An energetic president can promote steady administration of the laws through his regulatory agenda in several ways.

The Constitution gives the president great powers but also great obligations. Perhaps the most basic of these obligations is his duty to"take Care that the Laws be faithfully executed."1 The work of executing laws was challenging enough in George Washington's time. It is exponentially more challenging today, given the intervening accumulation of federal statutes and especially the past century's countless laws authorizing or commanding federal agencies to achieve innumerable federal policies.

Today's"administrative state wields vast power and touches almost every aspect of daily life," Chief Justice John Roberts wrote a few years ago."The Framers could hardly have envisioned today's vast and varied federal bureaucracy and the authority administrative agencies now hold over our economic, social, and political activities," he added."The administrative state with its reams of regulations would leave them rubbing their eyes."2

Fortunately, the framers' Constitution and the laws Congress passes give presidents tools for carrying out the weighty task of administering federal regulatory statutes. But to succeed in carrying out this part of his constitutional work, a president must draw on what Alexander Hamilton famously described as"Energy in the Executive"-an energy that is "essential to the steady administration of the laws."3

An energetic president can promote steady administration of the laws through his regulatory agenda in several ways. The president's executive orders should give his agencies clear policy guidance to channel their policymaking discretion toward specific goals. He should direct agencies to improve their policymaking efforts in terms of analytical rigor, factual accuracy, and legal viability. Finally, he should direct agencies to approach policymaking with an eye on the long view, recognizing that steady administration requires a time horizon longer than just a few years and thus longer than just one presidency

Seize the Initiative to Focus Agencies' Policymaking Discretion

Although the president and his agencies are tasked with executing the statutes that Congress has written, those statutes rarely specify what precise policy Congress wanted to be executed. Instead, as influential judge Henry Friendly once wrote, Congress too often legislates"a mood rather than a message."4 Because statutes leave such broad discretion to presidents and agencies, they create vast policymaking opportunities for each chief executive. However, using such opportunities threatens to undermine the principle of steady administration because regulation, like a pendulum, can swing wildly in its application from one administration to the next. Worse still, the panoply of statutes instructs agencies to vindicate general principles in conflict with one another, increasing the risk that agencies will work at cross-purposes over time and across the government.

Presidents have always directed or at least influenced agencies in the exercise of statutory discretion, formally or informally, by coordinating the agencies and charting an overall policy agenda for the administration. In recent decades the White House's central role in coordination has become much more formalized and regularized, as Elena Kagan described in her seminal article on"presidential administration" in 2001.5 Since then, the past two administrations have further formalized the president's leadership role by using executive orders that direct groups of agencies to exercise their statutory discretion in certain ways, to the greatest extent allowable by statute.

The use of executive orders, of course, has legal limits: A president cannot enforce executive orders that violate the Constitution or lawful statutes. But when the Constitution and statutes leave agencies with a wide field of discretion on a subject, the president can use executive orders effectively to narrow his agencies' range toward a particular policy outcome. President Barack Obama, for instance, achieved this through his executive order directing agencies to increase their support for certain forms of stem cell research. President Donald Trump followed that example and significantly increased the practice through overarching executive orders that announced the administration's core principles on financial policy,6 energy and environmental policy,7 health care,8 and other subjects.

This approach, when kept within its proper constitutional and statutory limits, promotes several principles of sound constitutional administration.9 First, it promotes political accountability by signaling to the people, political appointees, and civil servants that responsibility for a given policy traces all the way back to the president himself. ("The buck stops here.") Second, it promotes steadiness by pointing agencies toward more focused and better defined policy goals instead of leaving them to range among the arrays of potential goals left open to them by a broad statute. Third, it reduces regulatory uncertainty by telegraphing to the public in more certain terms, earlier in the administration, what policies the agencies will consider pursuing. The sooner a president's administration can telegraph what general policies it intends to pursue-once it has decided on them-the better.

