1 An Introduction to the Subject 1 An Introduction to the Subject
1.1 What Is Administrative Law? 1.1 What Is Administrative Law?
So you decided to take Administrative Law...
Good choice! One thing I often hear from attorneys is that Administrative Law is the single most relevant doctrinal class they took in law school. Likewise, my colleagues on the faculty often lament that students take their substantive classes (e.g., tax law, immigration law, securities law, environmental law, etc.) without having a firm grounding in background principles of administrative law. Both of these sentiments reflect the fact that administrative law is everywhere.
Why is it everywhere? Well, because, at root, administrative law is the body of law that structures, constrains, and enables the exercise of governmental power. Insofar as the government is interested in regulating things like environmental harms or consumer welfare, or in providing benefits like Social Security disability insurance, administrative law is always in the background, determining the ways that the government, usually acting through instrumentalities called administrative agencies, accomplishes those ends. Think of administrative law as a kind of extension of the Constitution: it fills in the institutional, procedural, and structural gaps that have been revealed in "constitutional law proper" over time as our society and the economy have modernized and the need for government regulation has grown.
This is all rather abstract, but don't worry too much--we'll soon get down to the dirty details. Administrative law is fun to learn and practice in part because it implicates high-minded concerns about things like democracy, rights, efficiency, and justice, but at the same time it forces administrative lawyers into a thicket of procedural rules and into the weeds of science and policy. Whether you've come to this class because you like the high-minded political theory of constitutional law, because you like the minutiae of procedural "how-to" classes like civil procedure, or because you simply like learning about high-stakes, real-world law (a sizeable proportion of the U.S. Supreme Court's docket involves administrative law), there's something for you in administrative law. (If you happen to like all of these things, you'll LOVE administrative law.)
Administrative law comes from many sources. At the highest level of generality, the U.S. Constitution provides important ground rules for the administrative state. Everything that the government does via the administrative state (the somewhat imperfect term we'll use to describe the collection of administrative agencies that comprise the government outside the three main branches of government you'll remember from "Schoolhouse Rock!") has to fit within the powers and rules delineated by the Constitution. Sometimes the Constitution is rather specific about what agencies can do or what they can look like. For instance, the Appointments Clause of Article II of the Constitution, which we will unpack later in the course, is specific: it requires that all Officers of the United States be appointed in very particular ways. Other times, the Constitution is not so specific, and it might even be downright silent on important questions. For instance, scholars and judges today are debating whether the Vesting Clause of Article I of the Constitution restricts Congress's ability to delegate certain of its powers to administrative agencies. Some say yes, but others say no. We will discuss this debate later in the class, but suffice it to say that while the Constitution spells out many important principles of administrative law, it runs out far before definitively answering many of the questions that this class will raise.
Enter the most important law you've probably never heard of: the Administrative Procedure Act of 1946 (APA for short). As many have noted, the APA is a kind of "quasi-Constitution" for the administrative state. See Christopher J. Walker & Scott MacGuidwin, Interpreting the Administrative Procedure Act: A Literature Review, 98 Notre Dame L. Rev. (forthcoming 2023). Some commentators even designate it as a "superstatute" alongside other framework statutes like the Civil Rights Act of 1964 and the Voting Rights Act of 1965. See William N. Eskridge & John Ferejohn, A Republic of Statutes: The New American Constitution (2013). So what, then, does it entail?
The APA sets out a basic set of rules governing 1) how, when, and to what effect agencies may do things with the power delegated to them by Congress and the President, and 2) how courts can be called upon to hold agencies to these rules. These two basic functions of the APA are the basis for two of the major units of this course--one on what agencies are, what they do, and how they do it, and the other on judicial review principles for when agency action (or inaction) is challenged.
So far in our survey of sources of administrative law we have focused on the Constitution and the APA. Notice what both of these sources have in common: they provide trans-substantive rules (i.e., they do not draw any real distinctions between different substantive subject matter areas in which the government might be acting). As far as these sources of law go, an agency action is an agency action is an agency action. Taken as a whole, administrative law is strongly committed to trans-substantivity and the idea that the rules should be the same whether we're talking about developing and enforcing immigration regulations or air pollution controls. But this is not the whole story.
It is extremely important that you keep in mind that sometimes agencies are subject to subject-matter-specific administrative law statutes. In effect, these laws displace the APA when the two conflict, so that the agency subject to such a statute may operate under very different rules than the trans-substantive ones that serve as a default. There's even a section of the APA that makes this explicit. See 5 U.S.C. § 559. In this course, we will often focus more on trans-substantive doctrine (in part because it is more the norm), but sometimes we will venture into areas where a subject-matter-specific administrative law statute governs. Keeping an eye on the source of the administrative law that governs in a particular case, and being analytically precise about what generally applies and what applies only if a certain statute requires it, is a very good practice.
Finally, it is sometimes easy to overlook what some scholars have called "internal administrative law." See Gillian E. Metzger & Kevin M. Stack, Internal Administrative Law, 115 Mich. L. Rev. 1239 (2017). So far, all of the sources of administrative law that we have discussed have been imposed on administrative agencies by other actors, such as Congress, the President, and the courts. Sometimes, however, agencies develop administrative law voluntarily. They might do this rather explicitly, as in codified regulations that spell out procedural rules that the agency commits to follow, or more organically, as in norms or practices that have the effect of eliding some of the discretion that agencies might have. In many cases, these forms of internal administrative law are agency-specific, although in some instances practices are widely shared. While we will not look too closely at many examples of internal administrative law in this course, it is helpful to consider that the administrative law rules that are imposed on agencies do not capture the full extent to which agencies are limited by administrative law.
In sum, administrative law is about what government can and cannot do. By studying administrative law, you will put yourself in a position to be an effective advocate whenever your client is affected by agency actions. Or, if you work for the government, it will give you a roadmap for how you accomplish things without running into legal troubles that can wreck your plans. No wonder, then, that it is so relevant. Once you start to look for administrative law, you begin to see it everywhere the law tries to solve social, economic, or political problems.
Before we move on, a word on this "book." One obvious reason for the book is to save students money. Many administrative law casebooks cost over $250, and that means that in a class of 50 students you will have spent about $12,500 on materials that are largely in the public domain. By putting together this book and offering it for free, I hope to contribute to a fairer and more accessible educational environment.
But I also have my own selfish reasons for putting this book together. Administrative law is a massive subject with endless nooks and crannies, complicated by the fact that it is a politically contested area of the law that is therefore constantly changing. I have been studying administrative law closely now for almost 20 years and I am still discovering new elements, and like everyone, I am on a treadmill, trying to keep up with the many changes to the field that are currently occurring. Many administrative law casebooks try to say a little bit about everything. As new cases come along and displace old ones, many casebooks just add them to an ever-growing list.
This book has a slightly different approach: we will cover many (but not all) of the highlights, and we'll go deeper on a select number of materials rather than trying to exhaustively cover everything. The basic theory is that you will benefit more from learning how to think through a select number of administrative law issues or problems than from developing an encyclopedic knowledge of everything that has ever been said or decided in administrative law. Practically speaking, this means that this casebook contains fewer cases than many administrative law casebooks, which should allow us throughout the semester to spend more time dissecting these materials. The casebook also avoids lengthy post-case notes. Instead, the extra context that you might need will generally be provided in lectures. The downside to these choices is that there is much that we will not cover that is still very much worth knowing about administrative law should you choose to pursue a practice area that involves a heavy dose of the subject. Consider this course to be just a first step towards mastery of the subject, but hopefully one that whets your appetite for more and that gives you the foundations you need to branch out beyond the topics and materials we'll cover here.
1.2 Unpacking the "Administrative State" 1.2 Unpacking the "Administrative State"
In section 1.1, we introduced the term administrative state to refer to the collection of agencies that are the focus of administrative law. In this section, we want to unpack that umbrella concept just a little bit. We'll do that in two ways: 1) by inquiring into the reasons that agencies, and by extension the administrative state, exist in the first place, and 2) by uncovering the variety of agencies and the work they do.
1.2.1. Why Agencies?
A threshold question you may have as you embark on your study of administrative law is why we have agencies at all. The simple answer to this question is that agencies help the more traditional branches of government accomplish their ends—that is, agencies are the agents (in the principal-agent sense) of our legal system. Agencies exist because legislation delegates responsibilities to agencies. In doing so, Congress often leaves agencies with substantial discretion in pursuing certain goals. For instance, Congress might task agencies via statute with serving the “public interest” in a particular subject-matter domain, but it might leave many or even most details about what is actually in the public interest for the agency to figure out. (Note that there may be constitutional limits to how much discretion Congress can leave to agencies, and we will explore how courts have dealt with that question later in the course.)
