In the first weeks of the Trump Administration, a Harvard Law Review article warned that attacks on federal regulatory agencies—and calls for their closer judicial supervision—heralded a revival of the “anti-administrativism” that animated attacks on the New Deal in the 1930s.¹Gillian E. Metzger, 1930s Redux: The Administrative State Under Siege, 131 Harv. L. Rev. 1, 4–6 (2017). That article invoked an earlier, more detailed study of debates in the 1930s, which argued that the Administrative Procedure Act (“APA”) was not the product of an evolving consensus but rather a “fierce compromise” between pro- and anti-New Deal forces.²George Shepherd, Fierce Compromise: The Administrative Procedure Act Emerges from New Deal Politics, 90 Nw. U. L. Rev. 1557, 1560 (1996).

This view has become conventional wisdom. In December of 2019, the Trump Administration’s Solicitor General, Noel Francisco, opened a conference on “modernizing the Administrative Procedure Act” with the observation that the Department of Justice was well-equipped to encourage new thinking, since a Department of Justice initiative at the end of the 1930s had helped persuade Congress to abandon the Walter–Logan bill which was “highly restrictive of federal agencies.”³Solicitor General Noel Francisco Delivers Remarks at Department of Justice Summit on Modernizing the Administrative Procedure Act, Just. News (Dec. 6, 2019), https://perma.cc/AN7J-MPPH. Instead, he claimed, the 1941 Final Report of the Attorney General’s Committee on Administrative Procedure (“Final Report”) prepared the way for the more accommodating provisions of the postwar APA.⁴Id.

Such accounts are misleading in three ways. First, they exaggerate the distance between the APA, as enacted after the war, and the proposals advocated by critics of regulatory practice in the 1930s. Second, such characterizations wrongly reduce the legal debates of the 1930s to partisan or political disputes about the substance of New Deal regulatory programs. Third, they leave out a range of other issues which the Attorney General’s Committee recognized as deserving the attention of reformers. That may well be because positions on these issues do not lend themselves to later interpretations focusing on partisan or political motivations supposed to have been at stake. Correcting the stories that have come down to us is not merely a service to historical accuracy. The actual debates of that era remain relevant to our time in ways that conventional accounts have failed to recognize.

This Article proceeds as follows. Part I recounts the origins of the Attorney General’s Committee and its Final Report. Each of the next four parts examines a specific topic debated in that era: the rule of law (Part II), the president’s role in the administrative state (Part III), rulemaking procedure (Part IV), and judicial review (Part V). Finally, Part VI reflects on the ways subsequent developments may help us reach a better understanding of the concerns that animated the drafters of the Administrative Procedure Act.

I.     Origins of the Attorney General’s Committee

The chairman of the Attorney General’s Committee was Dean Acheson.⁵When the Attorney General first called the Committee together, James Morris was designated as Chairman. Office of the Attorney General, Order No. 3215 (Feb. 16, 1939), in Final Report of the Attorney General’s Committee on Administrative Procedure, S. Doc. No. 77-8, 252, 252–53 (1st Sess. 1941) [hereinafter Final Report]. Acheson succeeded him as Chairman a few weeks later, when Morris was appointed to a federal judgeship. He had served as assistant secretary of the Treasury in the first year of Roosevelt’s first term, before resigning (over policy disagreements) and entering active law practice at Covington & Burling.⁶For background on Acheson’s early career, see Walter Isaacson & Evan Thomas, The Wise Men 125–39 (1986). The Attorney General’s Committee was sometimes called “the Acheson Committee.”⁷This was true at the time. see, e.g., John Dickinson, The Acheson Report: A Novel Approach to Administrative Law, 90 U. Pa. L. Rev. 757 (1942) (“the Attorney General’s Committee . . . commonly known as the Acheson Committee”), and is still true some seven decades later. Yair Sagy, A Triptych of Regulators, 44 Akron L. Rev. 425, 442 (2011) (“the famous Acheson Committee”). After the war, Acheson would serve as Undersecretary and then Secretary of State for President Truman, helping to launch the Marshall Plan, NATO, and the overall containment strategy that guided U.S. policy through decades of Cold War.⁸See Isaacson & Thomas, supra note 6, at 253–504.. Acheson titled his memoir of that era—whimsically, more than boastfully—Present at the Creation.⁹Dean Acheson, Present at the Creation: My Years at the State Department (1969). The book’s epigraph is a statement attributed to a medieval Spanish king: “Had I been present at the creation, I would have given some useful hints for the better ordering of the universe.” Id. at xvii..

Acheson also published a separate memoir about his prewar experience.¹⁰Dean Acheson, Morning and Noon (1965). He did not use that volume to highlight his role as chairman of the Committee on Administrative Procedure. He devotes less than a page to that episode in his memoir, and his account does not leave an ordinary reader thirsting for more detail. The Committee, he says, “labored hard and obscurely on this then unknown and now forgotten task.”¹¹Id. at 215. The full sentence adds that those who “labored” at this task can “take our satisfaction from Justice Holmes’s observation that ‘legal progress is often secreted in the interstices of legal procedure.’” Id. He speaks approvingly but very briefly of the postwar APA—in a way that confirms the sense that he did not want to revisit old debates: “The Administrative Procedure Act . . . has for nineteen years with minor amendments been accepted by regulators and regulated as providing workable and fair procedures.”¹²Id. The memoir devotes much more space to explaining how Acheson came to be appointed to the Committee—after provoking FDR’s displeasure by resigning his Treasury post in 1933 over a disagreement on the legality of Roosevelt’s decision to decouple the dollar from gold. After assisting Felix Frankfurter’s confirmation to the Supreme Court, Acheson was restored to Roosevelt’s favor but declined proposed appointments to the U.S. Court of Appeals for the DC Circuit and then to a high post in the Department of Justice and felt he could not decline the (temporary and part-time) appointment to the Attorney General’s Committee. The whole account implies—somewhat misleadingly—that there were no great issues at stake in the Committee’s work..

Officially, the idea for the committee had originated with Attorney General Homer Cummings. In mid-December of 1938, Cummings had written to President Roosevelt to propose a study of “procedural reform in the wide and growing field of administrative law.”¹³Homer Cummings, Letter from the Homer Cummings to Franklin D. Roosevelt, Dec. 14, 1938, in Final Report, supra note 5, at 251, 251–52. There is no account of who might have advised Cummings to pursue this proposal, though the letter begins by noting that it aims “to renew the suggestion which I have made publicly at different times, that there is a need for procedural reform in the wide and growing field of administrative law.” Id. The need was for administrative procedure “which affords quick and well informed action, grounded upon the fundamentals of fair play.”¹⁴Id. at 251.. Cummings noted that new rules of civil procedure had recently been adopted for federal district courts after extensive study and debate by a “representative group of the bar, appointed by the Supreme Court.”¹⁵Id. He called for a similar study of administrative procedure.

The analogy was slightly misleading, however, since the issue was not simply procedure but institutional authority—who could decide what, not just the order or process for decisions. President Roosevelt endorsed the idea but offered a more pointed rationale for the project. His official reply to the Attorney General in February of 1939 did not emphasize administrative efficiency or public trust but judicial acceptance. He noted the challenges faced by the “Department of Justice in endeavoring to uphold actions of the administrative agencies of the Government when the validity of their decisions is challenged in the courts.”¹⁶Franklin D. Roosevelt, Letter from Franklin D. Roosevelt to Frank Murphy, Feb. 16, 1939, in Final Report, supra note 5, at 252, 252. Murphy had succeeded Cummings as Attorney General in January of 1939.. President Roosevelt seemed to be thinking less about the recently adopted Federal Rules of Civil Procedure than the failure, a year earlier, of his proposal to “pack” the Supreme Court with more sympathetic justices.