In charting its overall policy agenda and coordinating and directing its agencies toward those ends, an administration will need to decide how much power to assert over the so-called"independent" regulatory commissions. For well over a century, Congress has created multimember regulatory agencies that enjoy at least some measure of nominal independence by limiting the president's control over the agencies' leadership. Usually this independence is embodied by statutes that expressly limit the president's power to fire agency heads in cases of, for example,"inefficiency,""neglect of duty," or"malfeasance."10 The precise meaning of those statutes has never been squarely adjudicated, and they might allow for the president to fire a commission member over policy disputes. But in general, they are taken as affording commission members some degree of independence from presidential control.11

Thus, even presidents who have asserted strong central control over an administration's agencies have generally stopped short of asserting similar power over independent agencies. The Ronald Reagan administration, for example, established the modern framework for centralized regulatory oversight with the White House's Office of Information and Regulatory Affairs (OIRA) yet excluded independent agencies from OIRA's oversight-not because the White House doubted the president's constitutional power to oversee those agencies but for prudential political reasons.12

Each new administration must strike its own prudential balance regarding White House oversight of traditionally independent regulatory commissions. While independent regulatory commissions' responsibilities might have been relatively sleepy in 1981, today many independent agencies are responsible for matters at the heart of the most consequential regulatory policies; for example, the Federal Energy Regulatory Commission's responsibilities for wholesale markets significantly influence not just energy markets but also environmental policy generally and climate policy specifically. The powers entrusted to the Federal Reserve Board of Governors, the Securities and Exchange Commission, and other financial regulators overlap significantly with the responsibilities of purely executive (that is, nonindependent) agencies such as the Treasury Department and the Consumer Financial Protection Bureau. Thus, any administration must decide not just whether to include independent agencies in its informal policy-coordination efforts but also whether to keep exempting independent agencies from the coverage of executive order 12866 and other regulatory-coordination orders.

Given the significance of those independent agencies' policymaking discretion and the impact of their policies on an administration's overall policy efforts, presidents should continue to narrow and eventually eliminate the independent regulatory commissions' sweeping exemptions from presidential oversight. This does not mean independent commissions need to be treated like executive agencies in every respect. Rather, the White House can tailor its oversight of each independent commission to suit the particular policies and sensitivities at issue for each agency, just as President Trump's OIRA agreed to carefully tailor its oversight of IRS regulations to befit the obvious political sensitivities of presidential involvement with tax enforcement.13

Bridget Dooling, a former OIRA official now at George Washington University's Regulatory Studies Center, aptly describes such an agency-specific approach as"bespoke regulatory review."14 The sooner an administration asserts some form of oversight over independent regulatory commissions' policymaking power, the better; after all, those agencies' assertions of power to execute laws must ultimately be aligned with the Constitution's vesting of the"executive power" and"take care" responsibilities to the president alone and with the people's own judgments as expressed through the election of the president.

Challenge Agencies to Improve Their Policymaking Analysis

The institutionalization of cost-benefit analysis for agencies' regulatory proposals is one of the most significant developments in modern regulatory history. It improved the quality of agency regulatory analysis and created an institutional counterweight to the administrative state's largely unchecked growth.

This was not an overnight achievement, as Andrew Rudalevige has explained in an essay for National Affairs. 15 Rather, it was a more organic and incremental process, beginning in the late 1960s through the 1970s, before being codified by the Paperwork Reduction Act's creation of OIRA in 1980 and President Reagan's executive order 12291 in 1981. The process of institutionalization and improvement continued through the Bill Clinton, George W. Bush, and Obama administrations. Today, OIRA's role is governed primarily by President Clinton's executive order 12866, supplemented most recently by President Trump's executive order 13771 on regulatory budgeting. As Rudalevige describes, OIRA's success required more than just the issuance of executive orders; it required the sustained investment of presidential political capital and a commitment to seeing the project work not just in theory but in practice.

