Get the Best Value in Transit Procurement | IFP

Content

This piece is part of IFP’s Transit Abundance Playbook, a collection of proposals for reducing American transit construction costs.

Summary

State and local procurement rules govern how federally funded transit contracts are awarded, and they overwhelmingly favor selection of the lowest-priced rather than the best-value bidder. Despite evidence against low-bid selection, many states and localities mandate the lowest-bid method, and even states with more flexibility still require most of a bidder’s score to be based on their initial bid amount. Common misconceptions about the low-bid method, including its supposed cost-effectiveness or ability to limit corruption, entrench its use. On the contrary, low-bid selection is linked to higher infrastructure costs, contributes to chronic cost overruns and delays, and is more vulnerable to bidder manipulation. These risks are greatest for high-complexity projects like transit construction, where technical mistakes can derail performance.

Best-value selection is an evidence-based alternative shown to result in more cost-effective projects. This method selects contractors on the basis of technical merit, experience, and lifecycle costs in addition to factors such as bid price. Federal intervention can accelerate state and local authorization and use of best-value selection by (1) conditioning major infrastructure grants on the use of best-value selection; (2) tailoring federal-grant scoring to incentivize the development of best-value expertise; (3) establishing clear guidance on appropriate cost and non-cost weights for selecting bids; and (4) charging federal agencies to develop guidance to assist implementation of best-value selection. These reforms would modernize transit procurement processes in line with low-cost models in other countries and curb waste.

Problem

“You get what you pay for” rings true in infrastructure procurement. Yet many infrastructure contracts are awarded to the nominally cheapest bidder, including roughly 80% of US highway construction contracts. This practice paradoxically makes projects more expensive once they get underway. By some estimates, annual losses tied to low-bid selection range into the hundreds of millions, if not more.¹ And prolonged construction can add years of disruption for communities it is meant to serve, feeding perceptions of government dysfunction. The primary alternative, best-value selection, assesses bids according to metrics beyond just bid price, taking into account past performance, technical rating, risk, and lifecycle costs. This more holistic assessment often results in more successful — and ultimately lower-cost — projects.

Federal agencies have deployed best-value selection with great success for decades, but they currently lack the latitude to encourage its use at the state and local levels.² While most transit construction relies on some amount of federal funding, states and localities manage the finer details of such projects. In total, federal spending accounts for $650 billion of the nation’s $2 trillion in annual procurement spending. The federal grants that supplement state expenditures often only require that contracts be awarded through an “open and competitive” process, leaving key selection details to the funding recipient. States and localities thus must choose which procurement methods to deploy based on the statutory requirements set by their legislatures.

As a result, states and localities often make selection decisions based on bid price alone. Sometimes low-bid selection is mandated by their procurement laws.³ The Model Procurement Code, a baseline framework for most state and local procurement law that around 60% of states and localities have at least partially adopted, mandates that “contract[s] shall be awarded…to the lowest responsible bidder.” And even if the law permits best-value selection, the transit agency or other government body managing the procurement may opt not to use it, operating under the misconception that low-bid procurement is cheaper or invites less corruption. Bid protests filed by disgruntled bidders to contest processes or awards can further deter best-value selection. Procuring officers, particularly those operating without broad best-value authority or clear guidance for its use, may worry that mistakes during implementation could lead to costly litigation. Limited state capacity reinforces this dynamic, entrenching low-bid selection because it is the simpler option to implement.

But low-bid selection leads to worse project outcomes in all forms of procurement (and transit in particular). San Francisco’s Central Subway Project, a 1.7-mile downtown subway line, used the selection method for its stations and systems work. By construction completion in 2022, the project’s total price tag was more than $1.2 billion above comparable projects completed in other advanced economies, where best-value selection is common.⁴ Compared to preconstruction estimates, the project ran four years behind schedule and $375 million over budget (on top of the already higher US baseline costs compared to international peers). A State of California study found low-bid selection substantially contributed to the project’s problems. The study’s authors identified that the procuring agency’s failure to consider factors like technical merit and risk of schedule growth was unsuited for a complex subway construction. Complications that arose once the project got underway interfered with other interdependent project components, multiplying the magnitude of added costs and delays.

