Proceedings of a Workshop—in Brief

Convened April 7–8, 2025

Implications of Law, Policy, and Federal Agency Decision-Making Under a New Judicial Standard
Proceedings of a Workshop—in Brief

On April 7–8, 2025, the Committee on Science, Technology, and Law (CSTL) of the National Academies of Sciences, Engineering, and Medicine (the National Academies) held a workshop in Washington, D.C., titled “Implications of Recent Supreme Court Decisions for Agency Decision-Making.” The major impetus for the workshop was the Supreme Court’s 2024 decision in Loper Bright Enterprises v. Raimondo, which overruled the principle of Chevron deference. This principle was the product of a 1984 case, Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., in which the Supreme Court directed courts to defer to reasonable interpretations made by executive agencies on ambiguous aspects of the laws they implement. Related to the Court’s Loper Bright decision is a principle known as the major questions doctrine, which holds that Congress may not delegate to agencies decisions that are unheralded, transformative, and/or have major political or economic consequences without clear authorization

The workshop was designed to explore how agencies will develop, revise, evaluate, and potentially rescind regulations within an evolving regulatory landscape; the future roles of agency scientists and lawyers; the drafting of legislation, especially legislation aimed at agencies charged with protecting human health and the environment; and how industry, non-governmental organizations (NGOs) and the public are positioning themselves to address the new situation. Recognizing the recent actions of the Trump Administration, the workshop provided participants with an opportunity to consider whether the Loper Bright decision will lead to diminished deference to agencies and to assess the implications of agency staff reductions for the agencies’ ability to respond to Loper Bright. As CSTL co-chair Martha Minow (Harvard Law School) said in her welcoming remarks, the workshop would allow for discussion “of the impact of judicial decisions on agencies, lower courts, nonprofit organizations, and regulated communities, as well as Congress, and to consider anew the role of scientific information in agency decision-making.” Moreover, the workshop would contemplate how courts and judges would interpret laws, facts, and science, and how Loper Bright might change their approach.

In setting the context for the workshop, the co-chairs of the workshop planning committee, Richard A. Meserve (Covington & Burling LLP) and William B. Schultz (Zuckerman Spaeder LLP), noted that we are holding the workshop under a dramatically different policy environment than when workshop planning began in late 2024. As Meserve said, “Those who have defended the traditional role of administrative agencies see the greater

judicial scrutiny demanded by Loper Bright as a possible means to limit some of the new regulatory directions that the Trump Administration seeks to pursue.” Schultz observed that courts can generally still be expected to defer to the science involved in a decision, but “over the next day and a half we’ll explore whether our approach to regulation needs to fundamentally change in light of recent Supreme Court decisions.”

“Over the next day and a half, we’ll explore whether our approach to regulation needs to fundamentally change in light of recent Supreme Court decisions.” – William B. Schultz

THE NEW PRESIDENTIALISM

In the first of two keynote addresses at the workshop, Thomas W. Merrill (Columbia Law School), who was introduced by workshop planning committee member Barbara Schaal (Washington University in St. Louis), pointed out that, particularly since the Obama Administration’s second term, the Court has turned “sharply negative” toward the administrative state. In various opinions, justices have indicated sympathy for the nondelegation doctrine, which would require Congress to legislate with considerable specificity about the policies agencies should pursue. The Supreme Court has also used the major questions doctrine to invalidate novel agency initiatives designed to deal with, for example, the COVID19 pandemic and climate change.

“What the Court seems to believe is that administrative government is antidemocratic,” said Merrill. “Agency administrators are not elected; they’re appointed by the executive, and the Court seems to believe that critical policy decisions should be made not by agencies but by Congress.” However, this conception “ignores the contemporary reality that Congress is not likely to step into the breach and start dictating public policy in any detailed fashion,” he observed. “Maybe the Court is guilty of naivete, or as Justice Kagan has suggested, in many cases it’s engaged in a power grab because it realizes that once agencies can’t do these things, Congress will not do them either.”

At the same time, the Court has abetted what Merrill termed “the new presidentialism.” It has adopted rulings that enhance the power of the president and make it harder to subject presidential decisions to judicial review. It has looked favorably upon the idea of unitary executive theory, in which the Constitution’s vesting of executive power in “a” singular president is thought to give the president authority to control the decisions and personnel of the executive branch—for example, to remove at will members of independent commissions notwithstanding statutory limits on such removal power. Other historical developments, including the 1921 creation of the Bureau of the Budget (now the Office of Management and Budget), the 1974 Budget Act, and various statutes granting emergency powers to the president, have further bolstered presidential prerogatives. In the process, Merrill said, the presidency has transitioned from a position with “very few explicit powers” that are granted by the Constitution to one that “has unilateral and inherent authority to act with the force of law, without regard to whether, particularly in the domestic context but also internationally, there has been a specific delegation of authority.”

The problem with vesting authority entirely in the president and other elected officials is that “it ignores other dimensions on which we should judge government as being good or bad,” Merrill said. These include the resolution of disputes in a fair and impartial manner, adherence by agencies to the rule of law under the Administrative Procedure Act (APA), and the use of expertise in agency decision-making. “One can think of political power as water running downhill,” Merrill concluded. “If agencies are under attack, Congress is unwilling to take things up, and the courts are not institutionally capable of making all these decisions involving critical technical and regulatory issues, we’re left with the president, which means that we have unchecked and unbalanced presidentialism.”