At all events, both the Department of Justice and the White House must have been aware that denunciations of administrative abuse were then echoing in the halls of Congress. Reform proposals had been urged by the American Bar Association (“ABA”) and taken up by congressional committees. Congress approved the Walter–Logan bill—sponsored by Representative Francis Walter (D-Pa) and Senator Marvel Mills Logan (D-Ky)—at the end of 1940.¹⁷See H.R. 6324, 79th Cong. (1st Sess. 1939). President Roosevelt blocked it with one of his rare vetoes. His veto statement, while criticizing the measure, also urged Congress to wait for the Attorney General’s Committee to complete its inquiries and make its own recommendations.¹⁸Franklin D. Roosevelt, The President Vetoes the Bill Regulating Administrative Agencies. Dec. 18, 1940, in The Public Papers and Addresses of Franklin D. Roosevelt 616, 616–20 (Samuel I. Rosenman ed., 1941). He did not need to mention that as Washington debated the fine points of administrative procedure, the Luftwaffe was bombing London every night and the German army was tightening its grip on France and the Low Countries.

Meanwhile, the Attorney General’s Committee pressed forward with its charge. The new Attorney General Frank Murphy seems to have recruited most members for their political reliability but others for their broader respectability. If Acheson is counted, a majority of the members had performed high-level legal service for the New Deal.¹⁹Robert Jackson, serving at the time as the U.S. Solicitor General, was among the original members but resigned when he was appointed Attorney General in January of 1940. James W. Morris, an assistant to successive attorneys general since the beginning of the Roosevelt administration, was initially tapped to serve as chairman, then relinquished that position to Acheson, when appointed to the U.S. District Court for the District of Columbia in the spring of 1939. Other New Deal lawyers included Golden W. Bell, who had been Assistant Attorney General for Legal Counsel since 1937; Carl McFarland, another Assistant Attorney General who had been at the Department of Justice since 1933; Lloyd Garrison who had been the initial chairman of the National Labor Relations Board (though by 1940 was a professor at University of Wisconsin); and Francis Biddle, who had served as chairman of the National Labor Relations Board and would subsequently serve as Robert Jackson’s successor as Attorney General (then as judge at the same Nuremberg International Tribunal where Jackson served as chief American prosecutor).. There were also two judges in the initial slate²⁰Duncan Lawrence Groner had been appointed to the U.S. Court of Appeals for the DC Circuit by President Hoover in 1931, then elevated to the Chief Justice position (as it was then called) by President Roosevelt in 1937. Arthur Vanderbilt, a reformer of courts in New Jersey as Chief Justice, had been president of the American Bar Association (“ABA”) in 1938, as the ABA began to weigh in on administrative law debates. and a number of relatively young law professors.²¹The list included Harry Shulman, professor at Yale Law School; Ralph Fuchs, Brookings Institution economist, then professor at Washington Law School in St. Louis; Blyth Stason, professor at University of Michigan Law School; and Henry Hart, professor at Harvard Law School. The staff director was also recruited from a distinguished law school: Professor Walter Gellhorn of Columbia Law School.

The eleven continuing members turned out to be divided on important issues. The Committee submitted its Final Report (the official if unimaginative title) on January 22, 1941. It comprised roughly 200 pages of text. But three members—Carl McFarland, Blythe Stason, and Chief Justice Arthur Vanderbilt—offered forty-five pages of “Additional Views,” advancing a more stringent analysis with more detailed proposals.²²See Additional Views and Recommendations of Messrs. McFarland, Stason, and Vanderbilt, in Final Report, supra note 5, at 203, 203–47 [hereinafter Views of McFarland, Stason, and Vanderbilt]. Chief Judge D. Lawrence Groner of the U.S. Court of Appeals for the DC Circuit also offered his own (even more stringent) “Additional Views.”²³Additional Views and Recommendations of Mr. Chief Justice Groner, in Final Report, supra note 5, at 248, 248–50.

The body of the Final Report emphasized the complexity of the issues. The minority had decided to spell out their own proposals for new legislation which they called, “A Code of Standards of Fair Administrative Procedure” (running to thirty pages).²⁴Views of McFarland, Stason, and Vanderbilt, supra note 22, at 217–47.. The majority responded with its own legislative proposal, demurely entitled “A Bill” and running to some twelve pages.²⁵Final Report, supra note 5, at 191–202. Given that the Justice Department chose so many of its own lawyers for the committee, a margin of 7–4 was not a ringing endorsement for the “majority” proposals.²⁶Shepherd, supra note 2, at 1594–95. What became the APA actually owed far more to the proposals of the Committee’s minority. From what appears in the Final Report, one might conclude the debate was not so much on what to do as how much to do. Even the APA did not adopt all that the minority members proposed.

But 1941 was not the time to sort through a debate on fine-grained questions of administrative procedure. In the summer of 1940, President Roosevelt had appointed Republican elder statesmen to head the Department of War and the Department of the Navy, and announced in December that America must gear up its production of war materials to make itself the “arsenal of democracy.”²⁷U.S. Dep’t of State, Peace and War: United States Foreign Policy 1931–1941, at 599–608 (1943) (reprinting a White House radio address from December 29, 1940). Congress agreed to military conscription—a “draft”—for one year in 1940, then renewed it without limit the next year, which also saw congressional approval for the “Lend-Lease” program of massive military aid to Britain and then to Soviet Russia.²⁸Acheson, supra note 9, at 27–35. A central participant in the negotiations over Lend-Lease assistance to Britain was Dean Acheson, by then Assistant Secretary of State for Economic Affairs (a position he took up in February of 1941—shortly after the Acheson Committee submitted its report to then Attorney General Jackson). The diplomacy (and domestic politics) of Lend-Lease receives considerable attention in his memoir, Present at the Creation. See id.. Congress was preoccupied with the challenges of impending war. War preparation consumed the United States in full before the end of the year. The Acheson Report was put on the shelf until the last months of that war.

II.     The Prewar Debate About the Rule of Law

The Acheson Committee was designed, at some level, as a response to gathering protests against the practices of regulatory agencies. Certainly it gained that charge after President Roosevelt, in vetoing the Walter–Logan bill, insisted that before new legislation could be adopted, the Attorney General’s Committee must be given time to complete its comprehensive study of the relevant issues.

In the first decades after the APA’s adoption, legal commentators embraced the claim that the Attorney General’s Committee had succeeded in building support for a broad consensus, so that Congress ultimately enacted the APA by large majorities. Some of the most prominent postwar commentators were bound to find this compelling, since they had served on the staff of the Attorney General’s Committee at the time (or in allied positions at the Department of Justice).²⁹Paul R. Verkuil, Walter Gellhorn & Kenneth Culp Davis, Present at the Creation: Regulatory Reform Before 1946, 38 Admin. L. Rev. 507 (1986). The title—borrowed from Acheson’s famous memoir of his State Department service—acknowledged the service of Kenneth Culp Davis and Walter Gellhorn on the staff of the Acheson Committee in 1940. Gellhorn went on to edit one of the leading case books in administrative law; Davis, the leading treatise. Davis also credits the influence of Louis Jaffe (professor of law at Harvard, coeditor of a leading case book) as a consultant to the Justice Department when the APA was adopted. Gellhorn offered a more extended account—emphasizing continuity with the Acheson Committee’s work. See Walter Gellhorn, The Administrative Procedure Act: The Beginnings, 72 Va. L. Rev. 219 (1986). A nonparticipant in the creation endorsed the general account. See Alan B. Morrison, The Administrative Procedure Act: A Living and Responsive Law, 72 Va. L. Rev. 253 (1986). In more recent decades, political scientists have argued that the debate over administrative procedure was really a proxy for larger debates about the New Deal.³⁰See Matthew McCubbins, Roger Noll & Barry Weingast (McNollgast), Political Origins of the Administrative Procedure Act, 15 J.L. Econ. & Org. 180 (Mar. 1999) (foremost in this literature). For more historical detail but to the same effect, see Shepherd, supra note 2, at 1683 (acknowledging advice from McNollgast, supra note 30); Martin Shapiro, The APA: Past Present, Future, 72 Va. L. Rev. 447 (1986).. Both interpretations assume that the debate in the 1930s was at best overblown. Quite prominent figures at the time, however, were quite insistent about the larger implications of the debate.