Maryland’s Purple Line construction similarly suffered the consequences of making a low-bid selection. The state abandoned a plan to weight non-price factors heavily after budgetary pressures led the government to “laser[] in on price.” The state’s chosen bid, despite being the cheapest upfront, ranked among the worst on technical merit. The project later overshot its original contract price, with the contractor claiming $800 million in cost overruns. The Purple Line, approved in 2016, is now approaching $10 billion in costs — almost five times the projected cost — and is not expected to be complete until at least the end of 2027. A former Maryland transportation official noted that the decision to overweight bid price “put an artificial lens on the review process,” as “[o]ne contractor’s price proposal was so below everyone else.” Selection of a higher-priced but more technically sound proposal could likely have avoided some of the project’s pitfalls.

Empirical research consistently links low-bid selection to worse outcomes, including higher overall costs, longer timelines, and reduced quality. The disparities are most glaring for the highest-complexity projects, like constructing new transit lines, which demand advanced technical expertise. Bidders with more accurately priced (and thus higher) bids, more experienced teams, and better risk-management plans are, by design, screened out. One study that surveyed global procurement practices found that stringent regulations, such as low-bid mandates, worsened procurement outcomes in developed countries like the United States.⁵ Indeed, greater use of low-bid selection may explain why the United States spends multiple times more than peer countries on comparable infrastructure.

Low-bid selection leads to worse outcomes because it misaligns bidder incentives. The exclusive focus on bid price rewards contractors for cutting corners to ensure their bids are competitive. In practice, this means contractors may underinvest in risk management and sacrifice project quality to win contracts. The selection strategy ignores project lifecycle costs, and only incentivizes innovation if it can reduce upfront costs and thereby lower bid price. This leaves projects less likely to succeed, and both construction and long-term maintenance are less cost-effective.

Low-bid selection also invites strategic underbids, wherein a bidder submits a proposal at an unattainably low price and expects to recoup actual costs after shovels hit the ground.⁶ Contractors often have the upper hand with these change-order requests — agencies undertaking a “too big to fail” project, or those wary of litigation, may not contest even unjustified change orders. Such risks are well known at the federal level, and are cited by federal agencies as a reason to instead use best-value selection. The prevalence of underbidding in low-bid selections has helped make cost overruns the norm in US infrastructure procurement. The practice may have derailed San Francisco’s Central Subway construction: Tutor Perini, the winning bidder and one of the largest general contractors in the United States, has faced repeated scrutiny for underbidding California state contracts procured via the low-bid method, only to file for change orders once projects got underway.⁷

One misconception is that low-bid selection reduces corruption and collusion, but the opposite may be true. A comprehensive study of public construction contracting found that the simplicity of the low-bid method makes bid-rigging by contractors easier and more common. Many construction contracts have a small bidder pool, so participants that are known to one another can collude to rig their bids and drive up the contract price. This practice is difficult to detect. Best-value selection, by contrast, permits agencies to gauge bids on their merits and is harder for bidders to rig.

Some states and localities have attempted incremental procurement reforms, but results have been correspondingly marginal. Many remain constrained by restrictive laws or outright bans on best-value selection. A roadblock to rapid reform is that the government responsible for conducting the procurement must possess its own best-value authority to use the method. For example, New York state first authorized best-value selection in 2011, but New York City did not adopt it until two years later and was unable to use it during the interim. After adoption, city agencies were still barred from using the method for key procurement categories such as public works. It was not until 2024 that New York City authorized best-value selection for major infrastructure projects. This patchwork process deters, and sometimes altogether prevents, state and localities from deploying best-value selection.

