Ten electric school buses are running from a depot in East New York, Brooklyn. The buses are real; the more ambitious claim—that their batteries will return power to New York's grid—is still a planned test. Reporting published September 23 from a visit to the depot makes the distinction tangible. A $9 million project has completed its first phase of buses and charging infrastructure. Its proposed second phase is where the grid benefit has to be demonstrated.
School buses make an unusually interesting candidate for this experiment. Their routes are scheduled, and vehicles spend substantial time parked. If a depot can predict when the buses must be charged for pickup, it can move some charging away from the most expensive or crowded hours. With bidirectional equipment and suitable agreements, a bus battery might someday send power back during a grid peak. But every kilowatt-hour sent out is one that needs replacing before the next route, and it may add to battery wear and operational complexity.
The first problem was under the ground
First Student and Con Edison say their East New York installation uses modular, above-ground charging rather than conventional buried conduits. The reported reason is striking: a jet-fuel line serving nearby JFK airport runs beneath the site, making trenching especially difficult. In their September 15 joint announcement, the companies claim this method reduced infrastructure construction costs by 13%. The figure is a project-party estimate, not an independent life-cycle audit. The later site report relays a nearly 20% estimate for project costs; those descriptions need not have the same denominator, so they should not be treated as competing measurements of one verified number.
Charging a bus from the grid is a one-way task. Vehicle-to-grid adds the other direction: the battery can discharge electricity when the grid needs it, then recharge before it carries students. The East New York depot has buses and chargers, but the public project description treats that return flow as a future demonstration. A bidirectional-capable setup is not itself proof of exported power.
The company says it plans to add 40 electric buses; Canary Media reports delivery between September 2026 and June 2027. That is a plan, not a current fleet count. For the ten already operating, the immediate gains are more modest and more concrete: no diesel exhaust at the tailpipe, a charging schedule managed around the route, and a depot layout designed around an awkward site. Con Edison and New York's energy research authority helped fund the pilot. Public subsidy and site constraints belong in a cost discussion alongside any operating savings.
What would count as a grid service?
A useful demonstration would publish the power exported, measured in kilowatts, and energy delivered over time, measured in kilowatt-hours. It would show when export occurred relative to a local grid peak, what the buses needed for the next morning's routes, how much charging cost before and after the intervention, and what additional equipment and battery wear cost. Otherwise a large combined battery capacity risks being mistaken for an available, profitable grid resource.
This is not an isolated novelty fleet. The World Resources Institute counted more than 8,500 electric school buses operating across the United States by June 2026, out of 14,146 committed buses. It also reports that an average new Type C electric bus cost around $376,000 in 2025, against roughly $140,000–$170,000 for a comparable new diesel bus. Those are broad national purchase averages, not this project's invoice, but they explain why grid revenues and operating savings matter to school systems deciding what they can afford.
They also warn against assuming that the financial case is already settled. The value of export depends on local tariffs, interconnection rules and whether a utility actually needs power during hours the buses can spare it. A pilot that shifts charging away from peaks could be worthwhile even if exporting energy proves uneconomic. Conversely, a profitable grid-service contract might make a more expensive bus easier to buy without erasing the upfront financing hurdle for a district.
The East New York project is therefore two stories on different clocks. The first is an operating fleet plus a pragmatic construction solution. The second is an unproven promise to turn scheduled transport batteries into useful grid assets. The next public result worth watching is a measured export event with route readiness, costs and grid timing disclosed together. Without all four, a yellow bus plugged into a charger remains a cleaner ride—not yet a verified power plant.
Keep exploring
- Geothermal’s next test starts with a hole — another energy claim where the next physical test matters.
- The grid’s missing component is trust — cooperation and economics behind shared electricity.
- AI’s power problem is becoming a scheduling problem — shifting demand can be a resource too.
Source-based analysis of company statements, field reporting and WRI data; no visit or bus test by Vastkind. AI-assisted. Sources checked.