The ABA framed the underlying contest as a struggle for “the rule of law” as against “administrative absolutism.”³¹See Report of the Special Committee on Administrative Law, 63 Ann. Rep. A.B.A. 331, 339–40 (1938) [hereinafter Report of the Special Committee (1938)].. In 1938, the ABA committee on administrative law recruited Dean Roscoe Pound of Harvard Law School to serve as its chairman. Soon after, he produced a report comparing the trend toward “administrative absolutism” among federal agencies to the methods of dictatorial rule in Europe and analyzed defenders of the trend in American universities as followers of a “Marxian idea.”³²Id. He did not intend the latter phrase as a mere rhetorical flourish. He tried to explain the charge with learned references to works by Soviet jurists. And he did not hesitate to deploy the same language in criticizing President Roosevelt’s veto of the Walter–Logan bill in 1940.³³See Roscoe Pound, The Place of the Judiciary in a Democratic Polity: An Examination of the Walter-Logan Bill Veto Message, 27 A.B.A. J. 133, 139 (1941). Pound did not talk in that way because he was a dogmatic formalist or a lifelong reactionary. Earlier in his career, he had coined the term “social engineering”—as a term of praise for what could be achieved by well-crafted adjustments to the common law.³⁴Linus J. Macmanaman, Social Engineering: The Legal Philosophy of Roscoe Pound, 33 St. John’s L. Rev. 1, 33 n.123 (1958).

Trends in the outside world seemed to show that the rule of law could no longer be taken for granted. It looked quite imperiled in the late 1930s, as Stalinist dictatorship intensified in Russia and Nazi dictatorship in Germany. As the Acheson Committee continued its deliberations, the slide to tyranny became an avalanche, as more and more countries fell under Soviet or Nazi control.³⁵Germany and Russia divided Poland between themselves in September of 1939. The Soviet Union swallowed Lithuania, Latvia, Estonia, and a slice of Romania in June of 1940, while Germany seized control of France and the Low Countries, following earlier seizure of Norway and Denmark.. We now have a term for the Soviet and Nazi regimes: totalitarian. It was not yet in general use before World War II. But people already saw the point—everything might be up for grabs because the rulers in these states saw nothing beyond their reach. Talk about the rule of law was, in some way, a symbol of a broader anxiety, which did not mean it was merely rhetorical.

The critics could point to disturbing trends at home. President Roosevelt had started his second term by proposing to expand the size of the Supreme Court to ensure that new appointees would provide a compliant majority. His veto statement on the Walter–Logan bill—which provoked Pound’s denunciation—blamed critics of administrative procedure for wanting to give a central role to lawyers.³⁶86 Cong. Rec. 13,942 (1940), reprinted in Logan-Walter Bill Fails, 27 A.B.A. J. 52 (Dec. 18, 1940 veto message) (“[A] large part of the legal profession has never reconciled itself to the existence of the administrative tribunal. Many of them prefer the stately ritual of the courts, in which lawyers play all of the speaking parts . . . . Many of the lawyers still prefer to distinguish precedent and to juggle leading cases . . . .”). It could easily be understood to mean administrators should not have to be much distracted by law. That sentiment was, in fact, proclaimed by prominent New Dealers. James Landis, former Chairman of the Securities and Exchange Commission, enthused about disregard for law:

One of the ablest administrators that it was my good fortune to know, I believe, never read, at least more than casually, the statutes that he translated into reality. He assumed that they gave him power to deal with the broad problems of an industry and, upon that understanding, he sought his own solutions.³⁷James M. Landis, The Administrative Process 75 (1938).

The critics focused their alarm on a charge that made agencies sound like lynch mobs: current practice allowed the same agency to make the rules, then serve as “prosecutor, judge, jury, and executioner” in enforcing them. Wasn’t this the antithesis of rule by law? The Acheson Committee’s Final Report grappled with the charge. In fact, it focused most of its attention on this very issue.

The Committee’s brilliant tactic was to diffuse the issue by conducting—and publishing—an elaborate survey of some three dozen agencies and programs to determine how they dealt with this issue. Specialized monographs published with the main report run to hundreds of pages. So the Committee acknowledged the concerns of critics and then responded in stupefying detail. Along the way, it changed the subject from the abstract meaning of rule of law to the technical tedium of administrative process. Early on, the Final Report warns that, “while much criticism is general in language, the thought behind it is specific. As a generality, it may not be sound, but as a specific criticism of a particular practice of a particular agency it may be justified.”³⁸Final Report, supra note 5, at 2.. The strategy of the Final Report was to counter “generality” with an avalanche of specifics.

Soon after the release of the Final Report, there was a program celebrating the scientific rigor of these studies. It was hosted by Columbia Law School—the home institution of Professor Gellhorn, the Committee’s staff director. Justice Felix Frankfurter participated. He praised the Final Report with high-table diction.³⁹See Felix Frankfurter, Foreword, 41 Colum. L. Rev. 585, 587–88 (1941).. There was even praise from John Foster Dulles, then a leading figure in the New York bar (and in the following decade, Acheson’s successor as Secretary of State).⁴⁰John Foster Dulles, The Effect in Practice of the Report on Administrative Procedure, 41 Colum. L. Rev. 617, 617–27 (1941). Other commentary also expressed respect for the Committee’s wide-ranging research.⁴¹Louis Jaffe, The Report of the Attorney General’s Committee on Administrative Procedure, 8 U. Chi. L. Rev. 401, 440 (1941).

But for all its accumulation of factual detail—sometimes actually interesting—the Final Report did not prove very much about the subject it purported to investigate. Staffers asked various agencies how they organized their operations and in particular, how they protected hearing examiners from pressure. Then they wrote up what the agencies told them. The specialized monographs on individual agencies or programs (published as a companion volume with the Final Report) offer little we would now recognize as data—such as how often commissioners (or other top administrators) disregarded the findings of the hearing examiner and how often courts, in turn, upheld the agency’s ultimate determination. The monographs have few numbers at all, beyond telling how many examiners had been employed in each agency. They certainly do not try to measure the effectiveness or even the actual work product of various agencies.

What the specialized studies do bring home to a dutiful reader is how much variety there was already in 1940 among different agencies and how differently the issue might appear from one agency to the next. But they also show that officials had given some thought to procedure and taken some steps to reassure regulated or affected interests. They even show broadly similar impulses, regarding the separation of examiners from presenters of evidence.

The appeal to the complicating differences among different programs was, by its own logic, a claim to leave administrators to make their own expert determinations, even on questions of procedure. It all might seem too complicated for a single rule of the sort familiar to judges, so it might best be left to experts in each agency to sort out. A law professor of a leftist bent made that precise point in response to the Final Report.⁴²See A. H. Feller, Administrative Law Investigation Comes of Age, 41 Colum. L. Rev. 589, 592 (1941). Feller warned that success of the National Labor Relations Board (“NLRB”) in protecting labor rights “could not have been accomplished as successfully if the powers of investigation and adjudication had not been combined.” Id. at 600. He criticized the proposed “Code” of procedure (proposed by the minority on the Acheson committee) as “unfortunate” since it “proceeds on the premise of the necessity of uniformity.” Id. at 615. After the war, Feller served as General Counsel of the United Nations. He committed suicide in 1952 when called to appear before a congressional investigating committee. His article remains an unusually cogent exposition of those most opposed to the ABA view.. But no one on the Acheson Committee would insist on that logic (at least in public).

