similar-project, and reference requirements on award probability, controlling for price and technical score, to isolate the screen's independent exclusionary effect.
  • Counterfactual threshold simulation: re-score historical solicitations under relaxed thresholds (lower years-in-business, allow comparable/aggregated experience, accept subcontractor past performance) to estimate how many additional ready M/WBEs would have qualified.
  • Quantify experience-requirement prevalence: code a sample of City of Boston solicitations for the presence/stringency of years-in-business, prior-similar-contract, and reference clauses; cross-tabulate stringency against observed M/WBE participation rates.
Qualitative
  • Business-owner availability and experience survey (NCHRP 644 standard): structured instrument capturing firm age, capacity, and specific instances where a years-in-business, prior-similar-project, or reference requirement disqualified the firm despite being able to perform; closed-ended items support quantification, open-ended items capture narrative.
  • In-depth one-on-one interviews with M/WBE owners across construction, professional services, and goods to document the track-record catch-22 (cannot get the contract without the reference; cannot get the reference without the contract), including verifiable specifics (solicitation numbers, dates) to meet H.B. Rowe's verification expectation.
  • Public hearings / sworn testimony before the City Council, the Supplier Diversity Council, or a study advisory committee, creating an on-the-record anecdotal corpus on past-performance exclusion that courts credit when corroborating statistics.
  • Trade-association and chamber input: structured sessions with Hispanic, Black, Asian-American, women's, and veteran business associations and regional contractor/AGC and professional-services groups to surface industry-specific experience norms and how primes apply higher track-record standards to minority subs (the H.B. Rowe finding).
  • Prime-contractor and procurement-officer interviews: how experience/reference language is set, scored, and waived, and whether comparable or subcontractor experience is accepted, identifying where the screen is discretionary and thus a remediable barrier.
  • Focus groups segmented by demographic and by firm age (start-up vs. established) to differentiate barriers facing newer M/WBEs (the population the requirement targets) from those facing mature firms.
  • Documented-instance verification protocol: follow up a subset of reported anecdotes against bid records to confirm them, addressing the Tippett caution that unverified anecdotes carry less weight.
Why it matters in Boston

Boston's 2020 baseline shows exactly the pattern an experience floor produces: BBC found MWBE firms available for 16.9 percent of the City's contracting dollars but paid 11.0 percent, an overall disparity index of 65, with Black American-owned firms at 12 and substantially under-used across every work type studied. Firms exist and are willing; the update's job is to identify what screens them out before price or merit is ever scored, and prior-similar-contract and reference clauses are directly codeable from the City's own solicitation record. The recovery in certified-MWBE spend since the study (5.2 percent of discretionary spending in FY2019 to 12.3 percent in FY2025, a certified-firms floor on true participation) shows the pool responds when the City opens doors, which strengthens the inference that track-record screens rather than capability were binding. The Sheltered Market Program (M.G.L. c. 30B, § 18, adopted August 2022) is a live corrective instrument: sheltered awards build precisely the past performance that experience requirements demand.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Establishes the 'qualified, willing and able' framing: a significant statistical disparity between the number of qualified minority firms willing and able to perform and the number actually used can give rise to an inference of discriminatory exclusion. Experience/track-record screens are the mechanism that can remove otherwise-qualified, willing, and able firms from the pool before that comparison is made; the study must measure 'qualified and able' net of barriers that are themselves products of discrimination.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in Massachusetts, where Croson and the post-SFFA strict-scrutiny climate govern. Requires statistical disparity be 'corroborated by significant anecdotal evidence of racial discrimination.' Upheld race-conscious remedies for African American and Native American subcontractors where surveys showed primes held minority subs to higher standards and viewed them as less competent, the experiential signature of track-record/past-performance barriers. Also a caution: the court noted unverified anecdotes are weaker, so the study must verify owner accounts.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, National Academies / TRB, 2010)NCHRP Rep. 644 (2010), NAP record 14346

    The canonical methodological authority (Jon Wainwright, NERA, and Colette Holt, co-PIs). Defines anecdotal evidence as qualitative data on owners' accounts of disparate treatment and barriers to business success, and treats personal experience with discriminatory policies/systems as probative. Supplies the disparity-index framework and the regression-based 'business formation, ownership, and earnings' analyses used to show that but for discrimination there would be more and larger M/WBE firms, the direct counter to an experience floor.

  • 49 C.F.R. 26.51(b)49 C.F.R. 26.51(b)(1)-(b)(9)

    Race-neutral remedy menu. (b)(1) arranging solicitation specifications/quantities/delivery to facilitate small-firm participation; (b)(2) easing bonding/financing; (b)(3) technical assistance; (b)(5) supportive-services capacity building; (b)(6) helping firms 'handle increasingly significant projects' toward self-sufficiency; (b)(7) assisting 'new, start-up firms, particularly in fields in which DBE participation has historically been low.' Directly targets experience/track-record barriers and frames the remedies the study should recommend.

  • Robert W. Fairlie & Alicia M. Robb, Race and Entrepreneurial Success (MIT Press, 2008)Fairlie & Robb (2008), ISBN 9780262514941

    Peer-reviewed access-to-capital/human-capital evidence using Census CBO data: lack of startup capital and limited pre-business experience drive smaller, younger, fewer Black-owned firms. Explains why later/smaller entry is itself a product of disadvantage, so an experience floor disproportionately excludes diverse firms regardless of present capability.

  • BBC Research & Consulting, City of Boston Disparity Study (final report February 2021)City of Boston Disparity Study (BBC, study period FY2015-FY2019)

    The City's own baseline study. Analyzed 47,000+ contracts and built a custom-census availability database (~800 firm interviews); found an overall MWBE disparity index of 65, a Black American-owned index of 12, and substantial disparities for Black-owned firms across all work types studied. Evidence that qualified, willing firms exist but are screened out or under-used, the pattern an experience floor produces, and the empirical baseline this update must deepen.

  • Keen Independent Research; MGT; NERA Economic Consulting (Jon Wainwright)named disparity-study practitioners

    Establish the field-standard methodology this barrier draws on: examining firm size, firm age ('how long they have been in business'), and contract-size capability as availability/utilization variables, and pairing utilization regressions with marketplace (business-formation/earnings) regressions and anecdotal evidence to isolate track-record screens from genuine capability gaps.

Boston data sources
  • Analyze Boston 'City of Boston Contract Award' dataset (FY2019-FY2026, contracts of $10,000 or more with the City's certification flags): award value, department, vendor, and dates, the utilization base.
  • Full text of City of Boston solicitations (IFBs/RFPs/RFQs) from the City's Supplier Portal and City contract files, coded for years-in-business, prior-similar-project, reference, and past-performance requirements and their stringency.
  • Bid tabulations and unsuccessful-bid records obtainable from the City's Procurement Department, to identify firms screened out before evaluation and to build win-probability models.
  • City of Boston MBE/WBE/SLBE/SBE certification directory and the Massachusetts Supplier Diversity Office (SDO) directory, to build the M/WBE availability universe with firm age and capacity attributes.
  • The 2020 BBC disparity study final report and appendices, including its custom-census availability database and anecdotal appendix, as the FY2015-FY2019 baseline.
  • U.S. Census Bureau Annual Business Survey (ABS 2022) and Nonemployer Statistics by Demographics (NES-D 2022) for the Boston-Cambridge-Newton metro area (CBSA 14460) and the five RGMA counties, for the marketplace/business-formation regressions.
  • County Business Patterns and Nonemployer Statistics for the Boston market area to anchor firm-age and firm-size distributions by industry.
  • Dun & Bradstreet / commercial business registry data for firm establishment year, employee size, and revenue to validate years-in-business and capacity measures.
  • Surveys, interview transcripts, public-hearing testimony, and trade-association session notes generated by the study (the anecdotal record).
06

Prequalification, Registration, and Approved-Bidder-List Regimes

Bench lead

Econometrician / disparity economist (NERA / Wainwright-style lead) for the funnel-attrition, disparity-index, approval-probability, and business-formation/earnings regressions, working in tandem with a survey methodologist who designs and fields the prequalification battery and codes anecdotal evidence; a legal scholar / MWBE-program counsel (Colette Holt-style) frames the M.G.L. c. 149 / c. 30B and Croson defensibility analysis.

This contract
Event ID EV00017483

The solicitation demonstrated the barrier in real time.

The City released this disparity-study RFP (Event ID EV00017483) on June 29, 2026 through its Supplier Portal, where the documents sit behind vendor registration: a firm must know the portal exists, create an account, and register before it can read a single page of an RFP about barriers to participation. That is ordinary procurement hygiene, and it is also a live, documentable instance of the registration gate this chapter studies. The study can measure how many available MWBE firms never clear that first step.

See the data each analysis needs

Prequalification systems are the gatekeeping mechanisms a public body uses to decide who is even allowed to bid before any specific solicitation is scored. They include (1) mandatory vendor registration in a procurement portal, (2) prequalification of bidders for a class of work or a specific construction project, and (3) standing approved-bidder, pre-approved-vendor, or pre-qualified-pool lists from which the agency draws. Eligibility typically turns on audited financial statements and working-capital ratios, demonstrated bonding/surety capacity, owned equipment and bonded backlog, insurance limits, and a track record of comparable prior public-sector work. Firms that cannot clear these screens are excluded from competition entirely, regardless of price or capability on the actual job. In Massachusetts, the analogous gates are vendor registration on the City's Supplier Portal, the responsible-and-responsive bidder standards of M.G.L. c. 30B, and, for public building construction, the Commonwealth's contractor certification program under M.G.L. c. 149, § 44D, which conditions bidding eligibility on evaluated project experience, references, and financial capacity.

Why it is a barrier

Prequalification criteria are facially race-neutral but screen on exactly the attributes where diverse and emerging firms are structurally behind, so they convert pre-existing capital and experience gaps into a hard bar at the front door. Croson itself catalogued these as the classic barriers facing any new entrant: "deficiencies in working capital, inability to meet bonding requirements, unfamiliarity with bidding procedures, and disability caused by an inadequate track record" (488 U.S. at 498-99). The access-to-capital literature shows why those criteria bite disproportionately: Fairlie & Robb (Race and Entrepreneurial Success, MIT Press 2008) find that low levels of startup capital are the single most important driver of weaker outcomes for Black-owned firms, that nearly half of Black families hold under $6,000 in total wealth, and that undercapitalized firms have lower sales, profits, and employment and higher failure rates. Thin balance sheets and limited net worth translate directly into failing a financial-ability screen and into inability to secure surety bonding, since surety underwriting leans on personal/business net worth and credit. The "appropriate experience" and "prior public work" screens are a circular trap: a firm cannot get prequalified without a public track record, and cannot build a public track record without being allowed to bid. Massachusetts makes this concrete: contractor certification for public building work under M.G.L. c. 149, § 44D turns on evaluated project experience, references, and financial capacity, the grounds most correlated with firm age, size, and owner wealth rather than with capability on a divisible scope. H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010), upheld race-conscious remedies in the Fourth Circuit precisely on a record of disparity-study subcontractor interviews documenting such barriers; it is not controlling in Massachusetts, but it remains a persuasive template for how this evidence can be assembled and analyzed under Croson.

How the full study investigates it
Quantitative
  • Build a prequalification funnel/attrition analysis: for each prequalification or registration gate, compute the share of registered/applicant firms that advance to prequalified status, disaggregated by minority-, women-, and small-business status and industry, and test whether diverse firms drop off at higher rates (cohort attrition rates with chi-square / two-proportion z-tests).
  • Compute disparity indices (utilization / availability, expressed 0-100, with the 80% substantial-disparity benchmark) separately among the prequalified pool versus among all ready-willing-and-able firms in the market, to isolate how much underutilization is produced upstream by the prequalification screen itself rather than by bid scoring.
  • Estimate logistic/probit regressions of prequalification approval (or list inclusion) on minority/women ownership, controlling for firm age, size, revenue, bonding capacity, and prior public contracts, to test for a residual ownership effect after capacity controls (the NCHRP 644 'marketplace conditions' approach applied to the gate).
  • Run the NCHRP 644 / NERA-style business-formation and business-earnings regressions on Census microdata (PUMS / Survey of Business Owners / Annual Business Survey) to quantify capital, net-worth, and credit-access gaps that predict failure of financial-ability and bonding screens, linking the macro disparity to the specific criteria.
  • Bonding-threshold simulation: model how lowering or unbundling bonding/financial-ability thresholds (per 49 CFR 26.51(b)(1)-(b)(2)) would change the count and diversity composition of the eligible pool, to quantify the marginal exclusionary effect of each criterion.
  • Benchmark the City's prequalification/registration criteria and renewal frequency against peer Massachusetts and regional jurisdictions, scoring each criterion for stringency and exclusionary risk.
Qualitative
  • Owner survey (the disparity-study availability/marketplace survey) with a dedicated prequalification battery: whether the firm attempted to register/prequalify, where in the process it stalled, which specific criterion blocked it (financials, bonding, experience, equipment, insurance), and time/cost burden - structured for quantification (NCHRP 644 anecdotal-evidence standard).
  • In-depth one-on-one interviews with minority-, women-, and small-business-owned firms that abandoned or were denied prequalification, capturing first-person accounts of disparate treatment and the circular experience trap, mirroring the subcontractor interviews the Fourth Circuit credited in H.B. Rowe v. Tippett.
  • Public hearings / listening sessions advertised through certification mailing lists and minority chambers, with testimony recorded and coded so anecdotes tie to specific barrier categories (a documented H.B. Rowe-style evidentiary record).
  • Trade-association and community input from regional minority and women contractor associations, the Massachusetts Supplier Diversity Office, and surety/bonding agents serving diverse firms, on how the City's criteria compare and where they exclude.
  • Key-informant interviews with City procurement and construction-management staff on how prequalification and responsibility determinations are issued and noticed, and how approved-vendor and registration lists are maintained and refreshed.
  • Structured document review of a sample of actual prequalification application forms, denial letters, and the criteria/weights used, to verify on paper which screens are binding and whether less-restrictive alternatives were considered.
Why it matters in Boston

A diverse firm pursuing City work clears at least two gates before any bid is scored: certification (City of Boston MBE/WBE/SLBE/SBE or Massachusetts SDO, recognized in parallel) and Supplier Portal registration, with DCAMM certification layered on top for public building construction. Each gate is defensible alone; the study's question is their cumulative attrition, and the 2020 BBC finding that availability (16.9 percent) ran well ahead of utilization (11.0 percent) says the pool thins somewhere between existing and winning. The FY2019-FY2026 rise in certified-MWBE spend from 5.2 to 12.3 percent counts certified firms only, so every firm that never finishes certification or registration is invisible to it; measuring that funnel end-to-end is the update's most direct extension of the baseline.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469, 498-99, 509-10 (1989)

    Identifies 'inability to meet bonding requirements,' 'deficiencies in working capital,' 'unfamiliarity with bidding procedures,' and 'inadequate track record' as barriers, and lists race-neutral cures including 'simplification of bidding procedures, relaxation of bonding requirements,' and elimination of 'formal barriers caused by bureaucratic inertia.' Establishes that a jurisdiction must consider these race-neutral alternatives, making prequalification mechanics a required subject of study.

  • H.B. Rowe Co. v. Tippett (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority; in Massachusetts the controlling frame remains Croson strict scrutiny. Upheld race-conscious M/WBE subcontracting goals for African American- and Native American-owned firms on the strength of a disparity study's statistical analysis plus interviews of minority subcontractors reporting barriers; rejected the program for groups (Asian American, Hispanic American, women) lacking that evidentiary base. Sets the local standard for the quantum and granularity of barrier evidence, including barriers like prequalification and access to work.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board / National Academies, 2010)NCHRP Rep. 644 (TRB 2010), Wainwright (NERA) & Holt (Colette Holt & Assocs.), PIs

    The National Academies methodological standard. Frames 'marketplace conditions' analysis and 'anecdotal evidence' (defined as business owners' accounts of disparate treatment and barriers to business success) as core study components, and prescribes how to combine custom-census/regression disparity analysis with structured qualitative barrier evidence in a legally defensible design - the framework for studying prequalification as a barrier.

  • 49 C.F.R. § 26.51 (USDOT DBE rule, race-neutral remedy menu)49 C.F.R. § 26.51(b)(1)-(b)(2)

    Enumerates the race-neutral remedies that directly target this barrier: (b)(1) arranging solicitations, quantities, specifications, and schedules to facilitate small-firm participation (i.e., unbundling/right-sizing prequalified scopes); and (b)(2) 'assistance in overcoming limitations such as inability to obtain bonding or financing... by such means as simplifying the bonding process, reducing bonding requirements, eliminating the impact of surety costs from bids.' Provides the federally sanctioned recommendation set the study should benchmark the City against.

  • Fairlie & Robb, Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (MIT Press 2008), analyzing Census Characteristics of Business Owners (CBO) data

    Peer-reviewed evidence that startup capital is the most important determinant of firm outcomes, that Black households hold dramatically lower wealth (nearly half under $6,000), and that undercapitalized firms underperform and fail more often. Explains the causal mechanism by which financial-statement, working-capital, and bonding-capacity prequalification screens disproportionately exclude diverse firms.

  • M.G.L. c. 149, § 44D (contractor certification for public building construction) and M.G.L. c. 30B (Uniform Procurement Act)M.G.L. c. 149, § 44D; M.G.L. c. 30B

    The local legal mechanisms. Public building construction above the statutory threshold requires state contractor certification based on evaluated project experience, references, and financial capacity, and M.G.L. c. 30B conditions award on responsible-and-responsive bidder determinations; together with the City's own Supplier Portal registration gate, these are the concrete screens the study can examine as applied.

Boston data sources
  • City of Boston Supplier Portal vendor registration records and solicitation files (the registration gate itself).
  • City of Boston and Massachusetts SDO certification directories, with application and renewal requirements, to measure the certification gate end-to-end.
  • DCAMM contractor certification records for public building construction under M.G.L. c. 149, § 44D, for the construction prequalification tier.
  • Analyze Boston 'City of Boston Contract Award' dataset (FY2019-FY2026, with certification flags) and City bid tabulations obtainable from the Procurement Department.
  • The 2020 BBC disparity study availability database and appendices, for the ready-willing-and-able universe and the FY2015-FY2019 baseline.
  • U.S. Census ABS 2022 and NES-D 2022 for CBSA 14460 and the five RGMA counties, plus ACS/PUMS microdata, for business-formation and marketplace-conditions regressions.
  • Surety/bonding industry and SBA bonding-program data, and FFIEC CRA small-business lending data, for the capital-access link.
  • Primary-collected disparity-study owner survey, interview transcripts, and public-hearing testimony.
  • Peer-jurisdiction prequalification ordinances and criteria for benchmarking.
07

Restrictive or Proprietary Bid Specifications

Bench lead

Quantitative econometrician/statistician (disparity-study methodologist) leads, paired with a public-procurement law scholar. The econometrician owns the restrictiveness-coded regressions, disparity-ratio decomposition, and the H.B. Rowe-compliant capacity controls; the procurement-law scholar owns the specification audit against Croson, 49 C.F.R. 26.51(b), and M.G.L. c. 30B, and the legal-sufficiency framing of the anecdotal record. A survey methodologist supports the owner survey and hearings instrument design.

A restrictive or proprietary specification is a technical requirement, qualification standard, brand-name designation, or scoring criterion written into a solicitation that is not essential to satisfactory performance of the work but that narrows the field of eligible competitors. Common forms include: brand-name-only or sole-source designations without an "or equal" provision; experience or past-performance minimums calibrated to an incumbent's resume (e.g., "must have completed five projects of identical type for this agency"); proprietary materials, software, or methods that only one supply chain carries; licensing, certification, or staffing requirements exceeding what the scope demands; and evaluation rubrics that weight prior-contract-with-this-agency or firm size/longevity so heavily that newer entrants cannot score competitively even at a lower price. The defining feature is a gap between what the specification requires and what the work actually needs. The unnecessary increment screens out otherwise qualified, willing, and able firms and, in practice, channels the award toward the established incumbent for whom the specification was effectively drawn.

Why it is a barrier

Diverse and emerging firms are disproportionately newer, smaller, and lack a track record of prior contracts with the specific awarding agency. Restrictive specifications convert each of those characteristics into a disqualifier before price is ever considered. Because the barrier operates at the eligibility/responsiveness stage rather than the award stage, it suppresses the pool of "willing and able" bidders that disparity statistics measure, producing low utilization that looks like a competitive outcome but is actually a design artifact. The mechanism compounds documented capital and human-capital gaps: Fairlie and Robb show minority-owned firms enter with substantially less startup capital and less prior family-business and managerial experience, so requirements that demand a long agency-specific resume, large bonded capacity, or proprietary tooling fall hardest on exactly the firms those baseline gaps already disadvantage. The U.S. Supreme Court in Croson expressly recognized that facially neutral procurement requirements can suppress minority participation and that their elimination or modification "would have little detrimental effect on the city's interests" while increasing opportunity, treating narrow requirements as a barrier remediable by race-neutral means. The MBDA's 2017 review of roughly 100 disparity studies found contracting disparities pervasive (a majority of disparity ratios below the 0.80 substantial-disparity threshold), with anecdotal evidence repeatedly identifying restrictive specifications, qualification requirements, and procurement procedures as marketplace barriers rather than firm-quality differences.