It might be possible to imagine a world in which Congress, through highly reticulated statutory language, obviates the need to rely on delegation of any tasks to agencies. Perhaps statutes could say with exacting precision everything private actors might need to know about what the law is, and all that remains is enforcement actions brought in a court. (Notice that even this barebones setup requires at least one agency—the Department of Justice—unless we require that the President personally undertake all enforcement actions.) Or perhaps Congress might pass statutes that leave many details unresolved but lodge responsibility for the implementation of the delegated authority in some other institutional actor other than agencies, such as the President or the courts.
In practice, Congress has rarely elected to follow this model. Even going back to the very earliest days of the nation, Congress has passed statutes creating all sorts of agencies and giving them substantial discretion. See, e.g., Nicholas R. Parrillo, A Critical Assessment of the Originalist Case Against Administrative Regulatory Power: New Evidence from the Federal Tax on Private Real Estate in the 1790s, 130 Yale L.J. 1288 (2021); Christine Kexel Chabot, The Lost History of Delegation at the Founding, 56 Georgia L. Rev. 81 (2020); Jerry L. Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law (2012). The Constitution itself seems to contemplate the existence of at least some agencies when it refers to “Heads of Departments” within the executive branch. See U.S. Const. art. II, § 2, cl. 2. Arguably, Congress has scaled up its reliance on this sort of arrangement over the past century or so as it has attempted to use public law and regulation to solve more social problems, but there is much historical practice behind delegation to agencies.
Why might Congress prefer to delegate responsibilities to agencies? This is a very complicated question, but watch the video linked above and think about the advantages of designing a system for implementation of Congress’s statutes this way. Then think about some of the disadvantages. We’ll discuss your thoughts in class. Understanding why Congress designates agencies as the repository of so much of the government’s power goes a long way towards an understanding of administrative law.
1.2.2 The Institutional and Legal Diversity of the Administrative State 1.2.2 The Institutional and Legal Diversity of the Administrative State
Although Congress's reasons for creating agencies and tasking them with delegated responsibility are fairly standard across the administrative state, the specific ways that Congress does this are highly varied. The result is an incredibly diverse administrative state.
To some extent, administrative law resists this diversity by seeking generally trans-substantive rules of law that apply across the administrative state regardless of specific regulatory subject matter (see section 1.1). But, acknowledging that Congress is in the driver's seat, administrative law also allows for Congress to design agencies and structure their operation in particular ways.
For instance, one choice that Congress often makes is between an independent agency and executive agency model. Many factors, such as a President's ability to remove the heads of agencies, term limits, and source of budget, matter to whether an agency is considered independent, and we need not concern ourselves too much with these classifications for now. The important point for now is just that agencies vary on this dimension: for instance, the Federal Trade Commission (FTC), the Federal Energy Regulatory Commission (FERC), the Securities and Exchange Commission (SEC), and the National Labor Relations Board (NLRB) are all widely considered to be independent agencies, whereas the Environmental Protection Agency (EPA), the Occupational Safety and Health Administration (OSHA), and the Department of Homeland Security (DHS) are considered to be executive agencies.
And although it is most common for Congress to give agencies a combination of functions (e.g., rulemaking, enforcing, and adjudicating), sometimes Congress tinkers with this model and gives agencies only very specific authorities. An example of this is the formal division of authority between OSHA and the Occupational Safety and Health Review Commission (OSHRC): the former agency makes rules interpreting relevant occupational safety and health statutes and investigates potential violations, but instead of adjudicating violations itself, it brings these cases to OSHRC, which independently determines whether OSHA is right. Both of these agencies are executive agencies, but they exercise different types of executive power.
You could spend your entire life trying to document the many different agency design choices that Congress makes when it decides to create and empower administrative agencies. Rather than doing that, I'll leave it to you to investigate one agency closely. We'll then spend time in class discussing what you find and examining some of the design choices that Congress made. As you fill out answers to the questions below for a particular agency of your choosing, think about why Congress might have wanted to design the agency the way it did.
Agency Meet-and-Greet
Pick an agency and answer the following questions:
What is your agency’s delegated mission?
Is your agency’s delegated mission clearly defined by statutes?
Does your agency have rulemaking responsibilities, enforcement responsibilities, adjudication responsibilities, or all of the above?
Who leads your agency?
How are the leaders of your agency appointed?
How are the leaders of your agency removed?
Who supervises the leaders of your agency?
What is the source and amount of your agency’s budget?
How many full-time employees (FTEs) does your agency have?
Does your agency have regional offices, or are all operations in Washington, D.C.?
Is your agency considered independent?
In choosing a federal agency to examine, you can consult the following list of agencies (though feel free to identify an off-list agency that piques your interest):
Environmental Protection Agency (EPA)
Food and Drug Administration (FDA)
Department of Health and Human Services (HHS)
Department of Labor (DOL)
Consumer Financial Protection Bureau (CFPB)
Federal Reserve Board (Fed)
Commodity Futures Trading Commission (CFTC)
Federal Energy Regulatory Commission (FERC)
Social Security Administration (SSA)
National Labor Relations Board (NLRB)
Securities and Exchange Commission (SEC)
Federal Trade Commission (FTC)
Department of Homeland Security (DHS)
Customs and Border Protection (CBP)
Citizenship and Immigration Services (USCIS)
Immigration and Customs Enforcement (ICE)
Board of Immigration Appeals (BIA)
Department of Justice (DOJ)
Patent Trial and Appeal Board (PTAB)
Department of Commerce (DOC)
Department of Energy (DOE)
Department of Education (DOE)
Merit Systems Protection Board (MSPB)
National Highway Traffic Safety Administration (NHTSA)
Department of Transportation (DOT)
Occupational Safety and Health Administration (OSHA)
Department of Defense (DOD)
National Endowment for the Arts (NEA)
Postal Service (USPS)
Department of Agriculture (USDA)
You may consult any sources, but you may find much of this information (and much more) in Jennifer L. Selin & David E. Lewis, Administrative Conference of the United States: Sourcebook of United States Executive Agencies (2018), available at https://www.acus.gov/sites/default/files/documents/ACUS%20Sourcebook%20of%20Executive%20Agenices%202d%20ed.%20508%20Compliant.pdf. This terrific source is perhaps the most complete compilation of information about the institutions that comprise the administrative state. Studying it is not required, but it will help you appreciate just how diverse the administrative state is.
1.3 Dealing with Indeterminacy in Administrative Law 1.3 Dealing with Indeterminacy in Administrative Law
Administrative law is a challenging subject to study. No doubt a big part of why it is challenging is because there are many unsettled issues. While this is considered a doctrinal course, and while there are, in fact, many hard rules and standards that you will learn over the course of the semester, there is much room for debate about what administrative law is or should be. In fact, you are coming to the study of administrative law at a particularly turbulent time when many are raising both little and big questions about administrative law. The titles of two recent pieces of scholarship tell you much of the story here. Compare Philip Hamburger, Is Administrative Law Unlawful? (2014) with Adrian Vermeule, No, 93 Tex. L. Rev. 1547 (2015) (reviewing Hamburger, supra). In recent years, the Supreme Court has chipped away at foundational doctrines in administrative law without explicitly overturning them, and it has hinted at its receptivity to critiques of the foundations of the administrative state.
With so much in flux, how are students (or even teachers!) supposed to make any sense of the subject? This section introduces two ways of anchoring administrative law: one focused on realizing and promoting the overall goals of administrative law, and one focused on ostensibly more objective foundations of text and original understanding. While these are likely familiar frameworks to you, they manifest differently in the context of administrative law. Throughout the semester, you will see these two approaches in the reasoning of judicial opinions and other materials. In some sense, practicing administrative law amounts to becoming adept at marshaling these approaches to make arguments about what the law should be when the answers are (as they often are) not clear on their face. Even where precedents seem clear, the flux in administrative law in our current age means that you need to think through these perspectives even when the law seems settled--often, the changes we eventually observe in court opinions got their start when advocates, scholars, and judges began to see weaknesses in that settlement and articulated arguments for change based on the approaches outlined in this subsection.
1.3.1 The Policy Machinery of Administrative Law: Wong Yang Sung v. McGrath 1.3.1 The Policy Machinery of Administrative Law: Wong Yang Sung v. McGrath
One way to deal with indeterminacy in administrative law is try to understand the purpose(s) of administrative law and align decisions with our best sense of those purposes. In fact, courts have been doing precisely this since the very early days of the Administrative Procedure Act. The following case—Wong Yang Sung v. McGrath—is a classic example of the Supreme Court’s efforts to sketch out what might be called the policy machinery of administrative law: that is, the purposes that should motivate administrative law’s development. Much of the background is omitted, but the dispute surrounded whether the APA should be interpreted as applying to deportation proceedings.