Instead, the Committee urged that officials taking evidence—the Final Report dignified them with the novel title “hearing commissioners”—should not participate at all in the gathering or presentation of the evidence at the hearing. That basic demarcation of roles was not enough for the minority, which discoursed at length on the remaining challenges. The “hearing commissioner” must not only have an independent role but a judicial-style seclusion from other agency officials. Allowing the “hearing commissioner” to seek advice from agency experts on the side would be “entirely subversive of every fundamental notion of fair procedure. . . . A secret conference of a few minutes [with other agency officials] destroys an entire trial.”⁴³Final Report, supra note 5, at 243 (note on section 309, subparagraph “m” of A Code of Standards of Fair Administrative Procedure). So the minority proposed elaborate safeguards against such ex parte communications by hearing examiners.

All members of the Committee agreed, in principle, that the answer to complaints about the mingling of powers in administration was an internal separation of investigating and prosecuting functions from judging. But the minority demanded this separation be safeguarded with explicit rules on relevant communication between “hearing commissioners” and other staff. It might look like a disagreement on whether the administrative process should be required to conform to judicial process. One could even imagine this as the central struggle over the fate of the New Deal—if one inclined to the view propounded by political scientists decades later.

If you emphasize the independence of the trial examiner, you embrace the idea that each case should be decided by a neutral umpire because the agency has no greater claim to coerce or punish than the regulated firm has to retain its liberty and property rights. That is how we think about judges or juries hearing contending claims in ordinary litigation between private parties (and all the more so in criminal trials). But a regulatory agency is not only an enforcer of rules but an engine of national policy.

It certainly looks like a policymaker if you think regulatory aims may be central to the overall health of the economy but not readily subject to rules, as was said of commissions engaged in regulating transportation sectors. The Interstate Commerce Commission, the Civil Aeronautics Board, and the Federal Maritime Commission were designed not only to arbitrate between shippers and carriers but to advance an overall economic vision. If you think the aim is to guide investment—to plan the main lines of economic development—then it makes less sense to constrain regulators with procedural safeguards for the return-on-equity interests of regulated firms.

There was gathering debate on how or whether “rule of law” could apply to the new responsibilities of regulatory management. Should Washington engage in overall economic planning? Was “planning” even a sensible aim for the economy overall? Could a state with such ambitions still provide security to private property? Would it end by making everyone feel too dependent on government to feel safe in criticizing or opposing public policies? One would not need to see this as a straight path to “serfdom” or “despotism” to feel qualms. Politicians (and later, professors) who used extreme language were talking about something that might be considered fundamental.⁴⁴See generally Gordon Hewart, The New Despotism (1929); see also Friedrich A. Hayek, The Road to Serfdom (1944).

Tellingly, though, there is scarcely a word about this larger debate in the Acheson Committee’s Final Report. The aim of the Acheson Committee was not to settle great questions. Nor even to clarify them. In effect, its aim was to push them off or bury them in procedural details.

Apart from tactical advantage and lawyerly instinct, the Committee did, in fact, have reason to resist the great ideological debates in the background. It is not at all obvious, even in retrospect, that the fate of any particular New Deal agency was at stake in these procedural questions. It was certainly not obvious to lawyers and judges at the time.

One reason to doubt that view (which we can call the “political science view”) is that even the Walter–Logan bill was content to call for “fair and independent” hearings without spelling out what this would require. The Walter–Logan bill did not even require that hearing examiners have a separate status within the agency. It prescribed that hearings would be conducted by a panel of three agency employees, only one of whom need be a lawyer. All three examiners might very well spend most of their time as investigators, prosecutors, or rule drafters, so long as they did none of those things on the particular case where they were enlisted as factfinders (even if that case had close similarities to cases they might otherwise be pursuing as prosecutors). In all of this, the Walter–Logan bill followed the earlier proposals of Dean Pound’s 1938 ABA Report.⁴⁵See H.R. 6324, 76th Cong. § 4(a) (1939) (“Every head of an agency shall from time to time designate three employees of his agency for such intra-agency boards . . . as may be necessary and desirable. . . . When the members of any board are not engaged in the hearing of administrative appeals as hereinafter provided, such employees shall be assigned to other duties in the service of the agency concerned.”) The same language (virtually word for word) appears in the proposed ABA bill. See Report of the Special Committee (1938), supra note 31, at 364 (Section 3 of the Proposed Act to Provide for the More Expeditious Settlement of Disputes with the United States and for Other Purposes titled “Statutory Approval and Authority for Administrative Boards and Prescribing Their Procedure”).

The Walter–Logan bill still assumed—as did all sides on the Acheson Committee—that the findings of the hearing examiners could be reviewed by agency heads (the secretary of an executive department or the commissioners of an independent commission). At that level, competing policy responsibilities might well distract or distort the judgement of officials, even if they purported to decide individual cases solely on the record compiled for each particular case.

Yet a more thoroughgoing separation had already been proposed. In 1937, the President’s Committee on Administrative Management had urged a comprehensive reorganization of federal agencies. It was known as the “Brownlow Committee” for its chairman, University of Chicago Professor Louis Brownlow, a leading figure in the new field of public administration. The Brownlow Committee’s report urged (among other things) that the rulemaking and prosecutorial functions of independent regulatory commissions should be given to executive agencies under direct presidential control—adjudication of particular cases would then be left to a “judicial section” in the agency that would remain entirely independent.⁴⁶Robert Cushman, The Problem of the Independent Regulatory Commissions, in Report of the President’s Committee on Administrative Management 41–42 (1937) [hereinafter Report]. Cushman was a professor in the Cornell Department of Government. In the 1950s, he hired an undergraduate student, Ruth Bader, to help with research. She would reach the Supreme Court as Bader Ginsburg and speak warmly of Cushman as a champion of civil liberties. Jon Craig, Ruth Bader Ginsburg Reminisces About Her Time on the Hill, Cornell Chron. (Sept. 22, 2014), https://perma.cc/N3QV-KRRU (“The jurist’s respect for the First and Fifth Amendments is rooted in college research for Professor Robert Cushman in which she tracked ‘black lists’ of the entertainment industry during the McCarthy era.”). Cushman’s proposal in the Brownlow Committee’s report was subsequently developed in a full length book. Robert Cushman, The Independent Regulatory Commissions (1941). The argument was not seen as placing him on the political right.. The Brownlow Committee’s report was published with a preface by President Roosevelt, endorsing its recommendations, expressing no concern at all about the proposed isolation of regulatory adjudication.⁴⁷Franklin D. Roosevelt, Message from the President of the United States, in Report, supra note 46, at iv, v (“I endorse this program [including placing commissions under presidential control] . . . . [I]n so doing [and implementing this program] we shall know that we are going back to the Constitution, and giving to the executive branch the modern tools of management and an up-to-date organization which will enable the government to go forward efficiently.”).

Congress was not at all receptive to this sort of reorganization, however. It favored the independence of regulatory commissions. The minority on the Acheson Committee noted the experience—and noted that their own proposal was more modest than the Brownlow Committee’s scheme.⁴⁸See Views of McFarland, Stason, and Vanderbilt, supra note22, at 206 (describing the proposal of the Committee on Administrative Management as a “complete and drastic program of separation” between adjudication and other agency functions).. It was, however, more demanding than what the majority sought.

