Dangerous Finds

EPA Administrator Lee Zeldin recently announced plans to rescind the agency’s so-called “endangerment finding,” a 2009 order that obligated the regulation of greenhouse gases, like carbon dioxide and methane, under the Clean Air Act. “The proposal would, if finalized,” said Zeldin, “amount to the largest deregulatory action in the history of the United States.”

“It’s possible that he’s right,” responded the New Yorker’s Elizabeth Kolbert.

Everyone is doing a little too much here. Zeldin’s EPA is trolling climate advocates by not merely curtailing emissions regulations, as they did during the first Trump Administration, but rescinding them entirely. And climate advocates, true to form, are feeding the trolls. “When the history of this era is written,” tweeted US Representative Sean Casten of the rescission, “Donald Trump will have been responsible for more deaths than Stalin, Mao and Hitler combined.”

This enthusiastic backlash to Zeldin’s announcement is consistent with the kind of fortress mentality that has characterized the climate research and advocacy community for a long time. But despite the banality of the endangerment finding itself—climate change really does pose risks to humanity—it’s not at all clear the order has had a significant practical effect on policy or emissions. Emissions regulations were not the driving force behind shifts in the US transportation sector over the past twenty years, and had nothing to do with the even larger shifts in the electric power sector.

Climate and clean energy advocates should take a beat, and use the rescission as an opportunity to come up with a regulatory framework for greenhouse emissions that actually makes sense. Because despite various court decisions and statutes conflating them, greenhouse gas emissions and conventional pollutants are actually different along multiple dimensions, and efforts to regulate both of them under the Clean Air Act—a law drafted to cover air pollutants, not emissions—were never going to be particularly successful.

Straightforward or Symbolic?

Climate researchers, as a group, were displeased with Zeldin’s decision. “There is no evidence that has emerged or been published in the scientific literature in the past 16 years that would in any way challenge the scientific basis of the 2009 endangerment finding,” said Stripe climate scientist and Breakthrough senior fellow Zeke Hausfather. In the New York Times, climate researchers Marshall Burke and Solomon Hsiang wrote that the rescission “runs counter to both basic logic and a growing mountain of science documenting direct harms from greenhouse gas emissions via climate change.”

And in a straightforward way, Zeldin’s critics are correct. The endangerment finding is remarkably easy to defend on literalist merits. Specifically, the order finds that “the elevated concentrations of the six greenhouse gases in the atmosphere…endanger both the public health and the public welfare of current and future generations.” That’s it.

But many of Zeldin’s critics make claims about the endangerment finding that get way ahead of the evidence. Burke and Hsiang, for instance, write that regulations based on the endangerment finding “have begun to make a dent in America’s contribution to climate change and the hazards it creates.” Climate scientist Michael Mann said the endangerment finding has “been the primary tool that we have had to actually regulate carbon emissions and meet our obligations under various global agreements to address the climate crisis.”

And while there may be some important symbolism in the regulations, those statements are just pretty hard to defend on the empirics.

When the endangerment finding was established, during Barack Obama’s first term in office, the authority to regulate emissions was relatively new, having been granted by the Supreme Court in Massachusetts vs. EPA (2007). Much more recently, the Inflation Reduction Act amended the Clean Air Act to statutorily define greenhouse gas emissions as pollutants, in addition to the traditional criteria pollutants including lead, sulfur- and nitrogen-oxides, ozone, particulates, and carbon monoxide.

The most concrete way this authority took effect was in the Obama Administration’s introduction of new greenhouse gas standards for the vehicle fleet. These regulations were integrated with long-standing CAFE standards, and continued to drive improvements in fleet-wide fuel economy. The problem, from an emissions standpoint, is that fuel economy improvements have been largely eaten up by gains in engine horsepower and vehicle size and weight, which helps explain why transportation-sector emissions have flatlined for 25 years despite efficiency gains. The best that could be said is that the emissions standards pushed vehicle emissions lower than they might otherwise have been. But CAFE standards already existed for this purpose, so it’s not clear how big this effect was. One now wonders if the CAFE standards on their own would have proved more durable without the endangerment finding.

