President Biden’s fiscal 2024 budget request proposes to incur $13.5 billion in unfunded promises for “New Start” mass transit systems, by far an all-time high.

“New Starts” are new transit systems, or new operable segments of existing transit systems. They have to be “fixed guideway” systems which means heavy rail, light rail, streetcar, or bus rapid transit with a dedicated right-of-way. And New Starts are so expensive that they require multi-year promises of federal funding, the schedule for which is laid out in a full funding grant agreement (FFGA) signed by FTA and the transit agency.

FFGAs are not really grant agreements in the traditional sense – they do not obligate the federal government to do anything. They contain language saying that the agreement is not legally binding on the federal government and does not represent an obligation to pay any money that has not already been appropriated. (Normal grant agreements legally obligate the federal government to provide a fixed amount of money if the non-federal partner meets the terms of the agreement, and the non-federal partner can sue the U.S., and win, if they fulfill the terms and the money is not forthcoming.)

The FTA helpfully archives the old annual reports of the CIG program that accompany the President’s Budget each year. We went back to fiscal 2006, the first year the program was in its current form. (Prior to that, “New Starts” were a subset of the section 3 capital grant program, along with bus grants and what is now the State of Good Repair formula program, and a substantial portion of each year’s New Start funding came from earmarks of multi-year contract authority in authorization acts, so the “unfunded promises” calculation was fundamentally different.)

Anyway, we went back to fiscal year 2006, and in Table 1 of each report, we looked up the total amount of the federal CIG share of all of the signed FFGAs that had not yet been paid off, and subtracted the total amount of existing appropriations made and allocated for those projects at the time the report was issued. We refer to this amount as “unfunded promises” that then have to be fulfilled by new appropriations made in the upcoming budget year and in future out-years.

For fiscal years 2006 through 2024, the amount of “unfunded promises” has averaged $3.5 billion, with a high of $5.2 billion (FY 2014) and lows of $1.4 billion (FY 2007) and $1.2 billion (FY 2024, a.k.a. the new budget released this month).

(We are excluding core capacity projects and “small start” projects from this calculation. Core capacity is out because they didn’t show up in the reports until fiscal year 2018 and they are currently back down to zero this cycle, and small starts are often paid off in a single year so they don’t really count as unfunded promises.)

Under the budget request, the entire $1.2 billion in unfunded promises in the current New Start portfolio would be paid off, which is the good news. But the Administration proposes to sign six new FFGAs for new start projects in the fiscal 2024 budget year, and the CIG share of those projects is an astounding $18.2 billion (the Hudson River Tunnel, Silicon Valley extension, and Second Avenue Subway extension are particularly large).

Between prior year allocations and the proposed FY 2024 total budget allocation, the budget assumes that $2.9 billion of that $18.2 billion will be provided before the end of the 2024 budget year, leaving $15.3 billion in unfunded promises for New Starts that would appear in the fiscal 2025 report.

But that’s not quite fair, because there will still be two years of IIJA advance appropriations remaining in 2025 and 2026. That is real funding, already on the books, and should be applied against future unfunded promises.

It is worth reminding people that “CIG” and “New Starts” are not synonymous – the latter is a subset of the former. Even though the IIJA provided $8 billion over five years for the CIG program, only $4.4 billion can to to New Starts:

So the last two years of the IIJA (FY 2025 and 2026) would provide $1.74 billion for New Starts ($871.2 x 2), reducing the $15.266 billion in unfunded promises for New Starts in that notional FY 2025 budget to $13.521 billion.

This is still a phenomenal amount by historical standards:

Such a large amount of unfunded promises would not been illegal prior to 2012. From 1978 through the MAP-21 law in 2012, the law provided a “contingent commitment ceiling” on the aggregate amount of New Start unfunded promises that could be incurred at any one time. From 1982-1998, when these projects were mostly funded from the Highway Trust Fund, the ceiling was a percentage of Mass Transit Account balances. Starting with TEA21 in 1998, the ceiling was the combined last two years of New Start program authorization levels, and this was increased to three years in 2000.

This limitation prevented FTA from signing a FFGA for the ARC tunnel in 2009, so the FY 2010 appropriations bill waived the ceiling for the tunnel (mooted when Chris Christie became governor and killed it later in 2010). MAP-21 then repealed the contingent commitment ceiling altogether.

Prior to its repeal, that language said “The total estimated amount of future obligations of the Government and contingent commitments to incur obligations covered by all outstanding letters of intent, full funding grant agreements, and early systems work agreements under this subsection for major new fixed guideway capital projects may be not more than the greater of the amount authorized under sections 5338(a)(3) and 5338(c) for such projects or an amount equivalent to the last 3 fiscal years of funding allocated under subsections (m)(1)(A) and (m)(2)(A)(ii) for such projects, less an amount the Secretary reasonably estimates is necessary for grants under this section for those of such projects that are not covered by a letter or agreement. The total amount covered by new letters and contingent commitments included in full funding grant agreements and early systems work agreements for such projects may be not more than a limitation specified in law.”

The current authorization level for the CIG program is $3 billion per year, so a contingent commitment authority ceiling based on the old one would limit unfunded promises for New Starts to $9 billion, far less than $13.5 billion. (Also, core capacity projects would also be subject to that ceiling as well.)