After the war, Congress was content to accept most of the recommendations of the Acheson minority. The APA laid down express restrictions on communications with hearing examiners outside the hearing.⁴⁹5 U.S.C. § 554. By then, it did not seem momentous. Whatever symbolic weight the issue might have had in the 1930s, it no longer seemed urgent.

The Congress which enacted the Walter–Logan bill in 1940 still had a solid majority of New Deal supporters. When Congress approved the APA in 1946, it was approved by voice vote in both chambers with no recorded dissents. President Truman signed it without fuss. It was not a compromise with the claims advanced by the ABA and the supporters of the Walter­–Logan bill. It was an extension or elaboration of those claims (regarding the independence of adjudicative proceedings from other agency functions). No one then protested that the APA would jeopardize the operations of established New Deal agencies.

So why did President Roosevelt think it necessary to veto Walter–Logan? If he saw the procedures it laid down as threatening to the New Deal, it may well be because he had a view of the New Deal as something much larger than the aggregate of already established agencies and programs. Certainly, it was not an idiosyncratic view. The leftwing National Lawyers Guild denounced the Walter–Logan bill as “an attempt to prevent the effective enforcement of the bulk of the New Deal legislation” and “an attack on democratic procedure.”⁵⁰Mortimer Riemer, Special Agencies Defended: Reply to Recent Criticisms of Federal Bureaus Assails Walter-Logan Bill, N.Y. Times, June 14, 1939, at L22.

If the real issue was whether we would have comprehensive economic planning, Roosevelt might have expected, by the end of 1940, that full wartime mobilization would make some version of that inevitable. But voters displaced more New Deal supporters at each wartime election and in 1946 they returned Republican majorities to both houses of Congress. Wartime planning (with rationing and price controls) left an enduring shadow over the more grandiose visions of prewar New Dealers. Barely a dozen years after the enactment of the APA—the statute that was supposed to have saved the New Deal⁵¹See Shepherd, supra note 2, at 1680 (“The fight over administrative reform was the major political battle for the life of the New Deal.”).—scholars of regulation derided the inertia and aimlessness of independent commissions.⁵²Notably, see Marver H. Bernstein, Regulating Business by Independent Commission (1955).

Yet the Final Report was not merely an effort to defuse the great background debate about the rule of law in the modern world. The Committee also grappled with a range of other issues barely noticed in subsequent accounts of its work.⁵³Joanna Grissinger provides one of the few scholarly contributions to focus on the work of the Committee. It offers useful background on the political climate of the era but focuses almost all its attention on the committee’s proposals for internal separation of adjudicatory functions within agencies. See Joanna Grissinger, Law in Action: The Attorney General’s Committee on Administrative Procedure, 20 J. Pol. Hist. 380, 408 (2008). That is particularly unfortunate, because the issues are still with us.

III.     The President as Administrative Manager

Those that believed in central planning should have supported central coordination for the jumbled array of regulatory commissions and agencies that had accumulated in Washington by the late 1930s. That was the argument of the Brownlow Committee’s report, dressed up in the management nostrums of public administration experts. Congress had firmly rejected that idea in 1937 (when it seemed a disturbing complement to FDR’s simultaneous, immensely controversial plan to enlarge the Supreme Court to pack it with pro-New Deal justices).

The Acheson Committee did not go there. Perhaps it was not really a question of “procedure,” anyway. But the Final Report did emphasize the range and complexity of challenges faced by different agencies, raising the question: Could there really be one statutory formula to cover all the agencies?

The Final Report proposed that Congress should establish an Office of Federal Administrative Procedure. It would have a director appointed by the president—and removeable by the president. It would “[c]onduct . . . inquiries into practice and procedures” and “[m]ake . . . recommendations” on “practices, procedures, and methods of organization which have proved most satisfactory.”⁵⁴Final Report, supra note 5, at 194. It might sound like the precursor to today’s Administrative Conference—itself a mid-1960s add-on to the APA. In fact, the original proposal was more than that.⁵⁵The Administrative Conference of the United States, established in 1964, has no authority other than to convene conferences, sponsor studies, and make recommendations. 5 U.S.C. §§ 591–596. Characteristically, the presiding official is the “Chairman” rather than the “Director” (though selected by presidential appointment, subject to Senate confirmation, and serving a five-year term)..

To begin with, the Final Report envisioned that the Director of Administrative Policy would “[r]eceive complaints regarding the procedure of particular agencies, investigate those which appear to be made in good faith, and report thereon to the complainants and to the agency concerned, recommending to the agency any measures which seem to the director desirable to correct deficiencies.”⁵⁶Final Report, supra note 5, at 194. The Final Report did not say the agency would be bound to accept the director’s assessment of the complaint, let alone heed his recommendation of how to fix the problem. But it gave agencies incentives to listen carefully.

One such claim on agency attention was that the director would be responsible for making an annual report to the president (as to Congress), accounting for the work of the Office of Federal Administration Procedure—such as assessing complaints—and then extend “recommendations relating to the practices and procedures of the agencies.”⁵⁷Id.. That report might be a good place to indicate that some agencies were in need of personnel changes at the top. Or a change in budget allocations. John Foster Dulles saw the proposal as “a projection, in permanent form, of the Attorney General’s Committee, with continuing authority to investigate agencies” and called it the “most important and most constructive proposal of the [Final] Report.”⁵⁸Dulles, supra note 40, at 618.

The proposal did not leave this office as a mere advisory body, however. The majority of the Acheson Committee proposed that “hearing commissioners”—what we now call Administrative Law Judges (“ALJs”)—should be dependent on approval of the director of the Office of Administrative Procedure. He would be responsible for deciding on the choice of individual “commissioners” (from nominees by the agency, but he could ask for further nominees if not satisfied with the first three). He would also have the last word on removing “hearing commissioners” accused of misconduct (after opportunity for a hearing on the charges against them).⁵⁹Final Report, supra note 5, at 196–97 (in contesting charges constituting “cause” for removal—brought by the agency or by the Attorney General—a “hearing commissioner” could “demand a hearing” either “before the Office of Federal Administrative Procedure,” presumably the director or designated representatives of the director, or “before a trial board consisting of the Director and two other individuals designated by the Office”).

The minority report was in broad agreement with these suggestions. It emphasized that when appointing and removing hearing commissioners, the director should act only with the agreement of his two associates on a three-person board (comprised of the director, a judge from the U.S. Court of Appeals for the DC Circuit, and an official from the Administrative Office of U.S. Courts).⁶⁰Id. at 222. This suggestion accepted the premise that hearing commissioners should be subject to central oversight but sought to ensure the oversight would be somewhat insulated from political pressure. This refinement implicitly acknowledged that the director would be seen as a political agent of the president and potentially quite powerful, hence the need to constrain (or at least, chaperone) his decisions on the sensitive question of appointing and removing hearing commissioners.

The minority report proposed another revealing refinement. It recommended that the president be authorized to suspend existing procedures in any particular agency for a period of two years.⁶¹Id. at 223–24. The minority claimed it was “obvious” that this might sometimes be necessary.⁶²See id. at 224. It did not question that decisions to suspend normal rules should be made by the president. In effect, this provision underscored something which the majority report may have thought implicit (and better left unsaid): agency heads would look for presidential approval before suspending their normal rules of procedure. The minority may even have made the change a bit more difficult by imposing a formal process for suspending normal procedure.

None of this corresponds to anything in the Walter–Logan bill.⁶³See id. As the minority proposal took the trouble to point out, “[n]either the Logan-Walter bill nor the committee proposal makes any provision such as suggested above” regarding the suspension of rules. Id. Barely a trace of these proposals remained in the bills that became the APA after the war. The APA does authorize suspension of normal rulemaking procedure in special circumstances—but at the decision of the agency (which might be an independent commission).⁶⁴5 U.S.C. § 553(b)(3)(B) (referring to notice-and-comment rulemaking: “this subsection does not apply . . . when the agency for good cause finds (and incorporates the finding and a brief statement of reasons therefor in the rules issued) that notice and public procedure thereon are impracticable, unnecessary, or contrary to the public interest”).