How the full study investigates it
Quantitative
  • Specification-restrictiveness coding: pull a stratified sample of City of Boston solicitations (IFBs/RFPs/RFQs) across departments and NAICS, and code each for restrictive features (brand-name without 'or equal'; sole-source justifications; agency-specific experience minimums; proprietary materials/software; staffing/licensing exceeding scope; longevity or incumbency-weighted scoring). Express prevalence as a share of solicitations and dollars.
  • Bidder-count and single-bid analysis: regress number of responsive bidders (and probability of a single/sole bid) on the restrictiveness code, controlling for contract size, NAICS, and procurement method, to test whether restrictive specs suppress competition.
  • M/W/DBE participation gradient: compare availability-weighted M/W/DBE bid and award rates on restrictive vs. non-restrictive solicitations to estimate the participation penalty attributable to specification design.
  • Disparity-ratio decomposition: compute disparity ratios (utilization/availability) overall and segmented by restrictive-specification status, flagging ratios below the 0.80 substantial-disparity threshold per MBDA/standard practice.
  • Regression controlling for legitimate capacity (per H.B. Rowe): logistic/linear models of award probability and award dollars including firm size, age, bonding capacity, prior experience, and minority/women ownership, isolating the residual associated with restrictive specifications net of bona fide capacity factors.
  • Responsiveness/rejection analysis: tabulate rates at which M/W/DBE bids are deemed non-responsive or fail to meet minimum qualifications, by specification-restrictiveness category, to locate where in the funnel firms are screened out.
  • Scoring-rubric simulation: re-score a sample of competitively evaluated awards with incumbency/longevity/size weights neutralized to estimate how many awards would change, quantifying the criteria's marginal effect.
  • Brand-name and proprietary-specification audit against M.G.L. c. 30B: rate of specs that restrict competition without a documented justification or a conforming 'or equal' allowance, as a statutory-baseline measure of remediable restrictiveness.
Qualitative
  • Owner survey of M/W/DBE and emerging firms with closed-ended items on whether specifications, qualification minimums, brand-name/proprietary requirements, or scoring criteria deterred or disqualified them from City of Boston solicitations, with frequency and dollar-impact estimates.
  • In-depth one-on-one interviews with a purposive sample of diverse-firm owners to capture verifiable accounts of specific solicitations whose specs they believe were unnecessarily narrow or wired to an incumbent, including solicitation numbers where they will disclose them (to support corroboration per H.B. Rowe).
  • Public hearings/community testimony sessions advertised to the diverse-business community, with sworn or recorded testimony on specification and qualification barriers, structured to elicit identifiable detail rather than anonymous generalities.
  • Trade-association and chamber input (e.g., regional minority and women business councils, AGC/A-E associations, and Hispanic, Asian American, and Black chambers) via roundtables and written submissions on industry-specific restrictive-spec patterns.
  • Interviews with City procurement officers, contracting officers' technical representatives, and using-department specification writers on how specs and evaluation criteria are drafted, who drafts them, and what review controls exist for restrictiveness and 'or equal' compliance.
  • Prime-contractor interviews on subcontract specifications and pass-through requirements that may exclude diverse subs even when the prime contract is open.
  • Structured case studies of specific solicitations flagged in the quantitative coding, triangulating the document, the bid tabulation, and owner/staff accounts.
Why it matters in Boston

M.G.L. c. 30B gives the audit a built-in statutory yardstick: purchase descriptions must define what the City actually needs without unreasonably restricting competition, and proprietary or sole-source specifications require documented justification, so every deviation the coding finds is a remediable departure from Massachusetts' own competition baseline. BBC's 2020 finding that substantial disparities concentrated in construction (index 34) and services (67 and 66) rather than goods and supplies suggests specification and qualification design, not product supply, is where the screens bind. The 47,000-contract baseline and the public FY2019-FY2026 award file give the update enough volume to correlate restrictiveness coding with bidder counts and certified-firm participation at statistical strength.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469, 507, 509-510 (plurality/O'Connor, J.)

    The Court identified race-neutral alternatives the City failed to consider, including simplification of bidding procedures, relaxation of bonding requirements, and elimination or modification of various requirements, whose modification 'would have little detrimental effect on the city's interests' while increasing minority opportunity. This makes unnecessary/restrictive requirements a recognized, remediable barrier and establishes that a jurisdiction must show race-neutral fixes are unworkable before resorting to race-conscious remedies.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233, 241-242, 248-253 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in the First Circuit. A jurisdiction may meet its evidentiary burden with a significant statistical disparity between availability and utilization of qualified, willing, and able minority firms, but it must be corroborated by significant anecdotal evidence and must control through regression for non-discriminatory factors such as firm size and experience. Directly relevant because restrictive specifications operate through size/experience proxies, so the study must both document them anecdotally and statistically distinguish them from legitimate capacity differences.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board, National Academies, 2010)NCHRP Rep. 644 (TRB 2010) (co-PIs Jon Wainwright, NERA, and Colette Holt)

    The National Academies methodological standard for legally defensible disparity studies. Directs studies to examine barriers to DBE/M/WBE formation and participation including marketplace and procurement-process conditions, to measure availability of qualified/willing/able firms, and supplies a model RFP scope of work. Grounds both the availability measurement and the barriers-analysis chapters in which restrictive-specification evidence sits.

  • 49 C.F.R. 26.51(b)(1)49 C.F.R. 26.51(b)(1) (USDOT DBE race-neutral means)

    The federal race-neutral remedy menu's lead item is 'arranging solicitations, times for the presentation of bids, quantities, specifications, and delivery schedules in ways that facilitate participation by DBEs and other small businesses.' Confirms at the regulatory level that specification design is both a recognized barrier lever and a primary race-neutral remedy, anchoring the study's recommendation chapter. Companion 26.51(b)(2) addresses simplifying/reducing bonding requirements.

  • MBDA, Contracting Barriers and Factors Affecting Minority Business Enterprises: A Review of Existing Disparity Studies (U.S. Dept. of Commerce, Jan. 18, 2017)MBDA (2017), review of ~100 disparity studies

    Federal synthesis finding contracting disparities pervasive (majority of disparity ratios below the 0.80 substantial-disparity threshold; ~65% statistically significant) across construction, professional services, A/E, and goods/supplies, with anecdotal evidence identifying restrictive specifications, qualification requirements, and procurement procedures as barriers. Corroborating multi-jurisdiction evidence base.

  • Colette Holt & Associates / CHA disparity study framework (e.g., WSDOT and municipal studies)CHA study scopes addressing 'bidding specifications, requirements, and procedures, anti-competitive practices, or other barriers'

    Leading practitioner framework expressly enumerates bidding specifications and requirements as a marketplace-barrier category investigated through paired statistical and anecdotal analysis, modeling the analytic structure this chapter adopts.

  • MGT of America, City of Fort Worth, Texas Disparity Study (2020); BBC Research & Consulting and Keen Independent Research disparity studiesMGT Fort Worth (2020); BBC (e.g., Caltrans, Indiana); Keen Independent (2017 Minnesota Joint Disparity Study)

    Established disparity-study firms whose marketplace-conditions and anecdotal chapters routinely test whether specifications, scope definitions, contract size/bundling, and qualification standards disadvantage M/W/DBEs, providing methodological precedent for the qualitative and quantitative approaches used here.

  • Fairlie, R. & Robb, A., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (2008), MIT Press (Census Characteristics of Business Owners data)

    Peer-reviewed evidence that minority-owned firms enter with less startup capital and less prior managerial/family-business experience. Explains why specifications keyed to firm size, bonded capacity, prior agency experience, or proprietary tooling fall disproportionately on diverse and emerging firms rather than reflecting genuine performance risk.

  • M.G.L. c. 30B, § 14 (purchase descriptions; proprietary specifications)M.G.L. c. 30B, § 14

    Massachusetts' local-procurement statute requires purchase descriptions that define what is actually needed without unreasonably restricting competition and confines proprietary or sole-source specification to documented necessity. Supplies the state-law benchmark against which City of Boston solicitations can be audited for impermissibly restrictive brand-name or proprietary specifications.

Boston data sources
  • City of Boston solicitation documents and amendments (IFBs/RFPs/RFQs and standard terms) from the City's Supplier Portal and the Procurement Department.
  • City bid tabulations and award records (bidder counts, responsiveness determinations, winning vendor, price) obtainable from City contract files.
  • Analyze Boston 'City of Boston Contract Award' and 'City of Boston Discretionary Spending' datasets (FY2019-FY2026) for vendor, department, and dollar-value spend patterns.
  • Sole-source and proprietary-specification justification files maintained under M.G.L. c. 30B, from City contract files.
  • City of Boston and Massachusetts SDO certification directories and the 2020 BBC availability database of qualified, willing, and able firms.
  • Supplier Portal vendor registration records to identify firms that registered but did not or could not bid.
  • Owner survey responses, interview transcripts, public-hearing records, and trade-association submissions generated by the study.
  • Firm-level capacity covariates (firm age, employment/size, bonding capacity, prior experience) from survey, D&B/commercial registry, and certification records to support the regression controls.
  • M.G.L. c. 30B and City procurement policy/manual text as the statutory and policy benchmark for restrictiveness coding.

Barrier family

Capital and the cost of entry

The financing, bonding, insurance, and cash-flow demands that a firm must absorb to compete at all.

02

Bonding requirements (bid / performance / payment)

Bench lead

Access-to-capital / finance economist (econometrician), leading the bonding-capacity and credit-access regressions, supported by a surety/construction-finance subject-matter expert and a survey methodologist for the bonding survey module; legal scholar reviews for Croson/H.B. Rowe defensibility.

A bond is a three-party financial guarantee, issued by a surety, that a contractor will honor its obligations: a bid bond guarantees the bidder will enter the contract at the quoted price; a performance bond guarantees the work will be completed per the contract; a payment bond guarantees that subcontractors and suppliers will be paid. Public owners (including the City of Boston) routinely require bid, performance, and payment bonds above statutory thresholds. To issue a bond the surety underwrites the firm against the "three Cs" - capital (working capital and net worth), capacity (track record and equipment), and character (credit history and references). The barrier is that many minority-, women-, and emerging-owned firms cannot obtain a bond at all, cannot obtain one large enough to bid the full contract, or can obtain one only at a premium that erodes their price competitiveness. The result is exclusion from bonded work before the bid is ever evaluated on the merits.

Why it is a barrier

Bonding is gated by exactly the resources that the discrimination disparity studies measure are unequally distributed: capital, credit, and a prior track record. Because surety underwriting keys on personal and business net worth and credit, firms whose owners hold less wealth are screened out before price competition begins, converting historical wealth gaps into a present-day contracting bar. The federal evidence is direct: a 1995 GAO study of minority-owned firms found that more than 1 in 4 firms that did obtain bonds between 1990-1993 had been denied a bond at least once in that period, and roughly half of surveyed minority construction firms had never obtained a surety bond at all (often because they did not bid on bonded work), with long waiting periods, opaque denials, large financial commitments, and unaffordable fees cited as obstacles (GAO RCED-95-244R). The access-to-capital root cause is documented by Fairlie and Robb, who find that low levels of startup capital - tied to a Black-white family-wealth gap where nearly half of Black families held under $6,000 in total wealth - are the single most important driver of weaker minority business outcomes; thin capital and credit are precisely what surety underwriting penalizes. Named disparity studies repeatedly identify bonding and access to capital as measured barriers (Keen Independent's New Orleans and Colorado studies found quantitative disparities for M/WBEs in access to capital, bonding, and insurance, and documented firms that could not meet bonding/qualification requirements to bid at all). The Supreme Court itself flagged bonding as the mechanism in Croson, noting that if MBEs disproportionately cannot meet bonding requirements, relaxing those requirements is a race-neutral path to greater participation. The disproportionate impact is the equity harm: a uniform bonding rule applied to firms with unequal access to capital produces unequal access to contracts.

How the full study investigates it
Quantitative
  • Bonding-capacity disparity ratios: compute the share of M/WBE and emerging firms able to obtain bonds at the dollar levels City of Boston contracts require, benchmarked against non-M/WBE firms of similar size/age, and express as disparity indices (observed/expected) consistent with the NCHRP 644 framework.
  • Threshold/exclusion analysis: tabulate City of Boston solicitations by required bond amount and measure the count and dollar value of contracts whose bonding thresholds exceed the maximum bond capacity reported by M/WBE firms - quantifying contracts those firms are screened out of before bid.
  • Multivariate (logistic/probit) regression of the probability of obtaining a bond on race/ethnicity/gender ownership while controlling for firm age, revenue, net worth, credit score, and prior bonded experience, isolating the residual disparity not explained by neutral firm characteristics (NCHRP 644 / Wainwright disparate-impact model).
  • Bond-cost (premium) regression: model surety premium rate as a function of ownership characteristics plus controls to test whether diverse firms pay more per dollar of bond coverage, quantifying the 'impact of surety costs from bids' that 49 CFR 26.51(b)(2) targets.
  • Business-formation and earnings regressions (Fairlie/Robb method) using PUMS/CBO/ABS microdata to estimate capital-access and credit-denial gaps that feed bonding eligibility, establishing the upstream capital constraint statistically.
  • Marketplace availability cross-tab: link bond-capable firm counts to availability estimates so the disparity is expressed relative to firms ready, willing, and able to bid (the Croson 'qualified pool' requirement), broken out by group to satisfy H.B. Rowe group-specific evidentiary needs.
  • Statistical significance testing (standard errors, confidence intervals, t-tests on disparity indices) so each bonding finding is defensible under strict scrutiny.
Qualitative
  • Business-owner survey with a dedicated bonding/credit module: structured questions on whether the firm has ever been denied a bond, the maximum single and aggregate bond it can obtain, premium rates paid, whether bonding requirements caused it to forgo or scale down a City of Boston bid, and experience with the SBA Surety Bond Guarantee Program (modeled on the GAO RCED-95-244R survey instrument).
  • In-depth one-on-one interviews with M/WBE and emerging contractors who report being shut out of bonded work, capturing specific anecdotes of denials, delays, and prime/subcontractor bonding pass-through demands (the anecdotal evidence Croson and NCHRP 644 require).
  • Public hearings / listening sessions where contractors testify on record about bonding barriers on City of Boston projects, creating the contemporaneous anecdotal record that strengthens the strong-basis-in-evidence showing.
  • Surety industry and agent interviews to document underwriting criteria (the three Cs), prequalification practices, and how those criteria differentially affect thin-capital firms - establishing the mechanism, not just the outcome.
  • Trade-association and chamber input (e.g., minority/women contractor associations, AGC, local Hispanic/Asian/Black chambers) on systemic bonding obstacles and on the effectiveness of any existing assistance programs.
  • Prime-contractor interviews on whether they require subs to be bonded and how that propagates the barrier down the subcontracting tier.
Why it matters in Boston

Construction is where Boston's 2020 numbers are worst: BBC measured MWBE availability at 19.7 percent of construction dollars against utilization of 6.7 percent, a disparity index of 34, the lowest of any industry studied. Bonding is the classic mechanism behind a construction-specific gap of that depth, so the update should code bond thresholds across the City's solicitation record and pair them with group-specific bonding-capacity evidence, disaggregated as strict scrutiny requires. The 2021 executive order's 25 percent MWBE goal and the Sheltered Market Program create demand-side pull, but neither touches surety underwriting; if the construction disparity persists in the FY2019-FY2026 data, a bonding-assistance program modeled on 49 C.F.R. 26.51(b)(2) and the SBA guarantee is the race-neutral complement the record would support.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469, 507, 509-510 (1989)

    The Court identified inability to meet bonding requirements, deficiencies in working capital, and unfamiliarity with bidding procedures as barriers, and stated that relaxing bonding requirements, simplifying bidding, and providing financial aid to disadvantaged entrepreneurs of all races are race-neutral means that would open public contracting; it also held that statistical and anecdotal evidence of a strong basis in evidence is required to justify race-conscious remedies. Justia: https://supreme.justia.com/cases/federal/us/488/469/

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in the First Circuit: a properly supported disparity study can sustain race-conscious M/WBE measures for some groups (African American and Native American subcontractors upheld) while striking them for groups lacking an evidentiary basis (Hispanic, Asian, women). A jurisdiction need not conclusively prove discrimination, only a strong basis in evidence - making group-specific bonding/capital findings essential. CourtListener: https://www.courtlistener.com/opinion/151335/hb-rowe-co-inc-v-tippett/

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board / National Academies, 2010)NCHRP Rep. 644 (TRB 2010)

    The National Academies methodological standard. Directs that disparate impacts of economic factors (including access to capital and bonding) on M/WBEs be shown with statistics and econometric models, and that anecdotal/qualitative evidence be gathered on the factors necessary for entrepreneurial success including equal access to business capital and bonding. National Academies: https://www.nationalacademies.org/publications/14346

  • 49 C.F.R. 26.51(b)(2) (USDOT DBE Program)49 C.F.R. 26.51(b)(2)

    Race-neutral remedy menu, exact text: 'Providing assistance in overcoming limitations such as inability to obtain bonding or financing (e.g., by such means as simplifying the bonding process, reducing bonding requirements, eliminating the impact of surety costs from bids, and providing services to help DBEs, and other small businesses, obtain bonding and financing).' This is the canonical remedy template for the bonding barrier. eCFR/Cornell: https://www.law.cornell.edu/cfr/text/49/26.51

  • GAO, Minority-Owned Firms' Access to Surety Bonds (GAO/RCED-95-244R, 1995)GAO/RCED-95-244R (1995)

    Federal empirical evidence of the barrier: among minority firms that obtained bonds 1990-1993, more than 1 in 4 had been denied at least once; about half of surveyed minority construction firms never obtained a surety bond; obstacles included long waiting periods, opaque denials, large financial commitments, extraneous paperwork, and unaffordable fees. Justia/GAO: https://gao.justia.com/small-business-administration/1995/7/minority-owned-firms-access-to-surety-bonds-rced-95-244r

  • Fairlie, R. & Robb, A., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (MIT Press 2008)

    Peer-reviewed access-to-capital root cause: using the Census restricted-access Characteristics of Business Owners (CBO) dataset, finds startup capital is the most important factor in business success and that the lack of startup capital for Black-owned firms (tied to family wealth where nearly half of Black families held under $6,000) drives weaker outcomes - the underlying mechanism that surety underwriting penalizes. MIT Press: https://direct.mit.edu/books/monograph/2508/

  • Keen Independent Research, City of New Orleans Disparity Study (2018) and State of Colorado Disparity Study (2020)Keen Independent, New Orleans (2018); Colorado (2020)

    Named disparity studies (Keen has conducted 200+ since Croson) finding quantitative disparities for M/WBEs in access to capital, bonding, and insurance, and documenting via survey/anecdote that some small/M/WBE firms cannot meet bonding and qualification requirements to bid at all. New Orleans: https://nola.gov/nola/media/Economic-Development/Supplier%20Diversity/KeenIndependentNOLADisparityStudySummaryReport04112018.pdf

  • SBA Surety Bond Guarantee Program (15 U.S.C. ch. 14B; CRS R42037)SBA SBG Program; CRS R42037

    The operative federal remedy instrument for 49 CFR 26.51(b)(2): SBA guarantees bid, performance, and payment bonds for small firms that cannot obtain them commercially, with an enhanced 90% guarantee for socially and economically disadvantaged, 8(a), HUBZone, and SDVOSB firms (vs. 80% otherwise); contract limit $9M ($14M for federal). Demonstrates a tested race-neutral mechanism the City can adopt or partner on. CRS: https://www.congress.gov/crs-product/R42037

Boston data sources
  • City of Boston solicitation and contract records: bond requirement amounts (bid/performance/payment) from Supplier Portal solicitation documents and City contract files, with award values and prime/sub structure.
  • Analyze Boston 'City of Boston Contract Award' dataset (FY2019-FY2026, with the City's certification flags) for award values and vendor identification.
  • Business-owner survey responses (Boston-area M/WBE, emerging, and majority firms) with the bonding/credit module.
  • Surety underwriting and premium data: surety company / agent records and prequalification criteria; National Association of Surety Bond Producers and Surety & Fidelity Association of America industry data.
  • SBA Surety Bond Guarantee Program participation and approval/decline data.
  • Credit and financial benchmarks: Dun & Bradstreet / commercial credit data and firm financial statements where obtainable, for net worth and working-capital controls.
  • U.S. Census Annual Business Survey (ABS 2022) and ACS/PUMS microdata for capital-access regressions in the Fairlie/Robb tradition.
  • FFIEC CRA small-business lending data, HMDA data, and the Federal Reserve Small Business Credit Survey for regional credit-access and financing-denial benchmarks.
  • The 2020 BBC disparity study final report and anecdotal appendix, which document the FY2015-FY2019 Boston baseline the bonding analysis must update.
  • GAO RCED-95-244R and the Washington State surety bond accessibility study as instrument and benchmark exemplars.
03

Access to Capital and Financing

Bench lead

Access-to-capital / small-business-finance economist (econometrician), leading the SSBF/KFS-style credit-outcome regressions, supported by a survey methodologist for the owner-survey capital battery and a legal scholar to align the evidentiary record with Croson and the Fourth Circuit's H.B. Rowe standard.

Diverse- and women-owned firms face systematically worse outcomes in the small-business credit and equity markets: higher loan-denial rates, smaller approved loan amounts, higher interest rates and collateral demands, greater reliance on costly or informal sources (owner equity, personal credit cards, family loans), and thinner starting capital. These gaps persist after controlling for firm size, industry, profitability, credit score, and owner creditworthiness, which means they are not fully explained by differences in business fundamentals. Because public contracts require a firm to front payroll, materials, equipment, and bonding for 30 to 90 days before the government pays, the ability to borrow working capital is a precondition for bidding on and carrying out a contract. A firm that cannot finance the gap between performing the work and getting paid cannot scale up to take on larger or more numerous City contracts, regardless of its technical competence.

Why it is a barrier

Capital is the binding constraint on minority and women business formation, survival, and growth, and the disadvantage compounds the public-contracting disadvantage. Three evidence layers: (1) Wealth/startup-capital channel. Fairlie and Robb (Race and Entrepreneurial Success, MIT Press 2008) find that the single most important factor explaining why Black-owned firms underperform and Asian-owned firms outperform is the level of startup capital, and that low Black startup capital traces directly to low family wealth (roughly half of Black families held under $6,000 in total wealth). Less startup capital means smaller, more undercapitalized, failure-prone firms entering the contracting market. (2) Credit-market denial channel. Federal Reserve econometric work on the National/Survey of Small Business Finances (the lineage relied on in BBC and NERA disparity studies) finds Black-owned firms are roughly twice as likely to be denied credit even after controlling for creditworthiness and other factors; among applicants, denial rates ran about 10.8% for white-male-owned firms versus 14.8% female-owned, 16.5% Hispanic-owned, and 66.6% Black-owned. Disparities also appear in interest rates, collateral, and discouraged-borrower effects (qualified firms not applying because they expect rejection). (3) Persistence in startups. Fairlie, Robb and Robinson ("Black and White: Access to Capital among Minority-Owned Startups," NBER w28154; Management Science 2022) and Robb (SBA Office of Advocacy 2013, Kauffman Firm Survey) show Black-owned startups raise far less external debt, rely more on owner equity, and the gap persists controlling for credit score and net worth, with both creditworthiness and perceived/actual bank treatment contributing. The contracting consequence: undercapitalized diverse firms self-select into smaller bids, cannot bridge progress-payment lags on large awards, fail bonding capacity tests tied to working capital, and grow more slowly into prime-eligible size, producing measurable utilization disparities that a Croson-compliant study must trace to this barrier rather than to firm quality.

