1. Executive Summary
The United States does not have a rural transportation problem so much as a rural transfer problem. Each mode is separately funded, separately planned, separately measured, and separately advocated for — and the trips people actually need to take cross all three. A resident of a county seat 200 miles from a hub airport can often reach the interstate, sometimes reach a bus, occasionally reach a train, and rarely reach any two of them in sequence within the same day with a single ticket.
This paper argues for reorganizing federal and state practice around connectivity outcomes rather than modal outputs, and offers five instruments to do it. The central recommendation is a mode-neutral rural connectivity block grant that pools the Essential Air Service appropriation, the rural intercity bus set-aside, and a modest state-flexible share, and lets states purchase whatever combination of bus, rail-feeder, and air service delivers a stated connectivity floor — under multi-year contracts long enough to finance vehicles and terminals against.
The timing is not incidental. Surface transportation programs authorized by the Infrastructure Investment and Jobs Act expire on September 30, 2026, and the law’s $156 billion in advance appropriations will not continue past FY2026 without further congressional action. A five-year reauthorization beginning in FY2027 faces a projected gap between Highway Trust Fund revenues and outlays of $166 billion; a six-year bill faces $199 billion. CBO projects the highway account balance approaches zero in FY2028. Whatever Congress does in the next several months sets rural mobility policy for the following five to six years, and it will do so under revenue pressure that makes new spending hard and reallocation comparatively easy.
That constraint is the argument’s friend. The program proposed here is small — low single-digit billions annually against a surface program measured in tens of billions — and its principal moves cost little because they are organizational rather than capital.
2. The Problem Is at the Seams
Four separate money streams govern non-automotive mobility for a small American community, and none of them talks to the others.
Air. The Essential Air Service program, funded through DOT appropriations plus overflight fees. FY26 provided $513.6 million in discretionary funding plus an estimated $173.9 million from overflight fees, for roughly $687.5 million total, alongside $15 million for the Small Community Air Service Development Program. Contracts typically run two years and are relitigated annually in the budget.
Rural intercity bus. FTA Section 5311(f). Each state must spend no less than 15 percent of its annual rural formula apportionment on developing and supporting intercity bus transportation, unless it certifies after consultation with providers that the state’s intercity bus needs are adequately met. IIJA raised 5311 contract authority from $673.3 million in FY2021 to $875.3 million in FY2022, stepping up to $959.6 million by FY2026, without explicitly expanding intercity bus funding.
Rail. Amtrak’s federal grant, plus the Corridor Identification and Development Program and state-supported service under the PRIIA cost-sharing regime. Federal law defining public transportation explicitly excludes Amtrak, intercity buses, and school buses, which means the transit program and the rail program are legally separate universes.
Local transit. FTA Section 5311 formula funds for rural areas under 50,000.
A traveler takes one trip. The money is administered in four unconnected pieces, on four different clocks, by agencies with four different performance metrics. The predictable result is that each mode optimizes its own service and none of them optimizes the handoff. This is the failure this framework addresses.
3. What Each Mode Actually Does Well
Bus is the cheap, fast, flexible layer — and it is the one currently in institutional collapse.
The service side is healthy and improving. State-supported networks require millions rather than billions, use existing highways without major capital works, and can draw on federal funds already flowing to states. Colorado’s Bustang, Oregon’s POINT, Vermont Translines, Virginia Breeze, and Travel Washington have rebounded faster since the pandemic than most urban transit, and intercity bus is more amenable to experimentation, faster to roll out, and cheaper per passenger carried than rail. Bustang carried nearly 280,000 riders in 2023 across express Front Range and I-70 service, the rural-focused Outrider brand, and the Pegasus shuttle. One Colorado official described the rural intercity set-aside as the bridge that lets local services coordinate with intercity buses to complete a rural trip to Denver.
The infrastructure side is the crisis. FlixBus acquired Greyhound in 2021 and the two brands now operate a combined network of more than 1,800 destinations — far fewer places, especially rural ones, than the roughly 4,750 stations Greyhound ran at its 1940s peak. Greyhound’s terminal properties were not part of the acquisition; they stayed with the former parent and were subsequently sold to other owners. Downtown depots have closed in Houston, Philadelphia, Cincinnati, Tampa, Louisville, Charlottesville, and Portland, with stops relocated away from city centers to locations often unreachable by local transit, or converted to curbside operation. Roughly three-quarters of intercity bus riders report annual incomes below $40,000, and more than a quarter would not make the trip at all without the bus.
The lesson is precise: the vehicles and the routes are not the binding constraint; the terminals are. Private real-estate liquidation removed the physical interchange points, and no mode can connect to another at a curb with no shelter and no posted information.