In response to a question about whether other cases, including the Supreme Court’s decision in Skidmore v. Swift & Co.,¹ might provide support for courts giving weight to agency interpretations of statutes, Merrill said that longstanding and consistent interpretations of a statute have been recognized as deserving weight in legal determinations. However, the situation remains uncertain and will depend on future court decisions.

__________________

“If agencies are under attack, Congress is unwilling to take things up, and the courts are not institutionally capable of making all these decisions involving critical technical and regulatory issues, we’re left with the president, which means that we have unchecked and unbalanced presidentialism.” – Thomas W. Merrill

CHARACTERIZING EVIDENCE AND UNCERTAINTY

In the second keynote address, Jonathan Samet (University of Colorado) contextualized Merrill’s comments with several real-life examples. To prepare the 1964 report Smoking and Health: Report of the Advisory Committee to the Surgeon General of the Public Health Service, a committee of experts conducted a systematic analysis of more than 7,000 scientific papers. Their work led to legislation and regulations that have made the report the nation’s single most influential government report on human health, Samet said. Similarly, a series of reports produced by the National Research Council’s Committee on Research Priorities for Airborne Particulate Matter, which Samet chaired, were based on quantitative summaries of the available evidence.

The gathering and analysis of evidence is the initial step in the translation of science to policy, Samet said. This includes defining what is not known, which introduces uncertainty. However, policy does not result simply from balancing evidence and uncertainty. Many other factors come into play, including costs, politics, and pressure from activists. Critical issues include whether evidence is fit for purpose, the large and rapidly expanding size of the scientific literature, and efforts by interested parties to manufacture doubt.

Frameworks and processes have been developed to move from a question to policy actions, Samet observed. For example, systematic reviews and meta-analyses integrate evidence to assess such factors as causation, exposure-response relationships, and susceptibility. It is not possible to specify these frameworks and related procedures in statute, he said. Perhaps this is an area in which courts could play a role.

Characterizations of evidence and uncertainty are critical for policy making, Samet observed. While quantitative approaches are central to such characterizations, different people may interpret the same meta-analysis differently, and two meta-analyses on the same subject may produce different results. Samet recommended that expert judgment itself be studied for its role in characterizing evidence and uncertainty. Scientific evidence can be outweighed by other factors, including politics, he concluded, but “scientific evidence does not change when the administration changes.”

“Scientific evidence does not change when the administration changes.” – Jonathan Samet

EFFECTS ON FEDERAL AGENCIES

In the workshop’s first panel, which was led by William B. Schultz, Ann Carlson (University of California–Los Angeles School of Law) emphasized the continued importance of civil service lawyers and agency staff in the uncertain post-Chevron administrative environment. As acting administrator at the Biden-era National Highway and Traffic Safety Administration, Carlson oversaw the adjustment of fuel efficiency (CAFE) standards and the adoption of automatic emergency braking as implementation of “broad and not terribly lengthy statutory delegations of authority.” With CAFE standards, the agency uses voluminous data and complex cost analyses to achieve legislatively specified “maximum feasible standards.” With safety standards, it must assess technical feasibility, practicality, economic feasibility, and other objectives. Because such standards are often challenged in court, agency lawyers tend to be conservative in their interpretations, and they are likely to be even more so in the future, she said.

Recent Supreme Court decisions suggest that courts are going to be more skeptical about agency interpretations and actions, though the situation is still in flux, Carlson noted. A more robust application of the Skidmore decision may retain some deference for agency rulemaking, which argues for agencies to build strong and careful records of their actions. In addition, the deregulatory actions of the Trump Administration generally run counter to the intent of the original legislation, which may affect court decisions, although the loss of expertise within agencies will make it harder for them to continue to engage in complex rulemaking.

The manner in which the federal government approaches and then calculates risk, said former Environmental Protection Agency (EPA) assistant administrator and science advisor George Gray (The George Washington University

Milken Institute School of Public Health), has long been a focus of legal activity. Most EPA regulations take the form of quantitative standards, such as limits on the amount of chemicals in food or in the air. The laws guiding standard setting tend to be “capacious,” in that they require agency interpretations in both the assessment and management of risks. Relevant laws usually direct an agency to protect the public health with an “ample margin of safety” or reduce “unreasonable” risks to public health. However, there is no agreed-upon definition of an acceptable risk, Gray observed, and flexibility in risk assessment and management means that risks are managed at different levels in different situations.

“How far into the technical risk assessment are we going to go with judicial review?” Gray asked. “That’s where the action is—in actually coming up with the numbers that make a difference.” Answering that question requires distinguishing science, law, science policy, and actual policy, he insisted. “Let’s be clear what parts of these are being done with which set of judgments.”

“Congressional staff isn’t equipped to write prescriptive laws,” said Brian Ronholm (Consumer Reports) who served as a legislative staffer in the mid-2000s and later as deputy undersecretary for food safety at the U.S. Department of Agriculture (USDA). Using a menu labeling law written in the late stages of the 2010 Affordable Care Act and a provision affecting catfish inspection in the 2008 Farm Bill as examples, he observed that legislators often rely on agency expertise to interpret a law and make it functional. Similarly, as a political appointee at USDA, he acknowledged that experts “have forgotten a lot more of an issue than I’ll ever know” and that they are the ones who have the knowledge to make rules and policies workable.

In the case of less well-known regulations, such as those promulgated by the National Marine Fisheries Service (NMFS)—the source of the Loper Bright case—a relatively small number of worldwide experts have the knowledge to formulate policy, observed Andrew A. Rosenberg (University of New Hampshire Carsey School of Public Policy). When he worked at NMFS, his goal was to implement the most effective conservation strategy without risking losing his authority because of an action taken by a court, Congress, or a political appointee. He, too, predicted that recent Supreme Court decisions will cause agency lawyers to be more cautious. He also warned that the “new presidentialism,” along with a strong current of antagonism to expertise, might override traditional approaches to rulemaking.

