Climate Change and the Supreme Court

In 2022, the US court system could largely determine the future of US climate policy and environmental regulation. Acting on two specific doctrines in recent cases, the Supreme Court may end up substantially limiting regulations on carbon pollution.

Professor Lisa Heinzerling joined from the Georgetown University Law Center. She previously served as Senior Climate Policy Counsel to the Administrator of EPA, as well as Associate Administrator of EPA’s Office of Policy. Heinzerling was the lead author of the winning briefs in the monumental Massachusetts v. EPA case of 2007, in which the Supreme Court held that the Clean Air Act gives EPA the authority to regulate greenhouse gases.

Professor Jonathan Wiener joined from Duke University; he’s also a university fellow at RFF. He previously served at the White House Council of Economic Advisers, the Office of Science and Technology Policy, and the US Department of Justice during the first Bush and Clinton administrations. Wiener also helped negotiate the United Nations Framework Convention on Climate Change.

Susan F. Tierney moderated the conversation. Tierney is the chair of the board of directors at RFF, a senior advisor at Analysis Group, and an expert in energy economics, environmental regulation, and many other topics covered by RFF research. Previously, she was the Assistant Secretary for Policy at the US Department of Energy and served on the US Secretary of Energy advisory board.

The transcript of this discussion has been edited for length and clarity. The event was held on March 22, 2022.

Lisa Heinzerling

Despite the apparent lack of any current effect on or injury from any existing EPA regulation, the Supreme Court took up the challenge to EPA’s authority under Section 111. And at oral argument, the conservative justices showed no sign that they’re contemplating ditching the case because of the lack of a current injury from a regulation in place right now. I think it’s reasonable to expect the court to decide the case on the merits.

Two legal doctrines may loom large in the court’s decision. The first is the nondelegation doctrine; this one is constitutional. The second is the major questions doctrine; this one is statutory.

Nondelegation Doctrine

Major Questions Doctrine

Moreover, the relevant statutes tend to speak in broad terms about the power of agencies. This broad-strokes approach requires the agencies to tackle new environmental risks as those risks emerge, without specifying the risks in detail in advance. The point of this legislation is to empower agencies to tackle risks that Congress either doesn’t foresee or didn’t know enough about at the time the statutes were passed.

I think the court’s major questions idea threatens to undo these statutory judgments. And in these recent cases that we’re discussing, the conservative justices have intertwined the major questions idea with the nondelegation idea—thus perhaps transforming a principle of statutory interpretation into a principle of quasi-constitutional significance. To return to West Virginia v. EPA: the challengers to EPA’s authority argue that, given the policy importance of what they characterize as a restructuring of the energy sector, Congress needed to speak more clearly if it wanted to give EPA the authority that EPA claimed in the Clean Power Plan.

Consequences of Court Cases

Jonathan Wiener

Interestingly, the major questions doctrine seemed, at one time, to be a kind of exception to the Chevron doctrine, saying that when there really are major questions, then the agency would not get deference for its interpretation, and the court itself would interpret the ambiguous statute. But the major questions doctrine has evolved over time, seemingly to stand apart from the Chevron doctrine, and even to operate in cases where that whole Chevron multi-step analysis doesn’t even enter into the case. In West Virginia v. EPA, for example, there was no discussion of Chevron deference in the oral argument. And in the NFIB v. OSHA case in January, the opinions did not rely on Chevron deference.

This independence of the recent cases from the Chevron doctrine suggests (as others have commented) that the major questions doctrine is being treated (at least by its advocates) as a distinct limitation on the delegations that Congress can make to agencies. The major question doctrine seems to have become not just an exception to the idea of deference to agencies when statutes are ambiguous, but perhaps a limitation on agency authority even when the statute is clear. For example, in the NFIB v. OSHA vaccination and testing rule case, the statute said that OSHA has the authority to regulate occupational risks, new dangers, and grave threats. But the majority of six justices—and especially a concurring opinion by Justice Neil Gorsuch for three justices—wrote that the agency needed to go back to Congress, seemingly because OSHA covers occupational risks at the workplace but not public health risks outside the workplace, or because OSHA covers occupational risks but the law did not name a pandemic or COVID as such a risk.