Separately, Presidents Obama, Trump, and Biden all proposed their own greenhouse emissions rules for the electric power sector: Obama’s Clean Power Plan (CPP), Trump’s Affordable Clean Energy (ACE) rule, and Biden’s CPP 2.0. None of these plans went into effect, let alone “made a dent in America’s contribution to climate change.”

As I’ve written before, that’s no accident of history. In addition to the typical swings in presidential power and ideology over the past decade, there are structural reasons for the Clean Air Act’s underperformance on emissions.

Yes, Trump threw out Obama’s rule, and then Biden threw out Trump’s. But even in between the political ping-ponging, the Judiciary picked each rule apart. The Supreme Court actually stayed the original CPP shortly after its announcement in 2016, and the DC Circuit court remanded Trump’s ACE rule in 2019. Then in 2022, the Supreme Court found that Biden’s Clean Power Plan exceeded its statutory authority by mandating “generation shifting” across power plants instead of limiting emissions reduction to those “within the fenceline” of the individual plants themselves.

In theory, the IRA’s statutory revision of the Clean Air Act should bolster the EPA’s ability to regulate emissions under the law. But the revision does not resolve the fenceline issues, nor how the agency should manage issues of jurisdiction and standing, or evaluate compliance costs of so-called “best systems for emissions control.” The Biden EPA seemed prepared to test these issues with the courts, making some ambitious assumptions about the best system for emissions control within both the power and transportation sectors in a seeming effort to tighten the stringency of the emissions standards. Of course now those standards are moot.

But even if these many legal issues were resolved (a rather large “if”), there remains the practical difficulties of regulating gases emitted at the gigaton-scale, that are non-toxic at any normal concentration, and, indeed, that are essential to all life on earth. Unlike with lead or mercury, it is legitimately difficult, on a scientific level, to identify the discrete harm of a gram or a ton of carbon emissions to a human body, ecosystem, or community. Greenhouse gases by and large are not incidental to the industrial process that emits them: they are, literally, elemental to them. The benefits are immediate and obvious while the costs are diffuse and distant.

Climate economists have attempted to resolve these practical difficulties by integrating centuries of estimated climate damages into a discounted marginal “social cost of carbon” (SCC) for input into regulatory decision-making. Again, this estimation has been subject to political ping-pong, and to crude gamesmanship on the part of activist academics. And as Columbia’s Noah Kaufman, a climate economist who served on President Biden’s Council of Economic Advisers, recently wrote, “the SCC was always a fragile foundation for policy…[spitting] out virtually any number by replacing one reasonable assumption with another.” Relying on inherently arbitrary SCC estimates, especially those produced for transparently ideological ends, to justify compliance costs of regulatory implementation of a law the judiciary plainly distrusts, seems likely to end, at best, in years or decades of fruitless litigation, not meaningful emissions reduction.

Move On

The last time Democrats regained political control after the Trump Administration, they rapidly set about restoring the ex-ante climate policy apparatus. The next time Democrats regain control, the risk is that they’ll run the same playbook over again. They would be wise not to. As Kaufman sagely advises, “It may be advantageous to do so without attempting to reestablish a federal government SCC estimate.” The same holds for the Clean Air Act.

It’s understandable that climate advocates are angered at the Trump Administration’s various assaults on their policy accomplishments. But an unreflective defense of the status quo ante is an unproductive path forward. Of course it’s easier said than done to call upon Congress to backfill the Clean Air Act emissions authority with a more tractable policy instrument, whether sector-specific technology standards, emissions fees, or something else. But the alternative is another round of ambitious Democratic rhetoric on climate action in defense of overwhelmingly symbolic regulatory action.