As you read the case, think about what purposes the Court is inferring from the history of the Administrative Procedure Act. Are these purposes comprehensive, or can you identify others that the Court is not acknowledging but that should play a role? What are the “opposing social and political forces” which have “come to rest” in the APA? How should the Court go about interpreting what the APA requires when the opposing purposes behind the Act conflict with each other? Where in the APA is the Court getting these purposes?
These are not stale questions. Recently, leading administrative law scholars Cass Sunstein and Adrian Vermeule wrote a book called Law and Leviathan: Redeeming the Administrative State, in which they argued that the administrative state, when it works well, promotes a coherent theory of law's "morality," and they explicitly ground their approach as a "faithful translation or interpretation of Justice jackson's project in Wong Yang Sung." Cass R. Sunstein & Adrian Vermeule, Law and Leviathan: Redeeming the Administrative State 9 (2020).
In class, we'll discuss whether there is an essential purpose, morality, or whatever you want to call it that should animate our interpretation of administrative law. Expect there to be some disagreement...
WONG YANG SUNG v. McGRATH, ATTORNEY GENERAL, et al.
No. 154.
Argued December 6, 1949.
Decided February 20, 1950.
*34Irving Jaffe argued the cause for petitioner. With him on the brief were Jack Wasserman, Gaspare Cusumano and Thomas A. Farrell.
Robert W. Ginnane argued the cause for respondents. With him on the brief were Solicitor General Perlman, Assistant Attorney General Campbell, Robert S. Erdahl, L. Paul Winings and Charles Gordon.
Wendell Berge, A. Alvis Layne, Jr. and John B. Gage filed a brief for Riss & Co., Inc., as amicus curiae, supporting petitioner.
delivered the opinion of the Court.
This habeas corpus proceeding involves a single ultimate question — whether administrative hearings in deportation cases must conform to requirements of the Administrative Procedure Act of June 11, 1946, 60 Stat. 237, 5 U. S. C. §§ 1001 et seg.
Wong Yang Sung, native and citizen of China, was arrested by immigration officials on a charge of being unlawfully in the United States through having overstayed shore leave as one of a shipping crew. A hearing was held before an immigrant inspector who recommended deportation. The Acting Commissioner approved; and the Board of Immigration Appeals affirmed.
Wong Yang Sung then sought release from custody by habeas corpus proceedings in District Court for the District of Columbia, upon the sole ground that the administrative hearing was not conducted in conformity with §§ 5 and 11 of the Administrative Procedure Act.1 *36The Government admitted noncompliance, but asserted that the Act did not apply. The court, after hearing, discharged the writ and remanded the prisoner to custody, holding the Administrative Procedure Act inapplicable to deportation hearings. 80 F. Supp. 235. The Court of Appeals affirmed. 84 U. S. App. D. C. 419, 174 F. 2d 158. Prisoner’s petition for certiorari was not opposed by the Government and, because the question presented has obvious importance in the administration of the immigration laws, we granted review. 338 U. S. 812.
The Administrative Procedure Act of June 11, 1946, supra, is a new, basic and comprehensive regulation of procedures in many agencies, more than a few of which can advance arguments that its generalities should not or do not include them. Determination of questions of its coverage may well be approached through consideration of its purposes as disclosed by its background.
Multiplication of federal administrative agencies and expansion of their functions to include adjudications *37which have serious impact on private rights has been one of the dramatic legal developments of the past half-century.2 Partly from restriction by statute, partly from judicial self-restraint, and partly by necessity — from the nature of their multitudinous and semilegislative or executive tasks — the decisions of administrative tribunals were accorded considerable finality, and especially with respect to fact finding.3 The conviction developed, particularly within the legal profession, that this power was not sufficiently safeguarded and sometimes was put to arbitrary and biased use.4
Concern over administrative impartiality and response to growing discontent was reflected in Congress as early as 1929, when Senator Norris introduced a bill to create *38a separate administrative court.5 Fears and dissatisfactions increased as tribunals grew in number and jurisdiction, and a succession of bills offering various remedies appeared in Congress.6 Inquiries into the practices of state agencies, which tended to parallel or follow the federal pattern, were instituted in several states, and some studies noteworthy for thoroughness, impartiality and vision resulted.7
The Executive Branch of the Federal Government also became concerned as to whether the structure and procedure of these bodies was conducive to fairness in the administrative process. President Roosevelt’s Committee on Administrative Management in 1937 recommended complete separation of adjudicating functions and personnel from those having to do with investigation or prosecution.8 The President early in 1939 also directed the Attorney General to name “a committee of eminent lawyers, jurists, scholars, and administrators to review the entire administrative process in the various *39departments of the executive Government and to recommend improvements, including the suggestion of any needed legislation.” 9
So strong was the demand for reform, however, that Congress did not await the Committee’s report but passed what was known as the Walter-Logan bill, a comprehensive and rigid prescription of standardized procedures for administrative agencies.10 This bill was vetoed by President Roosevelt December 18, 1940,11 and the veto was sustained by the House.12 But the President’s veto message made no denial of the need for reform. Rather it pointed out that the task of the Committee, whose objective was “to suggest improvements to make the process more workable and more just,” had proved “unexpectedly complex.” The President said, “I should desire to await their report and recommendations before approving any measure in this complicated field.”13
The committee divided in its views and both the majority and the minority submitted bills 14 which were introduced in 1941. A subcommittee of the Senate Judiciary Committee held exhaustive hearings on three proposed *40measures,15 but, before the gathering storm of national emergency and war, consideration of the problem was put aside. Though bills on the subject reappeared in 1944,16 they did not attract much attention.
The McCarran-Sumners bill, which evolved into the present Act, was introduced in 1945.17 Its consideration and hearing, especially of agency interests, was painstaking. All administrative agencies were invited to submit their views in writing. A tentative revised bill was then prepared and interested parties again were invited to submit criticisms.18 The Attorney General named representatives of the Department of Justice to canvass the agencies and report their criticisms, and submitted a favorable report on the bill as finally revised.19 It passed both Houses without opposition and was signed by President Truman June 11, 1946.20
The Act thus represents a long period of study and strife; it settles long-continued and hard-fought contentions, and enacts a formula upon which opposing social and political forces have come to rest. It contains many compromises and generalities and, no doubt, some am*41biguities. Experience may reveal defects. But it would be a disservice to our form of government and to the administrative process itself if the courts should fail, so far as the terms of the Act warrant, to give effect to its remedial purposes where the evils it was aimed at appear.
II.
Of the several administrative evils sought to be cured or minimized, only two are particularly relevant to issues before us today. One purpose was to introduce greater uniformity of procedure and standardization of administrative practice among the diverse agencies whose customs had departed widely from each other.21 We pursue this no further than to note that any exception we may find to its applicability would tend to defeat this purpose.
More fundamental, however, was the purpose to curtail and change the practice of embodying in one person or agency the duties of prosecutor and judge. The President’s Committee on Administrative Management voiced in 1937 the theme which, with variations in language, was reiterated throughout the legislative history of the Act. The Committee’s report, which President Roosevelt transmitted to Congress with his approval as “a great document of permanent importance,”22 said:
“. . . the independent commission is obliged to carry on judicial functions under conditions which *42threaten the impartial performance of that judicial work. The discretionary work of the administrator is merged with that of the judge. Pressures and influences properly enough directed toward officers responsible for formulating and administering policy constitute an unwholesome atmosphere in which to adjudicate private rights. But the mixed duties of the commissions render escape from these subversive influences impossible.
“Furthermore, the same men are obliged to serve both as prosecutors and as judges. This not only undermines judicial fairness;' it weakens public confidence in that fairness. Commission decisions affecting private rights and conduct lie under the suspicion of being rationalizations of the preliminary findings which the commission, in the role of prosecutor, presented to itself.” Administrative Management in the Government of the United States, Report of the President’s Committee on Administrative Management, 36-37 (1937).
The Committee therefore recommended a redistribution of functions within the regulatory agencies. “[I]t would be divided into an administrative section and a judicial section” and the administrative section “would formulate rules, initiate action, investigate complaints . . .” and the judicial section “would sit as an impartial, independent body to make decisions affecting the public interest and private rights upon the basis of the records and findings presented to it by the administrative section.” Id. at 37.