Rosenberg asked “what will happen on the other side? These agencies will either have to be rebuilt, or we will have to do something else. I don’t think we’ll be rebuilding them in the form that they were in, and judges will have to make decisions based on the statutes that Congress has passed, most of which in the environmental field are quite old and probably won’t be revised, because everyone’s afraid to open them. But how will they make those decisions without agency expertise?”

The reduction in federal staffing levels during the initial months of the Trump administration adds uncertainty to the policy environment, said Patricia Zettler (The Ohio State University Moritz College of Law). In addition, the Loper Bright decision is just part of a broader movement within administrative law that will affect agencies in uncertain ways. For example, the Food and Drug Administration (FDA), which Zettler worked for as deputy general counsel at the Department of Health and Human Services during the Biden Administration, has to regulate new and rapidly changing technologies, which requires adapting to new circumstances. Furthermore, staffing shortages and changes in administrative law may slow down the agency’s regulation of emerging technologies, drugs, and health threats.

In response to a question about interactions between agencies and legislators, Carlson noted that agencies are fairly limited in their ability to interact with Congress, though other panelists noted that a strong congressional relations infrastructure can improve communications and that the situation differs among agencies. However, they also observed that the loss of agency personnel has made such communication more difficult, even within agencies.

Building on this point, the panel concluded with a discussion of the effects on government and on the public of reductions in the federal workforce. Rosenberg suggested the widespread narrative about government wastefulness “merits a long and careful strategic discussion about how you change the narrative about what government does.”

Carlson suggested that former federal employees could constitute “an emerging set of voices that we don’t typically hear from.”

“How far into the technical risk assessment are we going to go with judicial review? That’s where the action is—in actually coming up with the numbers that make a difference.” – George Gray

THE ROLE OF THE OFFICE OF INFORMATION AND REGULATORY AFFAIRS

One federal body directly affected by the Loper Bright decision, said workshop planning committee member Susan E. Dudley (The George Washington University Regulatory Studies Center, Trachtenberg School of Public Policy and Public Administration), is the Office of Information and Regulatory Affairs (OIRA), which was created by the Paperwork Reduction Act of 1980 under the Office of Management and Budget and has about 50 career staff. Dudley, herself a former OIRA administrator under George W. Bush, was joined by Paul J. Ray (The Heritage Foundation), who was OIRA administrator during the first Trump Administration, and Howard Shelanski (Davis Polk & Wardell LLP, Georgetown Law), who was OIRA administrator during the Obama Administration.

As Dudley described its functions, OIRA reviews agency regulations and the analysis supporting those regulations, coordinates among agencies’ overlapping and sometimes divergent interests in designing regulations and regulatory policy, and provides “accountability to the president” in overseeing regulation. Also, Shelanski observed, OIRA verifies the integrity and neutrality of data generated and used by statistical agencies of the federal government.

In response to a question from Dudley about how OIRA’s role might change, Shelanski reasoned that it may be less involved in agency interpretations of statutory authority going forward. Under Chevron, OIRA could ask agencies to strengthen the evidence and logic supporting an interpretation of a law, whereas Loper Bright has partly reassigned that statutory interpretation function to courts and to Congress. Nevertheless, Shelanski insisted that cost-benefit analyses carried out within the executive branch will continue to be important. Ray added that resource constraints and an absence of authorization in many cases are likely to limit agencies’ ability to do such analyses.

The three panelists also discussed the altered role of the courts in separating questions that can be answered on the basis of scientific expertise from those that depend on legal interpretations or policy matters. Ray said that courts do this in cases involving private parties and can appropriately do the same with federal agencies. Shelanski called the distinction between law and policy “fuzzy on the boundaries,” offering the case of coal dust concentrations and silicosis as an example. A statistical analysis of parts-per-million dust concentration and silicosis incidents is science, whereas what constitutes a “dangerous” level is a matter of policy. “Would you rather have courts overreach into facts and expertise, or have agencies overreach into legal interpretation?” he asked, adding that a preference for either varies from administration to administration.

OIRA’s traditional role has been to ensure the provision of information with which policy determinations can be made, but it also has been a vehicle through which the president’s policy positions filter into the regulatory process, Ray observed. In addition, OIRA has the task of ensuring that agencies are not working at cross purposes, which can slow down regulatory change. This will continue to be important in a new regulatory environment.

Shelanski and Dudley pointed to several recent reports that have called on Congress to identify ambiguities in its statutes and be clearer about its intentions. That will require additional staff and, perhaps, new congressional bodies, all three panelists said. The same observation applies to OIRA, given its continued importance in the regulatory process, said Ray. “Someone should say it on every panel, OIRA should have more staff.”

“Would you rather have courts overreach into facts and expertise, or have agencies overreach into legal interpretation?” – Paul J. Ray

EFFECTS ON REGULATED ORGANIZATIONS AND NONGOVERNMENTAL COMMUNITIES

Workshop planning committee member Ann Rosenthal (U.S. Department of Labor, retired) led a panel of three nongovernmental experts in anticipating the consequences of recent Supreme Court decisions for regulated organizations and nongovernmental communities. Panel members also discussed the initial actions of the Trump

Administration and its effects on both federal agencies and administrative law.

