While politically popular with parts of President Trump’s political base, the EPA decision is legally dubious. At heart, it is a costly, symbolic gesture that will divert administration resources from more meaningful deregulatory initiatives and does nothing to advance a rational climate change policy.

Statutory interpretation / Under Section 202 of the Clean Air Act (CAA), the EPA administrator is required to set motor vehicle standards for emission of “any air pollutant…which in his judgment cause[s], or contribute[s] to, air pollution which may reasonably be anticipated to endanger public health or welfare.” In 2009, the Obama administration concluded that greenhouse gas emissions from motor vehicles contribute to climate change, which in turn can be reasonably anticipated to endanger public health or welfare, thus triggering the obligation to regulate. As other provisions of the CAA contain similar language, the EPA’s 2009 conclusion as to motor vehicles triggered more expansive regulatory obligations.

In initially proposing to rescind the 2009 finding, the EPA delivered a wide assortment of arguments, including that the agency had misunderstood or misrepresented the relevant science. In its final rule, however, the EPA took a narrower approach, wisely eschewing a fight over whether existing climate science satisfies the statutory standard. The CAA does not require that the EPA administrator conclude that climate change is catastrophic, highly consequential, or even net detrimental to human welfare. Nor does it require the administrator to conclude that greenhouse gas emission controls are preferable to other forms of mitigation or to an emphasis on adaptation. Instead, it merely requires that the EPA administrator “reasonably anticipate” negative effects on health or welfare, effects that the CAA defines to include effects on climate, “economic values,” and “personal comfort and well-being.”

This explains why, despite all the hoopla about the revisionist July 2025 Energy Department report arguing that carbon-induced warming appears to be less threatening than commonly believed, the EPA opted to focus on legal arguments for the repeal rather than scientific ones. Setting aside whether the DOE report was produced in violation of the Federal Advisory Committee Act (as a court found) or accurately represents current climate findings, demonstrating that climate change poses less of a threat than what climate change activists claim would not provide an adequate basis for rescinding the endangerment finding. As the EPA concedes in its final rule, “Whether CAA section 202(a)(1) authorizes the EPA to regulate in response to global climate change concerns by prescribing emission standards is a matter of statutory interpretation, not scientific analysis.”

Supreme Court clash / The EPA’s current position is that under the “best reading” of the CAA, Section 202 “does not authorize the Agency to prescribe emission standards in response to global climate change concerns.” I have sympathy for this view, having drafted an amicus brief for the Cato Institute making just this argument. The problem is that the Supreme Court concluded otherwise in Massachusetts v. EPA (2007). The EPA offers other arguments to circumvent this holding, but they are wholly unpersuasive.

Contrary to the EPA’s protestations, the Supreme Court in Massachusetts held squarely that the “EPA has the statutory authority to regulate greenhouse gas emissions from new motor vehicles.” Rejecting the George W. Bush administration’s arguments that Congress had not authorized the EPA to address climate change, the Court declared it had “little trouble concluding” that Section 202 “authorizes EPA to regulate greenhouse gas emissions from new motor vehicles in the event that it forms a ‘judgment’ that such emissions contribute to climate change.” So much for the agency’s claim that the statute can be read to preclude EPA regulation “based on global climate change concerns.”

Lest there be any doubt about what the Court said in Massachusetts v. EPA, it reaffirmed its express holding in Utility Air Regulatory Group v. EPA (2014), another decision on which the EPA would now like to rely. While noting that the agency might have “authority to exclude greenhouse gases from the class of regulable air pollutants under other parts of the Act where their inclusion would be inconsistent with the statutory scheme,” the Court reaffirmed that Section 202 authorizes the EPA “to regulate greenhouse gas emissions from new motor vehicles.” And for the record, the majority opinion was written by the late Justice Antonin Scalia, one of the dissenters in Massachusetts v. EPA.

It does not matter if the EPA (again) believes the Massachusetts decision was wrong and the CAA was never meant to apply to greenhouse gases. The Court has given the act a definitive interpretation to the contrary, and the Court rarely reconsiders its own prior statutory interpretations. If the Court got a statute wrong, that is a mistake for Congress to fix.

The EPA tries to claim that intervening Supreme Court decisions, such as Loper Bright Enterprises v. Raimondo (2024) and West Virginia v. EPA (2022), call for a new interpretation of the CAA, but those arguments also fall flat. In Massachusetts the Court found that the statutory text was clear and rejected the argument that it should pause before concluding Congress authorized the agency to regulate the most ubiquitous byproduct of modern civilization (what we would now call a “major question”). Were that not enough, in Loper Bright Chief Justice John Roberts, writing the majority opinion, went out of his way to make clear that the Court’s decision did not “call into question” statutory interpretations reached in prior cases. Rather, he explained, the holdings of such cases, and the conclusions reached about what statutes require or what agency actions are lawful, “are still subject to statutory stare decisis.”