Another study was made by a distinguished committee named by the Secretary of Labor, whose jurisdiction at the time included the Immigration and Naturalization Service. Some of the committee’s observations have relevancy to the procedure under examination here. It said:
*43“The inspector who presides over the formal hearing is in many respects comparable to a trial judge. He has, at a minimum, the function of determining— subject to objection on the alien’s behalf — what goes into the written record upon which decision ultimately is to be based. Under the existing practice he has also the function of counsel representing the moving party — he does not merely admit evidence against the alien; he has the responsibility of seeing that such evidence is put into the record. The precise scope of his appropriate functions is the first question to be considered.” The Secretary of Labor’s Committee on Administrative Procedure, The Immigration and Naturalization Service, 77 (Mimeo. 1940).
Further:
“Merely to provide that in particular cases different inspectors shall investigate and hear is an insufficient guarantee of insulation and independence of the presiding official. The present organization of the field staff not only gives work of both kinds commonly to the same inspector but tends toward an identity of viewpoint as between inspectors who are chiefly doing only one or the other kind of work. . . .
“. . . We recommend that the presiding inspectors. be relieved of their present duties of presenting the case against aliens and be confirmed [sic] entirely to the duties customary for a judge. This, of course, would require the assignment of another officer to perform the task of a prosecuting attorney. The appropriate officer for this purpose would seem to be the investigating inspector who, having prepared the case against the alien, is already thoroughly familiar with it. . . .
*44“A genuinely impartial hearing, conducted with critical detachment, is psychologically improbable if not impossible, when the presiding officer has at once the responsibility of appraising the strength of the case and of seeking to make it as strong as possible. Nor is complete divorce between investigation and hearing possible so long as the presiding inspector has the duty himself of assembling and presenting the results of the investigation. . . .” Id. at 81-82.
And the Attorney General’s Committee on Administrative Procedure, which divided as to the appropriate remedy,23 was unanimous that this evil existed. Its Final Report said:
“These types of commingling of functions of investigation or advocacy with the function of deciding are thus plainly undesirable. But they are also avoidable and should be avoided by appropriate internal division of labor. For the disqualifications produced by investigation or advocacy are personal psychological ones which result from engaging in those types of activity; and the problem is simply one of isolating those who engage in the activity. Creation of independent hearing commissioners insulated from all phases of a case other than hearing and deciding will, the Committee believes, go far toward solving this problem at the level of the initial hearing provided the proper safeguards are established to assure the insulation. . . .” Rep. Atty. Gen. Comm. Ad. Proc. 56 (1941), S. Doc. No. 8, 77th Cong., 1st Sess. 56 (1941).
The Act before us adopts in general this recommended form of remedial action. A minority of the Committee had, furthermore, urged an even more thoroughgoing *45separation and supported it with a cogent report. Id. at 203 et seq.
Such were the evils found by disinterested and competent students. Such were the facts before Congress which gave impetus to the demand for the reform which this Act was intended to accomplish. It is the plain duty of the courts, regardless of their views of the wisdom or policy of the Act, to construe this remedial legislation to eliminate, so far as its text permits, the practices it condemns.
III.
Turning now to the case before us, we find the administrative hearing a perfect exemplification of the practices so unanimously condemned.
This hearing, which followed the uniform practice of the Immigration Service,24 was before an immigrant inspector, who, for purposes of the hearing, is called the “presiding inspector.” Except with consent of the alien, the presiding inspector may not be the one who investigated the case. 8 C. F. R. 150.6 (b).25 But the inspector’s duties include investigation of like cases; and while he is today hearing cases investigated by a colleague, tomorrow his investigation of a case may be heard before the inspector whose case he passes on today. An “examining inspector” may be designated to conduct the prosecution, 8 C. F. R. 150.6 (n), but none was in this case; and, in any event, the examining inspector also has the same mixed prosecutive and hearing functions. The presiding *46inspector, when no examining inspector is present, is required to “conduct the interrogation of the alien and the witnesses in behalf of the Government and shall cross-examine the alien’s witnesses and present such evidence as is necessary to support the charges in the warrant of arrest.” 8 C. F. R. 150.6 (b). It may even become his duty to lodge an additional charge against the alien and proceed to hear his own accusation in like manner. 8 C. F. R. 150.6 (1). Then, as soon as practicable, he is to prepare a summary of the evidence, proposed findings of fact, conclusions of law, and a proposed order. A copy is furnished the alien or his counsel, who may file exceptions and brief, 8 C. F. R. 150.7, whereupon the whole is forwarded to the Commissioner. 8 C. F. R. 150.9.
The Administrative Procedure Act did not go so far as to require a complete separation of investigating and prosecuting functions from adjudicating functions. But that the safeguards it did set up were intended to ameliorate the evils from the commingling of functions as exemplified here is beyond doubt. And this commingling, if objectionable anywhere, would seem to be particularly so in the deportation proceeding, where we frequently meet with a voteless class of litigants who not only lack the influence of citizens, but who are strangers to the laws and customs in which they find themselves involved and who often do not even understand the tongue in which they are accused. Nothing in the nature of the parties or proceedings suggests that we should strain to exempt deportation proceedings from reforms in administrative procedure applicable generally to federal agencies.
Nor can we accord any weight to the argument that to apply the Act to such hearings will cause inconvenience and added expense to the Immigration Service. Of course it will, as it will to nearly every agency to which it is applied. But the power of the purse belongs to Congress, and Congress has determined that the price *47for greater fairness is not too high. The agencies, unlike the aliens, have ready and persuasive access to the legislative ear and if error is made by including them, relief from Congress is a simple matter.
This brings us to contentions both parties have advanced based on the pendency in Congress of bills to exempt this agency from the Act. Following an adverse decision,26 the Department asked Congress for exempting legislation,27 which appropriate committees of both Houses reported favorably but in different form and substance.28 Congress adjourned without further action. The Government argues that Congress knows that the Immigration Service has construed the Act as not applying to deportation proceedings, and that it “has taken no action indicating disagreement with that interpretation”; that therefore it “is at least arguable that Congress was prepared to specifically confirm the administrative construction by clarifying legislation.” We do not think we can draw that inference from incompleted steps in the legislative process. Cf. Helvering v. Hallock, 309 U. S. 106, 119-120.
On the other hand, we will not draw the inference, urged by petitioner, that an agency admits that it is acting upon a wrong construction by seeking ratification from Congress. Public policy requires that agencies feel free to ask legislation which will terminate or avoid adverse contentions and litigations. We do not feel justified in holding that a request for and failure to get in a single session of Congress clarifying legislation on a genuinely debatable point of agency procedure admits weakness in the agency’s contentions. We draw, therefore, no inference in favor of either construction of the Act — from the *48Department’s request for legislative clarification, from the congressional committees’ willingness to consider it, or from Congress’ failure to enact it.
We come, then, to examination of the text of the Act to determine whether the Government is right in its contentions: first, that the general scope of § 5 of the Act does not cover deportation proceedings; and, second, that even if it does, the proceedings are excluded from the requirements of the Act by virtue of § 7.
IV.
The Administrative Procedure Act, § 5, establishes a number of formal requirements to be applicable “In every case of adjudication required by statute to be determined on the record after opportunity for an agency hearing.” The argument here depends upon the words “adjudication required by statute.” The Government contends that there is no express requirement for any hearing or adjudication in the statute authorizing deportation,29 and that this omission shields these proceedings from the impact of § 5. Petitioner, on the other hand, contends that deportation hearings, though not expressly required by statute, are required under the decisions of this Court,30 *49and the proceedings, therefore, are within the scope of § 5.
Both parties invoke many citations to legislative history as to the meaning given to these key words by the framers, advocates or opponents of the Administrative Procedure Act. Because § 5 in the original bill applied to hearings required “by law,” 31 because it was suggested by the Attorney General that it should be changed to “required by statute or Constitution,” 32 and because it finally emerged “required by statute,” the Government argues that the section is intended to apply only when explicit statutory words granting a right to adjudication can be pointed out. Petitioner on the other hand cites references which would indicate that the limitation to statutory hearing was merely to avoid creating by inference a new right to hearings where no right existed otherwise. We do not know. The legislative history is more conflicting than the text is ambiguous.
But the difficulty with any argument premised on the proposition that the deportation statute does not require a hearing is that, without such hearing, there would be no constitutional authority for deportation. The constitutional requirement of procedural due process of law derives from the same source as Congress’ power to legislate and, where applicable, permeates every valid enactment of that body. It was under compulsion of the Constitution that this Court long ago held that an antecedent deportation statute must provide a hearing at least for aliens who had not entered clandestinely and *50who had been here some time even if illegally. The Court said:
“This is the reasonable construction of the acts of Congress here in question, and they need not be otherwise interpreted. In the case of all acts of Congress, such interpretation ought to be adopted as, without doing violence to the import of the words used, will bring them into harmony with the Constitution.” The Japanese Immigrant Case, 189 U. S. 86, 101.