Dan Troy (Berkeley Research Group), former FDA Chief Counsel during the George W. Bush Administration and general counsel at GSK Pharmaceuticals, observed that industry has been preparing for the Loper Bright decision for a long time, which is likely to dampen its impact. He also noted that deregulation requires rulemaking, and the Trump Administration does not seem very interested in rulemaking. Political appointees can stop actions in federal agencies, but lasting change requires new rules.

Troy said that pharmaceutical companies are generally not interested in large-scale deregulation of their products, because they need regulatory approval to bring products to market. In general, the industry is guarded about dismantling existing protections based on allowances or language used by one administration. “I’m not going to do that,” said Troy, “because some new administration is going to come in, and then I’m going to be sitting on all this liability.”

David C. Vladeck (Georgetown Law School) agreed that the immediate consequences of Loper Bright will be muted. “I don’t think [it] is going to matter much,” he said. “It’s unfortunate. There was some value of Chevron in the lower courts.” However, he worried about a loss of expertise at agencies as a consequence of recent administration actions and a lack of enforcement of existing rules. “Chaos is not a good way to implement government. And chaos is now embedded in what this administration is doing.”

Speaking for an organization whose members “benefit day to day from government that’s informed by science and other expertise,” Matthew J. Ginsburg (American Federation of Labor and Congress of Industrial Organizations) proposed the possibility, mentioned earlier by Meserve, that the Loper Bright decision “may serve as something of a judicial check on some extreme executive branch action.” In that sense, recent Supreme Court decisions could produce a more stable regulatory environment. However, like Vladeck, he pointed to the importance of keeping the basic structures of government afloat “so that, in the future, the infrastructure is still there.”

Panel members agreed that courts are key in keeping the administrative state alive and functional as well as in check. This is a role they have played in the past, noted Rosenthal. Even under Chevron, courts “did not reflexively defer to administrative agencies” but rather looked at the overall interpretation of a given statute. In that respect, “Loper Bright is going to make agencies stick more closely to what they’re clearly authorized to do,” predicted Troy. But scientific information can still be expected to guide their decisions. “Scientists are not lawyers,” he said, “but neither do I think lawyers, especially judges, want to be scientists.”

“Scientists are not lawyers, but neither do I think lawyers, especially judges, want to be scientists.” – Dan Troy

IMPACT ON THE LOWER COURTS

After three panels addressed the implications of recent Supreme Court decisions for agencies, workshop planning committee member Diane Wood (The American Law Institute, University of Chicago Law School) moderated a panel centered on potential effects on courts. David J. Barron (U.S. Court of Appeals for the First Circuit) began by breaking down “what we mean by deference from the perspective of judges.” He presented the case study of National Labor Relations Board v. Hearst Publications, Inc., a 1944 court case that asked whether newsboys were “employees” of a newspaper company. Before Chevron, said Barron, the definition of what an “employee” is could be a “pure law” question, with no deference to agencies. The definition of “newsboy,” in contrast, would be deferred to agencies based on the facts of their hours, wages, and role. Chevron, when in place, would have caused the definition of an employee to defer to agency expertise. Loper Bright returned these pure law definitions to the jurisdiction of the courts, separating them from questions of fact. That said, decisions that mix factual and legal determinations in complicated ways—such as whether a newsboy is an employee—will be more difficult to make, he added.

Retired U.S. Court of Appeals for the District of Columbia Circuit judge David S. Tatel (Hogan Lovells), who said that he heard more Chevron cases during his three decades on the court than any other category of cases, observed that the majority of these cases were determined under what he called Chevron 1: “If the statute is

clear, that ends the matter.” If a statute is ambiguous, the case proceeds to Chevron 2, in which the reasonableness of an agency’s interpretation is determined.² “Chevron wasn’t perfect,” said Tatel, “but it did a pretty good job of protecting agency expertise under a statute from judicial interference. It also preserved the ability of the courts to rein the agencies in when they exceeded their jurisdiction.”

Neomi J. Rao (U.S. Court of Appeals for the District of Columbia Circuit) said that it is usually possible to find the “best meaning” of a statute, even if that meaning is not immediately apparent. A court ultimately judges a statute based on what Congress did. However, changes resulting from the democratic process can alter the way that statutes are administered. “For many statutes, there’s a range of approaches that an agency can take, and I do think it’s appropriate that with elections, agencies may take different approaches to administering a statute, as long as those approaches are within the meaning of the statute, then that’s permissible.”

Legislation can have a singular best interpretation, Tatel added, but more often it has a range of reasonable interpretations. After Loper Bright, the way that these interpretations are derived—either by the courts or by the agencies—will depend significantly on the judiciary’s continued commitment to Skidmore deference. If longstanding and consistent interpretations continue to command respect, “the world might look very similar to the way it did before,” said Tatel. However, he thought that unlikely to happen, and “that’s where a lot of the battles are going to be.”

Wood specifically asked the panelists about Loper Bright’s relation to both the major questions doctrine³ and the nondelegation doctrine.⁴ Tatel expressed concerns about the nondelegation doctrine, because “the modern Congress is probably incapable of legislating with the specificity the nondelegation doctrine would require.” Rao agreed that Congress likely lacks the will to legislate in a technical manner but suggested that courts’ lack of enforcement of the nondelegation doctrine has enabled Congress to dodge important questions. She pointed to the date of the last major Clean Air Act Amendment in 1990. “Think about how much the science and political thinking around climate change has changed since then,” she said. “Congress has said nothing about it.” Congressional silence on major issues, said Rao, is a problem for the United States’ constitutional democracy. Tatel added that, throughout his career, judges often wondered why Congress had not resolved some of the issues that advanced to the court. Courts were “forced to decide case after case where we would feel much better if the elected Congress was deciding this question, not us.” At the same time, Tatel complimented the expertise of executive branch agencies. “The agencies aren’t perfect,” he said, but “I left with a profound respect for the quality of the work I saw.”

