Contracting and Innovation

The problem

The way America traditionally delivers transportation projects — design completed in isolation, then priced by competitive bid, then rebuilt through change orders when the design meets ground truth — systematically discovers its true cost after the contract is signed. The most comprehensive dataset is Flyvbjerg, Holm and Buhl’s study of 258 transportation projects across 20 countries: nine out of ten projects experienced cost overruns, with average escalation of 45 percent for rail, 34 percent for fixed links, and 20 percent for roads — and forecast accuracy did not improve across the 70 years the data span. The finding is not that estimators err randomly. Errors run one way, in every decade and every project class studied.

The annual August Redistribution is the domestic, administrative face of the same problem.

Each year, States return obligation authority they cannot use on schedule, and FHWA reallocates it to States that can: $7.62 billion in FY2025, against $8.20 billion requested by States able to absorb more, in a pattern repeated at similar scale in FY2023 ($7.9 billion) and FY2026 (just over $7.5 billion). Billions in annual churn is not a funding shortage; it is a delivery-capacity shortage, distributed unevenly across States. Alternative delivery methods exist and are documented (Part IV); what the system lacks is an incentive to bear the cost and risk of first use, and an institution charged with spreading the competence the methods require. Both absences sit where this subtitle aims: in the contracting method chosen at the start of a project, and in whether the agency choosing it has ever been taught how.

The way America traditionally delivers transportation projects, design completed in isolation, then priced by competitive bid, then rebuilt through change orders when the design meets ground truth, systematically discovers its true cost after the contract is signed.

The proposed solution

Pay more of the bill when a project is delivered the modern way. When a State or local agency builds a project using one of three proven methods — design-build, where one team designs and builds under a single contract; construction manager/general contractor, where the builder is at the table while the design is still being drawn; or a public-private partnership under State law — the federal government raises its share of the project’s cost by five percentage points, up to paying 95 percent of it. No agency is required to use any of these methods. The offer simply makes the better-tested route the financially easier one to choose.

And build the school. A new federal Center of Excellence for Project Delivery teaches agencies how the methods work: training, standard contract templates any agency can adopt, a public library of best practices and case studies, and direct, hands-on help for agencies using a method for the first time — the moment of highest risk. The arrangement then holds itself to account: every two years, a report to Congress counts the projects that took the incentive and assesses what the methods did to their schedules and costs, so the next evaluation is built into the policy itself.

Use a modern delivery method, and the federal government pays five points more

The methods stay optional and subject to State lawThe incentive is a price signal, not a mandate.

A federal center teaches the methodsTraining, templates, case studies, and hands-on help for first-time users.

Results are reported to Congress every two yearsThe incentive is measured, not assumed.

The legislative proposal

Subtitle C states its purpose (Sec. 507) in one sentence — to accelerate project delivery, improve cost-effectiveness, and foster private-sector innovation “by promoting the use of modern contracting and project management techniques by State and local transportation agencies” — and builds two instruments.

The incentive — Sec. 508

For any project carried out using an “innovative contracting method,” the Secretary must increase the federal share payable — as determined under Section 106 of the Act — by 5 percentage points, to a maximum total federal share of 95 percent. The term is defined exhaustively: design-build; construction manager/general contractor; and public-private partnership under applicable State law. The incentive is a price signal, not a mandate: the methods remain subject to State law, and no agency must use any of them.

The institution — Sec. 509

The Secretary must establish a Center of Excellence for Project Delivery within the Department, for technical assistance, training, and standardized contract templates; a public repository of best practices and case studies; and hands-on help to first-time users. The Center reports to the House Transportation and Infrastructure and Senate Environment and Public Works Committees within 2 years, and biennially thereafter, on the projects that used the Sec. 508 incentive and its impact on schedules and costs.

Section 507Purpose.

Section 508Incentives for innovative contracting.

Section 509Technical assistance and best practices.

Source: FRAME Act, Official Draft, December 15, 2025, Sections 507 to 509.

The data case

Why it works: What the Methods Evidence Shows

Against the legacy record set out in Part I, the federal government has run the alternative-contracting experiment for two decades through the SEP-14 program, and FHWA’s Design-Build Effectiveness Study is its principal evaluation. Across the program, managers estimated that design-build reduced project duration by an average of 14.1 percent (median 10.0 percent) and cost by 2.6 percent (median 0.0 percent), at unchanged quality, relative to design-bid-build; earlier studies reviewed in the same report found duration reductions of 4 to 60 percent. The mechanism is structural, not managerial: a single procurement instead of two, design and construction overlapped rather than sequential, and a contractual incentive to design what can be built. The study’s paired comparison is more conservative and more instructive — design-build projects showed negative schedule growth (−4.2 percent against plan) where design-bid-build projects grew (+4.8 percent), while design-build cost growth (7.4 percent) exceeded design-bid-build’s (3.6 percent) in that small sample. Read whole, the evidence supports a precise claim: the alternative methods’ most reliable dividend is schedule certainty; their cost effect is smaller, variable, and dependent on the owner’s sophistication.

That dependence is the case for Sec. 509. The research infrastructure around these methods is now mature — the National Academies’ NCHRP Research Report 939 devotes full guidebooks to post-award contract administration for design-build and for construction manager/general contractor delivery, and the CM/GC literature (Alleman, Molenaar and colleagues, Transportation Research Record) now covers the majority of federally funded CM/GC projects delivered between 2004 and 2015. Knowledge, in other words, exists; what is scarce is its distribution to the hundreds of State and local agencies that have never run these procurements and face their highest risks on first use. A five-point share premium without capability would subsidize method-shopping; capability without a premium leaves the first-mover bearing all the learning cost. The subtitle pairs them, and then — unusually — requires the pairing to be evaluated: Sec. 509’s biennial report must count the projects that took the incentive and assess their schedules and costs, building the next evaluation into the statute itself.

Against the legacy record set out in Part I, the federal government has run the alternative-contracting experiment for two decades through the SEP-14 program, and FHWA’s Design-Build Effectiveness Study is its principal evaluation.