We think that the limitation to hearings “required by statute” in § 5 of the Administrative Procedure Act exempts from that section’s application only those hearings which administrative agencies may hold by regulation, rule, custom, or special dispensation; not those held by compulsion. We do not think the limiting words render the Administrative Procedure Act inapplicable to hearings, the requirement for which has been read into a statute by the Court in order to save the statute from invalidity. They exempt hearings of less than statutory authority, not those of more than statutory authority. We would hardly attribute to Congress a purpose to be less scrupulous about the fairness of a hearing necessitated by the Constitution than one granted by it as a matter of expediency.
Indeed, to so construe the Immigration Act might again bring it into constitutional jeopardy. When the Constitution requires a hearing, it requires a fair one, one before a tribunal which meets at least currently prevailing standards of impartiality. A deportation hearing involves issues basic to human liberty and happiness and, in the present upheavals in lands to which aliens may be returned, perhaps to life itself. It might be difficult to justify as measuring up to constitutional standards of impartiality a hearing tribunal for deportation proceed*51ings the like of which has been condemned by Congress as unfair even where less vital matters of property rights are at stake.
We hold that the Administrative Procedure Act, § 5, does cover deportation proceedings conducted by the Immigration Service.
V.
The remaining question is whether the exception of § 7 (a) of the Administrative Procedure Act exempts deportation hearings held before immigrant inspectors. It provides:
“Sec. 7. In hearings which section 4 or 5 requires to be conducted pursuant to this section- — •
“(a) Presiding officers. — There shall preside at the taking of evidence (1) the agency, (2) one or more members of the body which comprises the agency, or (3) one or more examiners appointed as provided in this Act; but nothing in this Act shall be deemed to supersede the conduct of specified classes of proceedings in whole or part by or before boards or other officers specially provided for by or designated pursuant to statute. . . .” 60 Stat. 237, 241, 5 U. S. C. § 1006.
The Government argues that immigrant inspectors are “specially provided for by or designated pursuant to” § 16 of the Immigration Act, which, in pertinent part, reads:
“. . . The inspection ... of aliens, including those seeking admission or readmission to or the privilege of passing through or residing in the United States, and the examination of aliens arrested within the United States under this Act,33 shall be conducted by *52immigrant inspectors, except as hereinafter provided in regard to boards of special inquiry. . . . Said inspectors shall have power to administer oaths and to take and consider evidence touching the right of any alien to enter, reenter, pass through, or reside in the United States, and, where such action may be necessary, to make a written record of such evidence; . . 39 Stat. 874, 885, as amended, 8 U. S. C. § 152.
Certainly nothing here specifically provides that immigrant inspectors shall conduct deportation hearings or be designated to do so. This language does direct them to conduct border inspections of aliens seeking admission. They may administer oaths and take, record, and consider evidence. But these functions are indispensable to investigations which are concededly within their competence. And these functions are likewise necessary to enable the preparation of complaints for prosecutive purposes. But that Congress by grant of these powers has specially constituted them or provided for their designation as hearing officers in deportation proceedings does not appear.
Section 7 (a) qualifies as presiding officers at hearings the agency and one or more of the members of the body comprising the agency, and it also leaves untouched any others whose responsibilities and duties as hearing officers are established by other statutory provision. But if hearings are to be had before employees whose responsibility and authority derives from a lesser source, they must be examiners whose independence and tenure are so guarded by the Act as to give the assurances of neutrality which Congress thought would guarantee the impartiality of the administrative process.
We find no basis in the purposes, history or text of this Act for judicially declaring an exemption in favor of deportation proceedings from the procedural safeguards *53enacted for general application to administrative agencies. We hold that deportation proceedings must conform to the requirements of the Administrative Procedure Act if resulting orders are to have validity. Since the proceeding in the case before us did not comply with these requirements, we sustain the writ of habeas corpus and direct release of the prisoner.*
Reversed.
Mr. Justice Douglas and Mr. Justice Clark took no part in the consideration or decision of this case.
Particularly invoked are §5 (c), 60 Stat. 237, 240, 5 U. S. C. § 1004 (c), which provides in part:
“The same officers who preside at the reception of evidence pursuant to section 7 shall make the recommended decision or initial decision required by section 8 except where such officers become unavailable to the agency. Save to the extent required for the disposition of ex parte matters as authorized by law, no such officer shall consult any person or party on any fact in issue unless upon notice and opportunity for all parties to participate; nor shall such officer be responsible to or subject to the supervision or direction of any officer, employee, or agent engaged in the performance of investigative or prosecuting functions for any agency. No officer, employee, or agent engaged in the performance of investigative or prosecuting functions for any agency in any case shall, in that or a factually related ease, participate or advise in the decision, recommended decision, or agency review pursuant to section 8 except as witness or counsel in public proceedings. . . .”; and § 11, 60 Stat. at 244, 5 ■U. S. C. § 1010, which provides in part: “Subject to the civil-service *36and other laws to the extent not inconsistent with this Act, there shall be appointed by and for each agency as many qualified and competent examiners as may be necessary for proceedings pursuant to sections 7 and 8, who shall be assigned to cases in rotation so far as practicable and shall perform no duties inconsistent with their duties and responsibilities as examiners. Examiners shall be removable by the agency in which they are employed only for good cause established and determined by the Civil Service Commission (hereinafter called the Commission) after opportunity for hearing and upon the record thereof. Examiners shall receive compensation prescribed by the Commission independently of agency recommendations or ratings and in accordance with the Classification Act of 1923, as amended, except that the provisions of paragraphs (2) and (3) of subsection (b) of section 7 of said Act, as amended, and the provisions of section 9 of said Act, as amended, shall not be applicable. . . .”
See e. g., Blachly and Oatman, Administrative Legislation and Adjudication 1 (1934); Landis, The Administrative Process 1 (1938); Pound, Administrative Law 27 (1942); Carrow, The Background of Administrative Law 1 (1948); The Federal Administrative Procedure Act and the Administrative Agencies 4 (N. Y. U. 1947); Final Report of Attorney General’s Committee on Administrative Procedure 7 (1941), contained in S. Doc. No. 8, 77th Cong., 1st Sess. (1941); Cushman, The Independent Regulatory Commissions, cc. II-V (1941); Frankfurter, The Task of Administrative Law, 75 U. of Pa. L. Rev. 614 (1927); materials cited in n. 4, infra.
See e. g., Dickinson, Administrative Justice and the Supremacy of Law, passim (1927); Final Report of Attorney General’s Committee on Administrative Procedure, supra, at 11-18, 75-92; and see materials cited in n, 4, infra.
E. g., Root, Public Service by the Bar, 41 A. B. A. Rep. 355, 368 (1916); Hughes, Some Aspects of the Development of American Law, 39 N. Y. B. A. Rep. 266, 269 (1916); Sutherland, Private Rights and Government Control, 42 A. B. A. Rep. 197, 205 (1917); Address of President Guthrie, 46 N. Y. B. A. Rep. 169, 186 (1923). After 1933, when the American Bar Association formed a Special Committee on Administrative Law, the Bar’s concern can be traced in this Committee’s reports. E. g., 58 A. B. A. Rep. 197, 407 (1933) ; 59 A. B. A. Rep. 539 (1934); 61 A. B. A. Rep. 720 (1936); 62 A. B. A. Rep. 789 (1937).
S. 5154,70th Cong., 2d Sess. (1929).
S. 1835, 73d Cong., 1st Sess. (1933); S. 3787, H. R. 12297, 74th Cong., 2d Sess. (1936); S. 3676, 75th Cong., 3d Sess. (1938); H. R. 6324, H. R. 4235, H. R. 4236, S. 915, S. 916, 76th Cong., 1st Sess. (1939); S. 674, S. 675, S. 918, H. R. 3464, H. R. 4238, H. R. 4782, 77th Cong., 1st Sess. (1941); H. R. 4314, H. R. 5081, H. R. 5237, S. 2030, 78th Cong., 2d Sess. (1944); H. R. 1203, S. 7, 79th Cong., 1st Sess. (1945).
E. g., Benjamin, Administrative Adjudication in the State of New York (1942); Tenth Biennial Report of the Judicial Council to the Governor and Legislature of California (1944). See also Fesler, The Independence of State Regulatory Agencies (1942); Handbook of the National Conference of Commissioners on Uniform State Laws, 226 et seq. (1943); 63 A. B. A. Rep. 623 (1938).
Administrative Management in the Government of the United States, Report of the President’s Committee on Administrative Management 37 (1937).