Articles by two of the most recent OIRA administrators detail its central importance to a well-functioning executive branch. Cass Sunstein, who administered OIRA for President Obama, emphasizes OIRA's role in "oversee[ing] a genuinely interagency process, involving many specialists throughout the federal government" and considering not just regulatory cost-benefit analyses but also questions of law and policy that implicate more than just a single agency's interests and expertise.16 Susan Dudley, who administered OIRA for President Bush, writes that the"durability" of OIRA's procedures and principles reflects its invaluable role in"giv[ing] the democratically elected president some control over the diverse agencies that comprise the executive branch" and providing much-needed continuity in the technical expertise necessary for effective White House oversight of agencies from one administration to the next.17

The best thing a president can do to preserve OIRA's capacity to play these crucial roles is to expand its fiscal and personnel resources. OIRA has often suffered shortages of both. The president should also build on the OIRA framework by expanding its mission and the corresponding duties of agencies that OIRA oversees in two key ways above all.

The first essential priority is retrospective review. Executive order 12866 and scattered statutes require many federal agencies to undertake a cost-benefit analysis of proposed regulations before they are finalized, but there is no similarly broad mandate to reexamine a rule's costs and benefits after that rule is finalized. The Administrative Conference of the United States, the American Bar Association's Section of Administrative Law and Regulatory Practice, and other expert bodies and analysts have long called on agencies to undertake"retrospective" cost-benefit reviews to decide whether to repeal old rules.18 There is merit to that suggestion, but the more compelling reason for requiring agencies to regularly subject old rules to retrospective review is the process will improve the agencies' analyses going forward. 19

As former OIRA Administrator Dudley described in a 2012 study, federal agencies' cost-benefit analyses are habitually erroneous: Their forecasts tend to exaggerate the eventual benefits of rules,20 and, one suspects, agencies similarly tend to underestimate eventual costs.21

As Philip Tetlock and Dan Gardner explain in Superforecasting: The Art and Science of Prediction, the key is to ensure that forecasters are held meaningfully accountable by tracking their predictive accuracy over time.22 To that end, regularly required retrospective review would force agencies to confront their own past errors in cost-benefit analysis, and that exercise would help them avoid making such mistakes in the future. The president should require such a practice for all agencies that do cost-benefit analysis through an executive order that expands OIRA's mandate to cover retrospective review.

The second essential priority involves agency adjudication. Agencies are generally free to make policy through either regulations or case-by-case adjudication of regulatory disputes.23 But the OIRA framework for regulatory review covers only regulations, not adjudication. This creates a major imbalance in regulatory oversight and an incentive for agencies to prefer adjudication, rather than the much more transparent and participatory rulemaking process, as the main vehicle for policymaking.

Of course, there is good reason for treating agency adjudication somewhat differently from rulemaking in terms of White House oversight. The case-specific nature of agency adjudication, as embodied by its name, implies the need for a certain degree of independent judgment in the hands of the agency adjudicators (who are often called"administrative law judges" or"administrative judges," though they are not judges for purposes of Article III of the Constitution). Even the Supreme Court's most emphatic decision in favor of executive oversight of the administration, Myers v. United States (1926), conceded that presidents should tread lightly around agencies' case-specific adjudications for the sake of due process. Thus, the challenge for a president and his administration is to find ways to achieve some measure of cost-benefit analysis of regulatory policies made through adjudications without risking heavy-handed interference in individual adjudications.

This should be done by requiring agencies to inform OIRA when a finalized adjudication has created a new policy of economic significance equivalent to an OIRA-reviewed regulation and then undertaking a new cost-benefit analysis of that policy as if it were a rule. Unlike OIRA's regulations review, this process should not give OIRA a veto over the adjudication. But the process of after-the-fact cost-benefit analysis and reporting will improve the administration's, Congress', and the people's understanding of how much burden each agency's policies impose on the public.