As Meserve pointed out in the question-and-answer session, the nondelegation doctrine and mixed questions of law and science are pertinent issues when Congress decides to delegate authority to agencies. For example, when Congress passed the Atomic Energy Act of 1954, the legislative branch “had no idea what the regulatory regime should be, and recognized that,” he said. Broad authority was therefore delegated to the agency, first to identify the technical problems and then to create a regulatory regime that protected public health and safety. “Congress would never have been able to define the bounds when it initially passed the statute.”

“Chevron wasn’t perfect, but it did a pretty good job of protecting agency expertise under a statute from judicial interference. It also preserved the ability of the courts to rein

__________________

the agencies in when they exceeded their jurisdiction.” – David S. Tatel

ENVIRONMENTAL LAW IN THE NEW REGULATORY CLIMATE

The second day’s opening session, moderated by Meserve, focused on the possible effects of recent Supreme Court decisions on environmental regulation, the field from which Loper Bright—and Chevron deference—stemmed.

Paul Noe (American Forest & Paper Association) expressed enthusiasm for Loper Bright, saying, “It invites an evidence-based regulatory system grounded in benefit-cost analysis. I think that provides what our country needs, which is more influence by science, more influence by engineering and technical experts.” The benefit–cost default principle established by Michigan v. EPA (2015), said Noe, informs the “best reading” of a statute and “reasoned decision making” required under Loper Bright unless prohibited by clear statutory language. He called benefit-cost analysis “the only form of analysis that considers all welfare effects of regulation.” Other forms of regulatory analysis (such as feasibility analysis, cost-effectiveness analysis, etc.) are “fundamentally flawed” in that they either do not consider all welfare effects of regulation or may be “ad hoc and intuitive” balancing and therefore “bound to have major errors.”

David Doniger (Natural Resources Defense Council), a former EPA attorney, pushed back against this reading of Michigan v. EPA, pointing to language that leaves interpretation open to agency determinations. Doniger also observed that EPA Administrator Lee Zeldin recently announced his intention to cut EPA staff and budget significantly and his goal of repealing or weakening 31 EPA standards, including the 2009 endangerment finding for carbon dioxide, which found that atmospheric carbon dioxide endangers public health and the environment. However, the endangerment finding and other EPA standards are on the books, Doniger said, and can only be taken off the books through the same procedures that created them—rulemaking with detailed legal and scientific records that justify any decisions made. Reductions in the agency’s workforce and resources will inevitably hinder this planned deregulation.

As someone who represents industry clients, Jeff Holmstead (Bracewell LLP), a former EPA assistant administrator during the George W. Bush Administration, expected that even if environmental enforcement was publicly waived, his clients would continue to comply with regulations “for a variety of practical reasons.” Holmstead predicted that the Loper Bright decision “will constrain both aggressively pro-environmental administrations and aggressively deregulatory administrations, like the Trump Administration.” He provided support for this point using the same evidence as Doniger, the endangerment finding for carbon dioxide, noting that both for mobile and stationary sources the statute “has always been interpreted by EPA to be purely a scientific issue with no consideration of cost.” A recent EPA press release⁵ seemed to premise endangerment finding rollbacks on downstream cost analyses. Under Loper Bright, said Holmstead, “I can’t imagine how that can possibly be the best interpretation of the statute. I think they have very little chance of reversing the endangerment finding in a way that will stand up in court.” Doniger added that cost analysis does come into effect, separate from the endangerment finding, under sections 111 and 202 of the Clean Air Act.

Several existing environmental statutes, said Noe, inhabit a middle ground on benefit–cost balancing, neither expressly requiring nor prohibiting it. In Noe’s view, based on Supreme Court Associate Justice Scalia’s written decisions in both Michigan v. EPA and Entergy Corp v. River-keeper, Inc. (2009), “there’s room to interpret a vast swath of statutes as requiring benefit–cost balancing” under the Loper Bright framework. Holmstead said that the language on endangerment findings “makes it quite clear that it’s entirely a scientific public health finding,” though EPA will consider cost when deciding how to regulate.

In the Loper Bright decision, the Supreme Court maintained that environmental decisions decided in the era of Chevron deference would remain valid. But the panelists wondered whether courts might revisit regulations in the new legal context. Noe said, “Agencies could revisit old regulations if they believe the regulation was not the product of the best reading.” Doniger, however, thought this to be less likely, particularly because industry and industry-associated groups “want stability.” In terms of the endangerment findings rollback, he said, “It’s not the companies, primarily, who are agitating to revisit this. It’s the ideologues.”

__________________

Holmstead supported this argument. “Not only is industry not supporting reversing the endangerment finding,” he said, “I know of a lot of business groups who have been lobbying against it.” Companies have invested capital to come into compliance with existing regulations, and “they’re not looking to have those rules undone.” He used the Mercury and Air Toxics Standards (MATS) as one example. “Once the companies got in compliance, they didn’t want the rule to go away. It would create all sorts of problems.”

The Loper Bright decision “will constrain both aggressively pro-environmental administrations and aggressively deregulatory administrations.” – Jeff Holmstead

UPCOMING SUPREME COURT DECISIONS

Christopher J. Walker (University of Michigan Law School), an administrative law expert with experience in the Department of Justice, the American Bar Association, and the Senate Judiciary Committee, joined workshop planning committee member Diane Wood to discuss three recent and forthcoming Supreme Court decisions that shed light on the nondelegation doctrine, the major decisions doctrine, and agency statutory interpretation after Loper Bright.