The quoted statement is from President Roosevelt’s message to Congress of December 18, 1940, vetoing H. R. 6324, the so-called Walter-Logan bill. H. R. Doc. No. 986, 76th Cong., 3d Sess., 3-4 (1940). The origin and orders leading to the creation of the Attorney General’s Committee are set out in Appendix A of the Committee’s Final Report, supra.
S. 915, H. R. 6324,76th Cong., 1st Sess. (1939).
86 Cong. Rec. 13942-3 (1940), reprinted in H. R. Doc. No. 986, 76th Cong., 3d Sess. (1940).
86 Cong. Rec. 13953 (1940).
86 Cong. Rec. at 13943; H. R. Doc. No. 986, supra, 4.
These bills appear at pp. 192 and 217 of the Committee’s Final Report, supra. The majority bill became S. 675, 77th Cong., 1st Sess. (1941) and the minority recommendation was embodied in S. 674, 77th Cong., 1st Sess. (1941).
The hearings ran from April 2 to July 2, 1941, and, with an appendix, have been collected in four parts and over 1,600 pages. Hearings before Subcommittee of the Committee on the Judiciary on S. 674, S. 675 and S. 918,77th Cong., 1st Sess. (1941).
H. R. 4314, H. R. 5081, H. R. 5237, S. 2030, 78th Cong., 2d Sess. (1944).
S. 7 and H. R. 1203,79th Cong., 1st Sess. (1945).
See H. R. Rep. No. 1980, 79th Cong., 2d Sess. 14-15 (1946); S. Rep. No. 752, 79th Cong., 1st Sess. 4-5 (1945), reprinted in S. Doc. No. 248, 79th Cong., 2d Sess., at 233, 248-249, and 185, 190-191, respectively.
S. Rep. No. 752, 79th Cong., 1st Sess. 37-45 (1945); 92 Cong. Rec. App. A-2982-5 (1946).
92 Cong. Rec. 2167 (1946) (passage by the Senate); 92 Cong. Rec. 5668 (1946) (amended version passed by House); 92 Cong. Rec. 5791 (1946) (House version agreed to by Senate); 92 Cong. Rec. 6706 (1946) (approved by the President).
H. R. Rep. No. 1980, 79th Cong., 2d Sess. 16 (1946); Final Report of the Attorney General’s Committee on Administrative Procedure, 20 (1941); McFarland, Analysis of the Federal Administrative Procedure Act, in Federal Administrative Procedure Act and the Administrative Agencies 16, 22 (N. Y. U. 1947). See also Hearings before Subcommittee No. 4 of the House Committee on the Judiciary on H. R. 4236, H. R. 6198, and H. R. 6324, 76th Cong., 1st Sess. 14, 31 (1939); S. Rep. No. 442, 76th Cong., 1st Sess. 9 (1939); H. R. Rep. No. 1149, 76th Cong., 1st Sess. 2-3 (1939); S. Doc. No. 71, 76th Cong., 1st Sess. 5 (1939).
81 Cong. Rec. 187, 191 (1937).
See n. 14, supra.
See 8 C. F. R. 150.1 et seq.
The initial step in a deportation case is the investigation of an alien by an immigrant inspector. 8 C. F. R. 150.1. This is followed by issuance of a warrant of arrest, 8 C. F. R. 150.2-150.4, and incarceration, unless the alien is released under bond. 8 C. F. R. 150.5. The formal hearing follows.
Eider v. Clark (D. D. C. 1948), 77 F. Supp. 610.
S. 2755 and H. R. 6652, 80th Cong., 2d Sess. (1948).
S. Rep. No. 1588, H. R. Rep. No. 2140, 80th Cong., 2d Sess. (1948).
Section 19 (a) of the Immigration Act of February 5, 1917, 39 Stat. 874, 889, as amended, 8 U. S. C. § 155 (a), provides in part:
“. . . any alien who shall have entered or who shall be found in the United States in violation of this Act, or in violation of any other law of the United States . . . shall, upon the warrant of the Attorney General, be taken into custody and deported. ... In every case where any person is ordered deported from the United States under the provisions of this Act, or of any law or treaty, the decision of the Attorney General shall be final.” See Note 33, infra.
The Japanese Immigrant Case, 189 U. S. 86, 100, 101; Kwock Jan Fat v. White, 253 U. S. 454, 459, 464; Bridges v. Wixon, 326 U.S. 135, 160 (concurringopinion).
Section 301 of the bills proposed in the majority and minority recommendations of the Final Report of the Attorney General's Committee on Administrative Procedure, pp. 195, 232-233.
Hearings before a Subcommittee of the Senate Committee on the Judiciary on S. 674, S. 675 and S. 918, 77th Cong., 1st Sess. 1456 (1941).
The original Act, 39 Stat. 886, reads “under this Act,” although in the codification, 8 U. S. C. § 152, it reads “under this section.” , The former is controlling. 1 U. S. C. (Supp. II, 1949) §§ 112, 204 (a).
[For order modifying the judgment, see post, p. 908.]
dissenting.
The Court, it seems to me, has disregarded a congressional exemption of certain agencies, including the Immigration and Naturalization Service, from some of the requirements of the Administrative Procedure Act. Such judicial intrusion into the legislative domain justifies a protest. It may be useful to call attention to the necessity of recognizing specific exceptions to general rules. This protest is rested on the ground that immigrant inspectors performing duties under § 16 of the Immigration Act are within the exception provided by § 7 (a) of the Administrative Procedure Act. The Court's opinion discusses this point under subdivision V. The sections are there set out and can be examined by the reader.
In this case no one questions the constitutionality of the hearing Wong received before the immigrant inspector, with administrative review by the Commissioner and the Board of Immigration Appeals. The question on which I disagree with the Court is whether the Administrative Procedure Act permits an inspector of the Immigration and Naturalization Service to serve as a presiding officer at a deportation hearing.
*54Section 7 (a) of the Administrative Procedure Act provides that the official presiding at the taking of evidence shall be an agency, an agency member or an examiner appointed under that Act. There is an exception to this requirement. It reads as follows:
“but nothing in this Act shall be deemed to supersede the conduct of specified classes of proceedings in whole or part by or before boards or other officers specially provided for by or designated pursuant to statute.”
It is this exception that made it proper for an immigrant inspector to preside at this deportation hearing.
Under § 16 of the Immigration Act, 39 Stat. 874, 885, the
“inspection ... of aliens, including those seeking admission or readmission to or the privilege of passing through or residing in the United States, and the examination of aliens arrested within the United States under this Act, shall be conducted by immigrant inspectors, .... Said inspectors shall have power to administer oaths and to take and consider evidence touching the right of any alien to enter, reenter, pass through, or reside in the United States, and, where such action may be necessary, to make a written record of such evidence; . . . .”
It seems to me obvious that the exception provided in § 7 (a) covers immigrant inspectors dealing with the arrest of an alien for violation of the Immigration Act. The examination of arrested aliens at a deportation proceeding is surely a specified class of proceedings under § 7 (a) of the Administrative Procedure Act, and it is surely conducted by an officer “specially provided for by . . . statute.”
The reason for the exception in § 7 (a) was not spelled out in the legislative history or in the Act itself. The *55exception may have been made to retain smoothness of operation in the several agencies where there were officials specially provided for by statute or designated pursuant to a statute. When making exceptions from the requirements as to separation of the investigatory and adjudicatory functions, it was natural to include officers specially designated by statute to sit in judgment. Agency members are excluded from these requirements of the Administrative Procedure Act. They, too, have investigatory and adjudicatory duties. Since the members of the agency and the statutorily designated officers were specially selected for the functions they were to perform, Congress probably reposed confidence in their experience and expertness. It doubtless did not wish to disorganize administration until time showed whether that confidence was well placed.1
Since the Court does not accept my view of the reach of § 7 (a), it would be useless to undertake an analysis of the other questions presented by the petition for certiorari.
Thus the congressional committee warned that should the exception “be a loophole for avoidance of the examiner system in any real sense, corrective legislation would be necessary. That provision is not intended to permit agencies to avoid the use of examiners but to preserve special statutory types of hearing officers who contribute something more than examiners could contribute and at the same time assure the parties fair and impartial procedure.” S. Doc. No. 248, 79th Cong., 2d Sess., p. 216.
1.3.2 The Limits of the Policy Machinery of Administrative Law 1.3.2 The Limits of the Policy Machinery of Administrative Law
As we just saw with Wong Yang Sung, one approach to interpreting ambiguities in the "fierce compromise" that is the APA, see George B. Shepherd, Fierce Compromise: The Administrative Procedure Act Emerges from New Deal Politics, 90 Nw. U. L. Rev. 1557 (1996), is to refer to the underlying purposes of the law. However, as you may or may not have discovered in your legislation and regulation class, the idea of interpreting statutes according to their purposes is somewhat controversial. The ascendance of textualism in the federal courts has led to more suspicion of such purposivist approaches to interpretation.