Even when state or local agencies have authority to perform best-value selection, many overindex on the initial bid price (whether due to law or agency practice), so selection continues to operate like the low-bid method. For example, Delaware state law mandates that procuring agencies weight bid price at 70–90% of the overall score, with less emphasis on proposed schedule and a contractor’s past performance. Minnesota grants broad best-value authority without prescribing weight allocations, yet the procuring agencies often exercise that discretion to assign disproportionate weight to bid price. Despite attempted reforms to implement best practices in procurement, state and local agencies may perpetuate the same failure modes. Research indicates that unless non-cost factors receive sufficient weight, they will not affect the bidding outcome.

International experience with best-value selection indicates non-cost factors should be weighted higher than cost factors. Madrid’s 1999–2003 metro extensions, which had low construction costs and were built on schedule, weighted bid price at just 30% while allocating 70% to non-cost factors like technical rating and contractor qualifications. Greater use of best-value selection for transit projects procured at the state and local level would be in line with existing efforts at the federal level. A Government Accountability Office study of Department of Defense contracts procured using best-value selection found that the agency weighted non-cost factors, such as technical approach and past performance, more heavily than cost factors, regardless of the acquisition price. Of the 96 randomly sampled contracts, none weighted cost above non-cost factors.

Solution

Targeted federal intervention can help overcome barriers to best-value selection and accelerate its adoption among states and localities. Federal reforms should:

Condition federal grants on the use of best-value selection;
Tailor federal grant scoring to incentivize the development of best-value expertise;
Establish clear guidance on appropriate cost and non-cost weights; and
Assist state and local use of best-value selection by developing guidebooks containing best practices.

Private bodies also have a role to play. The American Bar Association’s ongoing revisions to the Model Procurement Code, which functions to modernize and coordinate procurement-law reforms across states and localities, should incorporate similar reforms with express best-value authorization. These changes would benefit a variety of state and local procurement activities, but would be especially consequential for complex transit construction.

The first, and most straightforward, federal reform is to condition federal transportation funding on authorization — and, when appropriate, the use — of best-value selection. The statute governing the Capital Investment Grants (CIG) Program, 49 U.S.C. § 5309, could require that grants go to projects in states that permit transit agencies to use best-value selection. Congress could add a proviso making best-value a condition for grant eligibility in § 5309 subsection (c), or in the project evaluation and rating provisions in subsection (g). Grants for high-complexity projects that would most benefit from best-value selection, like urban transit construction, could be made contingent on the state or localities agreeing to use the procurement method. Congress could also incentivize transit projects’ access to best-value selection by conditioning “expedited technical capacity review” in subsection (c)(3) upon a procuring entity’s access to best-value authority.

Although amending federal statutory language is important to promote broad adoption of best-value selection, the executive branch could also use its existing authority to issue binding regulations or strongly presumptive guidance on this topic. Subsection (c)(1)(B) of the CIG statute permits the US Secretary of Transportation to award grants after determining that an “applicant has, or will have…the legal, financial, and technical capacity to carry out the project….” USDOT could update the definition of “technical capacity” to include express authorization for best-value selection. The Federal Transit Administration (FTA) could incorporate the existence of best-value authorization as a factor in the Secretary’s project scoring for New Starts or Small Start Projects.

Second, Congress should incentivize the development of best-value expertise at transit agencies by including an applicant’s demonstrated best-value capacity in its project scoring for the CIG program. Best-value selection requires more upfront work to design an appropriate selection framework and evaluate bidders against it. Many states and localities also procure major infrastructure projects infrequently, meaning they may have less implementation experience. Inexperience raises the risk of mistakes that can result in costly bid-protest litigation. A clear procedure for how administrators should conduct best-value selection is thus critical, including step-by-step instructions on tailoring selection criteria and completing bidder evaluation. To incentivize capacity-building, the CIG statute’s project evaluation criteria under subsection (g)(2) could award higher ratings to applicants that demonstrate best-value procurement capability. The higher rating might be tied to having staff with demonstrated expertise or training in best-value selection, published plans for using best-value selection, or documented compliance with federal best-value guidance.