Undaunted EPA / Perhaps recognizing the obstacle Massachusetts v. EPA poses for its arguments, the EPA offers a few alternatives, yet they are also unpersuasive. While ostensibly representing the “best reading” of the statutory text, the alternative bases for endangerment rescission read like backfilled justifications for a predetermined conclusion, not an agency’s good-faith effort to get the law right.

The EPA argues that Section 202 only calls for regulation of emissions that contribute enough to an air pollution problem to make a meaningful difference. Yet, unlike other portions of the statute, the text of Section 202 makes no such qualification. Section 111 of the Clean Air Act, upon which the Obama and Biden administrations sought to base regulations governing greenhouse gas emissions from utilities, only applies to emissions from sources that “cause or contribute significantly” to the air pollution at issue, suggesting the EPA need not—indeed, perhaps, cannot—regulate sources of de minimis contributions under that provision. The absence of any such qualification in Section 202, however, is a mortal blow to the EPA’s position.

The EPA asserts that regulating the emission of greenhouse gases from new motor vehicles would be futile in that it would not have an appreciable effect on global climate change. But the agency offers no textual argument for why this would preclude an endangerment finding. Courts have long understood that not everything in the CAA must make policy sense. The futility of regulating new vehicle emissions is an argument for not adopting costly controls, and perhaps even for adopting standards that are no more stringent than what is driven by consumer demand. Yet it is no argument that no endangerment finding may be made, let alone an argument that such a finding is precluded. Here as elsewhere in its rulemaking, the EPA is reading into the CAA things it simply does not say. Such arguments may have had purchase in a world in which courts were required to defer to an agency’s reasonable statutory interpretations. They fall flat, however, as an attempt to discern the “best reading” of what the statute actually says, particularly in light of what courts have said about it.

It is difficult to read the EPA’s endangerment rescission final rule without concluding that the agency is hoping to prompt the Supreme Court to reconsider Massachusetts v. EPA. On this reading, it does not matter that the efforts to evade that decision’s reach are unconvincing, because the EPA’s current position is Massachusetts v. EPA was wrong, and it hopes it has the votes at One First Street to vindicate that position. It is one thing to adopt regulatory policies with an eye for how they may be received in court. It is quite another to stake a major rulemaking on the Supreme Court’s willingness to reject stare decisis. In this regard it may be worth noting that the Roberts Court overturns precedents less often than did its predecessors, and it has largely confined such decisions to cases involving constitutional questions, not statutory matters a functional Congress could fix.

Even if there are five votes to revisit Massachusetts v. EPA, those cheering endangerment rescission should be careful of what they wish for. Like it or not, it was the Supreme Court’s conclusion that greenhouse gases are subject to regulation under the CAA that dictated the outcome in American Electric Power v. Connecticut (2011), in which the Court held that federal common law nuisance suits against fossil fuel emitters are displaced by federal statute. As the Court explained in American Electric Power, existing doctrine requires the displacement of federal common law causes of action where Congress has entered the field by enacting a relevant statute. Should the Massachusetts holding be undone, however, placing greenhouse gases and climate change concerns beyond the scope of the law, there would no longer be any basis to bar such suits from federal court. And were Massachusetts to be undone in this manner, it is possible that federal preemption of state greenhouse gas emission standards—such as those sought by California—would be cast aside as well. Congress may have put an end to CAA waivers for California greenhouse gas emission standards for new motor vehicles through the Congressional Review Act, but if greenhouse gas emissions from motor vehicles are no longer subject to CAA regulation, it is not clear why any such waiver would be required, or why such standards could only be adopted by California—though other statutes may still preclude state standards that operate as de facto fuel efficiency requirements.

Great white whale / The legal risks of rescinding the endangerment finding might be justified if there was no other way to scale back federal regulation of greenhouse gas emissions. Yet that is not the case. The endangerment finding itself is no obstacle to the relaxation or rescission of existing greenhouse gas emission regulations, including those imposed on motor vehicles. The relevant statutory text provides ample basis for ending regulation of such emissions from stationary sources, and there are strong legal arguments having nothing to do with endangerment that the EPA lacks the statutory authority to push automakers toward the production and promotion of electric cars. Such deregulatory moves would be quite straightforward, but they do not have the symbolic appeal of rescinding endangerment.

Administrator Zeldin called the endangerment finding “the holy grail of the climate change religion.” It may be more accurate to describe it as the great white whale of the second Trump administration’s EPA.