How the full study investigates it
Quantitative
  • Replicate the SSBF/Federal Reserve logit/probit credit-outcome regressions on local and regional data: model loan-application, loan-denial, interest-rate, and loan-amount outcomes as functions of owner race/ethnicity/gender while controlling for firm age, size/revenue, industry, profitability, credit score, owner net worth, and local banking-market structure, isolating the residual race/gender effect.
  • Estimate a 'discouraged-borrower' model (firms that needed credit but did not apply because they expected denial) by demographic group, following the SSBF discouragement literature, to capture suppressed demand that pure denial-rate analysis misses.
  • Compute capital-structure ratios by owner group (share of startup and ongoing capital from external debt vs. owner equity vs. personal credit cards/family loans) using KFS-style measures to quantify reliance on costlier/thinner capital.
  • Build a startup-capital gap estimate by group and link it to firm scale (employees, revenue, average contract size pursued) to connect undercapitalization to bid capacity.
  • Bonding-capacity analysis: relate working capital and credit access to surety bonding limits, then test whether bonding/working-capital constraints predict non-participation or smaller bids on City solicitations above given thresholds.
  • Marketplace-conditions regressions per NCHRP 644: include local geographic-market controls (HMDA/CRA lending patterns, branch density, deposit concentration) to test whether local credit-market structure independently explains group disparities.
  • Sensitivity and robustness testing (alternative specifications, creditworthiness proxies, and matched-comparison/decomposition such as Oaxaca-Blinder) to show the disparity is not an artifact of model choice, meeting Croson's strong-basis-in-evidence standard.
Qualitative
  • Owner survey (statistically sampled across the City vendor and availability universe) with a dedicated capital-access battery: applied for credit yes/no, denied yes/no, amount requested vs. received, interest rate and collateral terms, discouragement (needed credit but did not apply), use of personal assets/credit cards, and experience financing a specific City or public contract.
  • In-depth one-on-one interviews with minority- and women-owned firm owners about specific financing episodes, including being unable to bid or scale because they could not bridge progress-payment lags or meet bonding working-capital tests.
  • Structured anecdotal-evidence collection meeting NCHRP 644 and H.B. Rowe verifiability standards: capture corroborable, dated, specific accounts (lender, terms, outcome) so the record survives the Tippett critique that unverified anecdote is weak.
  • Public hearings/community testimony sessions hosted with the City, with targeted prompts on lending and bonding experiences, recorded and transcribed for the evidentiary record.
  • Trade-association and lender-side input: interviews with minority/women chambers and contractor associations, plus CDFIs, community banks, and surety agents serving the region to triangulate firm-reported barriers against supply-side practices.
  • Group-specific qualitative sampling to build a distinct women-owned-firm evidentiary record (responsive to the Fourth Circuit striking the women's goal in H.B. Rowe for thin evidence).
Why it matters in Boston

BBC's 2020 study found Black American-owned firms available for 3.6 percent of the City's contracting dollars but paid 0.4 percent, a disparity index of 12, the deepest gap it measured, with substantial disparities for Black-owned firms across all work types. That is the contracting signature the capital literature predicts: wealth-gated credit converts into missing bids and missing scale before any evaluator scores a proposal. Boston's high-cost market raises the wealth threshold to self-finance a firm, so the local analysis should replicate the SSBF-style credit regressions on HMDA/CRA and SBA lending data for the five RGMA counties (Suffolk, Middlesex, Norfolk, Plymouth, Essex). It should also build a distinct evidentiary record for women-owned firms, whose certified spend share has sat nearly flat at roughly 2.4 to 3.0 percent for eight fiscal years.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Race-conscious contracting remedies face strict scrutiny and require a strong basis in evidence of identified discrimination in the relevant local market; generalized national statistics are insufficient and the jurisdiction must consider race-neutral alternatives first. Capital access is a primary race-neutral lever, so the study must document capital-market discrimination affecting the City's own contracting pool to justify any remedy and to design defensible race-neutral capital programs.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in the First Circuit. Upheld North Carolina's minority subcontracting goals where the state combined statistical disparity evidence with anecdotal evidence of discrimination, but struck the women's goal for an insufficient evidentiary basis. Counsels the study team to pair quantitative capital-access analysis with verified anecdotal capital-access testimony, and to build a group-specific record (including a distinct evidentiary showing for women-owned firms).

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, TRB/National Academies, 2010)NCHRP Report 644 (2010)

    The methodological standard for legally defensible disparity studies. Treats capital-market and credit conditions as a core marketplace-conditions element to be analyzed with regression/econometric methods and corroborated with anecdotal evidence, establishing the accepted blueprint for investigating access to capital as a barrier.

  • 49 C.F.R. 26.51 (USDOT DBE Program, race-neutral measures)49 C.F.R. 26.51(b)

    Enumerates the race-neutral remedy menu Croson requires a jurisdiction to consider, including assistance to overcome inability to obtain financing and bonding, services to help small/DBE firms obtain bonding and financing, and programs to assist new start-up firms. The capital-access barrier maps directly onto these statutory race-neutral remedies, so findings here drive the study's remedy recommendations.

  • Fairlie, R.W. & Robb, A.M., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)MIT Press, 2008 (ISBN 9780262514941)

    Peer-reviewed/academic-press finding that startup capital is the most important determinant of business outcomes and that low Black startup capital is rooted in low family wealth (about half of Black families under $6,000 net worth). Establishes the wealth-to-undercapitalization-to-underperformance causal chain underlying the barrier.

  • Robb, A.M., Access to Capital among Young Firms, Minority-owned Firms, Women-owned Firms, and High-tech Firms (SBA Office of Advocacy, 2013)SBA Office of Advocacy, 2013 (Kauffman Firm Survey)

    Documents that inadequate capital is the major constraint on women- and minority-owned firm growth, that these firms are bank-dependent yet rely disproportionately on owner equity, and that capital-structure disparities relative to white-male/nonminority firms persist among comparable young firms.

  • Fairlie, R.W., Robb, A.M. & Robinson, D.T., Black and White: Access to Capital among Minority-Owned Startups (NBER Working Paper 28154; published Management Science, 2022)NBER WP 28154 / Management Science (2022)

    Finds Black-owned startups raise substantially less external capital, especially external debt, with gaps persisting after controlling for creditworthiness and net worth; attributes the gap to both differences in credit quality and perceived/actual differential treatment by banks. Confirms the disparity is not merely a fundamentals story.

  • Cavalluzzo & Wolken / Federal Reserve analyses of the (National) Survey of Small Business Finances (NSSBF/SSBF)NSSBF/SSBF credit-market studies (Federal Reserve)

    The data-and-method foundation used in BBC and NERA disparity studies: regression analysis of loan application, denial, interest-rate, and discouraged-borrower outcomes showing Black-owned firms about twice as likely to be denied credit controlling for creditworthiness, with elevated denial rates for women-, Hispanic-, and Black-owned firms. Provides the national econometric template the local study replicates and corroborates.

Boston data sources
  • Analyze Boston 'City of Boston Contract Award' and 'City of Boston Discretionary Spending' datasets (FY2019-FY2026, with the City's certification flags) to define the local firm universe and link respondents to actual contracting behavior.
  • City of Boston and Massachusetts Supplier Diversity Office (SDO) certification directories for the certified-firm universe.
  • Federal Reserve Survey of Small Business Finances / National Survey of Small Business Finances (SSBF/NSSBF) for the national benchmark model and variable design.
  • Kauffman Firm Survey (KFS) for startup-capital and capital-structure benchmarks by owner demographics.
  • SBA 7(a) and 504 lending data for the Boston market.
  • Home Mortgage Disclosure Act (HMDA) and FFIEC CRA small-business lending data for local credit-market structure and lending patterns across the five RGMA counties.
  • FDIC Summary of Deposits branch/deposit data for geographic banking-market controls.
  • Federal Reserve Small Business Credit Survey (SBCS) for recent denial, terms, and discouragement rates by race and gender.
  • Surety/bonding capacity data and SBA bonding-program data to model the working-capital-to-bonding link.
  • Primary survey and interview data collected by the study (owner survey, interviews, hearing transcripts), alongside the 2020 BBC study's anecdotal appendix.
  • U.S. Census ABS 2022 and NES-D 2022 for CBSA 14460 and the five RGMA counties for owner demographics, firm counts, and baselines.
04

Insurance Requirements (general-liability, professional-liability, workers'-compensation, and auto-coverage minimums, often paired with bonding)

Bench lead

Access-to-capital economist (small-business finance / labor economist), leading the regression and premium-burden quantitative work in the Wainwright/Fairlie-Robb tradition, working jointly with the legal scholar/disparity-study counsel who ensures the anecdotal-evidence record and remedy-mapping satisfy Croson and H.B. Rowe; the survey methodologist designs and fields the owner-survey insurance items.

Insurance requirements are the coverage minimums a buyer attaches to a solicitation or contract that a firm must carry, and prove, to bid and to perform. In public construction and services contracting they typically include: commercial general-liability (CGL) limits (commonly $1M per occurrence / $2M aggregate, sometimes higher with the buyer named as additional insured); professional-liability / errors-and-omissions (E&O) coverage for design, A/E, IT, and consulting work; workers'-compensation and employer's-liability coverage mandated for any firm with employees; and commercial auto liability for firms operating vehicles. These are distinct from surety bonds (bid, performance, payment) but are routinely bundled with bonding in the same "risk-transfer / responsibility" clause of an RFP, so the two barriers appear together in the anecdotal record. A firm that cannot obtain the required limits, or cannot afford the premiums and the up-front/escrow cash they demand, is effectively non-responsive and cannot compete for the work regardless of price or technical merit. The barrier has two faces: an access face (an underwriter will not write the coverage at the required limit for a thin-balance-sheet firm) and an affordability/cash-flow face (the firm can obtain coverage but the premium, deposit, and additional-insured/waiver-of-subrogation endorsements consume scarce working capital that larger competitors do not have to tie up).

Why it is a barrier

Insurance minimums function as a fixed, capital-intensive cost of entry that is largely invariant to contract size, so they fall hardest on small, newer, and lower-net-worth firms, which disproportionately are minority- and women-owned. Premiums are priced on the firm's loss history, payroll, revenue, years in business, and balance-sheet strength; emerging diverse firms score worse on every one of those underwriting inputs not because they are riskier operators but because they are younger and thinner-capitalized, the same conditions documented in the access-to-capital literature. Fairlie & Robb (Race and Entrepreneurial Success, MIT Press 2008) show that Black-owned firms start with markedly less capital and lower owner wealth (nearly half of Black families hold under $6,000 in total wealth), and that low startup capital is the single most important driver of weaker business outcomes; that same capital and wealth gap is what makes the up-front premium, deposit, and additional-insured/escrow demands of an insurance clause a binding constraint for diverse firms while being a routine line item for incumbents. Disparity studies repeatedly capture this in the anecdotal record: a named Colette Holt & Associates study (Harris Health System, 2022) identifies "overly burdensome insurance and/or bonding requirements" (p.54) and "strict bonding and insurance requirements" (p.59) as marketplace barriers, and recommends "developing programs to address issues of financing, bonding, and insurance important to all small and emerging businesses" (p.55). BBC Research & Consulting and other practitioners document the same pattern of unequal access to financing, bonding, and insurance across jurisdictions. The federal DBE rule itself treats inability to obtain bonding/financing as a recognized, remediable limitation in 49 C.F.R. 26.51(b)(2). The disadvantage is also structural: because the limits are flat dollar thresholds rather than scaled to contract value, a $1M CGL requirement on a $50,000 job screens out a capable small firm while imposing no real friction on a large one, narrowing the diverse-firm-eligible pool before price competition ever begins.

How the full study investigates it
Quantitative
  • Disparity-ratio analysis on insurance-sensitive industries: compute availability vs. utilization (disparity index = utilization % / availability %, with 80% as the conventional substantial-disparity threshold) for the City of Boston's contracts, restricted to NAICS codes where insurance minimums are heaviest (construction, A/E, IT/professional services with E&O, transportation/auto-exposed services), to test whether high-insurance work shows larger diverse-firm shortfalls than low-insurance work.
  • Logistic/linear regression of business-formation and business-outcome models (following NCHRP 644 / Wainwright methodology): regress firm formation, self-employment, sales, and earnings on race, gender, and ethnicity while controlling for non-discriminatory factors (owner age, education, industry, firm age, capital, geography); a residual race/gender penalty after controls is the statistical signature of barriers including capital/insurance access.
  • Bid-threshold / responsiveness analysis: tabulate City of Boston solicitations by required CGL, E&O, workers'-comp, and auto limits, and measure how the certified MWBE/SBE-eligible bidder pool contracts as required limits rise; estimate the share of solicitations whose flat insurance minimums exceed what newly-certified or sub-$X-revenue firms can plausibly carry.
  • Premium-burden modeling: estimate insurance premiums plus deposits and endorsement costs as a percentage of contract value and of firm working capital across firm-size bands, to quantify how the fixed cost is regressive against small/diverse firms (parallels the surety-premium-as-share-of-contract-value approach used for bonding, where premiums commonly run ~0.5%-2%+).
  • Marketplace-conditions / credit-and-capital quantitative analysis: use Survey of Business Owners / Annual Business Survey and Small Business Credit Survey microdata to compare, by race and gender, the incidence of being denied or unable to afford required insurance and bonding, controlling for firm characteristics.
  • Decomposition (Oaxaca-Blinder-style) of the diverse vs. non-diverse outcome gap to apportion how much of the gap is explained by capital/insurance/bonding access versus unexplained residual discrimination.
Qualitative
  • Owner survey instrument with dedicated insurance items: a citywide survey of certified and bid-eligible firms asking whether the respondent was ever unable to bid because it could not obtain or afford the required CGL, E&O, workers'-comp, or auto coverage; how additional-insured, waiver-of-subrogation, and limit-escalation endorsements affected cost; and whether premiums/deposits strained working capital, with results disaggregated by race, gender, and firm age (the NCHRP 644 / Croson-compliant anecdotal-evidence approach).
  • In-depth interviews with diverse business owners (the Colette Holt / H.B. Rowe interview model) capturing first-person, attributable accounts of insurance-driven exclusion, walking through specific lost solicitations; these provide the group-specific anecdotal evidence H.B. Rowe found necessary.
  • Public hearings / community testimony sessions where owners, prime contractors, and trade groups give on-the-record statements about insurance and bonding obstacles, creating a contemporaneous evidentiary record for the City.
  • Trade-association and stakeholder input from Greater Boston construction, A/E, and minority and women business associations, plus the Mayor's Office of Economic Opportunity and Inclusion and Supplier Diversity Program stakeholders, on industry-standard limits and observed disparities.
  • Interviews with the supply side of the barrier: insurance agents/brokers, underwriters, and surety producers serving Boston-area small firms, to document underwriting practices and whether diverse firms face differential access or pricing.
  • Prime-contractor interviews on how they treat sub-tier insurance flow-down requirements, to test whether prime-imposed (not just City-imposed) insurance clauses screen out diverse subcontractors.
Why it matters in Boston

Boston's certified-MWBE spend share has risen from 5.2 percent in FY2019 to 12.3 percent in FY2025, but it still sits below BBC's 16.9 percent availability benchmark and well below the 25 percent executive-order goal, and the WBE share of certified spend has been nearly flat at roughly 2.4 to 3.0 percent for eight fiscal years. Flat-dollar insurance minimums on small-dollar work are one screen that can hold smaller certified firms out even as outreach improves, because the limits do not scale to contract value. The update should code required limits and endorsements across the City's solicitation record and test whether high-limit work shows wider group-specific gaps than low-limit work, then map confirmed findings to scaled limits and technical-assistance remedies before any race-conscious step, consistent with Croson and the post-SFFA climate.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Sets the strict-scrutiny / strong-basis-in-evidence standard for state and local race-conscious contracting remedies. The local jurisdiction must show specific evidence of discrimination in its own market, not generalized societal discrimination. Establishes the evidentiary architecture (the jurisdiction must consider race-neutral alternatives, and the program must rest on identified barriers in the relevant market) that requires a disparity study to document concrete barriers like insurance/bonding access rather than assert them, and to pair statistical disparity with market-specific evidence.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit precedent, not controlling in the First Circuit. Upheld North Carolina's race-conscious subcontracting goals for African American and Native American firms where the disparity study combined statistical disparities with interviews of minority subcontractors, but struck the program as to groups lacking that evidentiary support. Confirms that anecdotal/qualitative barrier evidence (owner interviews documenting obstacles such as bonding and insurance) is a necessary, group-specific complement to the statistical record, and that remedies must be tailored to where both kinds of evidence exist.

  • 49 C.F.R. 26.51(b)(2) (USDOT DBE program, race-neutral remedy menu)49 C.F.R. 26.51(b)(2)

    Race-neutral remedy directly on point: recipients must provide 'assistance in overcoming limitations such as inability to obtain bonding or financing (e.g., by such means as simplifying the bonding process, reducing bonding requirements, eliminating the impact of surety costs from bids, and providing services to help DBEs, and other small businesses, obtain bonding and financing).' Federal recognition that inability to obtain bonding/financing is a remediable barrier and that the first-line remedy is race-neutral. The surrounding (b) menu (unbundling solicitations, technical assistance, supportive-services programs) supplies the recommended race-neutral remedies the study should map insurance/bonding barriers to.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, National Academies / TRB, 2010)NCHRP Report 644 (2010)

    The canonical methodological standard for legally defensible disparity studies. Directs that barriers in capital, bonding, and insurance markets be examined as part of the 'marketplace conditions' analysis, using both statistical (regression of business formation and earnings/outcomes controlling for non-discriminatory factors) and anecdotal/qualitative evidence, so the study can satisfy Croson's strong-basis-in-evidence requirement. Co-authored by Jon Wainwright (NERA) and Colette Holt.

  • Colette Holt & Associates, Harris Health System Disparity Study (2022)Harris Health System Disparity Study (2022), pp.54-55, 59

    Named, recent disparity study that expressly identifies 'overly burdensome insurance and/or bonding requirements' (p.54) and 'strict bonding and insurance requirements' (p.59) as marketplace barriers, and recommends race-neutral programs 'to address issues of financing, bonding, and insurance important to all small and emerging businesses' (p.55). Direct precedent that insurance requirements (alongside bonding) are a recognized, study-documented barrier and a target for race-neutral remediation.

  • BBC Research & Consulting, disparity-study practice (e.g., Caltrans Availability and Disparity Study; City of San Diego 2020; City of Boston 2020)BBC Research & Consulting disparity studies (2020)

    Leading practitioner whose studies document quantitative and anecdotal evidence of unequal access to financing, bonding, and insurance for minority- and women-owned firms across jurisdictions, corroborating insurance/bonding as a cross-market barrier and modeling the marketplace-conditions and anecdotal-evidence methodology.

  • Fairlie, R.W. & Robb, A.M., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (2008)

    Peer-reviewed access-to-capital evidence: low startup capital and low owner wealth (nearly half of Black families hold under $6,000 in total wealth) are the dominant drivers of weaker minority business outcomes. Supplies the causal mechanism for why insurance premiums, deposits, and additional-insured/escrow demands bind diverse firms (thin balance sheets and weak loss-history/payroll/tenure underwriting inputs) while being routine for incumbents.

Boston data sources
  • City of Boston solicitation documents from the Supplier Portal and City contract files, including the insurance requirement clauses attached to each (to link required limits to bidder pools and award outcomes).
  • Analyze Boston 'City of Boston Contract Award' dataset (FY2019-FY2026, with certification flags) for award outcomes by certified status.
  • City of Boston and Massachusetts SDO certification directories to define the available pool of certified diverse firms by industry and size.
  • U.S. Census ABS 2022 and NES-D 2022 for CBSA 14460 and the five RGMA counties for firm counts, owner demographics, firm age, and revenue by industry (availability denominators and regression controls).
  • Federal Reserve Small Business Credit Survey for race and gender patterns in capital, bonding, and insurance access.
  • Primary-collected data: the owner survey responses, interview transcripts, and public-hearing testimony generated by this study, alongside the 2020 BBC study's anecdotal appendix.
  • Insurance/surety market data: industry premium-rate references and broker/underwriter interview notes to model premium burden and access.
  • The City's standard insurance and indemnification contract language and required limits, from City contract files, to document the actual limits and endorsements imposed.
10

Slow or Non-Prompt Payment and Cash-Flow Strain

Bench lead

Access-to-capital / small-business-finance economist, leading the quantitative payment-cycle, retainage-burden, and survival modeling and the cost-of-capital differential; partnered with a survey methodologist for the owner-survey payment module and a procurement/construction-law scholar to anchor the M.G.L. c. 30 payment statutes and 49 CFR 26.29/26.51 benchmarks. The finance economist leads because the barrier's mechanism (working-capital deprivation interacting with unequal credit access and wealth, per Fairlie & Robb) is fundamentally a capital-markets problem.

