Performance Management

The problem

The federal performance regime measures; it does not manage. Since MAP-21, States have been required to set performance targets for safety, infrastructure condition, congestion, and freight — and to report against them. But the targets are self-set and unreviewable: the Federal Highway Administration’s own guidance states that MAP-21 “does not provide FHWA the authority to approve or reject State DOT or MPO established targets.” A target may lawfully decline from one year to the next. Nothing in the regime tests the quality of the target itself, and nothing follows when a target is missed.

The audit record shows what that design produces. Reviewing 2014–2017, the Government Accountability Office found that States did not achieve most of the fatality-related targets they had set for themselves, and that nearly half of States did not provide the required assessment of progress at all (GAO-20-53). The weakness is structural, not clerical: targets that cannot be reviewed cannot be failed, and a measure that cannot be failed is not a management tool — it is a filing.

The federal performance regime measures; it does not manage.

The proposed solution

FRAME keeps the architecture that already exists — national goals, state targets, public reporting — and adds the two things the present system lacks: a federal check on the quality of the targets themselves, and a defined consequence when the safety target is missed year after year. The proposal works in five steps.

• Six national goals, stated plainly. The whole federal program is aimed at six results: far fewer deaths and serious injuries on every public road; a system kept in good repair; less congestion on the national highway network; a system that works efficiently and reliably for everyone who uses it; a freight network strong enough to support the economy; and better performance that protects, rather than degrades, the natural environment.
• Common yardsticks. Within eighteen months, the federal government publishes one set of clear, data-driven measures that every state uses, so that a result in one state means the same thing in another.
• Targets a state must defend with data. Each year, every state sets its own targets for each measure, working with its metropolitan planning organizations and transit providers. Washington then reviews each target. If a target is not supported by the best available data, or is not a reasonable measure of progress, the state can be required to adjust it. States still set their own targets — they must now be able to show their working.
• One public report card, and a yearly verdict. Progress is reported once a year in a single public document — the state’s infrastructure health report card — and the federal government determines annually whether each state is making significant progress. Where a state is falling short, it receives technical assistance and a concrete list of actions it can take.
• One consequence, for safety only. If a state misses its safety targets three years in a row, it gives up one percent of its main federal block-grant funding. The money does not leave the state and does not return to Washington: it is re-awarded inside the state, competitively, to the projects judged most likely to cut deaths and serious injuries. A separate bonus pool elsewhere in the Act rewards the states that beat their targets, so the framework pays for performance in both directions.

The design is deliberately narrow. There is no federal power to set a state’s targets, no penalty for missing a congestion or condition goal, and no forfeiture before three consecutive years of safety failure. What changes is that targets become commitments a state must defend with evidence — and that persistent failure on the one measure that counts lives carries a cost that is itself spent on fixing the failure.

Six national goals, stated plainly

Common yardsticks

Targets a state must defend with data

One public report card, and a yearly verdict

One consequence, for safety only

The legislative proposal

Section 401 — National Goals and Performance Measures

Section 401(a) declares it the policy of the United States that the federal surface transportation program be administered to achieve six national goals: a significant reduction in traffic fatalities and serious injuries on all public roads; a system maintained in a state of good repair; a significant reduction in congestion on the National Highway System; improved efficiency and reliability for all users; improved freight network performance in support of economic vitality; and enhanced system performance while protecting the natural environment. Section 401(b) directs the Secretary, not later than 18 months after enactment, to promulgate a rule under section 553 of title 5 establishing “clear, data-driven performance measures” for States to use under the subtitle.

Section 402 — The “Pay for Performance” Framework

Section 402(a) requires each State, not later than one year after the section 401 rule and for each fiscal year thereafter, to set data-driven performance targets for each measure, coordinating with metropolitan planning organizations and public transportation providers to ensure consistency to the maximum extent practicable. The decisive change is section 402(a)(3): the Secretary reviews each target, and if a target “is not supported by the best available data or is not a reasonable measure of progress,” the Secretary may require the State to adjust it. Under current law, FHWA’s own guidance states that MAP-21 “does not provide FHWA the authority to approve or reject State DOT or MPO established targets.” Section 402 supplies that authority.