Third, Congress should direct federal agencies — or they should act on their own authority — to develop regulations or guidance establishing appropriate cost and non-cost weights, compliance with which could be made a condition for receiving federal grants. Best-value selection involves consideration of many factors that can include technical merit, contractor qualifications, past performance, quality of personnel, and bid price. Adequate assessment of each criterion demands discretion for procurement officers, with bid protests and criminal statutes as guardrails against misuse. But as noted above, flawed weights can undermine selection. Research indicates that weighting bid price at or above 57% of the total ultimately functions like low-bid selection and loses the benefits from focusing on technical merit or other factors.

FTA should publish preferred weights to ensure best-value selection is effective. Recommended weights should account for project type (e.g., rail, bus rapid transit, highway, or bridge) and delivery method (e.g., design-build or design-bid-build), where the relative importance of factors like technical rating, past performance, or long-term maintenance may be different. Recommendations could draw on the successful weight schemes implemented at the federal level or internationally, where non-cost factors may be weighted as high as 70%, as shown in Chart 2 above.

Fourth, federal agencies could aid implementation by expanding best practices in existing guidebooks. FTA should update Circular 4220.1G (Third-Party Contracting Guidance), covering compliance with statutory requirements and agency guidance to “assist recipients in administering FTA-funded projects and in meeting the responsibilities and reporting requirements of FTA awards.” The current version permits recipients to use best-value selection — when authorized by their state or localities — but does not mandate it or provide detailed guidance about how or when to implement it. Additional guidance could supplement Circular 4220.1G and the FTA’s Best Practices Procurement & Lessons Learned Manual. While the FTA’s Best Practices Manual discusses best-value implementation, it does not provide sufficient guidance on evaluation criteria, weights, scoring methodologies, or defensible documentation practices specific to best-value selection.

For highway projects, the Federal Highway Administration could include similar resources in the Center for Innovative Finance Support. State and local officials in charge of procurement could reference these enhanced guidebooks to implement best-value selection with more consistency and success. And the guidance’s very existence could support officials defending their processes and ultimate evaluation decisions from frivolous bid protests.

Private bodies can help facilitate state and local adoption of best-value authority. The American Bar Association (ABA) should make best-value procurement authority a priority in its ongoing revisions to the Model Procurement Code (MPC). As discussed above, the MPC is a widely used framework for state and local procurement laws and regulation. Reforming the MPC thus presents an efficient way to speed broader adoption and use of best-value selection across the country. The current version provides that “contract[s] shall be awarded…to the lowest responsible and responsive bidder,” permitting deviations only in “very limited circumstances.” The ABA should remove the “lowest” terminology altogether. The procuring agency should still be empowered to use low-bid selection when it is deemed most efficient, but broader language would provide more solid footing for best-value selection. The MPC could also include language preserving procuring-officer discretion to make judgment calls, to the effect of:

“…the procuring entity must award the contract to the bidder that represents the best value to the public; if the award is made to a bidder who did not receive the highest objective score under the pre-bid solicitation, the decision must be accompanied by a written justification supported by a reasonable basis.”

The code revisions should not impose specific limitations on the relative weights of cost and non-cost factors, nor express a preference for price over other key factors, like technical merit.

A significant share of America’s infrastructure cost crisis stems from state and local laws and practices that prohibit or constrain best-value selection. Continued reliance on the alternative, low-bid procurement, drives up project costs and undermines infrastructure construction. This contributes to the perception that the American government cannot build the infrastructure the public needs. Changing federal funding requirements, revising federal guidelines, and reforming model code provisions can extend the benefits of best-value selection to state and local agencies. These solutions would modernize a crucial but oft-neglected area of government.

Summary

Problem

Solution

Further Reading