Where does all this bring us in West Virginia v. EPA? Clean Air Act Section 111 clearly authorizes EPA to regulate air pollutants by setting a standard based on the “best system of emission reduction,” taking into account several other factors such as cost, non–air quality impacts, energy requirements, and that the “best system” is adequately demonstrated. All of that is relevant to the interpretation of Section 111. And the Supreme Court already has held that the term “air pollutant” includes greenhouse gases, in Massachusetts v. EPA back in 2007. But even so, the Supreme Court majority nonetheless might still invoke the major questions doctrine to say that climate change is such a major issue—or that the regulation of the electricity system is such a major issue—that Congress must say so more clearly to delegate that authority.

Invoking the Major Questions Doctrine

In addition to the distinction that Lisa mentioned about action versus inaction, invoking the major questions doctrine like this also poses an irony about which types of risk the courts are willing to allow executive action to address first. In these health and environmental cases—like NFIB v. OSHA, Alabama Association of Realtors v. Department of Health and Human Services, and possibly West Virginia v. EPA—the major questions doctrine is being deployed to say that if it’s not extremely clear in the statute, then the agency has to go back to Congress for new authority. But for national security risks, the typical posture is the opposite: the executive branch has latitude to act first, and then Congress could restrict that executive action. And if the point of the major questions doctrine—or even the nondelegation doctrine—is to force Congress to take democratic accountability for major decisions, then it seems possible to work either way. That is, it could be that Congress has to act first—but it could also be that the executive gets to act first, and then Congress can, by statute, condition or restrain, or limit the duration of, the executive action.

There’s a risk-selection asymmetry here, to the extent that health and environmental risks are being sent back to Congress, but the executive gets more latitude to act in response to national security risks. One might say that this asymmetry may derive from the executive’s greater authority over military, national security, and foreign policy issues under Article II of the US Constitution, but one could then recognize that climate change and pandemic disease also are national security risks and foreign policy issues. This asymmetry at least needs to be addressed as a risk-management problem—and as a legal question.

Social Cost of Carbon

Judging the Cases

Susan Tierney: In many of these environmental regulatory and public health issues, the technical, scientific, and evidentiary considerations for the construction of public policy is very dense. We couldn’t expect every congressperson to be expert across so many different complicated domains as to be able to provide detailed statutory instructions to agencies that cover the variety of circumstances that may affect regulation across the many states. And we also can’t expect bills and laws to anticipate, cover, and address all the technical permutations that could exist in, let’s say, regulating power plant emissions—if that were true, we’d see 3,000-page decisions.

Please speak to this tension between the ability of policymakers in Congress to actually make decisions on 3,000-page laws with the fulsome consideration of all of these questions, versus drafting general instructions that deliberately delegate to more technical experts the job of figuring out how to execute on that policy.

Lisa Heinzerling: One of the basic reasons that Congress delegates authority to agencies is because people in Congress don’t have the expertise or the time to handle all these problems. We’ve seen pervasive delegations of authority to administrative agencies since early on in this country’s history. What we’re really talking about is a fight between the power of the court and the power of Congress.

To the extent that these decisions shift from an approach that grants or withdraws deference to agencies, to an approach that says, “Congress, you need to speak more clearly if you want to do this,” the Supreme Court is taking direct aim at the power of Congress to delegate authority to whomever Congress wishes. This shift is a threat to that choice of handing the decision over to others.

And it’s hard to predict at this point how clear the court will insist Congress should be—especially given that statutory ambiguity is inevitable, at least to some degree, and especially on big problems that already require hundreds of pages of bills. There’s going to be ambiguity, but the court essentially is saying, “You can’t do that.” These are cases in which Congress has not failed to act—Congress has acted and passed a statute—but the court can’t or won’t hear what Congress has said.

Jonathan Wiener: I’d like to make three quick points—about timing, sticking up for the executive branch, and limiting principles.

