The Behavioral Economics Battle Lurking In EPA’s Endangerment Finding Repeal

Government analysts have long questioned consumers’ ability to value fuel economy accurately. But that may be changing as the EPA’s proposed repeal of greenhouse gas standards is leading to a reevaluation of some assumptions about consumer irrationality.

Getty Images

The EPA’s 2009 Endangerment Finding was the agency’s formal determination that greenhouse gases endanger public health and welfare. Ever since, it has served as the legal foundation for EPA climate regulations. Without this finding, EPA lacks Clean Air Act authority to regulate greenhouse gas emissions.

The Trump administration’s EPA has now proposed to repeal the Endangerment Finding, along with the agency’s greenhouse gas standards for light, medium, and heavy-duty vehicles that depend on it. In the agency’s announcement, EPA justifies the repeal by citing the severe economic burdens of its existing rules, including over $1 trillion in compliance costs.

As the agency moves to dismantle the Endangerment Finding, another battleground has opened up that has received less attention. That fight is about whether regulators should trust consumers’ preferences or instead attempt to “correct” them. The outcome of this debate could swing the measured benefits of climate rules by trillions of dollars.

The Role of Regulatory Impact Analysis

Because repealing the Endangerment Finding would also remove the legal basis for existing greenhouse gas standards for cars and trucks, EPA is required under longstanding executive orders to analyze the economic effects of that policy change. This requires the agency to tally the costs avoided and the benefits forgone from the action.

To comply with these requirements, agencies prepare a regulatory impact analysis (RIA) whenever a rule or policy change is expected to have an annual economic effect of $100 million or more. RIA is a framework for identifying the expected consequences of a regulation, quantifying them where possible, and monetizing them when the data and methods allow. In the case of the Endangerment Finding repeal, that means examining how vehicle technology, fuel use, air pollution, and consumer welfare would differ with and without the greenhouse gas standards, and then converting any differences into dollar terms.

In its draft RIA for the repeal action, EPA’s core engineering-model estimate finds the repeal would yield net costs of roughly $260 billion (at a 3% discount rate, over the years 2027 to 2055). This traditional government approach counts fuel savings as a benefit to consumers, making the repeal appear costly since those savings would be lost. However, Appendix B of the RIA includes an alternative “revealed preference” analysis that estimates net benefits of the repeal ranging from $3.05 trillion to $8.18 trillion. This set of estimates assumes that if consumers aren’t voluntarily choosing more fuel-efficient vehicles, then forcing them to do so through regulations actually harms them. Any estimated savings, in that case, were pure fiction. By extension, so were many of the benefits of regulation.

The Assumption of Revealed Preference

Cost-benefit analysis aims to tally up the monetized social gains and losses from a policy. An economist adds up the “private benefits” to particular individuals to arrive at a cumulative “social benefit” estimate for society as a whole.

“Revealed preference” is a concept central to this endeavor. By examining what people buy and how much they are willing to pay for different items and features, economists can estimate dollar values for different types of benefits and costs. This approach assumes that the observed willingness to pay of an individual reflects the value of a benefit to that person.

This method has a major advantage in that it respects people’s choices and doesn’t involve analysts judging whether people’s choices are good or bad; they merely accept that the choice made was what the individual preferred. The downside of this approach is that people don’t always make decisions that accord with their own interests, or that of society.

Fuel Savings Violate Revealed Preference

For years, agencies writing fuel economy and energy efficiency rules have counted fuel and energy savings as a benefit of those rulemakings. When a consumer buys a more fuel-efficient car or appliance, they save money on gas or their utility bill. The government counts that as a significant benefit of a regulatory action phasing out less-efficient devices.

This approach is valid if consumers genuinely underappreciate those savings when they buy a car or appliance. But if they already weigh fuel economy and energy efficiency against other attributes of a product before making a purchase, the savings are not a windfall benefit of the rulemaking. They’re the flip side of losing other features the consumers value more.

Appendix B of EPA’s regulatory analysis relies on exactly that logic. If a consumer picks a gas-powered truck knowing it’ll burn more fuel, they’ve made a trade they prefer. Forcing them into an EV to “save” fuel costs is a net loss to them. Yet for many years, the government has treated this as a benefit.

