The Emperor’s New Endangerment Theory (Part III)
How did EPA get to the absurd conclusion that 1.5 billion tons of carbon emissions aren’t significant? Well might you ask.
EPA has managed to come to the conclusion that carbon emissions from the power sector do not significantly contribute to climate change. No, I’m serious — even though power plants emit a billion-and-half tons of CO2 a year (a quarter of all U.S. emissions), EPA’s solemn conclusion is that they’re no biggie.
My prior posts (here and here) show how EPA’s new endangerment requirement makes little sense given the language of the Clean Air Act and EPA’s prior implementation of the law. Today I’m going to look at how EPA applied this newfound requirement. How does EPA justify the absurd conclusion that 1.5 billion tons is insignificant?
Begin with EPA’s interpretation of the statute. You might think that whether an industry significantly contributes to harmful pollution would depend on what it emits and how harmful that is. But no. EPA says it has discretion over how to define “significant contribution.” It proposes to define “significant contribution” in terms of “policy issues” such as the effectiveness of controls, whether their cost is reasonable, whether the presidential has issues relevant executive orders, side-effects involving other industries like AI, and impacts on public health and welfare, along with general principles of proximate cause. All that in the simple phrase, “significantly contributes.”
Here’s how EPA applies those factors:
First, EPA surveys various strategies for controlling emissions and excludes all of them as nonstarters. It finds each possible strategy either (a) inconsistent with its (strained) reading of Supreme Court precedent, (b) unreasonably expensive, or (c) ineffective.
Although it doesn’t say so, one of the strategies it brushes aside as useless is the very strategy the Trump Administration itself adopted in a previous regulation, the Affordable Clean Energy rule. What it once adopted as the right way to go turns out to be barely worth a moment’s thought this time. Apparently the Trump Administration admits what its environmental critics said at the time: the Affordable Clean Energy rule was nothing but a sham.
Second, “only extraordinary emission reductions on a global scale would have any impact” on climate impacts. Or so says EPA. That’s scientifically unsupportable and inconsistent with economic modeling of the social cost of carbon.
Even at the first Trump Administration’s lowball estimate of the social cost of carbon, the roughly 1.5 billion tons of CO2 emitted by the U.S. power sector each year would cause over $10 billion in U.S. damages. Emission reduction measures would be permanent, so benefits of a single regulation would accrue year after year. That all seems, to pick a word at random, “significant.” And the Trump estimate of the social cost of carbon was a fraction of the level favored by economists.
Third, the Administration’s policy is that “continued and increased reliance on fossil fuels” is crucial for public welfare. Of course, that policy is dubious, and might be changed with a stroke of the pen by the next President. Thus, EPA says, whether coal-fired power plants would significantly contribute to harmful pollution has to take into account the risk that regulation might require some of them to close.
That’s utterly inconsistent with the ordinary meaning of the statutory language. For instance, suppose you applied this definition of significance to medical expenses. It would mean that a person’s medical expenses would be considered insignificant even if they were a quarter of a person’s income if the medical care was indispensable.
Fourth, EPA now says“the threshold for significant contribution … is heightened by the multiple intervening actors, uncertainties, and extrapolations necessary to draw a connection between emissions by a source category and dangerous air pollution in the form of adverse effects in the U.S. from anthropogenic climate change.” Scientists would disagree with EPA’s view about the severity of the uncertainty. In EPA’s view, these factors “would require a higher volume and percentage of contribution.” On the other hand, some of the uncertainties cut the other way — harm could be much higher than expected. EPA ignores that. So that would argue for lowering the bar for finding significance to avoid the risks of less likely but possible catastrophic outcomes.
Fifth, EPA stresses that the U.S. power sector creates only 3% of global emissions. In EPA’s view, that “suggests that the risks to public health and welfare attributed to anthropogenic climate change would not be meaningfully changed even if [the power sector were to cease all GHG emissions.” Thus, regulation would not have a “discernible impact on the potential danger.”
Putting everything else aside, this argument ignores the fact that a tiny percentage of an extremely large number can still be a big deal. For instance, the Supreme Court seems to consider a billion dollars in compliance costs economically significant — but a billion dollars is less than 0.01% of GDP. Similarly, Elon Musk owns a tiny share — a quarter of one percent — of all U.S. wealth, but most people still think he has “significant” wealth.
———————-
The weirdest thing about EPA’s current analysis is that EPA seems to be suffering from severe amnesia. It seems to completely ignore the fact that it reached a directly contrary conclusion in 2021. During the first Trump Administration, “EPA determined that 3 percent of the U.S. GHG emissions was the best threshold for determining significance.” Power sector emissions are currently 25% of total U.S. emissions, far over this threshold.
Why does EPA reject its past analysis? It doesn’t say. It doesn’t even seem to remember it.
There is a very good chance that a court would find all this “arbitrary and capricious.” EPA’s effort to explain its ultimate conclusion rests on a hodgepodge of poorly analyzed considerations, which obviously have been reverse engineered to lead to EPA’s preferred conclusion.
The malleability and policy-driven nature of the analysis makes it even less likely that the 1970 Congress expected EPA to jump through such hoops to decide whether to list an industrial category. Congress told EPA to create a simple list and gave it only months to complete the job across the entire economy. It surely didn’t expect each industry to require an intensive policy analysis covering a multitude of factors and requiring EPA to decide in advance on the merits of every possible control strategy.
Even if a court were to agree that a new endangerment finding is needed for every pollutant and involves vast agency discretion, a later Administration would have little trouble reversing its application of the standard. What discretion does, later discretion can undo.