Social disaster is becoming increasingly affordable. On February 12 the Trump administration rescinded the Endangerment Finding, a 2009 EPA determination that “the current and projected concentrations of the six key well-mixed greenhouse gases…in the atmosphere threaten the public health and welfare of current and future generations.” For more than sixteen years the finding had required the EPA, under the Clean Air Act, to track, report, and limit climate-heating pollution from cars and trucks. Its repeal removes the scientific and legal basis that allows the federal government to regulate greenhouse gases from vehicles, the largest source of emissions.  

Emissions from stationary sources like power plants and fossil fuel infrastructure are regulated under a different section of the Clean Air Act. But the Trump administration has already proposed that those should not be regulated either, and in all likelihood the arguments it cited to end vehicle regulation will soon be applied to emissions in general. Administration officials believe that “climate change zealots” and the previous presidential administrations used “legal fictions” to “backdoor their ideological agendas on the American people.” Meanwhile, the calendar year 2024 was the first in which global mean surface air temperatures exceeded 1.5 degrees Celsius relative to the preindustrial baseline, a threshold widely taken as a tipping point beyond which irrevocable climate disaster becomes practically impossible to avoid. The EPA had not been doing enough to prevent climate change before; now it is actively intensifying the crisis.

In a press release announcing the repeal, the EPA claimed that it was “the single largest deregulatory act in US history.” It would, according to the agency, “save Americans over $1.3 trillion” and “result in an average cost savings of over $2,400 per vehicle,” which, they say, is “improving affordability and expanding consumer choice and ultimately advancing the American Dream.” The EPA has already been sued by several environmental groups, and the agency’s numbers have immediately been challenged. An analysis by the Environmental Defense Fund argues that repealing vehicle standards will mean an additional 18 billion tons of emissions by 2055, generating something like $4.7 trillion in costs to Americans from health and environmental damage, plus 77,000 early deaths.

The administration’s move is clearly motivated by fossil fuel profits and ideological commitment. But these disputes over dollars, costs, benefits, taxes, and savings are focused on what economists call “externalities.” These are things, both good and bad, that are not represented by the price of a market transaction. The price of a scone doesn’t capture the pleasant smell of a bakery to its neighbors; the price at the pump doesn’t capture the environmental degradation produced by burning a gallon of gas.

The idea of externalities was first developed by the Cambridge economist A.C. Pigou, mostly in his Wealth and Welfare (1912) and The Economics of Welfare (1920). Smoke from factories, he wrote, “inflicts a heavy uncharged loss on the community in respect of health, of injury to buildings and vegetables, of expenses of washing clothes and cleaning rooms, of expenses for the provision of extra artificial light, and in many other ways.” As the historian Ian Kumekawa has noted, Pigou’s first evidence for external costs was from the Manchester Air Pollution Advisory Board, which estimated in 1918 that the city’s residents spent £290,000 per year more on laundry than they would have without smoke pollution. Especially in the form of uncompensated harms to third parties, externalities are one of the most common types of market failure, because they represent a gap between market costs represented in prices and costs actually borne by society—evidence all around us that, contrary to the hauntological melancholia endemic to late capitalism, markets and society are not coterminous, and that one can do violence to the other.

*

Pigou argued that the gap between market costs and social costs could be closed by imposing taxes. Now known as Pigouvian taxes, they appear on things like cigarettes, sugary drinks, and, in some cases, gasoline. Pigou hoped that externalities would be a way for the newly scientific discipline of economics to offer something tangible to the public. Experts had long understood the economic cost of factory smoke, which they called the “black smoke tax” paid by everyone in Manchester. But Pigou thought it was a clear case for redistribution—for an actual tax the government could impose on the small set of factory owners to mitigate the social ills they produced. 

In her recent book Free Gifts, Alyssa Battistoni shows how this notion came under attack when the economist Ronald Coase and his colleagues at the University of Chicago reconfigured the concept of externalities from a moral harm into another kind of market.¹ Coase argued that all economic activities are reciprocal between parties, so the social cost of factory smoke was the fault of both the factory owners and the people who chose to live nearby. The important thing, Coase claimed in his immensely influential 1960 paper “The Problem of Social Cost,” was the allocation of property rights. Once property was allocated, there was no inherent reason for economists to prefer whether the factory owner paid nearby residents compensation for pollution or if the residents paid the factory owner to stop polluting; what mattered was avoiding whichever harm was greater. According to Coase and his followers in the “law and economics” movement there was an “optimal” amount of pollution—meaning the amount equal to what people were willing to pay to cause it, or to mitigate it.

The Clean Air Act, passed in 1963, was the product of an earlier age: it spoke not of finding the “optimal” level of pollution but of “reducing or eliminating” it entirely. And yet the drift of policy went the other way: as the journalist Binyamin Appelbaum has recounted, starting in the early 1970s the government began assigning a dollar value to human life.² In 1972 the National Highway Traffic Safety Administration added the price of a funeral and the inconvenience to employers into its calculations and concluded that the price of a life lost in a car accident was $200,700. In 1974 the government rejected a regulation that would require protective metal bars to be installed on the rear of semitrucks to prevent cars from sliding under them—known as “Mansfield bars,” after the Hollywood actress who was killed in 1967 when her car slid under the back of a truck that was obscured behind a cloud of pesticide. The Ford administration concluded that such bars would save 180 lives per year, but at $200,000 a life, Appelbaum writes, they would need to save four times as many to be worth the cost of installing them. Mansfield bars were finally mandated in 1998, when life had become sufficiently valuable.