The barrier arises when payment to a firm lags far behind the work it has performed. It has three linked forms in public construction and services contracting. (1) Slow agency-to-vendor and prime-to-subcontractor payment: the public body pays the prime late, or the prime, once paid, does not pass funds down promptly. (2) Retainage: the owner or prime withholds a fixed percentage (commonly 5-10%) of every progress payment until final completion, so a firm never receives full value for completed work until the very end of a job that may run a year or more. (3) Pay-when-paid and pay-if-paid clauses: subcontract terms that make the prime's receipt of payment from the owner a condition of paying the sub, shifting the entire risk of owner delay or non-payment onto the smallest, least-capitalized firms. Because subcontractors must front labor, materials, equipment, insurance, and payroll taxes long before they are reimbursed, the gap between outlay and collection must be bridged with working capital. Firms with deep cash reserves or cheap lines of credit absorb the gap easily; thinly capitalized firms cannot, and a delayed payment can force them to stop work, miss payroll, default on a job, or decline future work they could otherwise perform.

Why it is a barrier

Payment delay is not race- or gender-neutral in effect because the ability to survive it depends on accumulated wealth and access to credit, both of which are sharply unequal. Fairlie and Robb (MIT Press, 2008) document that the single most important determinant of business survival and growth is startup and working capital, and that Black-owned firms are systematically undercapitalized because nearly half of Black families hold very little total wealth, leaving owners unable to self-finance a payment gap or to pledge personal assets the way better-capitalized competitors can. The access-to-capital literature shows minority owners are more likely to be denied conventional bank credit and to pay higher rates when approved, so the line of credit that lets a well-capitalized prime ride out a 60-day delay is either unavailable or far more expensive to a diverse firm. The result is a cascade documented in disparity-study anecdotal records: the U.S. Commission on Civil Rights found contractors intentionally delayed payment 60 days or more, that 10% monthly retainage starved firms of completed-work value, and that owners had to pledge personal credit or mortgage their homes to bridge the gap. When the gap cannot be bridged, the firm leaves the job, which both removes a present contract and damages the past-performance record needed to win the next one, compounding exclusion. Because the same delay that is a nuisance to a large prime is existential to an emerging firm, slow payment operates as a selective filter that removes diverse and emerging firms from the pool even where they are otherwise qualified and competitively priced.

How the full study investigates it
Quantitative
  • Payment-cycle-time analysis: pull agency accounts-payable/ERP records (invoice date, approval date, payment-issued date, payment-clearing date) for a multi-year study period and compute days-to-pay distributions; test for statistically significant differences in mean and median days-to-pay between certified diverse/emerging vendors and non-diverse vendors, controlling for contract type, dollar size, department, and CLIN/work type (regression with vendor-classification indicator).
  • Retainage-exposure modeling: from contract and progress-payment records, compute retainage withheld as a share of cumulative billings and the average duration funds were withheld per firm class; estimate the implicit working-capital cost (retained dollars x duration x cost-of-capital) and test whether that burden falls disproportionately on smaller/diverse firms.
  • Prime-to-sub pass-through timing: where subcontract payment data are available (DBE/prompt-pay reporting, certified payrolls, B2GNow or equivalent compliance system), measure the lag between prime receipt of payment and sub payment, and the share of payments exceeding the Massachusetts statutory and 30-day USDOT benchmarks, disaggregated by sub ownership classification.
  • Working-capital-adjusted availability/utilization: integrate a capital-constraint covariate into the standard availability and disparity-index calculation so the study can estimate how much of the measured utilization gap is associated with capitalization and payment-cycle exposure rather than availability alone.
  • Survival/attrition analysis: using vendor registration plus award histories, model whether diverse/emerging firms that experience longer payment cycles or larger retainage exposure are more likely to exit the bidder pool, stop bidding, or fail to re-bid (logit/hazard model with payment-delay as a predictor).
  • Marketplace cost-of-capital differential: combine survey-reported credit-access and interest-rate data with the measured payment-cycle length to quantify the dollar cost a payment delay imposes on a diverse firm versus a non-diverse firm of similar size, monetizing the disparate burden.
  • Benchmark/compliance gap: tabulate the share of City contracts and subcontracts that meet versus miss the M.G.L. c. 30, §§ 39F and 39K and 30-day 49 CFR 26.29 standards, establishing the size of the enforceable gap a remedy would close.
Qualitative
  • Diverse/emerging-firm owner survey with dedicated payment-experience module: structured questions on typical days-to-payment from the City and from primes, frequency of payments exceeding 30/60/90 days, retainage percentages and release timing, exposure to pay-if-paid/pay-when-paid language, and the operational consequences (missed payroll, declined work, stopped work, layoffs, personal credit pledged). Capture firm classification and size to permit cross-tabulation with the quantitative results.
  • In-depth one-on-one interviews with subcontractors and small primes that exited or never re-entered the Boston market, to capture attrition stories the award data cannot show and to test the survival-analysis findings against lived accounts (Croson-grade specificity: who, when, which contract, what dollar impact).
  • Prime-contractor and key-personnel interviews on pass-through payment practices, retainage policy, and how prompt-pay obligations are monitored in practice, to corroborate the prime-to-sub timing data and surface informal practices.
  • Public hearing / sworn-testimony sessions advertised through chambers, faith-based, and ethnic business networks, providing an on-the-record forum for owners to describe payment-driven cash-flow failures; this satisfies the Croson preference for investigated, attributable anecdotes.
  • Trade-association and chamber input sessions (e.g., regional AGC/ABC chapters, NAMC, women's and minority chambers, Hispanic/Asian/Black business associations, regional minority-supplier development council) to gather pattern evidence and identify whether payment terms vary systematically by firm type.
  • Lender and surety interviews to confirm the credit-access mechanism: whether diverse firms are offered fewer or costlier lines of credit to bridge payment gaps, linking the anecdotal payment strain to the access-to-capital barrier.
  • Anecdote verification protocol: for each significant payment account, attempt corroboration against contract records, payment logs, or third parties before relying on it, consistent with NCHRP 644 and the case law's demand for investigated anecdotes.
Why it matters in Boston

Massachusetts already supplies the enforceable yardstick: M.G.L. c. 30, § 39K sets periodic-payment deadlines with interest and caps retainage at 5 percent on public building construction, and § 39F gives subcontractors a direct-payment remedy against the awarding authority. The empirical base is unusually good: the public Discretionary Spending extract holds over one million City payment lines for FY2019-FY2026, so days-to-pay distributions by certified status are computable at scale rather than sampled. Because thin capitalization is precisely what the record associates with the groups showing the deepest disparities (Black American-owned index 12 in 2020), payment-cycle findings connect the City's own administrative behavior to the utilization gap, and enforcement-grade remedies (faster cycles, retainage discipline, mobilization payments) require no constitutional showing at all.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469, 509 (1989)

    Establishes that a race-conscious remedy needs a strong basis in evidence, and that evidence of a pattern of individual discriminatory acts can, if supported by appropriate statistical proof, support broader remedial relief. This is the doctrinal warrant for pairing payment-delay anecdotes with statistical proof of disparate impact, and for documenting payment barriers in order to justify and calibrate any remedy.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in the First Circuit. Upheld North Carolina DOT's minority subcontracting goals as to African American and Native American firms where supported by statistical disparity plus corroborating anecdotal evidence, while striking the program as to groups lacking such proof. Confirms that the study must build group-specific, evidence-backed records (including payment-related barriers) and cannot rely on generalized assertions.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board / National Academies, 2010)NCHRP Report 644 (TRB 2010)

    The National Academies methodological standard for legally defensible disparity studies. Directs studies to examine business-formation and capacity conditions in the marketplace, including access to capital, bonding, and financing, and to combine availability/utilization statistics with systematically collected anecdotal evidence. Frames working-capital and payment barriers as part of the marketplace conditions a defensible study must measure.

  • 49 C.F.R. 26.29 (USDOT DBE Program, Prompt Payment and Return of Retainage)49 C.F.R. 26.29(a)-(f)

    Federal mandate requiring prime-to-sub payment within 30 days of the prime's receipt of payment, prompt return of retainage within 30 days of satisfactory completion, proactive monitoring of compliance (express statement that reliance on subcontractor complaints is insufficient), enforcement with penalties, and application to all lower-tier subs. Supplies the regulatory benchmark against which the City's contract terms and actual payment timing can be measured and the remedy template.

  • 49 C.F.R. 26.51(b) (USDOT DBE Program, race-neutral means)49 C.F.R. 26.51(b)

    The race-neutral remedy menu. Lists assistance in overcoming inability to obtain bonding or financing, financing support, technical/business-development assistance, and contract structuring as race-neutral measures recipients must use to the maximum feasible extent. Directly supports race-neutral remedies for cash-flow strain: prompt-payment enforcement, retainage reduction, mobilization payments, and working-capital/financing assistance.

  • Fairlie, R.W. & Robb, A.M., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (2008)

    Peer-reviewed economic evidence that startup and working capital is the most important determinant of business outcomes, and that Black-owned firms are undercapitalized largely because of the racial wealth gap. Establishes the causal mechanism linking payment delay to disparate firm survival: undercapitalized firms cannot bridge the outlay-to-collection gap.

  • U.S. Commission on Civil Rights, Barriers Facing Minority- and Women-Owned Businesses in Pennsylvania (1999)U.S. Comm'n on Civil Rights, PA Advisory Committee (1999)

    Documents the specific payment cascade in M/WBE owners' own words: contractors intentionally delaying payment 60 days or more, 10% monthly retainage withholding completed-work value, and owners forced to pledge personal credit or mortgage homes to bridge cash-flow gaps, with bonding and capital constraints compounding the strain. A confirmable precedent for the anecdotal pattern this study will test in Boston.

  • Massachusetts public construction payment statutes, M.G.L. c. 30, § 39K (periodic payments; retainage) and § 39F (subcontractor direct payment)M.G.L. c. 30, §§ 39F, 39K

    The Massachusetts legal baseline for public building construction: § 39K sets enforceable periodic-payment deadlines with interest on late payment and caps retainage at 5 percent, and § 39F gives a subcontractor a direct-payment remedy against the awarding authority when a general contractor fails to pay for completed work. Sets the statutory standard against which the City's actual payment behavior and contract terms can be benchmarked.

Boston data sources
  • City of Boston accounts-payable and procurement payment records (invoice receipt, approval, and payment-issuance dates), obtainable from the City, with the Analyze Boston 'City of Boston Discretionary Spending' dataset (FY2019-FY2026, over one million payment lines) for payment-pattern analysis at scale.
  • Contract files and progress-payment/retainage schedules showing percentage withheld and release dates, from City contract files.
  • Subcontractor prompt-payment compliance records where the City maintains them.
  • City of Boston and Massachusetts SDO certification directories, with firm-size indicators, for ownership classification.
  • Award and bid-history data from the Analyze Boston contract-award dataset and City records, to construct the bidder pool and measure re-bid and attrition behavior.
  • Primary survey microdata from the diverse/emerging-firm payment-experience module.
  • Interview and public-hearing transcripts (owners, primes, lenders, sureties, associations).
  • M.G.L. c. 30, §§ 39F and 39K statutory text and the City's standard contract terms and conditions (to establish the contractual prompt-pay and retainage baseline).
  • External credit-access benchmarks (Federal Reserve Small Business Credit Survey; SBA lending data) to contextualize cost-of-capital differentials.
  • NCHRP Report 644 and USDOT DBE prompt-payment/retainage regulatory benchmarks (49 CFR 26.29, 26.51) as methodological and compliance reference points.

Barrier family

How the buy is structured

Contract size, bundling, and lower-tier rules that decide whether smaller firms ever see a piece they can win.

05

Contract Bundling and Large Contract Size

Bench lead

Econometrician / disparity-study quantitative lead (NERA/Wainwright-style applied microeconomist) owns the contract-size distribution, prime-vs-sub disparity ratios, capacity-conditioned availability, and regression/simulation work; partnered with the access-to-capital economist (Fairlie & Robb tradition) on the bonding and working-capital constraint analysis, and with disparity-study legal counsel (Colette Holt tradition) to keep the unbundling remedy narrowly tailored under Croson and H.B. Rowe.

The practice of consolidating multiple smaller scopes of work into one large solicitation, or structuring prime contracts at a dollar value, duration, geographic scope, or bonding/insurance level that exceeds the capacity of small, diverse, and emerging firms. When a jurisdiction aggregates requirements that could be procured separately (for example, combining design, construction, maintenance, and supply into a single megaproject, or rolling many municipal facilities into one umbrella services contract), it shrinks the universe of firms that can bid as a prime. Diverse and emerging firms are then structurally confined to subcontractor roles, if they participate at all. The federal right-sizing remedy is codified at 49 C.F.R. 26.51(b)(1), which directs recipients to arrange "solicitations, times for the presentation of bids, quantities, specifications, and delivery schedules in ways that facilitate participation by DBEs and other small businesses," cross-referencing 49 C.F.R. 26.39, whose program element requires eliminating "unnecessary and unjustified bundling of contract requirements that may preclude small business participation in procurements as prime contractors or subcontractors."

Why it is a barrier

Large contract size mechanically caps who can compete as a prime. Three reinforcing effects disadvantage diverse and emerging firms. First, scale exceeds working-capital and bonding capacity: a firm cannot bond a $20M contract on a balance sheet sized for $2M projects, and minority firms are systematically undercapitalized at formation, so they cannot float the cash flow a large prime contract demands before progress payments arrive (Fairlie and Robb document that lower levels of startup capital, traceable to large racial wealth gaps, are a primary driver of weaker minority-firm outcomes). Second, bundling forecloses the prime role and pushes diverse firms into lower-value subcontracting. The White House Council of Economic Advisers, drawing on Wainwright (2022) data for 2001-2019, found the average prime contract was worth roughly nine times the average subcontract, and that minority-owned firms are relatively less likely to receive prime contracts (disparity ratio 0.51) and relatively more likely to receive subcontracts (0.66). Confinement to subcontracting caps revenue, profit margin, past-performance accumulation, and the bonding history needed to ever scale up to prime work, producing a self-perpetuating ceiling. Third, large bundles raise non-price qualification hurdles (minimum revenue thresholds, multi-year past performance on contracts of similar size, large insurance limits) that emerging firms cannot yet meet, so they are screened out at responsibility review before price is even considered.

How the full study investigates it
Quantitative
  • Contract-size distribution analysis: build the size distribution of City of Boston prime contracts and subcontracts (count and dollars) by NAICS/industry, and compute the share of total procurement dollars carried by contracts above ascending size thresholds (e.g., >$100K, >$500K, >$1M, >$5M) to measure how much of the market is effectively closed to small-firm primes.
  • Prime-versus-subcontract disparity ratios: following NCHRP 644, compute separate availability-vs-utilization disparity ratios for prime contracts and for subcontracts, by minority group and by gender, and test whether diverse firms are systematically displaced from primes into lower-value subcontracts (replicating the CEA 0.51-prime/0.66-sub pattern at the Boston level).
  • Mean/median contract-value comparison: compare average and median award values won by diverse versus non-diverse firms, overall and within the same NAICS, to quantify a Boston-specific prime-to-subcontract value gap analogous to the ~9x national figure.
  • Capacity-conditioned availability: construct availability counts at multiple capacity tiers (firm revenue, employment, bonding limits) to show how the pool of qualified diverse primes collapses as contract size rises, isolating size as the constraint rather than firm scarcity.
  • Multivariate regression on win probability and award size: regress probability of winning a prime (and log award value) on firm race/ethnicity and gender controlling for firm capacity, age, NAICS, bonding, and past performance, to test whether diverse ownership predicts confinement to smaller awards after accounting for legitimate capacity differences.
  • Counterfactual unbundling simulation: model how prime-eligible diverse-firm availability and projected utilization would change if specified large bundles were decomposed into right-sized packages, quantifying the remedy's expected effect (supports the narrow-tailoring record under Croson).
Qualitative
  • Business-owner survey with targeted contract-size items: ask diverse and emerging firm owners whether specific City of Boston solicitations were too large to bid as prime, whether bundling pushed them into subcontracting, and what right-sized package they could have led; quantify responses for the anecdotal record.
  • In-depth owner interviews: structured interviews with diverse prime-capable firms and with firms stuck in subcontracting, capturing concrete instances of bundled scopes, unmet bonding/working-capital thresholds, and lost prime opportunities (NCHRP 644 anecdotal-evidence protocol).
  • Public hearings / sworn testimony: solicit on-the-record accounts from firm owners about specific oversized City procurements, building the jurisdiction-specific evidence Croson and H.B. Rowe require.
  • Trade-association and chamber input: structured input from minority/women contractor associations, regional Hispanic, Asian American, and Black chambers, and small-business/construction trade groups on bundling patterns in Greater Boston public work.
  • Prime-contractor and procurement-staff interviews: interview large primes and City procurement officers on why scopes are bundled, whether unbundling was considered, and what right-sizing is operationally feasible (tests the 26.39 'unnecessary and unjustified' standard).
  • Anecdote corroboration: cross-reference each contract-size anecdote against award records to corroborate it, per NCHRP 644's guidance on weighting anecdotal evidence.
Why it matters in Boston

BBC analyzed more than 47,000 City contracts for FY2015-FY2019 and found MWBE disparity indices of 65 on prime work and 64 on subcontract work, so the displacement question in Boston is not sub-versus-prime alone but why both tiers under-deliver. The public FY2019-FY2026 award file (22,054 contracts of $10,000 or more on Analyze Boston) supports a direct contract-size-band decomposition: if certified-firm share collapses as award size rises, bundling is doing measurable work. The Sheltered Market Program (M.G.L. c. 30B, § 18) is the City's existing right-sizing instrument, designating contracts that certified disadvantaged vendors can win as primes, and the counterfactual unbundling simulation can quantify how much certified prime participation additional designations would open.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Subjects state and local race-based contracting remedies to strict scrutiny; remedies must rest on specific, identified discrimination within the jurisdiction (not general societal discrimination) and must be narrowly tailored, including serious consideration of race-neutral alternatives. The Court faulted Richmond for, among other things, not investigating race-neutral methods. Right-sizing/unbundling is a leading race-neutral alternative a Croson-compliant study must evaluate, making contract-size analysis a strict-scrutiny necessity, not an option.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in the First Circuit. Upheld NCDOT's M/WBE subcontracting remedy for African American and Native American firms where a disparity study supported it, but struck it down for Hispanic, Asian, and women-owned firms for lack of evidence. Confirms a defensible remedy must be group-by-group and evidence-specific, and that disparities in subcontracting versus prime participation are legally cognizable and separately provable.

  • 49 C.F.R. 26.51(b)(1)49 C.F.R. 26.51(b)(1)

    The race-neutral remedy menu. Verbatim: arranging 'solicitations, times for the presentation of bids, quantities, specifications, and delivery schedules in ways that facilitate participation by DBEs and other small businesses and by making contracts more accessible to small businesses, by means such as those provided under 26.39.' This is the direct regulatory basis for the unbundling/right-sizing remedy.

  • 49 C.F.R. 26.3949 C.F.R. 26.39

    Cross-referenced 'Fostering small business participation' element. Requires recipients to structure contracting requirements to facilitate small-business competition and to take all reasonable steps to eliminate 'unnecessary and unjustified bundling of contract requirements that may preclude small business participation in procurements as prime contractors or subcontractors.' Cites concrete strategies: small-business set-asides for primes under a stated amount (e.g., $1M) and, on megaprojects/design-build, requiring bidders to specify subcontract elements sized for small firms.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board / National Academies, 2010)NCHRP Report 644 (TRB 2010)

    The National Academies methodological standard for legally defensible disparity studies (NERA prime; Colette Holt & Associates subcontractor; Wainwright and Holt co-PIs). Supplies the model scope of work, marketplace-conditions/business-formation analysis, regression on firm earnings/formation, availability/utilization disparity-ratio construction, and the anecdotal-evidence protocol that a Croson/Rowe-compliant barriers analysis (including contract-size effects) must follow.

  • Council of Economic Advisers, 'Racial Disparities in Government Contracting' (The White House, Dec. 20, 2024), citing Wainwright (2022)CEA, Racial Disparities in Government Contracting (2024)

    Quantifies the prime/subcontract size gap: average prime contract ~9x average subcontract (2001-2019); minority-owned firms relatively less likely to win primes (disparity ratio 0.51) and more likely to be subcontractors (0.66), with group breakdowns (Black 0.46 prime/0.58 sub; Hispanic 0.40/0.71; Asian 0.57/0.78; Native American 0.42/0.56).

  • Robert W. Fairlie & Alicia M. Robb, Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (MIT Press 2008)

    Foundational access-to-capital evidence. Using the Census Characteristics of Business Owners (CBO) data, finds low startup capital is the single most important factor in minority-firm underperformance, rooted in large racial wealth gaps. Explains why diverse firms cannot self-finance the cash-flow and bonding demands of large bundled prime contracts.

  • City of Fort Worth, Texas Disparity Study 2020 (Colette Holt & Associates) and 2014 State of Missouri Office of Administration Disparity Study (Colette Holt & Associates)CHA, Fort Worth (2020); Missouri (2014)

    Named CHA disparity studies establishing the firm's NCHRP 644-aligned method (prime vs. subcontract disparity ratios, regression, owner/anecdotal evidence) and its practice of recommending race- and gender-neutral remedies, including right-sizing and small-business contract structuring, where the data support them.

Boston data sources
  • Analyze Boston 'City of Boston Contract Award' dataset (FY2019-FY2026, contracts of $10,000 or more): award value, duration, department, and certification flags for the contract-size distribution.
  • City subcontract and prime-reported utilization records where maintained, obtainable through the Procurement Department, to measure prime-versus-sub participation and the value gap.
  • Bonding and insurance requirement records by solicitation from Supplier Portal documents and City contract files, to map capacity hurdles against contract size.
  • City of Boston Supplier Portal solicitation records, to reconstruct how scopes were bundled and whether unbundling alternatives were considered.
  • City of Boston and Massachusetts SDO certification directories, plus the 2020 BBC availability database, to build the availability pool by capacity tier.
  • U.S. Census ABS 2022, NES-D 2022, and County Business Patterns for firm counts, size, and formation by race, ethnicity, and gender in the Boston market area.
  • Surety/bonding capacity benchmarks and the access-to-capital literature (Fairlie & Robb; Federal Reserve Small Business Credit Survey) to contextualize working-capital and bonding constraints.
  • Anecdotal record: owner surveys, interview transcripts, and public-hearing testimony collected by the study team.
11

Subcontractor / lower-tier utilization and good-faith-effort enforcement

Bench lead

Econometrician / labor economist specializing in disparity-ratio and regression analysis (the NERA / Jon Wainwright model), leading the quantitative subcontract-dollar disparity and CUF/pass-through analysis, working in tandem with a disparity-study legal scholar (the Colette Holt model) who owns the Croson / H.B. Rowe narrow-tailoring and good-faith-effort legal-sufficiency framing, and supported by a survey methodologist for the owner-survey/interview instruments.