Rail is the high-capital, slow-to-build, high-coverage layer.
Amtrak’s fifteen long-distance routes serve nearly half of all stations in the system, and despite low frequencies form the backbone of the national network across 39 states and the District of Columbia, connecting to state-supported routes, Northeast Corridor service, Amtrak Thruway buses, and local providers. That last clause matters more than it looks: Thruway is already the country’s largest working example of scheduled air-and-rail-style intermodal ticketing, and it is a bus product.
The timelines are the caution. Corridor ID selection brings a $500,000 grant for a Service Development Plan, and participants describe a process that could take seven to ten years before service begins on new routes. The long-distance study’s final report acknowledges that there is currently no sustained financial support or program to build or operate its preferred route options, though some may qualify for Corridor ID planning funds. One observer noted that the long-distance study carried a 2060 planning frontier. Rail belongs in a 2040 plan. It does not belong in a 2028 plan.
Air is the fast, expensive, thinnest-market layer, and its constraints were addressed in prior work: a crew-cost floor that makes small aircraft uneconomic, a 40-to-50-seat product gap, and an annually contested subsidy.
4. A Connectivity Standard
The framework’s organizing device is a stated floor that any community can be measured against, replacing modal output counts with a single outcome.
The proposed floor: every incorporated community above a population threshold should be able to reach a Tier 1 interchange — an airport with connecting service, a rail station with daily service, or a metropolitan bus hub — with no more than one transfer, on at least five days a week, arriving in time for a same-day return.
Three features make this workable:
It is mode-neutral. A state may satisfy it with a bus, a rail feeder, an EAS flight, or a subsidized shuttle. The metric does not care.
It is measurable. A connectivity index of this kind already exists in federal practice; GAO uses a connectivity measure of access to the aviation system and has tracked its decline at small communities. Extending it across modes is an analytic exercise, not a new science.
It exposes the seams. A community with three weekly flights and a daily bus that arrives forty minutes after the flight departs scores as failing, which is the correct answer and one that no single-mode metric produces.
5. Five Instruments
Instrument 1 — A mode-neutral rural connectivity block grant.
Pool the EAS appropriation, the 5311(f) rural intercity bus set-aside, and a state-flexible share into a single grant that states administer against the connectivity floor. Preserve a hold-harmless for currently served communities to defuse the obvious political objection, but let a state that can meet the standard for a given community more cheaply by bus than by air redirect the difference into serving a community that currently has nothing.
This is the structural change. Today a state cannot trade air dollars for bus dollars even when the bus is manifestly the better answer, and the reverse is equally true. GAO’s own inventory of options for improving small-community service includes bus substitution alongside pilot supply and electric aircraft — but the funding architecture forbids acting on it.
Instrument 2 — Multi-year contracts across every mode.
Five to seven years, with aircraft- and vehicle-type commitments. This is the single highest-return administrative change available. It converts a subsidy stream into collateral, which lets an operator finance rolling stock; it makes small-aircraft and coach procurement bankable; and it removes the annual budget theater that has consumed the program’s political capital for a decade. Congress enacted $514 million for EAS in February 2026, rejecting a proposed 52 percent cut and adding language preventing DOT from abruptly terminating contracts — protecting service to 56 communities the request would have eliminated — and a further $372 million reduction has since been proposed. The pattern of executive proposals and congressional restoration spans administrations of both parties. Multi-year authority ends the cycle without settling the underlying argument.
Instrument 3 — Public ownership of intermodal interchanges.
The cheapest high-leverage capital item in the entire framework. A modest facility — enclosed waiting room, restrooms, real-time information, staffed hours matching service, physical adjacency between bus bays and the rail platform or the airport terminal — costs a few million dollars and is the physical precondition for every transfer the connectivity standard requires.
The intercity bus terminal collapse happened because these assets sat on private balance sheets and were worth more as redevelopment sites than as transportation infrastructure. After the Flix acquisition, the former parent sold all but two of its Greyhound properties within about a year, pushing stops to curbside or peripheral locations without waiting rooms. The remedy is straightforward: treat the intercity interchange as public infrastructure on the same footing as an airport terminal, eligible for the same capital programs. Analysts examining the Chicago case recommended that the city assess acquiring, renovating, and operating the existing terminal, and that the state set a goal of coordinating train and bus transfers on its ten largest intercity routes. That recommendation generalizes.
Instrument 4 — Through-ticketing and schedule coordination as a condition of subsidy.