For fifty years now the government has periodically adjusted the price of human life, which in turn adjusts the amount of death, disease, misery, loss, and sorrow that is “optimal.” This “optimal” amount of harm is a crucial pillar of the cost-benefit analysis that is the foundation of essentially all American public policy. Such analysis is often based on a “willingness to pay” model. When the Mansfield bars were installed in 1998 “society” was willing to pay $2.5 million to prevent an automotive death, so safety measures that cost less than that threshold were justified and those that cost more were not. Today the “value of a statistical life” used by most federal agencies is about $13 million—sometimes higher or lower, depending on a range of factors. If, in aggregate, spending on automotive safety exceeds the willingness-to-pay estimate, then the consequently lower number of deaths is suboptimal. In January the EPA changed its cost-benefit procedures to stop counting the benefits of lives saved from regulating stationary air pollution, only the costs. It was a step toward setting the value of human life to zero.

There are known problems with the cost-benefit framework. Benefits may be small and diffuse while harms may be concentrated and acute; costs and benefits may be differently visible and emerge over different time horizons. The benefits to having clean air in the 2050s were not fully appreciated when the Clean Air Act was passed more than sixty years ago.

More to the point, “willingness to pay” is a hollow joke in the age of oligarchy. Elon Musk and I both seem willing to pay hundreds of billions of dollars to have him fired away from Earth in a rocket, but he has the money to make it happen and I don’t. According to Oxfam, at the start of 2026 twelve people controlled as much wealth as the poorest half of humanity. Several billion of us might be willing to pay for them to spend the rest of their lives breaking large rocks into smaller rocks, but we can’t collectively outbid them. Instead, their ability to pay allows them to inflict their preferences on the rest of us.

As the philosopher Martin Sandel has argued, economic analysis—and thus public policy—struggles both to distinguish between preferences with different moral valences and to determine which social costs are valid and which are not.³ The ability to define certain types of harms as economic and others as political, moral, or criminal has long been essential to the occlusion of market power in everyday life. In such moments, the question is not what society is willing to pay. The question is what set of political institutions can allow for a collective rejection of the willingness-to-pay model in general.

*

Externalities are not an immutable fact about the world. They are an analytical concept, and others may prove more valuable in our current moment. Writing in 1911, the French anarchist Émile Pouget argued that the “the very life essence of modern society” was what he called “capitalist sabotage,” which he thought “strikes at the sources of human life, ruins the health of the people and fills the hospitals and the cemeteries.”

As the scholar R.H. Lossin has shown, Pouget’s ideas were taken up by the Industrial Workers of the World to characterize the “allegedly accidental violence visited upon workers and consumers by business owners.” By 1921 the sociologist Thorstein Veblen argued that mutual, social sabotage was an inherent feature of the market system. If everyone is a competitor, trust and solidarity are impossible, any advantage is justified, and sabotage is generalized rational behavior.

As descriptions of the Trump administration go, it’s hard to beat “capitalist sabotage.” The concept has a wide contemporary applicability, from the fetid dreams of RFK Jr.’s MAHA to the depredations of DOGE to the class solidarity of the Epstein guest lists. Pouget and Veblen understood both that capitalist sabotage was an expression of class conflict and that, in Pouget’s words, “It does nothing but whet the ravenous appetites of the exploiters, that are never satisfied. It is the expression of a loathsome voracity of an unquenchable thirst of riches which does not even stop at crime!”

That shift in register is a necessary corrective. It’s astonishing to realize that the EPA’s Endangerment Finding dated only to the Obama administration. Its short lifespan suggests that stopping fossil fuel emissions is not going to be a matter of EPA rules. Even a future Democratic administration with a zeal for a green transition would need to get laws past filibustering Senate Republicans and the openly reactionary majority on the Supreme Court—both groups of people with an apparently high willingness to be paid. Accepting the logic of market control of social life, of externalities and optimal harms, of costs and benefits, and then contesting them around the edges is exactly what has led us here, to this burning world ruled by empty little men.

Many of those men seem to be on a suicide mission. But the temptation to psychologize them—to seek meaning in their apparent collective death drive—is to ratify their own sense of cosmic self-importance. In his imperishable 1976 treatise The Basic Laws of Human Stupidity, the economic historian Carlo Cipolla distinguished usefully between someone who is stupid—causing harm to others without benefit to themselves—and a bandit, who benefits from harming others. This valuable reference text has gotten a lot of use lately, but it has also become clear that beyond a certain magnitude of damage the two archetypes become indistinguishable. The oligarchs are bandits, but they are also stupid, and although they do not realize it, their rule will not last forever. We’re worth more than this.