This barrier covers what happens to minority-, women-, and other diverse-owned firms below the prime-contract tier, where most public-works dollars actually flow. It has four linked failure modes: (1) weak enforcement of subcontracting participation goals, so primes treat goals as aspirational rather than binding; (2) relegation of M/WBEs to low-value, low-margin scopes (trucking, hauling, supply, flagging, landscaping) instead of skilled or high-dollar trades, so headcount or "number of firms" parity coexists with deep dollar disparity; (3) commercially-useful-function (CUF) and pass-through abuses, where a certified firm is listed for goal credit but a non-diverse firm performs the real work, the diverse firm acts as a broker or "rent-a-cert," or supply/lease arrangements are mis-counted; and (4) lax good-faith-effort (GFE) review, where a prime that misses the goal is excused on a pro forma showing without the agency scrutinizing whether the prime genuinely solicited, negotiated with, and selected available diverse subs. In plain terms: the program can look like it is working at the headcount level while diverse firms are systematically confined to the cheap end of the work and shut out of the dollars.

Why it is a barrier

The defining evidence is H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010): the Fourth Circuit sustained North Carolina DOT's race-conscious subcontracting goals for African American and Native American subcontractors precisely because the disparity study showed statistically significant disparities in subcontracting DOLLARS even where firms were available and participating, while striking the program for groups (Hispanic, Asian, women) lacking that dollar-level evidentiary showing. The case itself was triggered when a low bidder's bid was rejected for failing the state's "lenient" good-faith-effort standard, putting GFE enforcement squarely at the center. The mechanism is well documented across disparity studies: M/WBEs cluster in low-value subcontract scopes, so a jurisdiction can report many participating firms yet show large utilization-vs-availability dollar gaps (disparity indices well below the 0.80 substantial-disparity threshold). Croson, 488 U.S. 469 (1989), supplies the doctrinal hook, the Court faulted Richmond for adopting a 30% subcontracting set-aside with "no direct evidence" that prime contractors discriminated against minority subcontractors, making prime-to-sub discrimination the exact factual gap a defensible study must fill. The federal DBE rule operationalizes the abuse vectors: CUF and pass-through rules (49 C.F.R. 26.55) exist because credit-gaming is a known problem (the 30%-self-performance presumption and the rule that work a DBE subs to a non-DBE earns no credit), and the GFE framework (49 C.F.R. 26.53 and Appendix A) exists because rubber-stamped GFE waivers let primes evade goals. The downstream economic harm compounds: relegation to thin-margin scopes starves diverse firms of the retained earnings and bonding/credit track record they need to graduate to higher tiers, reinforcing the access-to-capital constraint Fairlie & Robb document (undercapitalization as the dominant driver of differential firm outcomes).

How the full study investigates it
Quantitative
  • Compute subcontract-tier disparity indices separately from prime-tier indices for each group (Black, Hispanic, Asian American, Native American, women, and combined MWBE), each NAICS-defined industry, and each contract category, applying the conventional 80% (0.80) substantial-disparity threshold and flagging where dollar disparity coexists with headcount/firm-count parity (the H.B. Rowe pattern).
  • Decompose utilization by scope/work-type within each industry to test the relegation-to-low-value hypothesis: cross-tabulate diverse-firm subcontract dollars by trade/CSI division or commodity code to show whether participation concentrates in low-margin scopes (hauling, supply, flagging, landscaping) versus skilled/high-dollar trades.
  • Run weighted least squares and logistic regressions of (a) the probability a subcontract dollar goes to a diverse firm and (b) average subcontract dollar value per diverse firm, controlling for firm size, capacity, industry, geography, and bonding, isolating the unexplained race/gender residual per NCHRP 644 econometric guidance.
  • Compute statistical significance (t-tests / confidence intervals) on each disparity index to meet the Croson 'strong basis in evidence' / statistical-significance standard the Fourth Circuit applied group-by-group in H.B. Rowe.
  • Analyze good-faith-effort waiver outcomes quantitatively: rate of contracts where the goal was set vs. met vs. waived; approval rate of GFE submissions; and whether GFE waivers correlate with lower diverse-firm subcontract dollars, testing whether lax GFE review drives the dollar gap.
  • Estimate the race-neutral baseline per 49 C.F.R. 26.51 logic: model what diverse-firm subcontract utilization would be under race-neutral conditions (unbundling, contract sizing) to size the residual that race-conscious measures would need to address, the narrow-tailoring predicate.
  • Conduct a commercially-useful-function / pass-through audit sample: cross-match listed (committed) DBE/MWBE subcontract participation at bid against actual paid amounts and actual work performed, flagging firms where paid dollars, self-performance, or material-purchasing responsibility fall below 49 C.F.R. 26.55 thresholds.
  • Compare committed-at-bid subcontract goals against final-paid subcontract amounts (commitment-vs-payment slippage) to quantify post-award erosion of diverse-firm dollars.
Qualitative
  • Diverse-firm owner surveys (statistically structured, NCHRP 644 model) with a dedicated subcontracting module: questions on being passed over despite low/competitive sub-bids, being confined to low-value scopes, late or short payment by primes, pressure to act as a pass-through/front, and experience with the City's GFE process.
  • In-depth one-on-one interviews with diverse subcontractors capturing specific, attributable anecdotal accounts (the 'anecdotal evidence' Croson and H.B. Rowe weight alongside statistics), including named instances of bid shopping, scope down-grading, and CUF/pass-through pressure.
  • Interviews with prime contractors and their estimators/compliance staff on how they solicit and select subs, how they document good-faith efforts, and what they perceive as barriers to using diverse subs, to test the 'available but not selected' mechanism.
  • Public hearings / community input sessions (sworn or recorded testimony) advertised through City channels and trade groups, building the contemporaneous public record Croson found missing in Richmond.
  • Structured input from trade and business associations (regional AGC/ABC chapters, NAMC, local Black, Hispanic, Asian, and women business chambers, and certified-MWBE firm networks) on industry subcontracting customs and observed disparities.
  • Interviews with City procurement and supplier-diversity / compliance staff on how subcontracting goals are set, monitored, CUF-verified, and how GFE submissions are actually reviewed and approved or rejected.
  • Targeted case file review and follow-up interviews on contracts where goals were waived via GFE, to surface qualitative detail on the adequacy of the prime's solicitation efforts.
Why it matters in Boston

The RFP's own scope covers utilization of MWBEs as contractors and as subcontractors, and BBC's baseline shows why: the 2020 indices were 65 at the prime tier and 64 at the subcontract tier, near-identical shortfalls that headcount-style compliance would never surface. The practical gating question for Boston is data: the public award dataset captures prime awards, so the update's first task is establishing whether subcontract-tier payments are recorded well enough to compute defensible sub-dollar indices, and recommending permanent capture if they are not. The Sheltered Market Program (M.G.L. c. 30B, § 18) and the 2021 executive-order goals (25 percent MWBE, split 15 percent WBE and 10 percent MBE) give the City live instruments whose enforcement quality, including any good-faith-effort review, the study should evaluate directly.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Strict scrutiny / 'strong basis in evidence' standard for state and local race-conscious contracting remedies. The Court struck Richmond's 30%-of-dollars MBE subcontracting set-aside because, among other defects, the record contained no direct evidence that the city or its prime contractors had discriminated against minority subcontractors, making prime-to-sub subcontracting discrimination the precise evidentiary gap a defensible disparity study must close.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority on subcontracting remedies, not controlling in the First Circuit. Upheld NCDOT's race-conscious subcontracting goals for African American and Native American subcontractors based on statistically significant SUBCONTRACT-DOLLAR disparities, and struck them for Hispanic, Asian, and women-owned firms lacking that showing. Suit arose from rejection of a low bid for failing NCDOT's lenient good-faith-effort standard, placing both subcontract-dollar disparity and GFE enforcement at the heart of the case.

  • NCHRP Report 644: Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board / National Academies, 2010)NCHRP Report 644 (2010); NCHRP Project 20-76; co-PIs Jon Wainwright (NERA) and Colette Holt (Colette Holt & Associates)

    The recognized methodological standard for legally defensible disparity studies. Includes a model scope of work, guidance on availability and utilization measurement, and a dedicated appendix, 'Importance of Comprehensive Subcontract Data Collection,' directing studies to capture subcontract-tier payment data so that subcontract-dollar disparities (not just prime-tier or headcount measures) can be quantified.

  • 49 C.F.R. 26.51, What means do recipients use to meet overall goals?49 C.F.R. 26.51

    Requires recipients to meet the maximum feasible portion of the overall goal through race-neutral means, and enumerates the race-neutral menu, including arranging solicitations, quantities, specifications, and delivery schedules to facilitate participation, and unbundling/structuring contracts so small and diverse firms can compete. Frames the remedy menu and the requirement to project and justify the race-neutral share before resorting to race-conscious measures.

  • 49 C.F.R. 26.53 and Appendix A to Part 26, Good-faith-efforts procedures and guidance49 C.F.R. 26.53; 49 C.F.R. Part 26, App. A

    Defines the good-faith-efforts regime: a bidder that misses a contract goal must document that it took 'all necessary and reasonable steps' to achieve it, including soliciting and negotiating with available diverse subs and providing each DBE and non-DBE quote where a non-DBE was selected over a DBE. Appendix A supplies the adequacy criteria. The benchmark against which lax/pro-forma GFE review is measured.

  • 49 C.F.R. 26.55, How is DBE participation counted toward goals? (Commercially Useful Function)49 C.F.R. 26.55

    The commercially-useful-function and credit-counting rules: a firm earns credit only if it actually performs, manages, and supervises the work and is responsible for negotiating price, quality, quantity, ordering, and paying for materials; a rebuttable presumption of NO CUF applies if the firm self-performs less than 30% of the contract cost or subcontracts more than industry norm; work a DBE subs to a non-DBE earns no credit. The legal definition of the pass-through / 'rent-a-cert' abuse this barrier targets.

  • Fairlie, R.W. & Robb, A.M., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (2008), MIT Press, ISBN 9780262514941

    Peer-reviewed economic foundation linking the relegation-to-low-value-scopes effect to long-run firm outcomes: startup capital and undercapitalization are the dominant drivers of differential business performance, lower sales, fewer employees, smaller payrolls, and higher closure rates among Black-owned firms. Confinement to thin-margin subcontract scopes deprives diverse firms of the retained earnings and track record needed to capitalize, bond, and graduate to higher-value work.

Boston data sources
  • City contract and subcontract payment records: prime award files plus subcontractor payment data where maintained, requested from the City as actual-paid amounts, not just bid-time commitments (the comprehensive subcontract data NCHRP 644's appendix stresses).
  • Bid tabulations and subcontractor-listing / participation commitment forms submitted at bid, from City contract files, for commitment-versus-payment comparison.
  • Good-faith-effort submission files and City disposition records where participation goals apply (goal set / met / waived; GFE determinations), including Sheltered Market Program award files.
  • Commercially-useful-function monitoring and verification records, where the City maintains them.
  • City of Boston and Massachusetts SDO certification directories to identify the universe of available certified diverse firms and to mark utilization records.
  • Supplier Portal vendor registration data, state corporation registry data, and Dun & Bradstreet data to build the availability denominator.
  • U.S. Census ABS 2022, NES-D 2022, and County Business Patterns for the regional firm population by race, ethnicity, sex, and industry.
  • Owner-survey and interview response data collected by the study team, alongside the 2020 BBC anecdotal appendix.
  • The 2020 BBC disparity study, which computed separate prime (65) and subcontract (64) MWBE disparity indices, as the methodological and empirical baseline.
  • Bonding, surety, and credit/financial-capacity data (where obtainable) to control for capacity in the regression models.
15

Bid shopping, bid peddling, and double-bonding by prime contractors

Bench lead

Disparity-study survey methodologist / applied microeconomist leading the anecdotal-evidence and utilization-gap analysis, working in tandem with a construction-procurement legal scholar (to keep the Croson/H.B. Rowe evidentiary linkage tight) and an access-to-capital/surety-bonding economist (Fairlie-Robb tradition) for the double-bonding and margin-erosion mechanism.

A cluster of prime-contractor practices that squeezes the margins of diverse and emerging subcontractors and pushes them out of public work. (1) Bid shopping: after winning (or while assembling) a contract, the prime discloses a sub's quote to that sub's competitors and solicits a still-lower number, using each quote as leverage to drive the price down. (2) Bid peddling: a competing sub makes a post-bid offer to undercut the listed sub in order to displace it. (3) Double-bonding: the prime, or the prime's surety, requires the sub to post its own performance and payment bond on top of the bond the prime already carries on the whole job, forcing the sub to absorb surety costs and tie up bonding capacity it often cannot get. Taken together, these practices erode the already-thin margins of small diverse firms, reward primes that lack a genuine commitment to use a listed DBE/MWBE firm, and let primes meet a participation goal on paper while extracting price concessions that make the work uneconomic for the diverse sub. They are routinely documented in the anecdotal-evidence chapters of disparity studies (BBC, MGT, Keen) and in subcontractor interviews.

Why it is a barrier

Diverse and emerging firms are uniquely vulnerable to these practices for reasons grounded in the access-to-capital literature. Fairlie and Robb (Race and Entrepreneurial Success, MIT Press 2008) document that Black- and other minority-owned firms start and operate with far less capital and wealth than white-owned firms (nearly half of Black families hold under $6,000 in total wealth), and that undercapitalization is the single largest driver of lower minority-firm survival and profitability. A thinly capitalized firm cannot absorb the margin erosion that bid shopping forces, cannot self-finance the surety premiums and collateral that double-bonding demands, and cannot wait out the cash-flow squeeze. The result: (a) diverse subs are listed to help a prime win, then shopped down to a price at which they lose money or decline the work, producing the listed-vs-actual-utilization gap disparity studies measure; (b) double-bonding either disqualifies the sub outright (it cannot obtain a bond) or strips its margin via surety cost, compounding the documented bonding barrier minority firms already face (the 2020 WSDOT Surety Bonding Accessibility Study found market-level bias and prequalification hurdles disproportionately exclude minority/HUB firms from bonding); (c) firms stop bidding to primes known for shopping, shrinking their effective market. Critically, this is precisely the "private discrimination by prime contractors against minority subcontractors" that Croson identified as evidence a jurisdiction may act on, and the kind of subcontractor experience the Fourth Circuit credited in H.B. Rowe to sustain a narrowly tailored subcontracting remedy.

How the full study investigates it
Quantitative
  • Listed-versus-actual utilization analysis: compare the DBE/MWBE dollars listed in winning bids (subcontractor listing forms / commitment schedules) against dollars actually paid to those subs at closeout; a systematic shortfall is the quantitative fingerprint of bid shopping and post-award substitution.
  • Substitution/termination rate analysis: measure how often a listed diverse sub is replaced, reduced, or has its scope renegotiated downward after award, by firm ownership type, controlling for project size and trade.
  • Margin/price-concession analysis where data allow: compare initial quoted prices to final subcontract prices for diverse vs. non-diverse subs on the same trades to detect disproportionate post-listing price reductions.
  • Bond-cost incidence analysis: quantify how often subcontracts require the sub to post its own performance/payment bond, by ownership type and contract tier, and estimate the surety-cost burden as a share of subcontract value (the double-bonding margin hit).
  • Bonding-access/disqualification analysis: rate at which diverse vs. non-diverse subs are unable to bid or are removed for inability to obtain a sub-level bond; pair with bonding-capacity data.
  • Regression/disparity-index estimation linking these practice metrics to overall subcontract utilization disparity indices (availability-weighted), per NCHRP 644, so the anecdotal pattern is statistically corroborated.
  • Repeat-prime concentration analysis: identify whether utilization shortfalls and bond-shifting cluster among a subset of primes (evidence of practice, not random variation).
Qualitative
  • In-depth one-on-one interviews with diverse subcontractors about specific incidents of having quotes shopped, being peddled against, being listed then dropped or squeezed, and being required to post their own bonds (the H.B. Rowe-validated method).
  • Structured business-owner survey with dedicated items capturing frequency and dollar impact of bid shopping, bid peddling, and sub-level bonding demands, by ownership group, to allow quantification of qualitative experience.
  • Confidential interviews with prime contractors and project managers to surface bidding-desk practices, sub-listing and substitution norms, and surety-driven sub-bonding requirements.
  • Public hearings / community testimony where subcontractors and association members can describe experiences on the record (builds the contemporaneous evidentiary record Croson favors).
  • Focus groups by trade and ownership group to identify which trades and primes the practices concentrate in.
  • Targeted input from trade associations and minority/women business organizations (e.g., regional AGC chapters, NAMC, minority contractor associations, the certified-MWBE firm community) on prevalence and remedies.
Why it matters in Boston

Massachusetts has legislated against this barrier once already: the filed sub-bid system (M.G.L. c. 149, §§ 44E-44F) exists precisely to stop primes from shopping subcontract prices on public building work, which makes Boston a natural test of where the protection ends: on non-building work, below thresholds, in unlisted trades, and at lower tiers. BBC's 2020 subcontract disparity index of 64 shows the sub tier under-delivering despite that statute, so the listed-versus-paid and substitution analyses should target the uncovered seams. Filed sub-bid records give the study unusually direct sub-tier price evidence most jurisdictions lack, and double-bonding findings would map straight onto a City bond-assistance remedy in the 26.51(b)(2) mold.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Race-conscious contracting remedies are subject to strict scrutiny and must rest on a strong basis in evidence of identified discrimination, but the Court expressly recognized that if a city has evidence that nonminority prime contractors are systematically excluding or disadvantaging minority subcontractors, it may act to end that exclusion. Documenting bid shopping/peddling and double-bonding by primes against diverse subs is exactly the predicate evidence of private prime-on-sub discrimination Croson contemplates; Croson also faulted Richmond for not first considering race-neutral alternatives, which is why this barrier must be paired with race-neutral remedies.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit precedent, not controlling in Massachusetts, where Croson and strict scrutiny govern. The court sustained NCDOT's subcontracting preference for African American- and Native American-owned firms based in part on disparity-study statistics on subcontracting dollars combined with interviews of minority subcontractors reported in the study, even though no subcontractor testified at trial. Directly validates using disparity-study subcontractor anecdotal evidence about prime-contractor bidding practices as legally probative.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, National Academies/TRB, 2010)NCHRP Report 644 (2010)

    The National Academies guideline directs studies to collect and assess anecdotal evidence of discrimination (including subcontractor experiences in the bidding process) and to corroborate it with statistical disparity findings so the combined record is legally defensible. Provides the canonical methodology for documenting bid-shopping/peddling and bonding-related mistreatment through interviews and surveys and tying them to utilization disparities.

  • 49 C.F.R. § 26.51 (USDOT DBE program, race-neutral means)49 C.F.R. § 26.51(b)

    Enumerates the race-neutral remedy menu a recipient must use to meet the maximum feasible portion of its goal. Items directly targeting this barrier: (b)(2) bonding/financing assistance (simplifying the bonding process, reducing bonding requirements, eliminating the impact of surety costs from bids) addresses double-bonding; (b)(1) structuring solicitations and (b)(4) information programs on subcontract opportunities reduce primes' leverage to shop. Establishes the remedial vocabulary the study should recommend against documented bid-shopping and double-bonding.

  • Fairlie, R.W. & Robb, A.M., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (2008)

    Peer-reviewed economic evidence that minority firms operate with substantially less startup capital and owner wealth, and that undercapitalization is the dominant driver of weaker minority-firm outcomes. Explains the causal mechanism: thinly capitalized diverse subs cannot absorb shopped-down margins or self-finance double-bonding, so these practices fall on them disproportionately.

  • WSDOT Surety Bonding Accessibility Study (OMWBE/HUB), 2020WSDOT Surety Bond Study (2020)

    Independent state study finding the prequalification process is the largest bonding barrier for minority/HUB firms and that market-level bias persists in surety bonding. Corroborates the double-bonding prong: requiring diverse subs to post their own bonds compounds an already-discriminatory bonding market.

Boston data sources
  • Bid/award files with subcontractor listing forms and participation commitment schedules, from City contract files (listed participation).
  • Contract closeout and payment records, including retainage data and final paid-to-sub amounts where maintained (actual participation).
  • Change-order and subcontractor substitution/termination logs from City contract files.
  • Subcontract agreements showing bonding and surety-cost flow-down requirements, where obtainable.
  • Filed sub-bid records under M.G.L. c. 149, §§ 44E-44F for covered public building work, which document sub-tier bidding directly.
  • Surety/bonding data and any City bond-assistance program records; the Washington State surety accessibility study as a benchmark.
  • City of Boston and Massachusetts SDO certification directories and Supplier Portal vendor records, to classify ownership and build the availability denominator.
  • Owner-survey and interview datasets generated by the study itself, alongside the 2020 BBC anecdotal appendix.
  • Trade-association and bid-protest/complaint records referencing shopping, peddling, or non-payment.

Barrier family

Information and networks

Who learns of opportunities, who is introduced, and who is mentored, the informal channel Loury calls discrimination in contact.

08

Information and Awareness Barriers

Bench lead

Survey methodologist / applied microeconometrician (the disparity-study survey-and-statistics lead), partnered with a procurement-process analyst. The survey methodologist owns instrument design, sampling, weighting, and the NCHRP 644-required nonresponse-bias testing for the information-access battery; the procurement-process analyst owns the Supplier Portal registration/commodity-code reach analysis and the by-channel disparity ratios. A legal scholar versed in Croson and H.B. Rowe reviews the anecdotal record for corroboration sufficiency, and an access-to-capital/entrepreneurship economist (Fairlie & Robb lineage) frames the upstream network/experience channel.