The first case, Federal Communications Commission v. Consumers’ Research, concerns the Universal Service Fund (USF). Telecom companies pay money to the USF in order for the FCC to ensure adequate internet service in rural and poor areas, libraries, hospitals, and schools. These USF fees are set by a private entity created by the FCC. A consumer research group said that the FCC’s action in setting up this system was a violation of the nondelegation doctrine. Walker used the case—which was argued in March 2025—to contend that the Supreme Court should ultimately embrace a proportionality approach to the nondelegation doctrine. When approaching small questions, agencies have broad authority for implementation details. At the other end of the scale, Congress cannot delegate away the most major value judgments—for instance, allowing an agency to criminalize conduct. For questions of varying importance in between the extremes, Congress must provide additional guidance and guardrails proportional to the importance of the delegated authority. Walker expected the USF to be upheld against nondelegation doctrine challenge but wondered if opinions from the judges might “articulate a more administrable approach to nondelegation.”

Wood responded that “it’s very hard to articulate exactly how much detail Congress needs to put in any given statute.” She thought the “granularity of guidance” was an issue best left for Congress to decide. Walker compared this idea of legislative “granularity” to the work of legal scholar John Hart Ely, who argued that Congress should remain involved in questions entailing major value judgments. “Congress needs to be in the game here,” said Walker. “We shouldn’t allow presidents to come in and flip the light switch on and off on a major issue, like student loan relief, or nationwide eviction moratoriums, or tariffs.”

Walker’s second case was Nuclear Regulatory Commission v. Texas. After Nevada Senator Harry Reid blocked the establishment of a permanent nuclear waste storage facility in Yucca Mountain, a number of private entities stepped forward to store the waste at “interim storage facilities” in Texas. The U.S. Court of Appeals for the Fifth Circuit held that their offer violated the major questions doctrine. “Some people have said the major question doctrine is dead after Loper Bright,” said Walker. “I think that’s wrong. The major questions doctrine is still going to play an important role. In fact, I think it’s going to be what the Court uses, instead of the nondelegation doctrine, when it wants to [constrain] agency authority on big value judgment questions.” In general, Walker said that the major questions doctrine will be a factor when a president tries to use old statutes to do new things. “It’s deeply problematic when a new president comes to town and wants to do something and looks around and says, ‘Congress, will you work with me?’ And Congress says, ‘Absolutely not.’ So, instead, they issue an executive order or otherwise direct agencies to find some old statute that will allow them to do something they want to do.”

Walker’s final case, mentioned frequently on the workshop’s first day, was FDA v. Wages and White Lion Investments, LLC (2025), in which the Supreme Court sided unanimously with an agency decision to block some flavored vaping products going to market. Walker noted that the case gave birth to the “agency change in position doctrine,” which had been used before but not named as such. He summarized the doctrine: “What an agency has

to do when they’re changing their position is, one, recognize they’re changing their position, . . . two, provide a reasoned decision for the new policy, and three, consider the reliant interests and reasonable alternatives that would be involved. What they don’t have to do is say that the new agency policy is better than the old one.” Nevertheless, changing positions should entail thorough review of prior agency rulings and factual records, Walker said. “The agency has to consider all significant aspects of the problem, which includes grappling with science and reality,” he said. “The Court is very concerned about agencies changing their positions based on politics alone.”

Wood thought that the White Lion case exemplified the Supreme Court’s willingness to accept agency-based factual data without citing Loper Bright. “What you don’t see in this case is any particular discussion about how you treat agency expertise.” She said that the Supreme Court seemed “comfortable relying on the factual record, the scientific record, that the agency compiled.” Walker agreed and added that the case seemed to allow for agency flexibility based on new realities and facts becoming available, as vaping products constituted a novel regulatory target. However, he cautioned against agencies regulating other emerging technologies, such as biotechnology and cryptocurrencies, with broad, open-ended statutes. “I don’t feel so comfortable about that, and I don’t think the Supreme Court would either,” he said. “As a public, we have to re-envision Congress’s role of being the legislative branch—the branch that makes the major value judgments.”

Assessing Loper Bright’s ultimate role in the current deregulatory environment, Walker said that, based on an exercise he ran with data from “every Chevron decision in the lower circuit courts,” the period during which Chevron was in effect tended to correspond with judges being removed from their policy preferences. Without Chevron, Walker said, “You’re loading the die for whatever the politics of the lower court are at the time, which, right now, are deregulatory.”

“As a public, we have to re-envision Congress’s role of being the legislative branch—the branch that makes the major value judgments.” – Christopher J. Walker

ADDRESSING CONGRESS’S LEGISLATIVE CAPACITY

“This can be thought of as a climactic moment of the conference,” said workshop planning committee member David Goldston (Massachusetts Institute of Technology Washington Office), a former chief of staff for the House Committee on Science, when introducing the workshop’s final panel. “The last day and a half, we’ve been hearing, ‘If only Congress did x, y, and z.’ Now we get to hear from people who have worked there.” Goldston asked a panel of experienced legislative staffers how the legislative branch might regain authority and how recent Supreme Court cases could affect Congress’s interactions with the administrative state.