Section 402(b) routes progress reporting through a single public instrument: the annual State Infrastructure Health Report Card required under section 406, with contents specified at section 407. Section 402(c) requires the Secretary to determine annually whether a State has made “significant progress” toward each target; where a State is not making significant progress, the Secretary provides technical assistance and identifies actions the State may take.

Section 403 — Safety Performance Forfeiture

Section 403 is the consequence, and it is confined to safety. For each fiscal year, the Secretary identifies any State that has failed to make significant progress toward its safety performance targets for three consecutive fiscal years (Sec. 403(a)). A State so identified forfeits an amount equal to 1 percent of the funds apportioned to it under subtitle B of title II, in the first fiscal year following identification (Sec. 403(b)). The forfeited funds do not leave the State: under section 403(c) they are made available as competitive grants to eligible applicants, as defined in section 305(c), within that State, for the projects the Secretary determines “will most effectively contribute to the reduction of traffic fatalities and serious injuries.” The reward side of the framework — the Performance Bonus Pool for States exceeding their targets — sits in title II (Sec. 105); subtitle A carries the standard, the review, and the sanction.

Section 401National goals and performance measures.

Section 402"Pay for Performance" framework.

Section 403Safety Performance Forfeiture.

Source: FRAME Act, Official Draft, December 15, 2025, Sections 401 to 403.

The data case

The case for the proposal is, first, the documented failure of the alternative. The existing regime has been audited, and the audit is the case: targets self-set, unreviewable, and permitted to decline produced exactly what GAO recorded — most fatality targets missed, required progress assessments unfiled by nearly half of States (GAO-20-53). Each element of subtitle A answers a documented defect in sequence: targets the Secretary can test against the data (Sec. 402(a)(3)), an annual public determination of progress (Sec. 402(c)), and — for safety alone, after three consecutive years — a forfeiture that is reinvested in the same State’s fatality-reduction projects (Sec. 403).

Where targets are instead bound to a managed program and published data, the numbers move. New York City’s Department of Transportation reports that, comparing 2023 with 2013 — the year before Vision Zero — overall traffic deaths fell by more than 12 percent (299 to 257) and pedestrian deaths fell by 45 percent, across more than 1,200 completed safety projects, with eight of the ten safest years on record occurring in the Vision Zero era. Sweden’s experience is the longer proof of the same pairing: Vision Zero, adopted in 1997 alongside system redesign, cut road deaths from 541 in 1997 to 192 in 2021 while traffic grew. The pattern in both cases is identical: a stated goal, measured publicly, with an institution answerable for the trend.

The spending side shows the same failure of management without consequence. In FY2025, $7.62 billion of obligation limitation had to be redistributed in August to States able to use it, against $8.198 billion requested (FHWA Notice N4520.291) — buying power idle in one State while another waits. A regime that cannot test a target also cannot see a program drifting until the annual redistribution exposes it; the report-card and determination machinery is what makes drift visible in time to correct it.

The sequence of the design is itself the mechanism. Review disciplines the target before the fact: a State that must show its working sets a target it expects to be measured against. The annual determination then creates a public record, year by year, that legislatures, press, and road users can read without waiting for an audit cycle. Technical assistance comes before any sanction, so a State that is failing is first given help to stop failing. Only the State that has been measured, published, and assisted — and has still missed on safety three years in succession — reaches the forfeiture. Information first, assistance second, consequence last: each step is cheap to comply with and expensive to ignore.

The three-year trigger is a statistical choice as much as a legal one. Annual fatality counts, particularly in smaller States, are volatile: a single year’s movement can reflect exposure and chance rather than trajectory. Requiring three consecutive years of missed significant progress confines the sanction to States whose trend, not whose luck, has failed — and it protects the instrument’s credibility, on which its deterrent effect depends.

Subtitle A is the standard-and-sanction half of a framework whose reward half sits in title II: the Performance Bonus Pool (Sec. 105) pays States that exceed their targets, drawn from the same measurement system. The pairing answers the obvious objection to reviewable targets — that States will protect themselves by setting targets low. A low target is exposed twice: by secretarial review against the best available data, and by the State’s own forfeiture of the bonus a more ambitious target would have earned.

The case for the proposal is, first, the documented failure of the alternative.