The Pace of Congress

Jonathan Wiener: We can point to some situations in which Congress has written an extremely long bill with lots of detail, yet the court says, “Even so, the agency does not have the authority, so it’s back to Congress.” An example is FDA v. Brown & Williamson Tobacco Corporation in 2000, in which the Supreme Court said, “No, FDA does not have the authority to regulate tobacco.” Congress responded by enacting a new statute to give the authority more clearly to the agency—but it was still nine years later that Congress enacted the Family Smoking Prevention and Tobacco Control Act in 2009.

A more recent example is a little bit different. The DC Circuit held in 2017, in an opinion by then-Judge Brett Kavanaugh (when he sat on the circuit court), that EPA lacked the authority to regulate hydrofluorocarbons under Section 612 of the Clean Air Act—declining to defer to EPA’s statutory interpretation under the Chevron doctrine. Congress then enacted new authority in the American Innovation and Manufacturing Act in late 2020, just about three years after the DC Circuit decision.

So, things can take a long time—Congress can be slow, and in some cases, new authority may never be enacted. This slow pace could result in the court sending “major questions” back to Congress as a blank slate, which can result in gridlock and inaction. That’s why I mentioned earlier the possibility of the presumption going the other way—as in the national security arena—in which the executive can act first, but then Congress can still exercise its role by ratifying or restraining that executive action. In that case, Congress has a policy to respond to, rather than dealing with a blank slate of no policy.

Executive Branch, Checks, and Balances

Jonathan Wiener: In addition to the courts and Congress, we also have the executive branch—for good reason. Yes, Congress should be democratically accountable for passing laws, and the executive branch should take care that those laws are faithfully executed, as the US Constitution says. But the executive branch also performs the executive function of thinking through problems and making difficult policy decisions. I’m reminded here of Alexander Hamilton’s advice in Federalist Paper 71, in which he says that a key reason we have an executive branch is that we don’t want the government to reflect only the immediate “inclinations” of the general public—some complicated problems need to be thought through with “cool and sedate reflection,” involving expertise and executive decisionmaking. So Hamilton says that, while the people “commonly intend the public good,” they don’t always know how best to do so—and we need the executive branch to think carefully, and at times to “save the people from the very fatal consequences of their own mistakes.” I think all this is quite applicable here: neglecting pandemic disease or climate change, or designing policies with adverse effects, can be fatal mistakes.

That’s not to say the executive can act without any authority. But if Congress already has delegated to OSHA explicitly in the statute the regulation of occupational risks, grave dangers, and new hazards, then OSHA could apply its expertise to a new hazard where its risk is heightened by workplace conditions —and Congress can then act to restrain OSHA—rather than the court disabling OSHA from regulating at all. And likewise, potentially, in West Virginia v. EPA.

Limiting Principles

Jonathan Wiener: The nondelegation doctrine used to be understood as saying it’s okay for Congress to delegate if Congress provides intelligible principles to the agency. A 3,000-page statute, as we’ve mentioned, would have lots of limits and conditions on what the agency can do—lots of instructions. In NFIB v. OSHA, Justice Gorsuch seemed especially displeased with OSHA’s remark in its rule that OSHA has “almost unlimited discretion” to regulate occupational risks. The courts look for limits on authority delegated to agencies, in order to avoid excessive delegations (as in the Benzene case back in 1980).

During the oral argument in West Virginia v. EPA, it was noteworthy that US Solicitor General Elizabeth Prelogar specifically pointed out the limiting principles in Clean Air Act Section 111, such as the “best system of emission reduction,” costs, non–air quality impacts, energy requirements, and the “adequately demonstrated” clause. And the word “best” in “best system of emission reduction” could include benefit-cost analysis, analogizing from the Supreme Court in 2009 in Entergy Corp. v. Riverkeeper.

Limiting principles might satisfy a court that sufficient constraints limit executive action, to ensure that the executive is being faithful to the legislation and is not going too far afield.

Endangerment Finding

Susan Tierney: Can we read anything in the tea leaves about this Supreme Court with regard to the “Endangerment Finding”?