Philip S. Barnett (Co-Equal), having served as staff on the House Energy and Commerce Committee and the Oversight and Government Reform Committee, noted that recent court cases, including Loper Bright, interpreted statutes, not the Constitution, “so this is something Congress has the power to deal with.” It could amend the APA and judicial review provisions, as in last year’s proposed Stop Corporate Capture Act, which would have effectively reinstituted Chevron by requiring a court to “defer to the agency’s reasonable, permissible interpretation” when encountering statutory ambiguity. Other legislation, such as a bill recently introduced by Senator Katie Britt, would amend in the opposite direction, codifying Loper Bright into the APA. Congress could be clearer about delegating authority, said Barnett, or it could address in specific laws the courts’ ability to do judicial review. “Congress has these tools but is unable to act, where does that leave us?” Barnett asked.

Elizabeth L. Horner (ArentFox Schiff LLP) former chief counsel of the Senate Committee on Environment and Public Works, recognized a tension in “finding the single best interpretation in Loper Bright and what tools the Court has to evaluate that.” For a variety of reasons, some recent, commonly used methods of legislating provide limited information to courts in evaluating changing statutes. Legislative provisions are often attached to larger pieces of moving legislation such as an appropriations bill or the National Defense Authorization Act; negotiated changes are developed quickly in consultation with jurisdictional committees; and the so-called Byrd Rule of the Congressional Budget Act limits inclusion of policy details in reconciliation legislation. To develop more detailed bipartisan legislation, Horner argued, staffers need to have trust in each other and a willingness to try new approaches. “When I started, we thought,

‘We can never amend that statute [the Clean Air Act], because if you open it up, then the whole house of cards is going to fall.’” In her work on amending section 103 of the Clean Air Act to address carbon capture research, for example, “there was a middle ground” reached in bipartisan discussions. “Legislation is such a personal process. How do we, as a country and as a society, promote that internally? That’s the only way these bills get done,” Horner added.

The term “ambiguity” had not been defined during the workshop, said Heidi R. King (Heidi R. King Analytic Consulting), former chief economist, House Energy and Commerce Committee, offering her own interpretation as “text that conflicts with itself.” As an example, King used the term “safe,” which is common in health-oriented legislation. “The word has an intuitive meaning to everyday people. But a statistician or a scientist finds that very difficult to implement.” Sometimes the staffers drafting legislation may not have had experience in fields outside of those in which they trained and worked professionally, so they may not recognize the limits of terms and concepts when drafting text. She suggested introducing more science, technology, engineering, and mathematics (STEM) resources into Congress—for example, adding STEM staff in the Office of the Legislative Counsel of the U.S. House of Representatives—to reduce the ambiguities in legislation, saying, “There are elements of thought missing in Congress. There is an insufficient staffing of science and economic issues, and we don’t have transparency in legislative development,” King said.

Looking at the last century of Congressional history, said Dane Stangler (Bipartisan Policy Center), “I don’t think we ever had a period where Congress was operating as an ideal institution.” In the 1950s and 1960s, Congress swung toward powerful committee chairpersons, creating dissatisfaction that led it to swing in the opposite direction in the 1970s. “Congress has always been a pendulum.” The Bipartisan Policy Center has crafted guidelines for modernization of Congress that could help overcome some of the current polarization and gridlock. “We’re optimistic, perhaps unreasonably, that at some point, whether it’s a constitutional crisis precipitating it or some other external force, the institution will engage in some of this modernization.”

The panel discussed a recent tool used by Congress to interact with agencies, the Congressional Review Act (CRA), which Horner likened to “a blunt force instrument.” Little utilized before 2017, the CRA can be used by Congress to question agency actions and has been called upon more frequently since the first Trump term. Goldston described the CRA as “Congress’s expedited, unblockable way to say, ‘No, we don’t like that regulation.’” Barnett thought that the CRA had its greatest impacts around changes in administrations, when partisan policies are likely to be questioned by incoming opponents. King called it more of an “escape valve” than a long-term fix for Congressional legislative capacity.

To strengthen Congress’s legislative capacity, Stangler, like King, proposed allocating more resources to Congress. Legislative support entities such as the Congressional Research Service (CRS) and the Government Accountability Office (GAO) have seen decreased funding over the last 30 years, and the Office of Technology Assessment (OTA) has not been funded since the 1990s. Goldston asked, however, if increased staffing would solve problems that can also be attributed to political polarization. “More of the same is probably not the solution to the problem,” admitted King, but adding more STEM-specific skills and information to Congress could help reduce the use of ambiguous terms resulting from a lack of information.

Barnett supported increasing congressional staff and resources but noted that more resources and staffing are not a cure-all. “I’m a believer in Congress,” he said, “and I think we’ll figure out ways through this. But for all that stuff to work, you also have to have people recognizing there’s a problem and wanting to solve it. And unfortunately, on some of the big issues today, that dynamic sometimes is missing.”

Asked by Goldston to sum up their takeaway messages from the panel, Horner pointed to the need for stakeholders to provide more detailed input, including on technical issues, to legislators, which would “perhaps leave fewer things for courts to evaluate down the road.” King noted that current actions by the president emphasize the importance of strengthening legislative institutions so that ambiguities in legislation are less of a problem 20 years from now. Stangler was impressed by the

high levels of concern and thought being devoted to the formulation and review of regulatory policy. Barnett recommended that an expert and distinguished body could provide input on how judicial review provisions should be restructured. For his part, Goldston proposed establishing greater demarcations between science questions and policy questions. “Obviously, sometimes that’s a blurry line, but I think those things are getting more and more conflated.” He also pointed to artificial intelligence as a classic example of how new issues have to be addressed either with existing statutes or with new legislation.