Improve the Quality of Agency Fact-Finding

In addition to making policy, agencies also find facts. Indeed, an agency's factual analysis is one of the key foundations for its policymaking. Often agencies' efforts to"find the facts" are conducted simultaneously with efforts to craft policies. This is perhaps inevitable, at least to some extent, but it creates significant problems in the effectiveness and the public legitimacy of an agency's factual findings.24

As Oren Cass observed three years ago in National Affairs, agencies claiming to do"evidence-based policymaking" often do something closer to"policy-based evidence-making."25 The public's-and the agency's- understanding of the facts is inevitably colored by their understanding of the policy at stake.

Paul Romer, after winning the Nobel Memorial Prize in Economic Sciences for his study of ideas and innovation, argued that one of the most valuable things government could do to improve the quality of regulation would be to separate fact-finding from policymaking-to separate"the federal government's scientific advisers, who must be uncompromising in their pursuit of the facts," from"regulators, who rely on facts but must compromise to balance competing interests when they write regulations."26

"The institutionalization of cost-benefit analysis for agencies' regulatory proposals is one of the most significant developments in modern regulatory history."

The president and his administration should heed this advice. Whether by executive order, agency self-initiative, or a combination, federal agencies should explore how to separate fact-finding from policymaking to the fullest extent allowed by law. Agency fact-finding, including its assessment of scientific information, should be separated into its own proceedings, in its own agency docket, whenever possible. Ideally, an agency should complete its fact-finding work before proceeding to the rulemaking stage, though in some cases it might need to proceed simultaneously on parallel tracks.

To be sure, agency fact-finding would eventually face judicial review in any litigation challenging the policies built on those facts. But separating these parts of the regulatory process into discrete proceedings would help highlight the real meaning of each proceeding: in one docket, the finding of facts and marshalling of factual analysis; in the other docket, the policy choices and value judgments that rest on facts but are not ultimately dictated by facts alone.

Agency fact-finding benefits from an interagency approach. From the outset, agencies should identify factual or scientific issues that cut across individual agency lines, so that the administration can arrive on common factual determinations for all agencies, avoiding conflicts between agencies on particular factual issues, and avoiding conflicts among factual issues.

This process would be further improved by drawing from another aspect of the long-standing OIRA experience. Just as executive order 12866 requires OIRA and the agencies to produce a semiannual"Unified Regulatory Agenda" and"Regulatory Plan" for the sake of transparency and coordination,27 agencies should also be required to produce"Unified Factual Agendas" to identify major factual and scientific agendas that the agencies plan to undertake in the months ahead. This process would notify other agencies, the rest of government, and the people as to which factual and scientific questions will soon be examined by agencies and thus would help position the public and other agencies and institutions to contribute to the process.

Fortunately, this agenda-setting process already has begun in earnest, pursuant to the Foundations for Evidence-Based Policymaking Act of 2018 and policy guidance from the Office of Management and Budget. The White House should dedicate maximum resources to the success of this crucial initiative, and, as mentioned, Congress and the president should look for ways to institutionalize the fact-finding process in agencies' structures and overall processes.

While OIRA's oversight and interagency review process is generally focused on cost-benefit analysis and similar concerns, executive order 12866 expressly directs OIRA and agencies to be mindful of legal issues too. Specifically, OIRA's oversight extends to regulatory proposals that "raise novel legal or policy issues arising out of legal mandates."28

This aspect of the regulatory-oversight framework is more important than ever. As agencies assert more sweeping and novel regulatory programs pursuant to statutes enacted many decades ago, they race far beyond the expectations of the Congress that originally enacted those laws,29 and they undermine the political incentive for Congress to actually do the hard work of modernizing those statutes.30 Perhaps most important in the short run, such novel interpretations of old statutes raise the risk that federal courts will strike down the regulatory programs as unlawful.