Lisa Heinzerling: There’s a strange and frustrating thing about the court’s recent turn toward looking for major questions and, upon finding them, looking for extreme statutory clarity: Once the court has done so, it doesn’t have to do anything else. It just says, “This is a big question. Congress didn’t answer it clearly enough to our satisfaction. We’re done here.” So, the extensive scientific and technical bases for policies that we mentioned before become irrelevant, almost with a single sentence. Those cases on the CDC’s eviction moratorium and on OSHA’s shot or test rule were incredibly brief. The Supreme Court got in, saw a major question, and got out.

I don’t think the court has to do anything with the Endangerment Finding in order to cut way back on authority to regulate greenhouse gases in an ambitious way. The reason is that the court is operating on a whole different plane. Justices may have their own views about climate change and the science behind it, but those views are not even relevant in these cases, because the court is operating at such a high level of generality and looking only at legal principles—not at scientific findings or even environmental risks.

I mentioned before that I think the court applies the major questions doctrine in a skewed way, by applying the major questions doctrine against regulatory programs that are ambitious, rather than against failures to regulate. Thus, in looking for a major question, they’ve always been looking at what the effect is on the regulated industry—and they don’t take into account the effect on the environment or on humans of denying regulatory authority. So, it’s even worse than you might have supposed: It’s not that the Supreme Court is going to undo the Endangerment Finding; I just don’t see that happening. What’s going on is the Endangerment Finding is irrelevant to the court in thinking through the statutory questions.

Reigning Supreme

Susan Tierney: About the nondelegation doctrine: A limit constrains Congress in its ability to delegate power, when that power is applied to control the actions of private actors. Lisa, I’m intrigued by your comment just now about considering the effects on private actors versus effects on the environment or humans or the public interest more broadly—can you talk more about this focus on private action? And are the actions of individuals as citizens implicated under the umbrella of private action, or are we mainly talking about corporations and other organizations?

Lisa Heinzerling: Based on my reading of things, the court recently has been quite careful to talk about its worries about regulation—not its worries about nonregulation, nor weak regulation—that addresses significant problems and has significant economic and political consequences.

That’s where the nondelegation and the major questions doctrines start to look alike—both have a skew in the current court. And what the conservative justices seem to be worried about is the intrusion on liberty posed by government regulation. What I think they are not worried about is the intrusion on liberty posed by private people acting in ways that hurt other people, which the government might be able to step in and address.

Susan Tierney: The United States has three equal branches of government. But from what I’m hearing you both say, the Supreme Court is “acting supreme,” in some sense. Do you have thoughts on that?

Lisa Heinzerling: I think your question cuts to the heart of the fear that many of us feel about the Supreme Court. In a sense, the court is “acting supreme,” but the harder the Supreme Court makes it for Congress to pass legislation that the court is satisfied with, then the more the Supreme Court appears to be its own “supreme” entity within the government. And the court has gotten very aggressive recently—we’ve seen this with its shadow docket, and we’re now seeing this with its regular docket. The court has been very aggressive about taking cases, keeping cases that might not have an injury to support justiciability, and pushing back on the power of Congress. Congress under the US Constitution is supposed to be the first and main branch, right? These are decisions that strike at the power of Congress.

Jonathan Wiener: I’ll just note that RFF University Fellow Nathan Richardson wrote a detailed historical article with a powerful critical argument, called “Antideference,” about exactly this point. He describes the history and reasons for the evolution of the major questions doctrine, in the way the doctrine has been deployed during the last year in the CDC eviction moratorium case, the OSHA vaccination and testing case, and perhaps in West Virginia v. EPA about the Clean Air Act. He argues that this evolution of the doctrine speaks more and more to a kind of judicial supremacy or judicial activism.

In NFIB v. OSHA, it’s interesting to see how, even though Justice Gorsuch concurs with the denial of OSHA authority, and Justice Stephen Breyer speaks for the dissenters who would have recognized the authority in OSHA, they both conclude on exactly the same point. Justice Gorsuch said, “The question before us is not how to respond to the pandemic, but who holds the power to do so. The answer is clear … that power rests with the states and Congress, not OSHA.” By contrast, Justice Breyer concluded his dissent, “Underlying everything else in this dispute is a single, simple question: Who decides how much protection, and of what kind?”

