The last day of the Supreme Court term unfolded as if it were a performance orchestrated by Chief Justice John Roberts to rebut the view that the Court has become an outpost of right-wing extremism. In the first of his two opinions for the concluding day, West Virginia v. EPA, Roberts wrote for a 6–3 conservative majority that the EPA lacked authority to promulgate the Obama administration’s 2015 Clean Power Plan. But, contrary to the fears (or hopes) of many, his opinion neither discarded the doctrine called “Chevron deference” nor suggested that Congress lacked the constitutional authority to delegate broad regulatory powers to the EPA. In Biden v. Texas, he upheld the Biden administration’s discretion under the Immigration and Nationality Act (INA) not to return to Mexico non-Mexican nationals arriving in the United States from Mexico in order to await the results of their removal proceedings. In the latter case, Roberts wrote for himself, Justice Brett Kavanaugh, and the three most liberal justices. (Justice Amy Coney Barrett, writing separately, said she agreed with Roberts’s statutory interpretation. She did not agree, however, that the Court had jurisdiction to decide the case at all.)
In both cases, the chief justice’s majority opinion went to some lengths to stress its narrowness. In the immigration case, Roberts declined to rule whether the Biden administration (or its predecessors) had discretion under the INA to release large numbers of foreign nationals into the U.S., rather than detaining them, prior to an adjudication of their legal admissibility. It was enough to decide the immediate dispute that the return-to-contiguous-territory provision in the INA was not a mandatory requirement. In the EPA case, Roberts’s opinion rejected the Obama administration’s reading of the Clean Air Act but did not endorse the specific narrow reading of the act issued by the EPA under Donald Trump.
One might have thought—as Justice Elena Kagan suggested in dissent—that the EPA case should have been dismissed because review had been “improvidently granted.” The 2015 regulation invalidated by the Court was not being enforced anyway, and the Biden administration had told the D.C. Circuit Court of Appeals that it would not reissue it. A modest judicial posture would have been to delay review of the EPA’s greenhouse gas regulations until a new rule had been promulgated.
Instead, the Court tackled a complex issue arising from the Clean Air Act. Section 111(d) of that act authorizes the EPA to regulate pollutants from existing sources, like coal-fired power plants, when those pollutants are not otherwise regulated by the agency under other programs. Carbon dioxide falls into this category of pollutants. The EPA is required, under 111(d), to specify the “degree of emission limitation achievable” through the “best system of emission reduction” (BSER) for the pollution source being regulated. In this case, that would be the best system for reducing carbon dioxide emissions from power plants. The Obama administration adopted a BSER that would include possible improvements in the power plants themselves. But it also determined that the “best system” for reducing CO2 emissions would be to shift away from coal plants to natural gas plants and to renewable sources. The emissions performance rates that would be implemented under the BSER would implicitly force owners of coal plants to seek other sources of electricity generation.
EPA supporters feared that the Court would rule that Congress could not constitutionally vest the agency with such discretion over the U.S. power system. The argument was that giving the EPA so much authority would amount to an impermissible delegation of legislative power into executive branch hands.
The Court, however, although it had invited briefing on that issue, never reached it. Instead, it used a version of the “major questions doctrine” (MQD) to hold that Section 111(d) did not give the EPA the authority to induce reliance on cap-and-trade to cut greenhouse gas emissions. How the energy market ought to be structured was, the Court found, a “major question” that Congress had not left to the EPA to resolve. Although the agency might have a plausible basis in the statute’s text for its regulation, the Clean Air Act did not provide sufficiently specific congressional authorization for such “extraordinary” administrative power.
As it happens, however, the agency’s reliance on the statutory text was not merely plausible. It was compelling. The D.C. Circuit opinion had been meticulous in its textual parsing. Kagan wrote scathingly in her dissent, “Some years ago, I remarked that ‘we’re all textualists now.’ It seems I was wrong. The current Court is textualist only when being so suits it. When that method would frustrate broader goals, special canons like the ‘major questions doctrine’ magically appear as get out-of-text-free cards.”
It could have been worse. The Roberts Court could have taken a wrecking ball to the administrative state. Progressives feared that the Court might use the EPA case to clamp down on the constitutional authority of Congress to empower administrative agencies to make the significant policy decisions often entailed in deciding how to regulate economic activity to serve the interests of public health and safety. Under the so-called nondelegation doctrine, the Court permits Congress to create such administrative powers if those powers are constrained in the statute by what the Court has called an “intelligible principle” limiting the agency’s discretion. In other words, Congress cannot simply tell an agency, “Do whatever you want,” because that would be abdicating the power of Congress to the agency. Various of the current justices, however, have expressed dissatisfaction with a doctrine so indulgent of Congress. For example, in his much-discussed concurring opinion in a case called Gundy v. United States, Justice Neil Gorsuch argued that Congress ought not to be able to ask the executive branch to do more than assist in “filling up details and finding facts.” A nondelegation doctrine this stringent would have made Congress’s work much more difficult and threatened its capacity to set up agencies with the flexibility to respond sensibly to major national challenges, such as climate change.
Cut from whole cloth, the MQD, although animated by the same separation of powers concerns as the nondelegation doctrine, is potentially a better alternative for the sound working of government. A case like West Virginia v. EPA at least allows a future Congress to grant major policy-making authority to an agency without violating the constitutional separation of powers. The Court likewise didn’t foreclose judicial deference to agency statutory interpretation when a statute is legally ambiguous, and the regulatory implications of the agency’s view are less than “extraordinary.”
To be sure, the practical consequences of the MQD can be profound. It effectively excuses courts from close statutory reading. Instead, judges need only dance a three-step waltz: The issue is major. The statute is ambiguous. Agency loses. Taking the first step on the dance floor is within a court’s purview. All lobbyists need to do is stir up enough opposition to persuade regulation-skeptical judges that the agency’s initiative is so controversial an extension of agency authority that it requires a more specific legislative license. Given America’s political polarization, the congressional power the Court has left intact to rewrite the law may be more theoretical than actual.
In his EPA concurrence, Gorsuch painted the difficulty of legislating as a constitutional feature, not a bug. To some extent, of course, he has a point. Regulatory policy with significant impact should proceed from a highly inclusive deliberative process. But Congress has already demanded such a process through the Administrative Procedure Act. What the MQD does is limit Congress’s capacity to empower agencies to be proactive as the country’s needs evolve. Gorsuch views the MQD as protective of liberty. But it may prove a serious wound to national governing capacity. The Constitution was written as much to create an effective national government as to limit that government to proper channels. Gorsuch’s approach implicitly discounts the former original aim.
Biden v. Texas might also have been thought to raise a major question of executive branch authority. But Roberts’s opinion was limited to text. Indeed, it rests predominantly on Congress’s use of the word may rather than shall in empowering the executive branch to return aliens “arriving on land … from a foreign territory contiguous to the United States” while awaiting the adjudication of their admissibility. Being textualist suited a majority in this instance.
This term will long be remembered chiefly for the Court’s elimination of constitutional protection for abortion and assertion of a Second Amendment right for individuals to carry guns outside the home. We can be grateful that dismantling the administrative state and hamstringing the Biden administration on immigration were not on the menu. But the final day’s cases made one thing clear: An agency’s ability to pursue even the most scrupulous reading of its statutory authority will remain vulnerable to the predilections of our very conservative justices.