The unequal distribution of timely, actionable knowledge about who is buying, what they are buying, when, on what terms, and how to position to win it. Concretely, this is the gap between firms plugged into the procurement information loop (registered vendors with correct commodity codes, members of established networks, incumbents, and firms with relationships to buyers and primes) and firms that are not. It covers not knowing a solicitation exists, learning of it too late to respond competitively, not knowing the unwritten norms of how to bid and get on a prime's subcontractor list, and not knowing about certification, registration, capital, and bonding programs that exist to help. 49 C.F.R. 26.51 treats this barrier as remediable through race-neutral information-and-communications programs: (b)(4) requires "[c]arrying out information and communications programs on contracting procedures and specific contract opportunities (e.g., ensuring the inclusion of DBEs, and other small businesses, on recipient mailing lists for bidders; ensuring the dissemination to bidders on prime contracts of lists of potential subcontractors; provision of information in languages other than English, where appropriate)," and (b)(8) requires "[e]nsuring distribution of your DBE directory, through print and electronic means, to the widest feasible universe of potential prime contractors."

Why it is a barrier

Procurement information does not flow neutrally. It travels disproportionately through established relationships, incumbency, and informal industry networks that have historically excluded minority- and women-owned firms, so the same notice that is "public" reaches well-networked firms early and emerging firms late or never. NCHRP 644 catalogs this directly as an economy-wide barrier: it lists "Exclusion from 'good old boy' networks" as a barrier to DBE formation and "Nonsolicitation of DBEs in the absence of DBE goals" and "Industry domination by informal, racially exclusionary business networks" as barriers to competition by existing DBEs. The Supreme Court in Croson anticipated the same dynamic when it directed governments to use race-neutral devices to remove "formal barriers caused by bureaucratic inertia." The Fourth Circuit in H.B. Rowe credited survey evidence that prime contractors "change their bidding practices when not required to hire minority subcontractors" and characterized the underutilization as "more than a mere byproduct of misguided yet color-blind cronyism," which is the network/information channel operating in the subcontracting market. Empirically, an MBDA review of existing disparity studies found that among minority businesses surveyed, 55% reported not having the right contacts (with contracts effectively "wired" for firms with early information access) and 49% cited difficulty getting information such as bid information sent too late or not at all. A 2020 Washington State surety bonding accessibility study (WWU Center for Economic and Business Research, for OMWBE) quantified the front-end version of the same barrier: among underrepresented respondents, 56% "did not know where to start looking," 33% found the relevant website confusing, 33% "couldn't find where/who to ask," and 22% reported no one answered their request for information. Fairlie and Robb show why the network gap compounds over time: Black entrepreneurs have far fewer opportunities than white entrepreneurs to acquire valuable prebusiness experience by working in a family business, which is exactly where firms absorb the tacit, relationship-based knowledge of how the contracting market actually works. The result is a self-reinforcing disadvantage: firms outside the loop bid less, win less, build less past performance, and stay outside the loop.

How the full study investigates it
Quantitative
  • Vendor-registration and notification-reach analysis: pull the City's Supplier Portal registered-vendor file and match registrations and selected commodity/category codes against the universe of available M/WBE and emerging firms (from the City and Mass SDO certification directories and the 2020 BBC availability database), then compute the share of available diverse firms that are (a) registered at all and (b) registered under the correct codes to actually receive solicitation notifications. The gap is a direct, defensible measure of who the information system structurally fails to reach.
  • Differential disparity-ratio test by award path: compute the NCHRP 644 disparity ratio (utilization/availability, where <100% signals adverse disparity) separately for solicitations that were broadly advertised/open-bid versus those awarded through limited-solicitation, informal, sole-source, micro-purchase, or quote-based methods. A markedly lower ratio in the low-information channels isolates the information/network barrier as a driver of underutilization.
  • Multivariate/regression decomposition: regression of bid submission and win probability on firm race/ethnicity and gender, controlling for firm size, capacity, age, bonding capacity, past performance, and solicitation type, to test whether diverse firms are less likely to bid (an awareness/notification effect) net of capacity, consistent with NCHRP 644's directive to rigorously isolate non-discrimination factors.
  • Bidder-density analysis: for each solicitation, count distinct diverse bidders relative to available diverse firms and test whether response density falls as advertising reach narrows; low and concentrated bidder pools are a quantitative signature of restricted information flow.
  • Economy-wide business-formation and earnings disparity analysis (per NCHRP 644 and the Croson/Part 26 evidentiary framework): use PUMS/ACS and Annual Business Survey microdata to estimate disparities in business formation rates and owner earnings for minorities and women in Boston-relevant industries, establishing the upstream effect of network/information exclusion on the supply of ready firms.
  • Commodity-code coverage and self-selection analysis: quantify how many available diverse firms have selected zero or too-few commodity/category codes relative to the lines of business they actually perform, measuring the structural under-notification baked into a code-matched alert system.
  • Nonresponse-bias testing on the owner survey (NCHRP 644 explicitly warns most studies fail this), weighting responses to the certified/available population so the information-barrier prevalence estimates are population-representative and litigation-defensible.
Qualitative
  • Stratified owner survey (mail + online, per NCHRP 644's emphasis on broad quantifiable coverage) with a dedicated information-access battery: how the firm learns about City of Boston opportunities; whether it is registered on the Supplier Portal and which commodity/category codes it selected; whether it has received solicitation notifications; whether bid information arrived with enough lead time to respond; and whether it knew of City and SDO certification, the Sheltered Market Program, bonding-assistance, and outreach programs. Mirror the Washington OMWBE item structure (did not know where to start; could not find who to ask; no response to inquiries) to produce comparable percentages.
  • In-depth individual interviews with diverse and emerging firm owners to capture the texture of network exclusion: how opportunities reach them (or do not), experiences of being shut out of informal channels, late notification, and 'wired' procurements, the precise dynamics NCHRP 644 labels 'good old boy networks' and Fairlie & Robb trace to missing family-business experience.
  • Prime-contractor interviews to document how primes assemble subcontractor teams and disseminate (or fail to disseminate) potential-subcontractor lists, probing the H.B. Rowe finding that primes change bidding behavior absent goals.
  • Public hearings/community testimony (an NCHRP 644-endorsed anecdotal format) held in Boston neighborhoods with translation support, generating on-the-record accounts of awareness and outreach failures.
  • Focus groups segmented by certification status, race/ethnicity, gender, and industry to surface shared patterns in how firms do and do not find out about City work.
  • Structured input from trade associations and chambers (e.g., regional minority/women business councils, Hispanic and Asian American chambers, regional Asian American, Hispanic, and Black chambers, AGC/ABC local chapters) on how procurement information circulates in their networks and where diverse members fall outside it.
  • Mystery-shopper / process walkthrough: attempt the City's vendor onboarding, Supplier Portal registration, commodity-code selection, and solicitation-finding path from a new diverse firm's perspective and document friction points, complementing the survey's self-reported confusion data.
Why it matters in Boston

Boston's opportunity flow runs through the Supplier Portal, and this RFP itself illustrates the gate: the documents were released June 29, 2026 behind portal registration, so a firm outside the loop never sees them. BBC's 2020 support-services finding (availability 4.8 percent, utilization 3.2 percent, index 66) sits in exactly the small-dollar, quote-driven territory where informal sourcing and late notice do their damage. The City already operates the outreach infrastructure a remedy needs, including the Supplier Diversity Program, departmental equitable procurement plans under the 2021 executive order, and the public Equity in City Contracting dashboard; the update can measure whether that infrastructure actually reaches unregistered and mis-coded available firms, which is the 26.51(b)(4) and (b)(8) question in Boston form.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co.488 U.S. 469, 509-511 (1989)

    Even absent proof of local discrimination, a jurisdiction has 'at its disposal a whole array of race-neutral devices to increase the accessibility of city contracting opportunities to small entrepreneurs of all races,' including simplification of bidding procedures, relaxation of bonding requirements, training, financial aid, and 'elimination or modification of formal barriers caused by bureaucratic inertia.' Information and outreach programs are the paradigm race-neutral first step the strict-scrutiny framework expects a jurisdiction to try.

  • H.B. Rowe Co. v. Tippett615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in the First Circuit. The court credited anecdotal/survey evidence that primes 'change their bidding practices when not required to hire minority subcontractors' and held the underutilization was 'more than a mere byproduct of misguided yet color-blind cronyism' - i.e., exclusion operating through informal network/information channels - sufficient to corroborate a strong basis in evidence for African American and Native American subcontractors.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board / National Academies, 2010)Ch. 2, Economy-Wide Evidence and Anecdotal Analyses

    Names the barrier as a recognized category: 'Exclusion from "good old boy" networks' (barrier to DBE formation); 'Nonsolicitation of DBEs in the absence of DBE goals' and 'Industry domination by informal, racially exclusionary business networks' (barriers to competition by existing DBEs). Prescribes the disparity ratio = utilization/availability, multivariate/regression analysis to isolate non-discrimination factors, and analysis of disparities in business formation and earnings.

  • 49 C.F.R. 26.51(b)(4) and (b)(8)49 C.F.R. 26.51(b)(4), (b)(8)

    The federal race-neutral remedy menu prescribes the cure for this exact barrier: (b)(4) information-and-communications programs (bidders' mailing lists, dissemination of potential-subcontractor lists to primes, non-English information) and (b)(8) distribution of the DBE/vendor directory by print and electronic means 'to the widest feasible universe of potential prime contractors.'

  • NCHRP Report 644 (Wainwright & Holt), Anecdotal Analyses - Recommended ApproachCh. 2, Anecdotal Analyses

    Methodological authority for the qualitative side: 'Anecdotal evidence has been collected in a variety of formats including mail surveys, individual interviews, group interviews or focus groups, and public hearings,' and 'High-quality studies often employ multiple approaches.' Mail surveys are 'particularly important to establish a broad base of coverage that is capable of being quantified,' and the report flags testing for nonresponse bias.

  • Fairlie, R.W. & Robb, A.M., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (2008)

    Peer-reviewed evidence for the human-capital/network channel: Black entrepreneurs have significantly fewer opportunities than white entrepreneurs to acquire prebusiness work experience through a family business, depriving them of the tacit, relationship-based knowledge and contacts that drive contracting success - the upstream root of the information/awareness gap.

  • Minority Business Development Agency, Contracting Barriers and Factors Affecting Minority Business Enterprises: A Review of Existing Disparity Studies (U.S. Dept. of Commerce / MBDA)MBDA, Contracting Barriers (review of disparity studies)

    Cross-study synthesis quantifying the barrier: 55% of surveyed minority firms reported not having the right contacts (contracts 'wired' via early information access) and 49% cited difficulty getting information, including bid information sent too late or not at all.

  • Surety Bonding Accessibility Study: OMWBE and HUB Contractors (Center for Economic and Business Research, Western Washington University, Oct. 2020)WWU CEBR for Washington OMWBE (2020), Information Barriers analysis

    Quantifies the front-end awareness gap among underrepresented contractors: 56% 'did not know where to start looking,' 33% found the website confusing, 33% 'couldn't find where/who to ask,' and 22% reported no response to requests for information - a usable model for measuring information-access barriers via survey.

Boston data sources
  • City of Boston Supplier Portal registered-vendor extract, including selected commodity/category codes and certification flags, with notification records where maintained.
  • Analyze Boston 'City of Boston Contract Award' and 'City of Boston Discretionary Spending' datasets (FY2019-FY2026), coded by procurement method and dollar band to separate broadly advertised from low-information channels.
  • City of Boston and Massachusetts SDO certification directories used to build the available-firms universe.
  • The 2020 BBC disparity study availability database and anecdotal appendix, as the prior-study vendor file and baseline.
  • Master firm/availability database the study builds (matched to certification rolls, business directories, and prior disparity-study vendor files) to compute notification reach.
  • U.S. Census ACS/PUMS and ABS 2022 / NES-D 2022 data for CBSA 14460 and the five RGMA counties for business-formation and earnings disparity analysis.
  • Primary survey, interview, focus-group, and public-hearing transcripts and coded response data generated by the study.
  • Trade-association and chamber membership/communication information and outreach-event attendance records.
  • Benchmark comparators: the MBDA cross-study review and the Washington State OMWBE surety/information-barrier instrument for question design and effect-size calibration.
09

Old-Boy Networks and Relationship-Based Award (Social-Capital / Network-Based Discrimination)

Bench lead

Survey methodologist / qualitative research lead (anecdotal-evidence team), working in tandem with an economic-sociologist or labor/network economist who models social-network discrimination (the Okafor/homophily framework) and a legal scholar to map the findings to the Croson / H.B. Rowe 'strong basis in evidence' standard. The survey methodologist owns instrument design, sampling, interviews, hearings, and the verification protocol; the network economist owns the causal interpretation that distinguishes relationship gatekeeping from neutral business judgment.

The practice of awarding prime contracts and (especially) subcontracts on the basis of pre-existing personal relationships, repeat-player familiarity, and word-of-mouth referral rather than open, broadly advertised competition. Buyers and prime contractors solicit bids from a closed circle of firms they already know and trust; opportunities circulate informally ("who you know") before or instead of through public channels. Because that circle was formed under earlier conditions of exclusion, it tends to reproduce a network that is disproportionately white- and male-owned. Diverse and emerging firms outside the network never learn of opportunities in time, are never asked to quote, or are treated as unknown/untested even when qualified. Disparity studies record this as the recurring "good old boy network" finding in anecdotal evidence. Critically, this is a structural mechanism: network-based exclusion can persist and disadvantage minority firms even with no overt bias by any individual actor, simply because of homophily (people refer to people like themselves) and the smaller size of minority business networks.

Why it is a barrier

Network-based award disadvantages diverse and emerging firms through several documented channels. (1) Information exclusion: opportunities and bid invitations move through informal referral before reaching public solicitation, so out-of-network firms get less lead time, fewer invitations to quote, and are absent from primes' "go-to" subcontractor lists. (2) Trust/relationship gatekeeping: primes default to known repeat subs; a firm with no prior relationship is treated as risky regardless of qualifications. Colette Holt & Associates (CHA) disparity-study interviews capture this directly, with an owner describing "entrenched relationships that leave minority firms out of the loop" and stating, "You're not in the frat. You didn't get the letter, you know? You didn't get the call." (3) Self-reinforcing exclusion: the academic mechanism (Okafor 2025) shows that when the minority group is smaller, members systematically receive fewer referral opportunities and lower expected returns over time even under identical ability and even under formally "colorblind" rules, because referral flows follow same-group lines. (4) Capital/experience transmission: Fairlie & Robb (2008) show minority owners have far less access to the family-business and prior-work exposure through which majority owners inherit both startup capital and the relationship networks that generate early contracts. Courts treat this barrier as legally cognizable: in H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010), the Fourth Circuit upheld race-conscious subcontracting goals for African American and Native American firms in part because telephone-survey anecdotal evidence "exposed an informal, racially exclusive network which systematically disadvantaged" those groups, holding that such anecdotal evidence properly supplemented the statistical disparities.

How the full study investigates it
Quantitative
  • Disparity-ratio (utilization vs. availability) analysis disaggregated by procurement channel, comparing M/WBE subcontract utilization on contracts where the prime selects subs informally versus contracts with formal solicitation or subcontracting plans, a wider gap on informally-sourced work is the statistical signature of network exclusion.
  • Marginal-probability / regression modeling of award likelihood controlling for firm capacity, experience, bonding, and size to isolate the residual 'unexplained' gap attributable to relationship/network factors (the NCHRP 644 / Croson framework for ruling out non-discriminatory explanations).
  • Repeat-award and concentration analysis: measure the share of awards (and subcontract dollars) flowing to a small set of recurring incumbent firms, and the M/WBE share within that repeat-player set, to quantify how closed the de facto network is (e.g., Herfindahl-type concentration of awards by vendor).
  • Bidder-list / solicitation-reach analysis: for a sample of contracts, count how many firms were actually invited or quoted versus the number of available qualified M/WBEs in the registry, and measure M/WBE presence on primes' invitation lists.
  • 'First-time vendor' penetration rate over time: track how often new and emerging firms (and M/WBEs specifically) break into the active vendor pool, as a proxy for network permeability.
  • Private-sector / custom-market benchmark comparison (NCHRP 644 disparity-vs.-marketplace) to test whether disparities widen in the relationship-heavy private market where no neutral-remedy obligations apply.
  • Quantitative coding of the anecdotal corpus: tabulate the frequency with which surveyed/interviewed owners report network-based exclusion (not invited to bid, not on lists, 'who you know'), reported by race/ethnicity/gender, to convert anecdote into a measured prevalence rate.
Qualitative
  • Statistically structured owner telephone/online surveys of M/WBE and non-M/WBE firms with targeted items on how they learn of opportunities, whether they are invited to quote, and experiences of being excluded from primes' go-to lists, the exact method credited in H.B. Rowe v. Tippett.
  • In-depth one-on-one interviews with M/WBE owners (the CHA model: 1,000+ owners interviewed across studies) to capture verbatim network-exclusion narratives ('you didn't get the call' / 'out of the loop').
  • Public hearings and sworn/written testimony before the City Council, the Supplier Diversity Council, or a study advisory committee, creating an on-the-record evidentiary base of network-exclusion accounts.
  • Focus groups segmented by race/ethnicity, gender, and industry (construction, A/E, professional services, goods/supplies) to surface channel-specific relationship gatekeeping.
  • Structured input from trade and business associations (regional minority/women contractor associations, Hispanic/Asian/Black chambers, AGC and similar prime-contractor groups) on referral and teaming practices.
  • Interviews with prime contractors and buyers about how they build subcontractor/vendor lists and source quotes, documenting the informal-referral default from the supply side.
  • Triangulation/verification of anecdotal accounts against the statistical record where feasible, anticipating the H.B. Rowe critique that unverified anecdotes are weaker (cross-reference named instances to bid/award files).
Why it matters in Boston

Two Boston facts make the network mechanism worth measuring head-on. First, the WBE share of certified discretionary spending has sat nearly flat at roughly 2.4 to 3.0 percent for eight fiscal years while the MBE share more than tripled; outreach and goals alone do not explain a flatline that specific, and repeat-player sourcing is a testable candidate. Second, BBC's 2020 anecdotal record (570+ businesses engaged, seven neighborhood forums) already contains network-exclusion accounts the update can re-interview and verify rather than starting cold. The award-concentration and first-time-vendor penetration metrics run directly on the public FY2019-FY2026 award and payment data, making this one of the most data-ready chapters in the study.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Establishes the 'strong basis in evidence' standard and requires that a race-conscious remedy be supported by particularized evidence of discrimination in the relevant market; the Court credited (and demanded) both statistical and anecdotal proof, making anecdotal evidence of exclusionary networks a recognized building block of the evidentiary predicate rather than 'generalized assertions.'

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit precedent, not controlling in the First Circuit. The court upheld minority subcontracting goals for African American and Native American firms because the State's telephone-survey anecdotal evidence 'exposed an informal, racially exclusive network which systematically disadvantaged' those groups, sufficiently supplementing the statistical disparity evidence, directly validating the old-boy-network barrier and the survey/interview method used to prove it.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Transportation Research Board, National Academies, 2010)NCHRP Rep. 644 (TRB 2010), NAP pub. 14346

    The national methodological standard for defensible disparity studies. It treats anecdotal evidence (in-depth interviews, surveys, public testimony) and analysis of barriers to DBE formation and participation as core, expected components alongside the statistical availability/utilization disparity analysis, establishing that documenting relationship/network barriers is part of accepted study methodology.

  • 49 C.F.R. § 26.51 (USDOT DBE Program, race-neutral means)49 C.F.R. § 26.51(b)

    The federal race-neutral remedy menu directly targets network exclusion: § 26.51(b)(4) requires information/communications programs that disseminate to bidders 'lists of potential subcontractors' and ensure DBE inclusion on bidders' mailing lists; (b)(1) restructures solicitations to facilitate participation; (b)(8) requires wide distribution of the DBE directory. These remedies presuppose that closed informal networks suppress participation, confirming the barrier as the policy target.

  • Chika O. Okafor, 'Seeing Through Color Blindness: Social Networks as a Mechanism for Discrimination,' Journal of Law and Economics, Vol. 68, No. 3 (2025)68 J.L. & Econ. (no. 3) (2025); chicagounbound.uchicago.edu/jle/vol68/iss3/2

    Peer-reviewed economic model proving that referral/network-based allocation produces persistent discriminatory outcomes for the smaller (minority) group even under equal ability and formally colorblind rules, driven by homophily. Supplies the causal theory that distinguishes network exclusion from taste-based or statistical discrimination and rebuts 'it's just relationships, not bias' defenses.

  • Robert W. Fairlie & Alicia M. Robb, Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (MIT Press 2008), ISBN 9780262514941

    Documents that minority (especially Black) entrepreneurs have substantially less access to family-business background and prior work experience in family firms, the channel through which majority owners inherit both startup capital and the established contracting relationships/networks that generate early business; grounds the social-capital deficit underlying network exclusion.

  • MBDA, Contracting Barriers and Factors Affecting Minority Business Enterprises: A Review of Existing Disparity Studies (U.S. Dept. of Commerce, Minority Business Development Agency, 2017)MBDA (2017)

    Federal synthesis reviewing the corpus of disparity studies; documents that MBEs win fewer and smaller contracts relative to availability across construction, professional services, and A/E, and catalogs recurring qualitative barriers including exclusion from established relationships/informal networks, corroborating that the network finding is consistent across jurisdictions, not idiosyncratic.

