There were once places called “donor states” that paid more money in taxes into the Highway Trust Fund than they received in funding coming out of the Trust Fund. But they don’t exist anymore.

A situation where some states subsidize others was always inevitable from the creation of the Trust Fund, in the same 1956 Act that funded construction of the Interstate Highway System. A simple glance at the Interstate map, as designated in 1947, showed anyone who cared to look that there was no way that large-area, low-population states like Montana or Nevada could afford to build and maintain those long east-west divided four-lane throughways on their own. But if those roads were not built, you would unable to get across the country consistently, defeating the whole purpose of the Interstate System.

The need to build the roads on the Interstate map kept states from complaining about their annual share in the early years, but by the 1980s, most of the Interstate construction was completed, and Congress began to argue over how to distribute the growing amount of non-Interstate funding.

States began to get minimum “rates of return” on Trust Fund tax payments written into law, first at 85 percent of their tax payments, then 90 percent, then 92, and finally 95 percent. Yet even after a methodology change, Texas still managed to trigger the 95 percent guarantee trigger every few years, due primarily to its large amount of truck traffic, not its level of passenger vehicle traffic.

But the funding increases provided by the 2021 Infrastructure Investment and Jobs Act (IIJA) were so large, and tax rates having not increased since 1993, that there are no more donor states for the time being. The Federal Highway Administration recently released the computation tables for the fiscal year 2025 highway formula apportionments, which contain updated state-by-state Highway Account tax payment estimates. In fiscal year 2023, total tax payments into the Highway Account of the HTF totaled $37.4 billion. But formula apportionments to states for fiscal 2025 totaled $55.7 billion, a rate of return of 148.8 percent for the U.S. as a whole.

(Why not compare same-year tax payments and apportionments? End-of-fiscal-year timing issues. A fiscal year ends on September 30, but it takes Treasury two or three more weeks to calculate the total amount of Trust Fund tax payments made during that year, and it takes the Federal Highway Administration weeks or months longer than that to do the modeling to attribute payment levels to states. But formula apportionments have to be made by FHWA on October 1 using tax data from the most recent fiscal year where information is available. The tax data from FY 2024 was not available until several months after the FY 2025 apportionments had to be given out, so the FY 2023 tax data was used.)

The whole table of dollars in, dollars out, and rates of return is at the end of this article but here are the rate-of-return highlights.

The combination of funding increases, declining driving, more hybrids and EVs, and the closure of more gas stations in the District itself (forcing residents and commuters to fill up in Virginia or Maryland and pay the taxes there before driving in) has goosed the District’s rate of return to an absurd 1,170 percent. The tax payments here are so small (just $19.2 million to the Highway Account in FY 2023, versus $224.3 million in apportionments) that Congress should consider moving D.C. into the territories-and-Puerto-Rico non-formula program if the larger program ever becomes user-pay again.

Alaska, with its huge land mass, has always been a problem for any formula program that tries to connect places on the ground with physical infrastructure. The opposite is true for Vermont and Rhode Island, but those small land masses and small populations have a lot of expensive bridges that were built into the formula decades ago.

It is odd to see states this dissimilar crowded together around the rate-of-return median. Two low-population states, one with a massive land area, and two states that each have more than 10 million residents fit into medium-sized areas. Georgia was once part of the donor state coalition is now doing better.

Even mighty Texas, long the complainer-in-chief amongst donor states, now gets 22 percent more highway formula money than they pay in Highway Account taxes. The only state that might come close to dipping below 100 percent any time soon might be Colorado, because apparently they have a particularly bad formula share compared to their population and automotive intensity.

Not all federal-aid highway spending from the Trust Fund goes out via formula. Someday, if FHWA ever populates the now-vacant Highway Statistics 2023 webpage with tables (we are not holding our breath, because Elon), we will get an updated Table FE-221, which adds “allocations” (non-formula spending) in each state to the formula apportionments. Allocation levels aren’t known at the beginning of the fiscal year, while formula apportionment levels are. Every state’s rate of return based on apportionments and allocations is guaranteed to be higher than the rate of return based solely on apportionments.

Some people will say, but what about the Mass Transit Account? Well, to start with, the Transit Account is best viewed as a toll paid by rural areas to urban areas that was the quid pro quo for the 1982 gasoline tax increase. Second, since transit aid flows directly to cities and their transit providers, not to state governments, spending-side data by state is basically impossible to find going back prior to 2022. But since then, any state with a significant rail system or high bus utilization, including Texas and Colorado, also gets back more funding from the Mass Transit Account than they pay in taxes to that account.

(Hey, FTA – when you archive apportionments, it has the effect of deleting the state-by-state summary total spreadsheet from your website. Maybe bring that back somewhere?)