In the excerpt that follows, the authors survey how courts and scholars have started to move away from the policy balancing approach of Wong Yang Sung and to incoporate more of the textualism and originalism that is in vogue in contemporary jurisprudence.
As you read the following excerpt, think about whether textualism and originalism can provide clear enough answers to entirely obviate the need for the kind of policy balancing that Wong Yang Sung and the trichotomy endorse. As the authors conclude, APA originalism and APA textualism have many virtues, but they are not quite willing to follow it all the way to its logical conclusions. Why not? You likely will not have enough information at the start of this course to completely evaluate these questions, but it helps to keep the question in mind, especially since many of the most avulsive changes in administrative law today are motivated by originalist/textualist commitments on the part of judges.
Christopher J. Walker & Scott MacGuidwin
Interpreting the Administrative Procedure Act: A Literature Review
98 Notre Dame L. Rev. 1963 (2023)
Courts and scholars have employed numerous techniques in interpreting the APA, each stressing to different degrees the text, history, structure, purposes, and practical effects of the APA. These varied interpretative approaches, however, can roughly be grouped into four theories: (A) APA pragmatism; (B) administrative common lawmaking; (C) APA textualism; and (D) APA originalism. In many ways, each subsequent interpretive theory evolved in response to the prior one(s), though all are still active and evolving theories today. . . .
The predominant approach judges have historically employed when interpreting the APA has been a textually constrained purposivism. Under this approach, the text of the APA provides the ceiling but where the text, structure, or context is ambiguous, courts adopt a reading of the text that best advances rule-of-law values in administrative law, such as accountability, consistency, efficiency, expertise, predictability, stability, and transparency. In that sense, perhaps the best label for this predominant approach is APA pragmatism. APA pragmatists are likely to find more provisions ambiguous than APA textualists or originalists. But unlike administrative common law jurists, those exercising this approach choose to work within the framework of the text of the APA. . . .
So who are the administrative pragmatists when it comes to interpreting the APA? The answer is probably most administrative law scholars and many, many judges across the country. . . . Although there will no doubt be disagreements on particulars, these are the widely shared views of the leadership and membership of the American Bar Association (ABA) Section of Administrative Law and Regulatory Practice and the Administrative Conference of the United States (ACUS). Indeed, in 2016, the ABA Administrative Law Section sponsored a resolution adopted by the ABA House of Delegates that urged Congress to formally codify a number of these pragmatic judicial interpretations, including the Portland Cement doctrine and the agency rulemaking record requirement.
We also see administrative pragmatism embraced by the Roberts Court. Two prominent examples from the Trump Administration come immediately to mind. First, in Department of Commerce v. New York (the census citizenship question case), the Court held that under APA judicial review, an agency must “offer genuine justifications for important decisions, reasons that can be scrutinized by courts and the interested public.” In other words, reasons that are pretextual are not sufficient. Second, in Department of Homeland Security v. Regents of University of California (the DACA immigration relief rescission case), the Court held that arbitrary-and-capricious review under the APA requires the agency, when changing an existing agency policy, to consider reasonable regulatory alternatives and to demonstrate that it has adequately considered the reliance interests at stake in changing the regulatory baseline. Both decisions provide additional pragmatic gloss on the judicial review provisions of the APA. . . .
In the 1970s and 1980s, the D.C. Circuit not only recognized administrative law doctrines that seem to have at least some plausible textual basis in or gloss on the APA, but it also developed expansive administrative common law based on broader constitutional and APA quasi-constitutional values. As Kenneth Culp Davis famously observed in 1980, “Most administrative law is judge-made law, and most judge-made administrative law is administrative common law.” Today, the latter observation may be less true, with the rise of APA textualism and originalism. But Jack Beermann’s observation in 2011 strikes us as still accurate: “The most that one can confidently say today is that administrative law contains elements that appear to be highly statutorily focused alongside elements in which courts exercise the discretion of a common law court.”
Administrative common law jurists argue that the APA should be viewed more like a common law statute—like the Sherman Antitrust Act—where Congress is inviting courts to adopt additional practices and procedures to improve administrative governance. Or in the Eskridge-Ferejohn terminology, these framework statutes are “super- statutes,” which are interpreted and “applied in accord with a pragmatic methodology that is a hybrid of standard precepts of statutory, common law, and constitutional interpretation.”
Those exercising the administrative common lawmaking approach are less constrained by the text than APA pragmatists, textualists, or originalists. Common-law jurists see the text of the APA as a floor and an invitation for courts to innovate. Even if a decision contradicts the text of the APA, the benefits of stare decisis, agency reliance, and effective administrative governance exceed any benefits of confining administrative law to the APA’s text or express purposes. Under this administrative common lawmaking approach, Jack Beermann explains, “doctrinal systems governing important areas of administrative law become so well-developed that it becomes virtually unnecessary to refer to the text of the APA when deciding cases concerning APA provisions.”
Consider, for instance, Alan Morrison’s classic 1986 articulation of this “dynamic” or “living” approach to interpreting the APA:
[T]he APA is more like a constitution than a statute. It provides for flexibility in decision-making; it can be changed through interpretation without the need for amendment; its movements are more pendulum-like than linear. Its fundamental role is to shape the relationship between the people and their government, giving the government considerable leeway in carrying out the substantive laws that Congress has enacted, while at the same time providing the governed with a considerable degree of procedural protection. Considering the vast expansion of governmental regulation since 1946, it is a real tribute to the drafters of the APA that it remains the centerpiece of administrative procedure with virtually no amendments.
To commemorate the APA’s fiftieth anniversary, Peter Strauss returned to this theme of a living, common-law APA, bemoaning the increasingly textualist Supreme “Court’s new and static approach” to interpreting the APA, which “denies the possibilities of accepted accommodation and of adjustment that are such central elements of the common law.”
In recent years, Gillian Metzger has been perhaps the most forceful advocate for administrative common law. Professor Metzger argues that the difficulty of amending the APA puts more pressure on courts to ensure the APA “functions in a way that serves Congress’s interests.” She argues that administrative common law is effective as a constitutional avoidance technique—permitting courts to avoid potential separation of powers problems. She believes that common law doctrines are necessary to prevent “statutory obsolescence.” . . . Professor Metzger further argues that Section 559 of the APA provides a textual hook for administrative common law. That is because it states that the APA provisions “do not limit or repeal additional requirements imposed by statute or otherwise recognized by law.”Given that “statute” is expressly listed, she argues that “otherwise recognized by law” is likely judge-created law.
. . . It is fair to wonder how to draw the line between textually constrained, yet purposivist APA pragmatism and the more dynamic or living interpretive approach that embraces and encourages administrative common lawmaking. Some, no doubt, would view the examples of APA pragmatism provided in Part I.A as really the more dynamic or living category. Others may view examples of administrative common law as more appropriately falling within APA pragmatism, having at least some plausible APA textual hook. Indeed, as Ron Levin has observed, “the line of division between creative statutory interpretation and overt common law is blurry and frequently inconsequential (although it does matter in some contexts, such as for purposes of the Vermont Yankee doctrine).” This is further complicated by the fact, as John Duffy chronicles, that many of the APA pragmatism interpretations were originally developed as administrative common law, and then APA pragmatists and textualists subsequently attempted to ground them in the text and structure of the APA. . . .
Textualism has become the increasingly dominant approach to interpreting the APA in recent years. APA textualists start (and often end) with the text of the statute, focusing closely on the text, employing the various canons of statutory interpretation, and considering structural arguments. The first wave of APA textualists seems to have consisted mostly of conservatives and libertarians. In the legal academy, Professor Duffy is often viewed as one of the earliest advocates of a comprehensive textualist approach to the APA. . . . In recent years, APA textualism has become a more bipartisan enterprise. This largely tracks general trends in statutory interpretation, in which liberals and progressives have begun to embrace textualism—either as a matter of first principles or as a matter of strategy. . . .
When it comes to progressive textualism in administrative law, Professor Kovacs has led the charge. In 2015, she argued that the entrenched nature of the APA qualifies it as a “superstatute” in Eskridge-Ferejohn terms. . . . William Eskridge and John Ferejohn view super-statutes as opportunities for dynamic interpretation and the creation of accompanying administrative common law. Professor Kovacs, by contrast, argues against such dynamism when it comes to interpreting the APA. “[G]iven the extraordinary legislative process that led to the APA’s enactment and the relative paucity of agency-based deliberative feedback since then,” she argues, “courts should be particularly cautious about interpreting the APA’s text in a way that shifts the balance Congress reached through the political process.” Like Professor Duffy but from the other side of the ideological aisle, Professor Kovacs argues that much of “administrative common law is suspect”—for her because it “contradicts that deliberation-focused approach.” In subsequent work, Professor Kovacs has called on the Court to eliminate numerous atextual and dynamic interpretations of the APA, including the judge-made procedural requirements for notice-and-comment rulemaking discussed in Part I.A. She has also argued that a textualist interpretation of the APA would include the President as an agency for APA purposes.