But then Justice Breyer goes on to say that the methods we use to address a pandemic should be a choice between the expertise of the agency authorized by Congress, versus the court and the judges. And Breyer cites a previous opinion by Chief Justice John Roberts when he says, “And then, there is this court. Its members are elected by, and accountable to, no one. And we ‘lack the background, competence, and expertise to assess’” these complex risks. And Breyer goes on to say, when we (the court) are wise, we know when to defer to experts. “Today, we are not wise.”

I think in that question about the judicial branch versus Congress, the majority seems to be saying, “Congress has to make these decisions. Force Congress to be democratically accountable.” But the near-term result is to disable the executive, and Justice Breyer would have recognized the authority in the executive.

Catch-22 for Congress in Court Cases

Susan Tierney: I want to call everyone’s attention to Nathan Richardson’s question in the audience. He essentially asks whether all of this is setting up a catch-22 situation: The court is saying that Congress has to be very clear—but any delegation that’s clear enough to satisfy the major questions doctrine could be a violation of nondelegation criteria. Do you think the Supreme Court’s requirement for a clear statement from Congress in the major questions cases is something Congress could realistically satisfy? Or is any delegation that’s clear enough to satisfy the major questions doctrine itself a violation?

Jonathan Wiener: In NFIB v. OSHA, Justice Gorsuch’s concurring opinion has a sentence in it that says, “This lone statutory subsection does not clearly authorize OSHA’s mandate”—meaning that the delegation of authority is not clear enough, so under the major questions doctrine, the question has to go back to Congress.

But even if the delegation were clear, he hints that it would be excessive delegation. As Gorsuch wrote: “On the one hand, OSHA claims the power to issue a nationwide mandate on a major question but cannot trace its authority to do so to any clear congressional mandate. On the other hand, if the statutory subsection the agency cites really did endow OSHA with the power it asserts, that law would likely constitute an unconstitutional delegation of legislative authority.” And in the climate change context, what would that look like? What kind of law could Congress write to satisfy at least this one point of view on the court?

Hypothetically, if the Supreme Court majority in West Virginia v. EPA were to decide that Clean Air Act Section 111 has to be interpreted narrowly, such as inside the fence line, and if they go further to decide that Congress did not clearly delegate the control of greenhouse gas emissions to EPA, because that’s such a major question (which would, by the way, be overturning the Massachusetts v. EPA case that Lisa won)—then what? Could Congress write a new law, which might run into the same catch-22 we’re mentioning? Or is there a way that Congress could regulate greenhouse gases without delegating to an agency—such as by enacting a carbon tax directly in the statute? (In the oral argument in West Virginia v. EPA, Justice Kavanaugh alluded to Congress having enacted the acid rain trading program directly in the 1990 statute.) I don’t know how to estimate the likelihood of that kind of law, but I’d be curious to know what Lisa thinks.

Lisa Heinzerling: I think there are two separate questions. One: Is there any clarity that’s good enough for the Supreme Court, regardless of whether there’s a catch-22? I really don’t know. Because if you look at the OSHA statute, it feels like it fits the pandemic really well. And there’s no doubt that the standard applied only to workplaces. The notion that it wasn’t clear enough in allowing workplace standards during the pandemic strikes me as signaling that the court isn’t really hearing what Congress is trying to say.

On the catch-22 question: If Congress speaks clearly and says, “We see the following major decision that needs to be made, and we are going to hand it over to the agency with complete crystal clarity,” then Congress could get into trouble for being so clear in handing the decision over. I think the opinions in nondelegation cases that haven’t yet matured into a holding from the majority suggest that Congress needs to be clear—not just in handing over power to the agency, but actually in making the major decisions itself.

Susan Tierney: Listening to you in this conversation is giving me pause as I think about the new proposed regulations from the US Securities and Exchange Commission on climate disclosures.