Boston data sources
  • Analyze Boston 'City of Boston Contract Award' dataset (FY2019-FY2026): award concentration, repeat-vendor patterns, and certified-firm share, by department and dollar value.
  • Analyze Boston 'City of Boston Discretionary Spending' dataset (FY2019-FY2026 payment lines) to trace repeat-payee concentration below the formal award threshold.
  • City of Boston Supplier Portal solicitation records, and bidder/quoter lists obtainable from City contract files, for invitation-reach analysis.
  • City of Boston and Massachusetts SDO certification directories to define the available qualified pool.
  • Subcontractor utilization reports and compliance filings where collected by the City.
  • Owner survey and interview datasets generated by the study (the primary anecdotal corpus), alongside the 2020 BBC anecdotal appendix (570+ businesses engaged; seven neighborhood forums).
  • Public-hearing transcripts, written testimony, and advisory-committee submissions.
  • Trade- and business-association membership/referral information and association testimony.
  • U.S. Census ABS 2022, NES-D 2022, and County Business Patterns for the Boston market area to construct availability and capacity benchmarks.
  • Dun & Bradstreet / state corporation registry data for firm formation dates and to identify emerging firms and repeat incumbents.
12

Mentor-Protege and Capacity-Building Gaps

Bench lead

Small-business-development / entrepreneurship economist (human- and financial-capital formation), partnering with the survey methodologist who fields the anecdotal owner survey and interviews; legal scholar advises on tying group-specific capacity evidence to the Croson / H.B. Rowe narrow-tailoring standard.

The absence (or under-resourcing) of structured programs that build the management, financial, bonding, and bidding capacity diverse and emerging firms need to compete for and perform on public contracts. The relevant tools are the race-neutral menu in 49 C.F.R. 26.51(b): mentor-protege relationships in which an established prime transfers know-how and capacity to a developing firm; technical assistance; supportive-services programs that build long-term business management, recordkeeping, and accounting capability; and business-development services that help firms handle increasingly significant projects and reach self-sufficiency. The barrier is structural and remediable rather than a deficiency of the firms themselves: when a jurisdiction does not operate these programs, capable minority-, women-, and small-owned firms are left to acquire scale, systems, and relationships on their own, while incumbent majority firms benefit from informal mentoring, generational business experience, and established networks. The gap operates upstream of every other barrier (bonding, prime relationships, contract size) because it determines whether a firm ever develops the capacity to overcome them.

Why it is a barrier

Capacity gaps disadvantage diverse and emerging firms through compounding human-capital, financial-capital, and network deficits that capacity-building programs are specifically designed to close. Fairlie & Robb (MIT Press 2010), using the Census restricted-access Characteristics of Business Owners dataset, find that the success of business ownership turns on three inputs: human capital, financial capital, and family-business background. Black entrepreneurs have far fewer opportunities than white entrepreneurs to acquire valuable pre-business work experience by working in a family business, which deprives them of the informal, generational management mentoring that majority owners absorb at no cost. That is precisely the deficit a mentor-protege or supportive-services program exists to substitute for. The MBDA review of existing disparity studies and the U.S. Commission on Civil Rights both document that small minority firms repeatedly lack the management systems, business-planning capability, and accounting/recordkeeping capacity that lenders and primes require, and that bank officials cite these firms' inability to draw up business plans. Disparity studies consistently measure this as a scale/availability problem: minority- and women-owned firms cluster at the small end of the size distribution and are under-utilized relative to availability on larger and more complex contracts, the exact gap that 26.51(b)(6) (helping firms handle increasingly significant projects) targets. Croson itself makes the inquiry legally load-bearing: a jurisdiction must show race-neutral alternatives were tried or considered before a race-conscious remedy is narrowly tailored, so the demonstrated under-provision of these capacity programs is both a barrier and the predicate for any remedy.

How the full study investigates it
Quantitative
  • Disparity-index analysis by firm size/scale: compute utilization-to-availability ratios for diverse firms broken out by contract dollar bands (e.g., under $50K, $50K-$250K, $250K-$1M, over $1M) to test whether under-utilization widens as project size grows, the empirical signature of a capacity ceiling that 26.51(b)(6) addresses.
  • Firm-size distribution analysis: compare the size distribution (receipts, employees, years in business) of available diverse firms against majority firms in the same NAICS/PSC codes to quantify the scale gap mentor-protege/supportive-services programs would close.
  • Business-formation and survival analysis: measure entry, growth, and exit (firm age and survival rates) of diverse firms in the market area versus comparison firms to identify where in the lifecycle capacity attrition occurs.
  • Inventory and uptake analysis of existing capacity programs: count and characterize City of Boston and Massachusetts mentor-protege, technical-assistance, supportive-services, and small-business training offerings, and measure diverse-firm enrollment, completion, and post-program contract-award rates (program-effectiveness regression where data permit).
  • Regression decomposition (Fairlie-Robb / Blinder-Oaxaca style) of business outcomes on human-capital, financial-capital, and experience variables to isolate the share of the diverse/majority performance gap attributable to capacity inputs that programs can supply.
  • Bidder-to-awardee funnel analysis: track diverse firms from registration to bid to award to estimate where capacity-related drop-off (no-bid, non-responsive bids, failed performance) concentrates.
Qualitative
  • Owner survey (NCHRP 644-style anecdotal instrument) with dedicated capacity items: whether owners had access to mentoring, what management/bonding/bidding skills they lacked, whether they declined or lost work for lack of capacity, and whether available programs reached them.
  • In-depth interviews with diverse and emerging firm owners probing the management-experience and mentoring deficit (Fairlie-Robb family-business-experience theme) and concrete instances of being unable to scale to a larger contract.
  • Interviews with prime contractors and program administrators on the supply and quality of mentor-protege relationships, why protege placements succeed or fail, and barriers to participating as mentors.
  • Public hearings / community listening sessions to gather sworn or on-the-record anecdotal accounts that meet the Croson and H.B. Rowe evidentiary standard, group-disaggregated to support group-specific findings.
  • Trade-association and ethnic/women's chamber input (e.g., regional minority and women business councils, AGC/ABC chapters, NAWBO, regional Hispanic/Asian/Black chambers) on capacity-building needs and program gaps.
  • Key-informant interviews with capacity-providers (the City's Supplier Diversity Program, the Massachusetts Supplier Diversity Office, the regional APEX Accelerator, and local small-business development and CDFI partners) on which services are oversubscribed, underfunded, or missing for Boston diverse firms.
Why it matters in Boston

Boston's post-2020 record is a capacity story: certified-MWBE spend rose from 5.2 percent (FY2019) to 12.3 percent (FY2025) under the 2021 executive order and the Sheltered Market Program, but it remains below BBC's 16.9 percent availability benchmark and well below the 25 percent goal, and Black-owned firms entered that period with a disparity index of 12. Closing the remaining gap is less about finding firms than growing them: the support-services and construction indices (66 and 34) mark industries where certified firms exist but cannot yet carry City-scale scopes. The study should inventory what contract-linked capacity-building the City and its partners actually operate, measure uptake and post-program win rates, and size the 26.51(b)(5)-(b)(7)-style program the evidence supports.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Race-conscious contracting remedies are subject to strict scrutiny and must be narrowly tailored; the jurisdiction must show that race-neutral alternatives (capacity-building, technical assistance, simplified bidding) were unworkable or considered before resorting to race-conscious measures, which makes the presence or absence of mentor-protege/supportive-services programs a legally load-bearing element of the record.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority (not controlling in Massachusetts) applying Croson to a state DOT M/WBE program; upheld race-conscious relief only for groups (African American, Native American) for which the disparity study supplied a strong basis in evidence, and struck it for groups (Hispanic, Asian, women) where the evidence was insufficient, confirming that a defensible study must tie group-specific barriers and remedies to group-specific evidence.

  • 49 C.F.R. 26.51(b) (DBE race-neutral means)49 C.F.R. 26.51(b)(2),(b)(3),(b)(5),(b)(6),(b)(7)

    The federal race-neutral remedy menu: (b)(2) assistance overcoming bonding/financing limitations; (b)(3) technical assistance and other services; (b)(5) supportive-services program to develop and improve immediate and long-term business management, recordkeeping, and financial/accounting capability; (b)(6) services to help firms improve long-term development, handle increasingly significant projects, and achieve eventual self-sufficiency; (b)(7) program to assist new, start-up firms in fields where DBE participation has historically been low. Appendix D to Part 26 sets out the mentor-protege program guidelines.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (National Academies / TRB, project directed by Jon Wainwright, NERA)NCHRP Report 644 (2010)

    The canonical methodological guideline for legally defensible disparity studies; prescribes the combination of quantitative availability/utilization (disparity-index) analysis with anecdotal evidence, and frames designing defensible programs including the race-neutral business-development remedies that capacity-building findings support.

  • Fairlie, R.W. & Robb, A.M., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2010)Fairlie & Robb (2010), MIT Press, ISBN 9780262514941

    Peer-reviewed analysis (Census CBO dataset) establishing that human capital, financial capital, and family-business background drive business success, and that Black entrepreneurs have fewer opportunities to gain pre-business management experience through family firms, the human-capital deficit that mentor-protege and supportive-services programs are designed to remedy.

  • MBDA, Contracting Barriers and Factors Affecting Minority Business Enterprises: A Review of Existing Disparity StudiesMBDA (2016/2017 review of disparity studies)

    Federal synthesis of disparity-study findings documenting recurring capacity, management, business-planning, and access barriers for MBEs across jurisdictions, supporting the cross-study generalizability of capacity-building findings.

  • U.S. Commission on Civil Rights, Disparity Studies as Evidence of Discrimination in Federal Contracting (2006)USCCR (2006)

    Documents that small minority firms encounter difficulty obtaining loans because they lack experience drawing up business plans, tying the management/capacity gap directly to downstream financing and bidding failure.

  • Colette Holt & Associates disparity studies (e.g., State of Illinois Goods and Services Disparity Study 2022; Cook County 2022; State of Missouri 2014)Colette Holt & Associates (CHA), various jurisdictions

    Leading practitioner methodology recommending race- and gender-neutral remedies including technical assistance to small/new entrepreneurs, mentor-protege programs, bonding assistance, and collaboration with organizations providing management and technical assistance, demonstrating the standard remedial framework that capacity-gap findings support.

Boston data sources
  • Analyze Boston 'City of Boston Contract Award' dataset and Supplier Portal bidder/registration records, to build utilization and the bidder-to-award funnel.
  • City of Boston and Massachusetts SDO certification directories to build the availability universe and firm characteristics.
  • U.S. Census ABS 2022 and NES-D 2022 (and, where accessible via FSRDC, restricted-access owner-characteristics microdata) for firm size, owner human capital, and capital inputs.
  • County Business Patterns and Nonemployer Statistics for the Boston market-area firm-size distribution by industry.
  • Program records from the City's Supplier Diversity Program and Sheltered Market Program, the Massachusetts Supplier Diversity Office, the regional APEX Accelerator, and local small-business development and technical-assistance providers (enrollment, completion, outcomes where kept).
  • Primary-collected owner survey and interview data, public-hearing transcripts, and trade-association submissions.
  • SBA 8(a), HUBZone, and mentor-protege program participation data for Boston-area firms as a benchmark for capacity-program reach.

Barrier family

Upstream and the wider market

Business formation, private-sector treatment, and labor-market access that shape the supplier pool before the City buys anything.

13

Private-Sector and Price Discrimination

Bench lead

Access-to-capital / entrepreneurship economist (labor and small-business economics), leading the formation, earnings, and credit-market regressions and the Blinder-Oaxaca/Fairlie decompositions; partnered with the survey methodologist who fields the private-sector vendor survey and the legal scholar who maps findings to the Croson 'passive participant' and H.B. Rowe corroboration standards.

Discrimination that diverse- and women-owned firms encounter in the broader private (non-governmental) economy, before and apart from any public bid: being quoted higher prices or worse terms by suppliers and material vendors; being denied loans, lines of credit, and bonding, or getting smaller amounts at higher cost; being shut out of private customers and informal referral networks; and being treated worse by prime contractors (higher performance standards, dropped after award, solicited only when an MWBE goal forces it). These private-market disadvantages suppress the formation, capitalization, scale, and price-competitiveness of diverse firms so that, by the time they reach the public market, fewer exist, they are smaller, and they bid from a higher cost base. The barrier is "upstream": it shapes the pool of firms and their capacity rather than operating inside the City's own procurement rules.

Why it is a barrier

The private market is where firms are built, financed, and priced; discrimination there compounds into every later disadvantage. Credit: Blanchflower, Levine & Zimmerman found Black-owned small businesses roughly twice as likely to be denied credit even after controlling for creditworthiness, and those approved paid higher interest. The Federal Reserve Small Business Credit Survey continues to show large firm-level gaps in full-approval rates between firms of color and white-owned firms with comparable profiles, plus smaller amounts and higher costs. Capital/wealth: Fairlie & Robb show startup capital is the single most important determinant of business success, and that the lower personal/family wealth of Black entrepreneurs (a large share of Black families hold minimal net worth) translates directly into undercapitalization and lower survival, earnings, and scale. Supplier pricing: undercapitalized firms cannot buy in volume, pay cash, or float receivables, so they face worse input prices and terms, eroding their margins and bid competitiveness. Prime treatment: in H.B. Rowe v. Tippett the Fourth Circuit credited survey evidence that primes held minority subs to higher standards, viewed them as less competent, changed bidding when no MWBE goal applied, and dropped them after award. The cumulative result is the recurring disparity-study signature BBC measured in Boston in 2020: minority- and women-owned firms available for 16.9 percent of the City's contract dollars were paid 11.0 percent, with Black-owned firms under-used across every work type, a downstream fingerprint of upstream private-market exclusion.

How the full study investigates it
Quantitative
  • Business-formation / self-employment regression: model the probability of business ownership (and of incorporating/hiring employees) as a function of race, sex, and ethnicity controlling for age, education, industry, and net worth using PUMS/American Community Survey and Census Annual Business Survey microdata; a residual race/sex effect after controls is direct evidence that diverse residents form and scale firms at depressed rates - the upstream output of private-market discrimination.
  • Business-earnings / scale (capacity) regressions: regress firm receipts, employment, and owner business income on race/sex with human-capital and industry controls (Fairlie-Robb specification) to quantify how much of the size gap survives controls and is attributable to capital access.
  • Credit-outcome regression replicating Blanchflower-Levine-Zimmerman on regional data: model loan approval, amount, and interest rate as a function of owner race/sex while controlling for firm age, revenue, credit score, and collateral, using Small Business Credit Survey microdata and any HMDA-adjacent small-business CRA data, to estimate a discrimination-attributable denial gap for the Boston-Cambridge-Newton MSA market.
  • Bonding and surety access analysis: tabulate bonding capacity, denial, and cost by firm ownership from the custom vendor survey and surety data; test whether bonding constraints predict non-bidding on City work.
  • Cost-of-inputs / price-disparity analysis: from survey data, compare reported supplier prices, payment terms (net-30 vs. cash-in-advance), and trade-credit access across ownership groups within the same NAICS/material categories to estimate a pricing penalty.
  • Decomposition (Blinder-Oaxaca / Fairlie nonlinear decomposition): partition the formation, scale, and credit gaps into an 'explained' portion (endowments) and an 'unexplained' residual consistent with discrimination, the standard defensible quantification.
  • Linkage analysis tying private-market metrics to public-market disparity indices: correlate the capital/capacity deficit measures with the study's contract-utilization disparity indices to evidence the 'but-for' causal chain Croson requires.
Qualitative
  • Custom owner/vendor survey (per NCHRP 644 'private sector analysis') with dedicated modules on: supplier pricing and trade-credit terms; loan/line-of-credit/bonding applications, denials, amounts, and rates; private customer access; and prime-contractor treatment (higher standards, post-award dropping, solicited-only-when-a-goal-applies), structured to mirror the H.B. Rowe survey items that the Fourth Circuit credited.
  • In-depth one-on-one interviews with diverse firm owners capturing first-person accounts of being quoted higher prices, denied financing/bonding, or mistreated by primes, with enough specificity (who, when, terms) to be probative anecdotal evidence.
  • Public hearings / community forums (in-person and virtual, multilingual) for sworn or recorded testimony of private-market discrimination, scheduled across Boston neighborhoods for geographic and language reach, following the 2020 study's seven-forum model.
  • Targeted interviews with the supply side: lenders, CDFIs, surety/bonding agents, and majority prime contractors, to test and triangulate owners' accounts and surface practices (e.g., relationship-only sourcing) that disadvantage diverse firms.
  • Trade-association and chamber input: structured input sessions with regional minority/women business chambers, contractor and supplier associations, and CDFIs to corroborate patterns and identify race-neutral remedies that fit the local market.
  • Anecdotal-evidence coding and corroboration protocol: systematically code accounts by group, market, and discrimination type, and align each statistical finding with corroborating anecdote group-by-group, as Rowe requires.
Why it matters in Boston

At the time of the 2020 study Boston operated only race- and gender-neutral measures, yet BBC still found an overall MWBE disparity index of 65 and a Black American-owned index of 12: the Croson passive-participant frame fits, because the City spends into a regional private market whose credit, bonding, and referral practices shape who can even reach its solicitations. The update should run the formation, earnings, and credit regressions on the five RGMA counties BBC defined (Suffolk, Middlesex, Norfolk, Plymouth, Essex) so the private-market record aligns with the availability geography. Since 2021 the City has added the executive-order goals and the Sheltered Market Program; documenting the private-market deficit is what justifies keeping and sharpening those instruments under strict scrutiny in the post-SFFA climate.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469, 491-92 (O'Connor, J.)

    Supplies the doctrinal hook for remedying private-market discrimination: a government becomes a 'passive participant' when it spends in a discriminatory marketplace, and 'any public entity... has a compelling interest in assuring that public dollars... do not serve to finance the evil of private prejudice.' A state subdivision 'has the authority to eradicate the effects of private discrimination within its own legislative jurisdiction.' This converts private-sector and price discrimination from background noise into a constitutionally cognizable basis for a remedy.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit precedent, not controlling in the First Circuit. Upheld race-conscious subcontracting preferences for African American and Native American firms based on statistical disparity 'corroborated by significant anecdotal evidence,' including survey findings that prime contractors held minority subs to higher standards, viewed them as less competent, changed bidding when no MWBE goal applied, and dropped them after award - i.e., private prime-firm discrimination is a proper, judicially-credited evidentiary basis. Also confirms preferences for Hispanic, Asian, and women-owned firms failed for lack of corroborating anecdotal evidence, underscoring that this barrier must be documented group-by-group.

  • NCHRP Report 644: Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (TRB/National Academies; Wainwright & Holt et al.)NCHRP Rep. 644 (2010), private-sector analysis and anecdotal-evidence chapters

    Establishes 'private sector analysis' as a distinct, expected study component and instructs that anecdotal evidence 'should include economic factors and opportunities in the private sector.' Frames the methodology for linking private-market conditions (credit, bonding, supplier terms, prime treatment) to public-market disparity and for collecting probative owner accounts of disparate treatment.

  • USDOT DBE Rule, 49 C.F.R. 26.5149 C.F.R. 26.51(b)

    The race-neutral remedy menu this barrier maps onto: assistance overcoming inability to obtain bonding or financing (e.g., simplifying bonding), and structuring solicitations/quantities to ease small-firm participation. Croson/Rowe require a locality to consider race-neutral alternatives first; documenting this barrier shows which 26.51-style remedies (bonding pools, financing assistance, mobilization payments) target the actual private-market deficit.

  • Blanchflower, Levine & Zimmerman, 'Discrimination in the Small-Business Credit Market,' 85 Rev. Econ. & Stat. 930 (2003)85 Rev. Econ. & Stat. 930-943 (2003)

    Peer-reviewed evidence that Black-owned small businesses are about twice as likely to be denied credit after controlling for creditworthiness and other factors, and pay higher rates when approved. Anchors the credit-discrimination prong with a methodologically rigorous, widely-cited source.

  • Fairlie & Robb, Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008)Fairlie & Robb (MIT Press 2008), using Census Characteristics of Business Owners data

    Establishes that startup/financial capital is the most important determinant of business success and that lower Black personal and family wealth (rooted in historical discrimination) is a primary cause of lower formation, survival, and scale - the mechanism by which private-market capital exclusion becomes a public-market capacity gap.

  • Federal Reserve Banks, Small Business Credit Survey (Report on Employer Firms)Fed Small Business Credit Survey, recent annual reports on employer firms

    Current, firm-level survey evidence of persistent racial gaps in full loan-approval rates, loan amounts, and borrowing costs between firms of color and white-owned firms with comparable credit profiles. Provides contemporaneous national benchmarks against which Boston-region survey results can be compared.

  • BBC Research & Consulting, City of Boston Disparity Study (final report February 2021)City of Boston Disparity Study (BBC, FY2015-FY2019 data)

    In-jurisdiction baseline: BBC paired 47,000+ City contracts with ~800 custom-census availability interviews and anecdotal evidence from 570+ businesses, finding an overall MWBE disparity index of 65 and a Black American-owned index of 12, disparities that persisted across work types under wholly race-neutral programs. Supports the inference that private-market capital, credit, and network conditions, not an absence of firms, drive public-market underutilization.

Boston data sources
  • U.S. Census ABS 2022 and NES-D 2022 for CBSA 14460 and the five RGMA counties (firm counts, ownership, receipts, employment by race, sex, and industry).
  • American Community Survey / Census PUMS microdata (self-employment, business ownership, owner demographics and controls for formation and earnings regressions).
  • Federal Reserve Small Business Credit Survey microdata and reports (firm-level credit approvals, amounts, costs by ownership).
  • FFIEC CRA small-business lending data, HMDA data, and SBA 7(a)/504 loan data for the Boston market.
  • Surety/bonding industry data and SBA Surety Bond Guarantee program data.
  • Custom Boston-region vendor/owner survey (the study's own primary instrument, with private-sector pricing, credit, bonding, and prime-treatment modules).
  • Interview and public-hearing transcripts (study-generated anecdotal record), alongside the 2020 BBC anecdotal appendix.
  • Analyze Boston contract-award and discretionary-spending data, to link private-market metrics to City contract outcomes.
  • The 2020 BBC City of Boston disparity study and its availability database, for in-jurisdiction marketplace baselines.
  • Dun & Bradstreet / commercial registry data for firm size, age, and creditworthiness controls and availability-pool construction.
14

Business-Formation and Self-Employment Barriers (Upstream Availability)

Bench lead

Access-to-capital / labor economist (small-business and entrepreneurship economics), supported by a survey methodologist for the availability survey and custom census, and reviewed by the disparity-study legal scholar for Croson/H.B. Rowe defensibility

Diverse entrepreneurs (minorities, women, and other socially/economically disadvantaged groups) start, own, and survive in business at lower rates than comparable non-minority men. These lower rates are driven by gaps in personal and family wealth, human capital (education and prior management/industry experience), exposure to a family business while growing up, and access to startup credit. The cumulative effect is a smaller upstream pool of diverse-owned firms that ever form and persist long enough to bid on public contracts. In disparity-study terms, this barrier shrinks the "availability" denominator: it depresses the share of ready, willing, and able diverse firms in the marketplace before any contract is even advertised, so even a perfectly neutral procurement system will award diverse firms less. It is the foundational, pre-market barrier that all downstream barriers (bonding, capital, prime relationships, bid procedures) compound.

