One Overlooked Number Could Flip the Vehicle Pollution Rollback

THE SPREADSHEET says scrapping America’s vehicle pollution rules is a $790 billion bargain for the country. A single accounting error in how regulators value the fuel savings of efficient cars could turn the rollback of federal vehicle pollution rules from a $790 billion gain into a net loss, ten economists argue in a new Policy Article in the journal Science.

The rollback eliminated the Environmental Protection Agency’s (EPA) greenhouse gas (GHG) standards for new cars and trucks entirely, taking effect in April 2026 and leaving only a token fuel-economy rule that industry is unlikely to notice. The error, the authors say, sits in a single number: how much a car buyer is really willing to pay for a dollar of fuel savings, and it is the same kind of policy analysis question that comes up whenever regulators try to price a consumer benefit.

Vehicle efficiency rules have gyrated with almost every change of administration since 2000, tightened under Bush and Obama, loosened under the first Trump term, tightened again under Biden, and now eliminated outright. Each swing has cited its own formal cost-benefit analysis, as if the underlying evidence, rather than the politics, had changed.

Light-duty vehicles produce over half of America’s transportation greenhouse gas emissions and 16% of the country’s total emissions, which is why the rules carry so much weight. Two changes drove the 2026 rollback’s math. One openly erased $1.8 trillion in environmental and public-health benefits that the same standards were credited with under the 2024 analysis, a change that has already drawn heavy criticism. The other, subtler change concerns how the analysis treats what a car buyer actually gets for a more fuel-efficient vehicle, and it is worth roughly the same amount of money.

Decades of research agree that new-car buyers will not pay a full dollar today for a dollar of fuel savings tomorrow, a pattern one widely cited study of used-car prices confirms even outside this particular regulatory dispute; by the government’s own working assumption here, buyers pay something closer to 21 to 25 cents on the dollar. What divides analysts is why: whether buyers are simply inattentive to fuel costs, or whether efficient cars quietly sacrifice something else drivers value, from acceleration to cargo space.

The Missing Piece of the Dollar

The Science team built a framework that forces every dollar of fuel savings into three buckets that must add up to a whole dollar: what buyers genuinely value, what they undervalue through inattention, and what they lose in traded-off features. Slot the EPA’s own 2026 numbers into that framework, and the agency credits buyers with valuing just $0.22 of that dollar, which the press release accompanying the paper rounds to 23 cents, then assumes the entire remaining $0.75 to $0.79 reflects hidden trade-offs rather than inattention. The 2024 analysis made the opposite assumption, blaming inattention alone and crediting trade-offs with nothing at all.

Both extremes, the authors argue, contradict evidence that already exists. “Application of our framework to deal with valuation could correct errors from multiple administrations, providing analytical consistency that could help improve and stabilize policy,” Bento and his co-authors write.

Running the Numbers Again

A review of the published literature puts genuine inattention at somewhere between $0.54 and $0.62 of every fuel-saving dollar, not the $0 or $0.78 that the two administrations have alternately assumed. A separate study the authors cite estimates that fully electric vehicles (EVs) carry attribute trade-offs of $0.17 to $0.31 per dollar of fuel savings compared with a similar gasoline car, a gap that narrows as EV range improves. Splitting the difference, the authors propose their own values: $0.22 for genuine willingness to pay, $0.55 for inattention, and $0.23 for trade-offs, a set that happens to sum almost exactly to a full dollar. Run those numbers through the EPA’s own scenario, which currently shows $1,290 billion in fuel savings against $500 billion in costs for a $790 billion net gain, and the accounting shifts hard: correcting how driving reductions are valued adds $234 billion to the true cost of the rollback, and correcting the fuel-savings split adds $684 billion more. By the time one more fix, to how technology costs are estimated, is applied, the same scenario the EPA used to justify a $790 billion win turns into a $670 billion loss.

The rollback’s math has other soft spots too: the 2026 analysis pairs a low gasoline-price forecast with no corresponding high-price case, at a time when pump prices are already running above the central forecast, which mechanically shrinks the value of any efficiency gain. It also assumes drivers are better off taking 1.4 trillion fewer miles under the old standards, banking roughly $250 billion in avoided trip costs, without subtracting what those forgone trips were worth to the people who wanted to take them.

The authors are careful about what they cannot show: no study has directly measured what buyers would pay in a regulatory setting that mixes inattention and trade-offs together, so their $0.21 to $0.25 starting point is borrowed from the EPA’s own assumption rather than independently observed. Two of the ten authors disclose paid consulting or expert-witness work for the California Air Resources Board, Toyota, and other energy-related clients, though they say none of it funded this analysis.

Add up the driving, fuel-split and technology-cost fixes alone, and the authors calculate the rollback’s true net benefit falls by $1.5 trillion, before the $1.8 trillion in erased environmental benefits is even restored. That reversal matters well beyond the spreadsheet: agencies are required to justify a rescission with a defensible cost-benefit case, and the authors note the rollback is likely to be challenged in court, where a flawed analysis could undercut the agency’s own legal footing.

The paper is less interested in refighting each administration’s number than in ending the swings altogether, and it points to alternatives such as a straightforward tax on fuel consumption or a feebate, which taxes inefficient vehicles to reward efficient ones. Unlike fuel-economy mandates, which take decades to work through the vehicle fleet, a fuel or mileage tax would touch every car on the road immediately.

Whether any of that persuades a court, or the next administration, is an open question the paper does not attempt to answer. What it does offer is a single spreadsheet trick any future regulator could apply, in either direction, the next time the numbers need explaining.

FAQ