This last consideration is particularly acute in light of recent Supreme Court decisions. As the Supreme Court warned in an opinion by Justice Antonin Scalia in 2014,"When an agency claims to discover in a long-extant statute an unheralded power to regulate a significant portion of the American economy, we typically greet its announcement with a measure of skepticism." Therefore, when an agency asserts a new policy"of vast economic and political significance," the Supreme Court directs the lower courts to require the agency to show that its policy reflects Congress'"clearly" stated intentions.31

With a strong majority of Supreme Court justices (and a new generation of lower-court judges) now holding that view, it would be reckless for an administration to begin onerous regulatory processes without planning to satisfy that analytic requirement. The president should structure his OIRA process accordingly, directing OIRA and the agencies to place much greater emphasis on executive order 12866's requirement for heightened review of regulations involving"novel legal or policy issues."

This approach would be best achieved by involving another expert body in the executive branch: the Department of Justice. Once a regulatory agency or OIRA identifies a novel legal issue in a proposed rule, the Justice Department's Office of Legal Policy (or another existing or new unit in the Justice Department) should be responsible for reviewing the agency's legal analysis to ensure that its proposed regulation is genuinely supported by Congress' clear statutory mandates. OIRA can facilitate an interagency dialogue for each matter, as it does for agencies' cost-benefit analyses.

This process will never eliminate completely the risk of an agency producing an unlawful regulation. But it would help improve the agencies' work at the outset of the regulatory process, long before the regulatory process' completion in the courts.

Take the Long View

As noted at the outset of this chapter, Hamilton recognized that the executive's"steady administration of the laws" was a crucial goal of good government. He also warned that the changeover from one presidential administration to the next would threaten steady administration. This is not entirely bad, to say the least: The whole point of presidential elections is to allow the people to chart the course of governance through their election of the legislators who write laws and the executive who will faithfully enforce them. But Hamilton rightly observed in Federalist 72 that administration changes raise the risk of"mutability of measures"-a constant making and unmaking of policy that undermines the people's attempts to plan for the long run.

This, too, has been a point of increasing emphasis in the Supreme Court. In the 2019 case Kisor v. Wilkie, for example, the justices (in an opinion written by Justice Kagan) warned that agencies' interpretations of regulations would get less deference from the courts when the agency's latest interpretation"conflict[s] with a prior one."32

Thus, the administration must take extra care before departing from an agency's prior interpretation of a regulation or statute and should be prepared to particularly justify this interpretive change. This kind of legal issue could be made part of the aforementioned process for OIRA and Justice Department oversight of agencies' novel legal positions, but that may be unnecessary. At the very least, the president should order agencies to pay special attention to this kind of issue.

Additionally, agencies should be mindful the next administration could reverse their regulatory plans and should try to erect frameworks that minimize the costs of regulatory uncertainty. Agencies should affirmatively plan for the possibility that their proposed regulations could be undone, and they should explain why the benefits of short-lived regulation would still outweigh the short-term costs, including the costs of regulatory uncertainty. This aspect of agency planning could be improved by"scenarios analysis" to game-plan for future regulatory disruption.

Finally, and in all this, an administration must understand the fundamental lesson taught by Rudalevige's aforementioned study of OIRA: that a new administrative institution cannot be established simply through announcements and executive orders. Those mark the beginning of the process, not the end. A new institution's success requires more than just a year's effort; indeed, it requires more than just one presidency. It requires the investment of political capital and administration resources for many years and over multiple presidencies. Only through such effort can a president build what Clarence Thomas, then chairman of the Equal Employment Opportunity Commission, called"good institutions that protect and reinforce good intentions."33


View Citations Hide Citations

1. US Const., art. II, § 3.

2. Interior quotation marks and citations are omitted. City of Arlington v. FCC, 569 US 290 (2013).

3. Federalist, no. 70 (Alexander Hamilton).

4. Henry J. Friendly,"The Federal Administrative Agencies: The Need for Better Definition of Standards," Harvard Law Review 75, no. 7 (May 1962): 1263, 1311.