“Legislation is such a personal process. How do we, as a country and as a society, promote that internally?” – Elizabeth L. Horner

CONCLUDING REMARKS

To conclude the workshop, members of the workshop planning committee briefly summarized their own takeaway messages. Wood observed that “the scientific questions are not going to go away, whether it’s questions of environment or food safety or pharmaceuticals.” She was heartened, she said, to hear about places where stability exists and about “open pathways” to progress.

Goldston was struck by the uncertainty in the legal environment after the Supreme Court’s recent decisions, “which is probably a positive thing.” He noted that the scientific uncertainty around large questions frequently matches uncertainty in the law, rarely culminating in one agreed upon answer. “The ambiguity is legal and political and scientific,” he concluded. “The way all those things mix creates the situation we’re trying to figure out.”

The legislative branch, observed Rosenthal, is unlikely to provide satisfactory answers with the world changing at such a terrific rate. “Even in an ideal Congress, we would not be creating new legislation fast enough to deal with climate change, with viruses, with various other things.” She was encouraged to learn that the major questions doctrine may be used more as an interpretive tool than as a way to radically shift power among branches of government.

Dudley pointed out that observers in the 1980s could not have predicted the consequences of the Chevron decision, and the same observation applies to recent Supreme Court rulings. She left the workshop hoping the courts would provide a bulwark against Merrill’s “new presidentialism,” as well as the “wild swings in policy between administrations.” Agencies can help by distinguishing questions of law from questions of fact and science from questions of policy.

According to Meserve, the workshop underscored the present as “a time of great turbulence,” particularly given the concerns raised by recent Supreme Court decisions. After the last day and a half, he said, “I come away with a much greater sense of the uncertainty about where we are headed.” Nevertheless, the final panel left him hopeful that a pathway “through this storm” exists via Congress.

Schultz identified two administrative law tracks that seem to be playing out in tandem. The first concerns how much power the president has to shutter agencies, fire experts, and dismiss employees. “These raise fundamental questions of constitutional law. When Congress creates an agency, is it up to Congress or the president to dismantle it? When Congress funds an agency, is it up to the president not to fund it?” The second track involves the Supreme Court decisions and their lasting effects. “I, for one, can live with the idea of judges having the primary role in deciding questions of law.” But Schultz worried about the use of the major questions doctrine and the nondelegation doctrine, particularly in the lower courts. “The most important thing,” he said, “is science.” The recent mifepristone and vaping cases⁶ gave Schultz hope that agency deference would hold on matters concerning broad standards related to public health. However, “courts second guessing the agencies on science—and I think we’re going to see the lower courts doing this somewhat, relying incorrectly on Loper Bright—is a Rubicon that hopefully we won’t cross.”

As CSTL co-chair Harold Varmus (Weill Cornell Medicine) said, “We did not imagine the chaos that would follow the election of 2024: the paralysis of Congress, the increased dependence on the judicial branch, the terri-

__________________

fying attacks within the executive branch, with elements like DOGE⁷ destroying agencies that are science dependent to the point where they might be unable to provide expert opinion to the courts,” he said. The workshop “is an appropriate reappraisal of the past, but I remain uncertain about whether it will be a preparation for a calmer future.”

__________________

DISCLAIMER This Proceedings of a Workshop—in Brief was prepared by Eric Olson, Steve Olson, and Anne-Marie Mazza as a factual summary of what occurred at the workshop. The statements made are those of the rapporteurs or individual workshop participants and do not necessarily represent the views of all workshop participants; the planning committee; or the National Academies of Sciences, Engineering, and Medicine.

PLANNING COMMITTEE Richard A. Meserve (NAE), Covington & Burling LLP; William Schultz, Zuckerman Spaeder LLP; Susan E. Dudley, George Washington University; David Goldston, Massachusetts Institute of Technology; Adam Jaffe, Brandeis University; Goodwin Liu, Supreme Court of California; Ann Rosenthal, Occupational Safety and Health Administration (ret.); Barbara Schaal (NAS), Washington University in St. Louis; Daniel Walters, Texas A&M University School of Law; and Diane Wood, The American Law Institute and University of Chicago Law School

REVIEWERS To ensure that it meets institutional standards for quality and objectivity, this Proceedings of a Workshop—in Brief was reviewed by Gillian Metzger, Columbia University; Richard Murray (NAE), California Institute of Technology; Richard Revesz, New York University; and Lexi Shultz, American Geophysical Union. Marilyn Baker, National Academies of Sciences, Engineering, and Medicine, served as the review coordinator.

SPONSORS This activity was supported by grants to the National Academy of Sciences from the Rita Allen Foundation, the Burroughs Wellcome Fund (grant number 1439600), the Kavli Foundation, and the Gordon and Betty Moore Foundation (grant number GBMF12999). Any opinions, findings, conclusions, or recommendations expressed in this publication do not necessarily reflect the views of any organization or agency that provided support for the project.

STAFF Anne-Marie Mazza, Senior Director; Steven Kendall, Senior Program Officer; Renee Daly, Senior Program Assistant (until April 2025)

Suggested citation: National Academies of Sciences, Engineering, and Medicine. 2025. Implications of Law, Policy, and Federal Agency Decision-Making Under a New Judicial Standard: Proceedings of a Workshop in Brief. Washington, DC: National Academies Press. https://doi.org/10.17226/29169.

For additional information regarding the workshop, visit https://www.nationalacademies.org/our-work/implications-of-recent-supreme-court-decisions-for-agency-decision-making-a-workshop.

Policy and Global Affairs Copyright 2025 by the National Academy of Sciences. All rights reserved.