Behavioral Economics and the “Energy Efficiency Gap”

Economists use the term “energy efficiency gap” to describe the puzzling difference between the level of energy efficiency that appears cost-effective in theory and the lower level people actually choose in real life. For example, engineering calculations might show that spending $1,000 on better insulation, more efficient appliances, or a higher-MPG vehicle would pay for itself in a few years through lower utility or fuel bills. Yet, many consumers routinely forgo those investments.

What explains the gap? One interpretation is that buyers are making biased, short-sighted decisions. This is the classic territory of “behavioral economics,” a field focused on how real-world decisions often deviate from the assumptions of rational, optimizing behavior found in economists’ models. Cognitive biases like hyperbolic discounting (placing too much weight on present rewards relative to future ones) or inattention (failing to notice or process fuel cost information) could lead people to under-invest in efficiency and leave money on the table. This perspective justifies counting the full value of “missed” fuel savings as a regulatory benefit to the consumer, because the regulation is correcting their mistake.

But there’s another possibility, which is that the gap isn’t a sign of bias at all, but instead a reflection of genuine trade-offs. A consumer might choose the lower-MPG car because they care more about acceleration, cargo space, style, or any number of attributes that are not captured in the fuel-savings calculations. An analyst who misinterprets the gap as a bias, when in fact the choice was based on a rational calculation, could force consumers into a less-preferred option and make them worse off.

What’s at Stake

Separating bias from legitimate preferences is exceedingly difficult, and some would argue impossible. From the outside, the decision looks the same whether it’s the product of error or preference. If we can’t reliably distinguish between bias and preference, then the case for “correcting” consumer choices becomes more about paternalism than empiricism.

The stakes in this debate go beyond the Endangerment Finding. In many energy-efficiency rulemakings, 80 to 90 percent of the total monetized benefits come from the government’s calculations of consumers’ avoided energy costs. Environmental benefits to Americans are often in the low single-digit percentages. This means the overwhelming majority of the official benefit calculation hinges on the assumption that regulators can improve consumer welfare by steering people toward more efficient—and more expensive—products, even when buyers themselves would freely choose otherwise if left to decide on their own.

This leaves economists in a quandary. Do they assume that observed market behavior is the best available measure of welfare, even if it sometimes reflects mistakes? Or do they override those choices based on models of what they think people should want if they made careful choices using all the available information? Or do they seek a middle ground, acknowledging that their models are often accurate but may also ignore important context-specific trade-offs? The answer to these questions determines whether a regulation’s calculated benefits can be trusted.

Private vs. Social Benefits

Another complication relates to the difference between private and social benefits. Even when consumers make perfectly rational choices, what is in the interests of an individual doesn’t always benefit society as a whole. When one person’s gain imposes external costs on others, this can reduce, or even reverse, the net benefit for society.

One obvious group affected by our purchasing decisions is future generations. It is easy to imagine future people might prefer that today’s consumers forgo some luxuries in favor of greater savings and investment, which would improve living standards in the long run. But those intergenerational considerations are typically not reflected in market prices or, similarly, in economists’ measures of revealed preference.

In the context of energy and fuel economy, a dollar saved at the pump can be invested elsewhere in the economy, compounding to boost growth and future welfare. The enjoyment from a car feature like more horsepower or a panoramic sunroof can’t be reinvested in the same way. So while a consumer may be better off paying more for those amenities, future generations probably will not be. From society’s perspective, fuel and energy savings likely do represent social benefits for this reason, even when they don’t compensate for their drawbacks from an individual’s standpoint.

A Rulemaking Worth Watching

EPA’s Endangerment Finding RIA pushes this debate forward by putting the revealed preference framework front-and-center, challenging the government’s conventional inclusion of full lifetime fuel savings as a benefit. Whether that approach gains traction will matter well beyond this rulemaking. It’s a core issue for how government evaluates climate and energy efficiency regulations generally. And it’s another reason to watch closely how this already-high-stakes rulemaking unfolds.