Probably the main reason Congress was so resistant to embracing presidential management proposals was that it did not want to acknowledge a central role for the president (or some immediate appointee of the president) in all regulatory policy. Congress sought to constrain the agencies within procedural requirements. Congress was not attracted to the idea of putting a presidential agent in charge of monitoring, guiding, and enforcing these procedural norms. Of course, that did not put an end to the impulse of presidents to stretch a guiding hand toward regulatory agencies.

IV.     Rulemaking Procedure

President Roosevelt vetoed the Walter­–Logan bill on the grounds that it would slow down the work of agencies and open too many avenues to unproductive legal wrangling. So it is notable that even the Walter–Logan bill imposed hardly any procedures for rulemaking. One of the few requirements was an obligation to issue interpretative regulations within one year of a new statute’s taking effect—presumably to encourage policy development by rules rather than adjudication and so stabilize expectations for regulated firms.⁶⁵H.R. 6324, 76th Cong. (1939) (“Administrative rules under all statutes hereafter enacted shall be issued as herein provided within one year after the date of the enactment of the statute subject to the adoption thereafter of further rules from time to time . . . .”). Dean Pound’s Committee at the ABA also had very little to say about rulemaking procedure.⁶⁶Report of the Special Committee (1938), supra note 31, at 362 (noting that section 1 of the Proposed Act to Provide for the More Expeditious Settlement of Disputes with the United States and for Other Purposes titled “Implementing Rules and Regulations” offers less than one page of explanation and says rules should be adopted “after publication of notice to and hearing of interested parties” but does not say anything further about the contents of the notice or procedure of the “hearing”—which does not seem to mean a trial proceeding).

Given how little was demanded by outside critics, the majority on the Acheson Committee also decided there was no need to propose new procedures here. In the model bill accompanying the majority report, provisions on rulemaking are less than one page (compared with seven pages on adjudication). Almost all the provisions on rulemaking actually deal with ancillary matters—such as requirements to publish rules after their adoption and to report new rules to Congress on an annual basis.⁶⁷Final Report, supra note 5, at 195.. There is no prescribed procedure at all for developing rules.

It was the minority members who thought about procedures here. Their proposed “Code” offered seven pages of requirements for rulemaking. Most notably, it required advance “notice” in the Federal Register and specified that the notice should provide an account of “the issues or scope of the proposed rules . . . given with as much particularity and definiteness as deemed practicable.”⁶⁸Views of McFarland, Stason, and Vanderbilt, supra note 22, at 228.. That is more demanding than the requirements for informal rulemaking ultimately specified in the APA.⁶⁹5 U.S.C. § 553.

The minority’s proposed “Code” also insisted on the primacy of rules: “administrative agencies . . . shall, as a fixed policy, prefer and encourage rule making in order to reduce to a minimum the necessity for case-by-case administrative adjudications.”⁷⁰Views of McFarland, Stason, and Vanderbilt, supra note 22, at 224–25. Accordingly, it went on to warn against evading rulemaking procedures with interpretive rulings: “Each agency shall issue, in the form of rules, all necessary or appropriate rules interpreting the statutory provisions under which it operates . . . .”⁷¹Id. at 226 (emphasis added). The Code particularly cautioned against displacing rulemaking with “rulings” such as “opinions of general counsel,” not adopted by public procedure.⁷²Id. at 231. But see 5 U.S.C. § 553(b)(3)(A) (“[T]his subsection [on rulemaking procedures] does not apply to interpretive rules.”)..

None of these strictures or cautions can be found in the APA. Nor does the APA embrace the admonition in the proposed Code against rules that “merely repeat legislative provisions.”⁷³Final Report, supra note 5, at at 226.. On rulemaking procedure, as on much else, the APA ended up offering a compromise that leans toward the proposals of the Acheson Committee’s minority, but stops well short of codifying all those proposals.

V.     Courts as Enforcers

President Roosevelt had acknowledged, when endorsing his attorney general’s proposed study of administrative procedure, that his main concern was to satisfy courts.⁷⁴Roosevelt, supra note 16, at 252.. Or was it to guidecourts and encourage them to back off?

The Walter–Logan bill, reflecting the aims of the ABA, was supposed to ensure that administrative agencies remained subject to reliable judicial supervision.⁷⁵For respectful summary of the debate, see Paul R. Verkuil, The Emerging Concept of Administrative Procedure, 78 Colum. L. Rev. 258, 268–74 (1978). It was most aggressive in authorizing immediate review of new regulations, but it limited such review to basic procedural, constitutional, and jurisdictional questions.⁷⁶See S. 915, H.R. 6324, 76th Cong. § 3 (1939). It also made clear that this initial review would not necessarily bind a subsequent court viewing the regulation as applied to a particular regulated party.⁷⁷See id. Even when reviewing agency adjudications, the Walter–Logan bill acknowledged that review should respect an agency’s initial determinations as a factfinder.⁷⁸Id. § 5(a) (“Any decision of any agency or independent agency shall be set aside if it is made to appear to the satisfaction of the court (1) that the findings of fact are clearly erroneous; or (2) that the findings of fact are not supported by substantial evidence; or (3) that the decision is not supported by the findings of fact . . . .”).. It also sought to discourage abusive appeals by authorizing courts to impose penalties and fees for frivolous claims.⁷⁹Id. § 6 (“The courts shall have jurisdiction and power to impose damages in any case where the decision of the agency or independent agency is affirmed and the court finds that there was no substantial basis for the petition for review.”).

The Final Report took considerable trouble to explain to readers that courts could not, under the Constitution, simply give judgments outside a concrete case and that courts could not give judgments when they could not speak with finality.⁸⁰Judicial Review of Administrative Adjudication, in Final Report, supra note 5, at 75, 75–95. But then it had very little to say about actual standards of review. It was the Committee’s minority which urged more clarification of the judicial role. A particularly notable passage in their “Additional Views” included this caution:

It is unsatisfactory to the citizen and unfair to the courts to provide for judicial review without defining its scope. In effect the courts are asked to choose between themselves and other public agencies, they are asked to assume or deny themselves power of review, and they are made a party to the result of conflicting statutory interpretations. Under these circumstances, it is natural that the courts should lean backwards to deny themselves powers which Congress has not clearly conferred upon them.⁸¹Views of McFarland, Stason, and Vanderbilt, supra note 22, at 212..

Elsewhere, the minority deplored that “[t]he courts have been given no legislative directions, or almost none [in existing statutes]. The structure of judicial review, therefore, rests upon the precarious basis of judicial decisions rendered in scattered and dissimilar cases.”⁸²Id. at 245. But the minority also recognized that “[t]he Logan–Walter bill in its final form stated no greater scope of review than is now widely recognized by the courts” though it might “sweep into the judicial hopper” disputes not previously subject to judicial review.⁸³Id..

On the other hand, the minority report itself indicated that review should be available to “any person adversely affected” by an agency decision—seemingly encouraging a wider view of standing than the APA.⁸⁴Id. at 246. Compare 5 U.S.C. § 702 (“A person . . . adversely affected or aggrieved by agency action within the meaning of a relevant statute, is entitled to judicial review thereof.” (emphasis added)). Famously, the Supreme Court seemed to disregard the APA’s qualification (“within the meaning of a relevant statute”) in its broadening of standing in Assoc. of Data Processing Serv. Orgs., Inc. v. Camp, 397 U.S. 150, 153, 157 (1970).. It also stipulated that review should be available for the “lawfulness and adequacy of procedure,”⁸⁵Views of McFarland, Stason, and Vanderbilt, supra note 22, at 246 (emphasis added). a formula that seems to invite courts to take a wider role in defining which procedures would be “adequate.”