5. Elena Kagan,"Presidential Administration," Harvard Law Review 114, no. 8 (June 2001): 2245-383.

6. For instance, Exec. Order No. 13772.

7. For instance, Exec. Order No. 13783.

8. For instance, Exec. Order No. 13813

9. Adam J. White,"Executive Orders as Lawful Limits on Agency Policymaking Discretion," Notre Dame Law Review 93, no. 4 (2018): 1569-98,

10. See, for example, 15 USC § 41.

11. See Adam J. White,"Can the Supreme Court Adjudicate the CFPB's Independence Without Determining How Independent the CFPB Actually Is?," Yale Journal on Regulation, January 28, 2020,

12. See Adam J. White,"Reining in the Agencies," National Affairs 11 (Spring 2012),

13. US Department of the Treasury and Office of Management and Budget,"Memorandum of Agreement: Review of Tax Regulations Under Executive Order 12866," April 11, 2018, Treasury%20OIRA%20MOA.pdf.

14. Bridget C. E. Dooling,"Bespoke Regulatory Review," Ohio State Law Review (forthcoming).

15. Andrew Rudalevige,"Regulation Beyond Structure and Process," National Affairs 34 (Winter 2018), See also Andrew Rudalevige,"Beyond Structure and Process: The Early Institutionalization of Regulatory Review," Journal of Policy History 30, no. 4 (2018): 577-608.

16. Cass R. Sunstein,"The Office of Information and Regulatory Affairs: Myths and Realities," Harvard Law Review 126, no. 7 (May 2013): 1841, 1858.

17. Susan E. Dudley,"The Office of Information and Regulatory Affairs and the Durability of Regulatory Oversight in the United States," Regulation & Governance (July 20, 2020),

18. Administrative Conference of the United States,"Retrospective Review of Agency Rules: Recommendation 2014-5," December 17, 2014, research-projects/retrospective-review-agency-rules; and American Bar Association, Section of Administrative Law and Regulatory Practice,"Improving the Administrative Process: A Report to the President-Elect of the United States," 2016, POTUS%20Report%2010-26-16.authcheckdam.pdf

19. See Adam J. White,"Retrospective Review, for Tomorrow's Sake," Yale Journal on Regulatio, November 28, 2016,

20. Susan E. Dudley,"Perpetuating Puffery: An Analysis of the Composition of OMB's Reported Benefits of Regulation," Business Economics 47, no. 3 (July 2012): 165-76.

21. See Susan E. Dudley,"OMB's Reported Benefits of Regulation: Too Good to Be True?," Regulation 36, no. 2 (Summer 2013): 26-30.

22. Philip E. Tetlock and Dan Gardner, Superforecasting: The Art and Science of Prediction (New York: Crown Publishers, 2015), 87.

23. SEC v. Chenery Corp., 332 US 194 (1947).

24. See Hilary Putnam, The Collapse of the Fact/Value Dichotomy and Other Essays (Cambridge, MA: Harvard University Press, 2002).

25. Oren Cass,"Policy-Based Evidence Making," National Affairs 32 (Summer 2017),

26. Paul Romer,"Government Can Do More to Support Science and Innovation," Wall Street Journal, November 20, 2018,

27. Exec. Order No. 12866, 58 C.F.R. 51735 (1993) § 4(b), (c).

28. Exec. Order No. 12866, 58 C.F.R. 51735 (1993) § 3(f)(4).

29. Jonathan H. Adler and Christopher J. Walker,"Delegation and Time," Iowa Law Review 105, no. 5 (July 2020): 1931-93,

30. Adam J. White,"Democracy, Delegation, and Distrust," Hoover Institution, March 12, 2019,

31. Interior quotation marks and citations are omitted. Utility Air Regulatory Group v. EPA, 573 US 302, 324 (2014).

32. Interior quotation marks and citations are omitted. Kisor v. Wilkie, 588 US __ (2019).

33. Clarence Thomas,"Toward a 'Plain Reading' of the Constitution-the Declaration of Independence in Constitutional Interpretation," Howard Law Journal 30 (1987): 983, 989.