Any operator receiving connectivity grant funds must publish schedules in a common feed, guarantee connections at designated interchanges within a defined window, and participate in interline ticketing with other subsidized operators in the state. The Amtrak Thruway model demonstrates that this is administratively solvable; what is missing is the requirement.
Two-thirds of the perceived service gap in small communities is not missing service. It is service that exists but cannot be found, booked, or connected to in one transaction.
Instrument 5 — A connectivity performance report replacing modal reporting.
Require each state to publish an annual connectivity assessment against the floor, by community, with the modal mix used. This does for rural mobility what pavement-condition reporting did for highways: it makes failure legible and comparable across states, which is the precondition for anything else changing.
6. Paying For It
The framework’s sums are small enough that the funding argument is about allocation rather than revenue.
The scale. EAS plus the rural intercity bus set-aside currently runs on the order of $800 million to $1 billion annually. A connectivity program that closed most of the identified gaps might require $2 billion to $3 billion — against a public transportation program that averaged $21.4 billion annually in FY2022–FY2026 and a highway program several times that. The rounding error on a reauthorization funds the entire proposal.
The revenue context works against new spending and for reallocation. The projected five-year HTF shortfall is $166 billion, which means the coming bill will be an exercise in triage. Programs that can demonstrate outcome measurement and low cost survive triage better than programs that cannot.
Legislative vehicles exist now. The House Transportation and Infrastructure chair introduced a surface reauthorization bill in May 2026, and a reauthorization measure was ordered reported by the committee that same month. The window for shaping this is measured in weeks, not years.
Three revenue mechanisms deserve consideration: extending the overflight-fee model that already supplements EAS to other connectivity purposes; dedicating a small share of any new road-usage or EV-registration revenue to non-automotive rural access, on the grounds that the constituency losing the gas-tax subsidy is the one least able to drive; and permitting states to flex a defined percentage of highway formula funds into connectivity operating assistance, which is currently prohibited in most categories.
7. Sequencing
2027–2029: administrative and bus-led. Multi-year contracts, the connectivity standard, through-ticketing conditions, terminal acquisition, and state intercity bus expansion. All of this is achievable inside one reauthorization cycle, requires no new technology, and delivers measurable coverage gains within thirty-six months. The state-branded bus model has already proven it can be rolled out quickly at low cost.
2029–2034: air stabilization. Multi-year EAS contracts make the 40-to-50-seat aircraft program financeable, and small-aircraft Part 135 capacity expands into the thinnest markets under stable contracts.
2034–2040: rail delivery. Corridor ID projects that entered planning in the mid-2020s reach service on a seven-to-ten-year horizon, and the bus network built in phase one becomes the feeder system that makes them viable. This is the correct sequence — rail corridors fail when they open without the surrounding collection network, and buses are how you build one for a fraction of the cost while the rail money moves through its own timeline.
8. Honest Limits
This does not compete with driving for most trips, and should not try. For a licensed adult with a functioning car and no parking cost at the destination, no combination of bus, rail, and air will win on a 150-mile trip. The target population is people without that option: households without a vehicle, second and third adults in single-vehicle households, people who have stopped driving with age, people whose medical procedures preclude driving home, students, and people for whom a winter mountain pass or a night trip is genuinely unsafe. Framing the program as mode shift invites a comparison it loses; framing it as basic access is both more accurate and more defensible.
Bus substitution is politically radioactive in exactly the places that need it. A community that loses subsidized flights for a subsidized coach experiences it as demotion regardless of how much better the service is. The hold-harmless provision in Instrument 1 exists for this reason, and any state pursuing substitution should do it by adding bus service first and letting communities observe the frequency advantage before any air service is withdrawn.
Some communities are correctly served by nothing scheduled. The framework should be honest that a floor is a floor. Where population, distance, and demand all fall below any reasonable threshold, on-demand and volunteer-driver models are the appropriate answer, and pretending otherwise wastes money that other communities need.
Coordination is harder than it sounds. Every state that has attempted schedule coordination across independent operators has discovered that a guaranteed connection means someone holds a departure, and holding a departure costs money that nobody budgeted.
9. Conclusion
The instruments proposed here are unglamorous: a pooled grant, longer contracts, publicly owned waiting rooms, a common schedule feed, and an annual report. None of them requires a technology that does not exist, and together they cost less than a single urban rail extension.
Their justification is that the American non-automotive network already contains most of the pieces it needs. There are buses, there are trains, there are airplanes, and there are federal programs paying for all three. What is missing is the requirement that they meet each other — at a specific place, at a specific time, on one ticket. The authorization that governs most of this expires on September 30, 2026. The question in front of Congress is whether the next bill continues to fund four disconnected systems or begins funding one connected outcome.