The minority report also proposed that attorneys be licensed to practice before agencies and subject to disbarment there for “good cause.”⁸⁶Id. at 219.. It recommended this licensing function could be assigned to the proposed new Office for Administrative Procedure (which the majority report already made responsible for the selection and discipline of agency hearing examiners).⁸⁷Id. at 219–20.. Presumably this licensing arrangement would give courts more confidence when reviewing appeals by such certifiably knowledgeable and experienced advocates.

The minority report then itemized the claims courts should resolve: questions of conformity with “constitutional right,” “statutory authority or jurisdiction,” support for “findings, inferences, conclusions of fact,” or actions “otherwise arbitrary or capricious.”⁸⁸Id. at 246. The list is very close to what finally appeared in section 706 of the APA.⁸⁹5 U.S.C. § 706.. Where the minority report also invites courts to review the “adequacy of procedure” used by the agency, the APA speaks more narrowly of the agency’s “observance of procedure required by law.”⁹⁰Id. § 706(2)(D).

On the other hand, the final text of the APA also omitted the minority’s admonition, which the minority proposed for the conclusion of the “scope of review” provision in its model code: “Provided, however, That upon such review due weight shall be accorded the experience, technical competence, specialized knowledge, and legislative policy of the agency involved as well as the discretionary authority conferred upon it.”⁹¹Views of McFarland, Stason, and Vanderbilt, supra note 22, at 246–47.. The APA also omitted the clarification that “substantial evidence” should mean “support of all findings of fact, including inferences and conclusion of fact, upon the whole record” and “if it means a more restricted review, should be clarified by more precise language.”⁹²Id. at 211–12..

VI.     The Aftermath—And After

The Attorney General’s Committee was interested in assuring fair process for regulated interests. The minority was particularly interested in that, or particularly interested in assuring public trust in the administrative process. The Committee was also concerned with improving administrative procedures in other ways. In fact, the minority was particularly fertile in suggestions for improvements in other areas.

Compared with proposals advanced in the Final Report of the Attorney General’s Committee, the legislation enacted by Congress in 1946 was more limited. It adopted many proposals of the minority (such as insulation of hearing examiners and notice-and-comment rulemaking) but left out many things suggested in the minority’s proposed bill and even left out some proposals in the main Final Report. The APA offered nothing about a Federal Office of Administrative Procedure, with a director answering to the White House while demanding answers from the agencies. The APA said very little about rulemaking procedure (for “informal” rulemaking). It neglected several admonitions of the Committee’s minority on this topic, along with its suggestion that the president be authorized to suspend otherwise required procedures.

The same could be said of minority suggestions regarding judicial review. The APA gave very little guidance to courts about scope of review and declined to clarify vague terms like “substantial evidence.” Nor, of course, did it adopt the Walter–Logan bill’s provisions, authorizing immediate review of new agency regulations by any party “who may be aggrieved.”⁹³H.R. 6324, 76th Cong. § 5(a) (1939) (“Any party to a proceeding before any agency or independent agency . . . who may be aggrieved by the final decision or order of any agency . . . may at his election file a written petition . . . for [judicial] review of the decision.”). It does not impose restrictions on who may become a “party to a proceeding”—and agencies tended to allow a range of “parties” to participate..

On the other hand, many of these things have come to pass: elaboration of rulemaking requirements, White House-sponsored review of rulemaking procedure, expansion and intensification of judicial review (along with periodic calls for relaxation of judicial review), and immediate review of new rules. The status of hearing examiners (to use the original APA terminology) has been further enhanced with an exalted new title—“Administrative Law Judge.” The new term would probably have struck even the minority on the Acheson Committee as somewhat too grand. They embraced the term “hearing commissioner,” which still seems to give priority to the administrative aspect, rather than the judicial aspect, of the role. But the term ultimately adopted is certainly in the spirit of the minority’s recommendations for safeguarding the independence of the official presiding over formal hearings.

In the decades since its enactment, the requirements of the APA have been considerably enlarged or altered by judicial interpretation, sometimes also by administrative or executive practice—but rarely (on matters of substance) by legislative amendment. That suggests that the Congress which originally enacted this set of compromises was not riven with demands for more or fewer or different procedural constraints on agencies. It is not persuasive to depict the APA as emerging from a “pitched political battle for the life of the New Deal” in which “[c]onservatives sought to use administrative reform as a means to stop the New Deal,” as political scientists have contended.⁹⁴Shepherd, supra note 2, at 1676. To make the claims seem plausible, Shepherd assigns numerical scores—on a one-to-ten scale—to indicate the level of procedural burden in various proposed measures, where Walter–Logan earns an eight and the APA a five. The basis of such characterizations is not explained; see also McNollgast, supra note 30, at 183 (noting that “the very future of the New Deal was at stake” so “political preferences over economic outcomes as well as prosaic political strategizing and coalition building played major roles” in the acceptance of the APA). The main evidence offered is that New Deal supporters were more likely to oppose Walter–Logan in 1940 but there were fewer of them by 1946—which might explain why they were more open to compromise but not why their opponents were, still less why the compromise stuck..

Nor is it plausible to see the APA as a merely superficial gesture toward (disguising a substantial retreat from) the ABA’s demands in the 1930s.⁹⁵Shapiro, supra note 30, at 447–52. The successive positions of individual participants seem to bely this. Carl McFarland, who joined the minority wing of the Acheson Committee, had served the New Deal as Assistant Attorney General only a few years before; he would serve in a senior position in the ABA’s administrative law section a few years later. In 1946, he applauded the final version of the APA as satisfying the ABA’s main concerns.⁹⁶Grissinger, supra note 53, at 405.

Larger political trends seem to confirm the same view. Republican majorities gained control of both houses of Congress only a few months after enactment of the APA. They did impose new controls on the National Labor Relations Board in the Taft–Harley Act, which they sustained against President Harry S. Truman’s veto. They did not seek to alter the terms of the APA. Nor did they seek to do so after the 1952 elections, when renewed Republican congressional majorities could rely on a Republican president not to threaten a veto on wider measures. A dozen years later, Democrats won commanding majorities in both houses of Congress, while a Democratic president supported legislation to establish new regulatory agencies and programs. No one seems to have thought it would be logical to protect or enhance these programs by rewriting the APA.

This subsequent history undermines the notion that the APA was, at the time, a painful compromise between supporters and opponents of the New Deal. Another, related reason for doubting such claims is that liberals and conservatives—or regulatory enthusiasts and regulatory skeptics—have changed sides on many fundamental elements of regulatory control. New Dealers preached trust in agency expertise, while liberals of the 1970s and 1980s, welcoming challenges from public interest legal advocates, urged courts to challenge administrative torpor and ensure that agencies take a “hard look” at policy alternatives. Conservatives saw courts as essential guarantors for the rule of law in the 1930s; by the 1980s, as courts took up claims from environmental and consumer advocacy groups, conservatives preached judicial restraint and deference to agency expertise. Conservatives had distrusted presidential authority under FDR in the 1930s but under Republican presidents in the 1970s and 1980s embraced it with enthusiasm. Particular institutional arrangements do not have settled political valiance. The institutional loyalties of parties and constituencies will vary as the surrounding political context changes. That lesson of subsequent history would no doubt have occurred to members of Congress even in 1946.

