DOT Is Sitting on $160.7 Billion It Never Obligated. That Pile Is Now the Offset for the Next Highway Bill.

Two numbers tell the entire story of federal infrastructure funding this quarter.

The first is $36.8 billion — the annual IIJA Division J advance appropriations that lapsed with FY2026 when the continuing resolution extended surface transportation authorities through December 11, 2026 without extending their money. We covered that gap when it was still prospective, in Congress Extended Transit Authority and Deleted Its Money.

The second is $160.7 billion — the amount of IIJA funding the Department of Transportation had not obligated as of April 30, 2026, according to National Taxpayers Union Foundation analysis. Of that, $74.9 billion is Highway Trust Fund-backed, representing 20.8 percent of overall HTF authority, and $85.8 billion is non-HTF.

Almost all the coverage has been about the first number. The second is the one that will decide what happens to your award.

Here is why. Congress is now trying to write a five-year surface transportation reauthorization — the BUILD America 250 Act, H.R. 8870, authorizing roughly $580 billion — in a fiscal environment where every dollar needs an offset. The House Transportation and Infrastructure Committee approved its portion in May 2026. Other House committees have not finished. The Senate has not released text. And sitting in plain view, in every quarterly report, is a $160.7 billion balance of money Congress already provided and DOT has not yet put under obligation.

That balance is the most politically available offset in the federal budget. NTU Foundation's recommendation is explicit: Congress should "account for existing funds" and "assess whether higher funding levels are justified" before approving a $580 billion authorization.

"Unobligated" does not mean what most grantees think it means

This is the single most expensive misunderstanding in infrastructure grant management, so it is worth being precise.

Federal money moves through four distinct states, and the gap between them is where awards die:

1. Appropriated or authorized — Congress has provided the budget authority.
2. Announced or selected — the agency has published your name in a press release or sent a selection letter.
3. Obligated — a binding grant agreement is executed and the government is legally committed.
4. Outlaid — cash has actually moved.

A selection announcement is state 2. It is not a contract. Until a grant agreement is signed, the funds backing your project remain in the unobligated column, which means they remain fully exposed to rescission, reprogramming, reallocation, and policy-based cancellation.

Grantees routinely treat the press release as the finish line. It is roughly the halfway mark, and in the current environment the second half has become the dangerous half.

We have already documented what the exposed half looks like in practice. The SMART Grants program was terminated outright, with $204.9 million in unobligated funds reallocated — 122 Stage 1 and seven Stage 2 awards honored, no new NOFOs, and selected-but-unexecuted projects left holding nothing (details here). EDA Tech Hubs Phase 2 went through a rescission and re-award cycle that reshuffled which hubs got money (details here). The mechanics by which already-appropriated money gets killed are not hypothetical; we catalogued five of them.

What is new is the scale of the available target.

The trend line is the argument

If unobligated balances were falling steadily — money flowing out as projects matured — the case for rescission would be weak. You do not claw back a pipeline that is clearing.

The NTU Foundation figures do not show a clearing pipeline. The unobligated share of IIJA transportation funding ran 27.3 percent in January 2026, 30.4 percent in March, and 29.5 percent in April. Essentially flat, across a quarter, four and a half years into a five-year bill.

A flat unobligated share is a structural claim, not a timing artifact. It says the constraint on getting infrastructure money into the ground is not appropriation volume. It is project readiness, environmental review, local match capacity, staffing at state DOTs and transit agencies, and the obligation-processing throughput of federal program offices. Appropriating more money does not relieve any of those constraints — which is precisely the argument deficit hawks will make against the $580 billion number, and it is an argument with real evidence behind it.

For grant seekers, the uncomfortable implication is that the readiness problem is partly yours. Every month between selection and executed grant agreement is a month your project contributes to the statistic being used to justify cutting programs like yours.

What the competitive programs lost, and what it reveals

The National League of Cities' accounting of the Division J lapse covers roughly 25 competitive programs. Three figures from it are worth isolating, because they are not about scarcity — they are about demand:

• BUILD grants: about $1.5 billion in annual funding, against applications exceeding available funds by $12.8 billion.
• Safe Streets and Roads for All: about $1 billion annually, with $3.4 billion more requested than available. We flagged the FY2026 round as the program's last at the time: the final round of Safe Streets for All.
• Rail — RCE and CRISI: support for 186-plus rail projects, likewise oversubscribed.

Formula programs took losses too: roughly $5.5 billion a year for bridges, in a country where one in three bridges still needs repair or replacement; about $1 billion a year in transit for vehicle replacement and services for seniors and people with disabilities; $68.4 million a year for ferry boats.

Put the two datasets side by side and you get the paradox that defines this moment. The competitive programs are oversubscribed by nine or ten times available funding. And nearly 30 percent of the overall IIJA transportation appropriation is unobligated. Both are true, and they are not in contradiction — competitive discretionary programs with strong, shovel-ready applicant pools clear their money fast, while large formula and mega-project categories with multi-year environmental and design timelines carry enormous balances for years by design.

The policy risk is that the second fact gets used to cut the first. Rescinding unobligated balances is easiest where balances are largest, but reauthorizing at a lower level hits the programs where demand is highest. The accounting and the demand are concentrated in different places, and nothing about the legislative process guarantees they get matched up correctly.

What to do between now and December 11

If you hold a selection letter that is not yet an executed grant agreement: this is your priority above all other grants work. Call your program office this week and ask one question: what is outstanding on our side before obligation? Then clear it. Environmental documentation, match certification, DBE program, title VI assurances, buy-America documentation, real estate certifications — whatever the item is, every week it sits is risk you are choosing to carry. Do not wait for the agency to chase you.

If you are mid-execution on an obligated award: you are substantially safer, but not immune, and your exposure is concentrated in unspent later-year increments and in options or phases not yet obligated. Know, in dollars, which portion of your award is obligated versus merely programmed. If you cannot produce that split from memory, your finance office should produce it this month.

If you were planning to apply to a Division J-funded competitive program in FY2027: assume no NOFO until there is either a full-year appropriation or a reauthorization, and build your timeline around the December 11 date rather than a historical posting month. The SMART precedent is instructive — the program did not just go unfunded, it went away, and "no new NOFOs" was announced rather than implied.

If you depend on Highway Trust Fund contract authority programs — Bridge Investment Program, INFRA, Rural Surface Transportation, Wildlife Crossings — you are in the insulated tier, and that tier is where FY2027 planning effort should concentrate. Contract authority draws on the trust fund rather than annual discretionary appropriations, and the CR treated it very differently from Division J. That said, $74.9 billion of the unobligated balance is HTF-backed, so "insulated from the appropriations fight" is not the same as "invisible to the offset hunt."

If you work in advocacy: the NLC guidance is to contact congressional offices with specifics — name project types, request direct staff contacts for follow-up when Congress returns in November. The generic ask performs poorly; the ask that names a bridge performs better. Note also NTU Foundation's second recommendation, which is independent of the spending-level fight: replace flat EV registration fees with a vehicle-miles-traveled charge, so that highway revenue tracks highway use. That revenue question is the one that determines whether any reauthorization is sustainable, and it is where the real bipartisan negotiating space sits.

The thing to watch

December 11 is not a cliff for authorities — it is a decision point about whether Congress passes full-year appropriations, another extension, or a reauthorization. Watch for one specific signal: whether any vehicle moving through either chamber includes a rescission of unobligated IIJA balances as an offset. If it does, the relevant question stops being how much is authorized and becomes which program's unobligated column gets zeroed.

And if you are holding an unexecuted selection letter when that happens, you will find out that the press release was never the award.