Nicholas Bagley is another scholar who has leveraged textualist arguments to advance progressive (or at least pro-regulatory) ends in administrative law. He has argued that the APA’s harmless error provision should more regularly excuse agencies’ procedural errors from judicial invalidation. When it comes to a variety of agency mistakes in notice-and-comment rulemaking and inadequate agency reason giving, Professor Bagley argues that “there is often a mismatch between the underlying violation and the harshness of the conventional remedy.” In a subsequent article, provocatively titled The Procedure Fetish, Professor Bagley expands on those arguments to call for the potential reconsideration of “[t]he judicially imposed rigors of notice-and-comment rulemaking, the practice of invalidating guidance documents that are ‘really’ legislative rules, the Information Quality Act, the logical outgrowth doctrine, [and] nationwide injunctions against invalid rules.” In a somewhat similar vein, he has questioned the presumption of reviewability in administrative law.
Unsurprisingly, Professor Kovacs applauds Professor Bagley’s The Procedural Fetish, which she believes “is destined to be a classic” as it “systematically dismantles the arguments that procedure is necessary to legitimize the administrative state and avoid agency capture.” She then more directly connects Professor Bagley’s work to progressive textualism, declaring that “[t]t is time for progressive administrative law scholars to claim the APA as their own.” It remains to be seen whether more liberal and progressive judges and scholars take up Professor Kovacs’s call to action.
Regardless, textualism will likely remain a major theory for interpreting the APA for years to come. . . .
Over the past decade, many scholars have begun to employ a more originalist approach to interpreting the APA. This approach looks to the original understanding of the text and statutory provisions of the APA. As Emily Bremer describes APA originalism in her contribution to this symposium, “much of the statute’s meaning is apparent only if one reads the brief text [of the APA] against the rich context that produced it,” which includes “pre-APA administrative common law” as well as “a wide variety of pre-APA statutes and the agency precedents and practices that had developed and crystallized to varying degrees before the APA’s enactment.” The emergence of APA originalism is perhaps best exemplified by HeinOnline making publicly available a massive database of APA historical sources, edited and compiled by Professors Bremer and Kovacs.
APA originalism seems to mirror originalist constitutional interpretation methodologies, and may reflect the quasi-constitutional status that the APA has achieved. Like the distinction between APA pragmatism and administrative common lawmaking, the line between APA textualism and APA originalism is not easy to draw—and many APA textualist approaches engage in at least some originalist methods. Notwithstanding, we think it is important to separate out these two types of APA interpretive theories, as the more originalist versions of APA textualism venture beyond a more formalist textualist inquiry . . . .
Moreover, some APA innovations may be consistent with the statutory text, but not with the original understanding of the APA. Evan Bernick is one of the pioneers of APA originalism. In Envisioning Administrative Procedure Act Originalism, he argues that APA originalism, if adopted, would encourage interpreters to focus on: (1) evidence contained in text and structure of APA, including definitions; (2) “contemporaneous evidence of word usage and patterns and regularities in grammar and syntax”; and (3) judicial development in the years preceding the APA. . . .
Professor Bernick argues that APA originalism is more legitimate as law, and thus provides more legitimacy to the administrative state from a rule-of-law and democratic standpoint than other approaches. But he acknowledges that downsides of this theory are that it rarely yields clear answers, that the original APA was normatively bad, that common-law approaches have improved the APA, and that his approach will not yield legitimacy benefits in a nonoriginalist world. Professor Bernick ultimately seems to vacillate on whether courts should fully embrace APA originalism, but not on whether it should grow and thrive as a scholarly enterprise. Among other benefits, he argues, “insights derived from APA originalism may inspire efforts to replace the APA with a written administrative constitution that is better suited to today’s administrative state.”
APA originalism projects have proliferated in recent years. Perhaps the most ambitious and sweeping is Jeffrey Pojanowski’s Neoclassical Administrative Law—a theory that “is more likely to see the text’s original meaning, statutory context and structure, linguistic canons, and perhaps historical intent as appropriate tools for interpretation, rather than normative canons or legislative purpose at a high level of generality.” Similarly, using APA originalist and related historical methods, Professor Bremer has advanced pathbreaking reconceptions of agency adjudication and rulemaking under the APA. As one of us has observed, “[i]t is not often that an article requires a field to fundamentally reconsider its foundations. Yet, Bremer’s Rediscovered Stages is such an article for administrative law (and agency adjudication in particular).”
APA originalism arguments will no doubt continue to develop in the years to come. Unlike APA textualism, we have not seen an overt progressive turn—at least not yet—when it comes to APA originalism. But some administrative law scholars have argued for progressive interpretations of the APA based on text, context, and history, with Blake Emerson perhaps most prominently among them.
These four competing theories—pragmatism, administrative common lawmaking, textualism, and originalism—will all continue to influence how judges and scholars approach interpreting the APA. It would not surprise us, with the current Supreme Court majority’s approach to statutory interpretation, to see a further rise of APA textualism and APA originalism, and perhaps a further cutting back of administrative common lawmaking. [H]owever, the Roberts Court continues to dabble in APA pragmatism. Its more searching review of agency reason giving in the census citizenship question case and the DACA rescission case come immediately to mind. Indeed, some may well consider these precedents more constistent with administrative common lawmaking than mere APA pragmatism. And, of course, there is the new major questions doctrine . . ., which suggests administrative common lawmaking is not dead at the Supreme Court today.
When it comes to open statutory questions, we urge courts to interpret the APA based on the statute’s text, structure, context, and original understanding. In other words, APA originalism is the best path forward for open questions. We are sympathetic to more formalist textualism in certain statutory contexts, in large part because textualism (compared to originalism) sometimes may lead to more certainty and predictability in the law and can set clearer rules of the road for Congress’s legislative action (and reaction to judicial statutory interpretations). In other words, it is true at least some of the time that, as Professor Bernick argues, APA originalism may not yield as many clear, predictable answers compared to more formalist textualism. To do this originalism well involves a “daunting task of examining the historical, original meaning of the terms Congress included in the APA.”
When it comes to a framework statute like the APA, however, we believe a more originalist approach to open questions better respects Congress’s role enacting the APA—a seldom-amended framework statute that emerged as a political compromise after more than a decade of congressional deliberation. In that sense, Professor Kovac’s arguments against administrative common lawmaking have a lot of force here:
Given the extraordinary legislative process that led to the APA's enactment and the relative paucity of agency-based deliberative feedback since then, courts should be particularly chary of interpreting the APA’s text in a way that shifts the balance Congress reached through the political process. Courts should look more closely at the context and history of the APA’s individual provisions, including Congress’s treatment of each provision in the original legislative process and the quality of deliberation the provision has seen since enactment.
Moreover, for reasons similar to those advanced by Professors Duffy, Bernick, and Kovacs, APA pragmatism and administrative common lawmaking are less ideal than APA originalism (or APA textualism), in terms of respecting Congress’s Article I legislative role, especially when dealing with a seldom-amended framework statute like the APA.
For similar reasons, when it comes to APA statutory interpretation questions courts have already answered, we urge courts and litigants (and agencies) to emphasize the strong pull of statutory stare decisis. Statutory stare decisis arguably is already part of APA originalism, as stare decisis is part of “our law.” In that sense, our approach is, again, more in line with APA originalism than APA textualism, the latter of which in at least some forms may be less willing to adhere to the doctrine of statutory stare decisis. When it comes to judicial precedents interpreting statutes, the Supreme Court has made clear that “stare decisis carries enhanced force” because those who think the judiciary got the issue wrong “can take their objections across the street, and Congress can correct any mistake it sees.” . . . In 1946, the APA established the default rules for agency procedure and judicial review of agency actions; Congress has departed from those APA defaults in countless statutes governing hundreds of agencies in the federal bureaucracy. For decades, Congress has been legislating against the backdrop of APA statutory precedents when it authorizes—and reauthorizes—the hundreds of statutes that govern federal agencies today. In that sense, Congress is reenacting the statutory scheme that the Supreme Court considered. . . .
In sum, when dealing with a framework statute like the APA, statutory stare decisis should almost always—if not always—compel the Supreme Court to adhere to its prior interpretations. In so doing, the judiciary respects the separation of powers and administrative law’s important rule-of-law values.