Why it is a barrier

Fairlie and Robb (Race and Entrepreneurial Success, MIT Press 2008, using the Census Characteristics of Business Owners dataset) document that Black business-ownership rates run roughly 45% of white rates and that the single largest driver is lack of startup capital, traceable to wealth gaps (median Black household wealth they report around $6,000, under one-tenth of white levels). They further isolate two upstream channels independent of capital: (1) lack of prior work experience in a family-owned business, and (2) human-capital differences, both of which lower self-employment entry and survival. Lofstrom and Fairlie show the same low-wealth channel suppresses Hispanic/Latino self-employment entry. In follow-on work (Fairlie, Robb & Robinson, NBER w28154, "Black and White: Access to Capital among Minority-Owned Startups"), Black-owned startups begin smaller and stay smaller over their first eight years, and receive credit on worse terms, meaning fewer survive to become "ready, willing, and able" bidders. Because public contracting can only draw from firms that exist and survive, these formation and survival gaps mechanically reduce the number of diverse firms available to win City of Boston work; a low utilization figure can therefore reflect a depleted upstream pool, which is exactly the dynamic Croson warned must be analyzed (the City there pointed to "the small number of minority entrepreneurs" and barriers like "deficiencies in working capital" and "inability to meet bonding requirements"). NCHRP 644's framework requires the study to measure this availability denominator rigorously so the disparity ratio (utilization divided by availability) is defensible rather than a raw population comparison, which Croson rejected.

How the full study investigates it
Quantitative
  • Build the availability denominator per NCHRP 644 using a custom census/marketplace approach (master vendor lists, certification rosters, bidders lists, supplemental directories, and a telephone/web availability survey) to count ready-willing-and-able diverse firms by industry and size in the City's geographic and product markets, rather than a raw population comparison Croson forbids.
  • Compute business-formation and ownership rates for the Boston-Cambridge-Newton MSA market by race, ethnicity, and sex and benchmark them against national and Massachusetts rates using Census ABS and NES-D (and legacy SBO) tabulations to quantify how much the upstream pool is depressed.
  • Estimate the 'expected availability' of diverse firms but-for the formation gap by re-weighting local rates to national non-minority-male formation rates; the difference quantifies the upstream availability shortfall feeding the disparity ratio.
  • Run regression decompositions (Oaxaca-Blinder style) of self-employment/business-ownership entry and survival on owner wealth, education, prior managerial/industry experience, and family-business background using PUMS/SIPP/ABS microdata to isolate how much of the formation gap is attributable to each mechanism (wealth, human capital, family background, credit).
  • Survival analysis (hazard/Cox models) on firm persistence by owner race and sex to show diverse firms exit before reaching contract-readiness, controlling for size and sector.
  • Sensitivity tests on market-area and industry definitions and on the willing-and-able screen, with confidence intervals, so the availability estimate withstands the strict-scrutiny review applied in Croson and H.B. Rowe.
Qualitative
  • Owner survey of diverse and non-diverse business owners in the Boston market capturing startup capital sources, personal/family wealth at founding, family business exposure, education, prior industry experience, and reasons for not forming or expanding, to corroborate the quantitative formation gap with first-person evidence.
  • In-depth one-on-one interviews with diverse business owners, including would-be and recently-exited owners, probing why they did or did not start, why they could or could not survive to bidding readiness, and the role of wealth/credit/experience.
  • Public hearings and listening sessions (in-person and virtual) for diverse business owners to give sworn or recorded testimony on formation barriers, building the anecdotal record courts require alongside statistics.
  • Structured input from trade associations, minority and women chambers of commerce, the City's economic-development and supplier-diversity offices, SBDCs, and lenders/CDFIs on patterns of formation, capital access, and attrition among diverse firms.
  • Targeted interviews with capital providers (banks, CDFIs, microlenders) on differential approval and terms experienced by diverse startups, linking the credit channel to the formation/survival gap.
Why it matters in Boston

Upstream availability is where Boston's baseline is starkest: BBC's custom census put Black American-owned availability at 3.6 percent of contract dollars, Hispanic American-owned at 1.2, and Asian American-owned at 0.7 in a majority-minority city, so the formation gap precedes any procurement decision. The update's ABS 2022 and NES-D 2022 work for the metro area and the five RGMA counties can test whether that pool has grown since FY2019 and how much of any remaining gap survives wealth and human-capital controls. Because the City's certified-spend gains (5.2 to 12.3 percent) draw on the certified pool, formation-side findings determine how far demand-side tools alone can go, and they anchor the race-neutral formation remedies (capital, bonding, supportive services) that Croson expects the City to try and measure first.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co.488 U.S. 469 (1989)

    The Court rejected the City's reliance on the small number of minority entrepreneurs and identified non-racial, formation-stage barriers facing any new business (deficiencies in working capital, inability to meet bonding requirements, unfamiliarity with bidding procedures, inadequate track record), holding that a low minority-firm presence cannot by itself prove discrimination and must be analyzed against the pool of qualified firms; it endorsed race-neutral devices (financial aid, training, bonding relaxation, credit-discrimination prohibitions) to expand small-business access for entrepreneurs of all races.

  • H.B. Rowe Co. v. Tippett615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in the First Circuit: a properly conducted disparity study that compared utilization to the availability of qualified, willing, and able firms supported narrowly tailored remedies for African American and Native American subcontractors, while race-conscious measures unsupported by the availability/disparity evidence were struck down, confirming that the availability denominator (which formation/survival gaps shrink) must be measured firm-by-firm, not by population.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board / National Academies, 2010)TRB/NAP Pub. 14346

    The national methodological standard (co-PIs Jon Wainwright of NERA and attorney Colette Holt). It defines availability as the share of ready, willing, and able firms in the relevant geographic and product markets, prescribes a custom-census/marketplace approach to build the availability denominator, and treats upstream barriers to business formation and capital access as evidence supporting both the disparity finding and race-neutral remedies.

  • 49 C.F.R. § 26.51USDOT DBE Program, race-neutral means

    Codifies the race-neutral remedy menu that directly targets formation/survival barriers: technical and supportive-services programs to build business-management, recordkeeping, and financial/accounting capability; assistance obtaining financing and bonding; and making contracts more accessible to small businesses, the regulatory toolkit a study recommends to enlarge the upstream pool of diverse firms.

  • Fairlie, R.W. & Robb, A.M., Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United StatesMIT Press, 2008 (Census Characteristics of Business Owners data)

    Core peer-reviewed evidence that lower Black business-ownership/success rates are driven principally by low startup capital tied to the wealth gap, plus lack of family-business background and human-capital differences, the four mechanisms named in this barrier.

  • Fairlie, R.W., Robb, A.M. & Robinson, D.T., Black and White: Access to Capital Among Minority-Owned StartupsNBER Working Paper No. 28154 (2020)

    Shows Black-owned startups begin and remain smaller across their first eight years and face worse credit terms, evidencing the survival half of the formation/availability gap (fewer diverse firms persist to become bidders).

  • Lofstrom, M. & Fairlie, R.W., research on Hispanic/Latino self-employment entrye.g., A Dynamic Analysis of Hispanic Business Ownership

    Extends the wealth-to-entry channel beyond Black entrepreneurs: low household wealth depresses Hispanic/Latino self-employment entry rates, supporting a multi-group upstream-availability finding for Boston's diverse population.

  • U.S. Census Bureau Annual Business Survey (ABS) and Nonemployer Statistics by Demographics (NES-D); legacy Survey of Business Owners (SBO, through 2012)Census business-ownership-by-demographics programs

    The authoritative federal data on business ownership and formation by race, ethnicity, sex, and veteran status used to benchmark local formation rates against national/regional norms for the availability analysis.

Boston data sources
  • U.S. Census ABS 2022 and NES-D 2022 (and the legacy Survey of Business Owners) for business ownership and formation by race, ethnicity, sex, and veteran status in CBSA 14460 and the five RGMA counties.
  • Census/ACS PUMS and Survey of Income and Program Participation (SIPP) microdata for owner wealth, education, prior experience, and self-employment entry/exit by demographic.
  • City of Boston Supplier Portal vendor registration records and the City and Massachusetts SDO certification directories for the custom-census availability denominator.
  • The 2020 BBC availability database (~800 firm interviews) as the prior custom-census baseline.
  • County Business Patterns / Nonemployer Statistics for industry and firm-size benchmarks in the Boston market area.
  • Analyze Boston contract-award and discretionary-spending data for the utilization numerator.
  • Bureau of Labor Statistics and BEA regional data for industry composition and the geographic/product market definition.
  • Primary-collected data: availability survey responses, owner survey responses, interview transcripts, public-hearing testimony, and trade-association/lender input.
16

Workforce, Union, and Labor-Market Access Barriers

Bench lead

Labor economist (human-capital and self-employment specialist) leads, running the self-employment/business-formation and outcomes regressions and the Oaxaca-Blinder decomposition; paired with the study's econometrician/survey methodologist for the workforce-availability, apprenticeship-pipeline, and work-hours disparity analyses, and supported by the legal scholar to tie group-specific anecdotal evidence to the Croson / H.B. Rowe evidentiary standard.

Diverse- and emerging-owned firms face restricted access to the inputs a contractor needs to perform: a pool of skilled tradespeople to staff a crew, entry into union halls and hiring halls that control much of the experienced construction labor supply, seats in registered apprenticeship and pre-apprenticeship pipelines that produce the next generation of journey-level workers, and the supervisory, estimating, and project-management experience an owner must accumulate before credibly running larger jobs. This barrier operates on two linked planes. First is the firm-as-employer plane: a minority- or woman-owned firm that cannot reliably recruit and retain skilled labor (often because it lacks the union affiliation, signatory relationships, or reputation that route experienced workers to established competitors) cannot scale crews to bid larger or more complex work, so its bidding capacity is capped. Second is the owner-as-human-capital plane: minority and women owners are, on average, less likely to have acquired the "business human capital" that comes from prior paid work experience in a similar firm or in a family-owned business in the trade, which is the informal, apprenticeship-style training that teaches estimating, bidding, crew supervision, and the practical know-how of running a contracting business. Both planes restrict who can form a viable contracting firm and how large that firm can grow.

Why it is a barrier

The barrier disadvantages diverse and emerging firms through measurable, evidence-backed channels. Human-capital channel: Fairlie and Robb, using the Census Bureau's Characteristics of Business Owners (CBO) microdata, find that prior work experience in a family member's business has a large, statistically significant effect on outcomes, with the probability of business closure 0.042 lower, the probability of large profits 0.032 higher, the probability of having employees 0.055 higher, and sales roughly 40 percent higher when the owner had worked for a self-employed family member before starting the business; they characterize the mechanism as informal "apprenticeship-type training." Black owners are substantially less likely than white owners to have had this prior exposure (Black entrepreneurs have fewer opportunities to acquire valuable pre-business experience through family businesses, and the lack of such experience and of "specific business human capital" acquired by working in similar firms contributes to worse outcomes among Black-owned businesses), so the very experience that predicts firm survival and growth is unequally distributed by race. Labor-supply and pipeline channel: registered-apprenticeship and union pathways, the dominant pipelines into the skilled construction trades, are themselves stratified. EPI's analysis of U.S. Department of Labor RAPIDS data finds women and workers of color participate and complete at higher rates in union (joint labor-management) apprenticeships than in nonunion programs (women's completion 37.6 percent in union-backed programs versus 23.4 percent nonunion). A diverse firm that cannot access union halls or signatory relationships is therefore cut off from the most diverse and best-trained labor pool, while one that operates nonunion draws from a thinner, less-completed pipeline; either way its crew-staffing capacity, and thus the size and complexity of jobs it can bid, is constrained relative to incumbents. Workforce disparity studies (Keen Independent's Kansas City Construction Workforce Disparity Study) document concrete "barriers that depress the number of women and people of color trained and employed in specific construction trades" and disparities in union and training-program participation, confirming the pipeline is not race- and gender-neutral. The net effect: emerging diverse firms are structurally limited in capacity (the workforce-availability constraint) and in the owner-level know-how that converts a small firm into a competitive bidder.

How the full study investigates it
Quantitative
  • Self-employment / business-formation regression (NCHRP 644 core test): using PUMS/ACS microdata, regress the probability that an individual is a self-employed business owner in relevant construction and professional-services industries on race, ethnicity, and sex while controlling for human-capital covariates (education, age/experience, prior industry work experience, hours, region). A residual minority/women shortfall after controls is the quantitative signature that human-capital and labor-market access barriers, not differences in qualification, suppress firm formation.
  • Business-outcomes regression: replicate the Fairlie-Robb specification on Survey of Business Owners / Annual Business Survey (and CBO-style) data, regressing firm survival, employment, sales, and profitability on owner race/sex with controls for prior work experience in a similar/family business, to test whether the human-capital experience gap, not owner characteristics, drives capacity differences in the relevant industries.
  • Workforce availability and utilization analysis: build availability benchmarks for the construction workforce (share of skilled tradespeople and apprentices who are minority/women by trade) and compare to utilization (actual minority/women work hours on City and City-assisted projects); compute work-hour disparity ratios by trade, mirroring the Keen Kansas City and Prosper Portland work-hours methodology.
  • Apprenticeship and union-pipeline analysis: tabulate registered-apprenticeship enrollment, completion, and journey-worker counts by race/sex/trade from DOL RAPIDS and the Massachusetts Division of Apprentice Standards for the Greater Boston market; compute participation and completion disparity ratios and compare union vs nonunion programs (EPI method).
  • Firm-capacity benchmarking: scale availability and utilization by firm size/capacity (employees, payroll, bonding-supported job size) so the disparity estimate isolates whether diverse firms are smaller because of constrained skilled-labor access, and run capacity-weighted disparity indices to avoid overstating availability of firms that cannot staff larger crews.
  • Marginal-effects and decomposition (Oaxaca-Blinder style) on firm outcomes to partition the diverse/non-diverse gap into an explained portion (measured human capital, prior experience) and an unexplained/residual portion consistent with discrimination or access barriers.
Qualitative
  • Business-owner survey (NCHRP 644 anecdotal instrument): stratified survey of Boston-area minority-, women-, and emerging-owned firms in construction, A/E, and professional services with a dedicated module on access to skilled labor, ability to recruit/retain crews, union/hiring-hall access, apprenticeship sponsorship, and the owner's own pre-business work and supervisory experience.
  • In-depth in-person/telephone interviews with diverse firm owners about concrete instances where inability to staff a crew, lack of union signatory status, or thin supervisory experience caused them to no-bid, downsize a bid, or lose a job, capturing the specific, group-linked anecdotes Croson and H.B. Rowe require.
  • Public hearings / public meetings inviting testimony from diverse contractors, workers, and apprentices on workforce-access barriers (the Keen Kansas City model of interviews, surveys, and public meetings).
  • Trade-association, union, and apprenticeship-sponsor input: structured interviews with building-trades councils, local union halls, NABTU-affiliated joint apprenticeship committees, nonunion (ABC) apprenticeship sponsors, AGC, minority contractor associations, and community-college/pre-apprenticeship programs in Greater Boston on diversity of intake, referral practices, and barriers to entry.
  • Worker-level interviews/surveys (where a workforce component is included) with minority and women tradespeople and apprentices on hiring, retention, advancement to supervisory roles, and pathways into the trades.
  • Key-informant interviews with City construction-management and compliance staff on how apprenticeship-utilization, prevailing-wage, and Boston Residents Jobs Policy requirements interact with diverse-firm crew availability.
Why it matters in Boston

Boston's RFP puts this barrier squarely in scope: the study must examine utilization of qualified female and minority workers on City construction projects and large City-assisted construction projects, not just firm-level utilization. The construction numbers explain why: BBC found MWBE construction availability of 19.7 percent against utilization of 6.7 percent (index 34) in a union-dense, hall-mediated labor market where crew access runs through signatory relationships and apprenticeship pipelines. The Boston Residents Jobs Policy already generates project-level work-hour compliance data by trade and demographics, giving the update a ready utilization base to pair with Division of Apprentice Standards pipeline data and to tie firm-side capacity findings to worker-side pipeline findings in a single evidentiary record.

Authorities, data sources, and the Boston angle
Legal and academic authorities
  • City of Richmond v. J.A. Croson Co., 488 U.S. 469 (1989)488 U.S. 469 (1989)

    Established strict scrutiny for state/local race-conscious contracting programs and the predicate that remedies must rest on a strong evidentiary basis of discrimination. Croson expressly identified training and financial aid for disadvantaged entrepreneurs of all races as among the array of race-neutral devices a locality should consider, directly authorizing the workforce/training-access frame as both a barrier to document and a race-neutral remedy lane.

  • H.B. Rowe Co. v. Tippett, 615 F.3d 233 (4th Cir. 2010)615 F.3d 233 (4th Cir. 2010)

    Persuasive Fourth Circuit authority, not controlling in the First Circuit. Upheld North Carolina's MWBE program for Black and Native American subcontractors but struck it for Hispanic, Asian, and women-owned firms for insufficient evidence, relying on a 2004 MGT disparity study. Confirms that group-specific, evidence-rich proof (including the marketplace/capacity conditions that constrain firms) is required and that anecdotal and statistical evidence must be tied to each group.

  • NCHRP Report 644, Guidelines for Conducting a Disparity and Availability Study for the Federal DBE Program (Wainwright & Holt, Transportation Research Board / National Academies, 2010)NCHRP Report 644 (2010)

    The National Academies methodological standard for defensible disparity studies (co-PIs Jon Wainwright of NERA and Colette Holt). Prescribes availability analysis, utilization/disparity ratios, regression analysis of business-formation and outcomes (to test whether minority status depresses self-employment and firm capacity after controlling for human capital), and structured anecdotal evidence via business-owner surveys, in-person interviews, and public hearings, the framework this barrier's investigation follows.

  • 49 C.F.R. 26.51 (USDOT DBE Program, race-neutral means)49 C.F.R. 26.51

    Defines the race-neutral remedy menu directly responsive to this barrier: assisting DBEs to improve long-term development and increase capacity, establishing programs to assist new start-up firms, and developing firm capability. Documenting workforce/human-capital barriers justifies race-neutral capacity-building, training, and mentorship remedies and frames the maximum-feasible-race-neutral analysis.

  • Fairlie & Robb, Race and Entrepreneurial Success: Black-, Asian-, and White-Owned Businesses in the United States (MIT Press, 2008); and Robb & Fairlie, 'Families, Human Capital, and Small Business: Evidence from the Characteristics of Business Owners Survey'Fairlie & Robb (MIT Press 2008); Robb & Fairlie, CBO Survey study

    Peer-reviewed empirical foundation for the human-capital/prior-work-experience channel. Quantifies that working in a family member's business (informal apprenticeship-type training) lowers closure probability by 0.042, raises large-profits probability by 0.032, raises employment probability by 0.055, and raises sales ~40%, and that Black owners' lesser access to this experience and to 'specific business human capital' from prior work in similar firms helps explain worse Black-business outcomes.

  • Keen Independent Research, City of Kansas City Construction Workforce Disparity Study (2019)Keen Independent (2019), Kansas City Construction Workforce Disparity Study

    Named, on-point workforce disparity study. Documented pathways into construction work and 'barriers that depress the number of women and people of color trained and employed in specific construction trades,' and examined minority/women participation in construction trade unions and training programs using interviews, surveys, and public meetings with contractors, unions, trade associations, and workers, a direct methodological template.

  • Economic Policy Institute, analysis of U.S. DOL Registered Apprenticeship (RAPIDS) data on construction apprenticeship diversityEPI, RAPIDS construction apprenticeship analysis

    Documents that women, Hispanic workers, and workers of color participate and complete registered apprenticeships at higher rates in union (joint) programs than nonunion programs (women's completion 37.6% union vs 23.4% nonunion), evidencing that the skilled-labor pipeline is stratified by union access, which in turn correlates with firm and owner access to skilled labor.

Boston data sources
  • U.S. Census Bureau ACS/PUMS microdata for the Boston-Cambridge-Newton metro area (self-employment, occupation, industry, education, experience, by race and sex).
  • U.S. Census ABS 2022 / NES-D 2022 and the legacy Characteristics of Business Owners survey for owner human-capital and prior-experience variables.
  • U.S. Department of Labor RAPIDS apprenticeship data and Massachusetts Division of Apprentice Standards records for enrollment and completion by race, sex, trade, and union versus nonunion sponsorship.
  • BLS Occupational Employment and Wage Statistics and Current Population Survey for skilled-trade and supervisory occupation counts by demographics in the local labor market.
  • Certified-payroll and Boston Residents Jobs Policy compliance records for City and City-assisted construction projects (work hours by trade and worker demographics) for the workforce-hours utilization analysis.
  • Union/joint-apprenticeship-committee and building-trades-council membership and intake data; nonunion apprenticeship-sponsor data.
  • Analyze Boston contract-award data and City subcontract records (to link firm capacity and bidding to workforce access).
  • Primary survey, interview, public-hearing, and trade-association/union-interview records generated by the study team.
From barrier to remedy
Every barrier here is addressable, and most by race-neutral means: lowering experience floors to the level the work requires, easing bonding and insurance, advertising informal opportunities, prompt payment, anti-exclusivity rules, and capacity building. The study's recommendations lead with those, and reach race-conscious tools only where the evidence shows a group-specific gap that neutral measures cannot close.

One barrier, worked in full

The firm-experience requirement is the clearest example of a facially neutral screen that converts past exclusion into present disqualification. It has its own module, with the federal precedent for crediting personnel experience and a scan of how often procurements actually allow it.