Still, changes over time do confirm that the text of the APA left many procedural questions open to subsequent development. The drafters of the APA would not have been surprised by the accumulation of such interpolations in ensuing decades. One of the ways the enacted text of the APA does embrace the spirit of the Acheson Committee’s Final Report (apart from its adoption of many particular proposals) is its acknowledgement that there cannot be the same procedure for all agencies or all programs or functions. The text repeatedly recites exceptions or acknowledges cross-cutting statutory directives for particular settings.⁹⁷See, e.g., 5 U.S.C. § 553(a) (containing exceptions for “military or foreign affairs functions” and “agency management, personnel, public property, loans, grants, benefits or contracts”); id.§ 554(a) (containing exceptions for, among others, “selection or tenure of an employee,” “conduct of military or foreign affairs functions,” “decisions [which] rest solely on inspections, tests, or elections,” and “certification of worker representatives”).. The APA reads like the handiwork of lawyers, always mindful of exceptions and possible complications.

The APA has been viewed as the “Constitution” of the administrative state.⁹⁸See, e.g., Morrison, supra note 29, at 253 (“[T]he APA is more like a constitution than a statute.”); Keith Werhan, The Neoclassical Revival in Administrative Law, 44 Admin. L. Rev. 567, 583 n.115 (1992) (“[T]he APA . . . functions more like our Constitution than a statute.”). That is surely misleading in important ways, not least because the APA was not meant to displace the original Constitution which still outranks it and occasionally overshadows it. There is also this difference: unlike the original Constitution, the APA does not start with a stately preamble, reciting its underlying aims.

That is, in fact, another departure from the proposals of the Acheson Committee and perhaps a telling one. The minority was particularly emphatic on this point. They urged that a reform statute should “identify the few basic considerations and express them in legislative statements of policy, of principles, or of standards for the guidance of administrators.”⁹⁹Views of McFarland, Stason, and Vanderbilt, supra note 22, at 214. They noted that statutes commonly included such generalized statements of policy, then echoed Dean Pound’s warnings against a reversion to unbounded government: “To say that man can be so governed [by legal limits] but that the agents of the state cannot or should not be so governed, is a recognition of rejected forms of government.”¹⁰⁰Id. at 214–15. So their proposed bill begins with this “Declaration of general policy”:

The exercise of all powers of government through administrative officers and agencies, so far as such exercise affects rights or withholds or confers benefits or privileges, shall be conducted according to established and published procedures and practices which shall assure the adequate protection of such rights, the impartial conferring of authorized benefits or privileges, and the effectuation of the declared policies of Congress, and shall be adapted to the reasonable necessities and differences of legislation and subject matter involved.¹⁰¹Id. at 217..

Even the majority agreed to place a similar statement at the start of its own proposed bill.¹⁰²Final Report, supra note 5 at 192.. It was a tacit acknowledgement of the concerns expressed by the American Bar Association in the 1930s on the need to reaffirm the “rule of law.” The drafters of the APA dispensed with such hortatory language. There is no general “Declaration of policy” in the APA.¹⁰³There may be an echo of the proposed “Declaration of policy” in the APA provision authorizing reviewing courts to “hold unlawful and set aside agency action, findings, and conclusions found to be . . . arbitrary, capricious, an abuse of discretion or otherwise not in accordance with law.” 5 U.S.C. § 706(2)(A). It is characteristic that this appears as a detail at the end of the APA’s text rather than a clarion call at the outset. Perhaps it would not have made much difference in practice. But this absence left the central issue in the prewar debate buried in the statute’s technical provisions on particular elements of procedure. That did not, of course, suppress the concern to protect rule of law.

One year after enactment of the APA, the Supreme Court had occasion to cite the authority of the Attorney General’s Committee. SEC v. Chenery Corp. (Chenery II)¹⁰⁴332 U.S. 194 (1947). turned on the legality of an SEC ruling, requiring managers of a public utility holding company to surrender their personal holdings of stock in the company, though the Commission had neither issued a general rule on the subject nor indicated that its ruling derived from a general doctrine it would apply in future cases. In upholding the SEC’s decision, the Supreme Court cited the Attorney General’s Committee for the claim that the administrative process requires flexibility.

As it happened, the Court reached this conclusion only by overturning the contrary ruling of the U.S. Court of Appeals for the DC Circuit. That decision had been written by Chief Judge Groner, who had served on the Acheson Committee and sided with the minority.¹⁰⁵Chenery Corp. v. SEC, 154 F.2d 6 (D.C. Cir. 1946). Professor Richard Epstein praises Groner’s opinion for its confident and sensible reading of the relevant statute: “Groner’s view is no crabbed reading of the powers of the administrative state” but a recognition that administrative agencies should not twist words out of context to “make otherwise lawful conduct unlawful.” Richard Epstein, The Dubious Morality of Modern Administrative Law 166 (2020).

The Supreme Court ruling then provoked a scalding dissent from Justice Robert Jackson—who, as attorney general in 1941, had been the first recipient of the Acheson Committee’s Final Report. In his Chenery dissent, Jackson protested the Court’s decision in terms that might have gratified Dean Pound:

The Court’s averment concerning this order, that ‘It is the type of judgment which administrative agencies are best equipped to make and which justifies the use of the administrative process’ is the first instance in which the administrative process is sustained by reliance on that disregard for law which enemies of the process have always alleged to be its principal evil. It is the first encouragement this Court has given to conscious lawlessness as a permissible rule of administrative action. This decision is an ominous one to those who believe that men should be governed by laws that they may ascertain and abide by, and which will guide the action of those in authority as well as of those who are subject to authority.¹⁰⁶Chenery II, 332 U.S. at 217. Cf. Pound, supra note 33, at 134 (“[W]e are told that not . . . to leave administrative agencies at large in this way is technical legalism. Note what it is that is thus stigmatized. It is legalism to require such tribunals to keep within the limits of the jurisdiction and powers given them by the statute creating them. It is legalism to require them to take the policies they apply from the Act of Congress under which they sit and not from their own ideas of particular cases.”).

Justice Jackson ended his dissent with a reminder of his past support for “fostering” the “administrative process.”¹⁰⁷Chenery II, 332 U.S. at 217. He did not need to say he had supported the New Deal, both at the Department of Justice (as Solicitor General, then Attorney General) and as Supreme Court justice.

The debate about the rule of law in the formative period of the APA was not actually a debate about whether to encourage or resist particular regulatory programs. Nor was it primarily about such secondary issues (discussed in the Final Report) as presidential oversight or judicial review or procedures for rulemaking. The debate was principally about reconciling administrative authority with the rule of law. None of those most engaged by this debate in the 1930s—at least on the side of critics—would be surprised to find that this debate is still with us. The rule of law is a challenging and disputed ideal. But it was surely more than a mere slogan in the debates of the 1930s.

One way in which the APA does justify the term “constitution of the administrative state” is that we still argue about what it requires and what it allows—as we do with the more authoritative Constitution. Part of the reason is that proposals that were rejected in 1946 remained in the stream of debate—or at least, in the logic or repertoire of regulatory reform proposals. Formulas affirmed in the APA sometimes proved inadequate to satisfy ongoing impulses—as with the review provision for persons “adversely affected or aggrieved . . . within the meaning of the relevant statute.”¹⁰⁸5 U.S.C. § 702; see also Ass’n of Data Processing Serv. Orgs. v. Camp, 397 U.S. 150, 153–58 (1970). For the most recent survey of ensuing debates, see Caleb Nelson, ‘Standing’ and Remedial Rights in Administrative Law, 105 Va. L. Rev. 703 (2019). The APA sought to settle no more than seemed necessary at the time. That has probably helped explain its durability. It codified what its drafters could agree upon at the time as a guide to sound or fair procedure. For all the limitations in what they set down, they were not simply negotiating a partisan truce in a set of partisan disputes about contending policy priorities. They recognized that debate would continue on many issues but did not dismiss such debate as mere partisan posturing.