Social Cost of Carbon Holding Court

Susan Tierney: Is the social cost of carbon likely to come up in a case before the Supreme Court? If so, what do you anticipate would be the result, given the makeup of this court?

Jonathan Wiener: I would have thought that it would depend on the statute and the specific posture in which the case comes to the court. I think the decision by the Louisiana court was premature, because there was not a rule under a specific statute yet that might cause harm to the plaintiffs and for the court to review—and that’s effectively what the US Court of Appeals for the Fifth Circuit said.

The case in Louisiana, at least in the first round, was based in part on whether the global social cost of carbon can be used by federal agencies, as opposed to a domestic-only social cost of carbon. The court cited the initial section of the Clean Air Act, which refers to its purpose of protecting “the nation”—so, some people have argued that means EPA can consider only impacts in the United States. But on the other hand, other provisions of the Clean Air Act speak more broadly and apply to international and global environmental impacts. We could have a debate about what the Clean Air Act says. The National Environmental Policy Act, in Section 102(2)(F), tells all federal agencies to consider “worldwide” impacts and “international cooperation.”

The court also looked to OMB Circular A-4, which gives guidelines to federal agencies on how to do benefit-cost analysis. Issued in 2003 and still in force today (but possibly soon to be revised), OMB Circular A-4 contains two sentences about domestic versus global impact—but those sentences don’t say to focus only on domestic impacts; they say to focus on domestic impacts and to report impacts outside the United States. And a recent op-ed by RFF Visiting Fellow Arthur Fraas—alongside John Graham, Randall Lutter, and Jason Shogren—says the federal government should look at both domestic and global impacts of the social cost of carbon.

All of this poses an interesting question for those at RFF and elsewhere, who estimate the social cost of carbon: Can a global and domestic set of impacts be readily estimated in the social cost of carbon calculations?

One might have thought—under the Administrative Procedure Act of 1946 and a whole series of cases that goes back to at least 1983 in a case about automobile airbag rules called Motor Vehicle Manufacturers Association of the United States, Inc. v. State Farm Mutual Automobile Insurance Company—that agencies would be obligated to consider the full impacts of their decisions, and it would be arbitrary to omit an important impact. And indeed, just two years ago, regarding the citizenship question in the census case of Department of Commerce v. New York, Justice Breyer wrote an opinion saying that, in his view, it’s arbitrary under the Administrative Procedure Act to omit an important impact. So, it would be curious for the Supreme Court to say that the agency has to neglect an important impact as a general matter of agency regulation. From this point of view, an effort to estimate an important impact, like the damages from climate change, ought to be in—not out.

Lisa Heinzerling: The Louisiana district court decision is a little bonkers on the social cost of carbon, for a variety of reasons. There’s something a little crazy about saying that there’s justiciability now, talking about OMB Circulars as if they can’t be undone by later interagency documents that address the same questions, and for all sorts of other reasons. But there’s a nub of sense, which I worry about going forward. It is the kind of case that the justices might be interested in, but we’re a long way from any real controversy over the social cost of carbon and its injuries to people.

The estimate of the social cost of carbon is a weird kind of decision. It’s supposed to be used in all rulemaking proceedings, and it has the power, potentially, to change the direction of the agency regulatory proceedings—or at least the magnitude of the regulation that the agency embraces. And in that case, I do think the government would be well advised to take administrative procedures seriously and to effectively treat this as a rule by going through notice and asking for comments.

Jonathan Wiener: I’ll just quickly corroborate what Lisa just said, by adding that the Trump administration lost many cases in court—a much higher percentage than other administrations—often because they skipped steps in the administrative process. And courts may well be watchful for that now as well.

Susan Tierney: Thank you, Lisa and Jonathan, for sharing your expertise. This is a time of huge uncertainty. As a student of federal policy, state public policy, and regulation in this area, I feel as though we’re gravitating toward a place we’ve often been—where states do their own form of regulation in a patchwork of approaches that may cause the regulated industries to ask for regulation nationally.