Presidential Administration and the Traditions of Administrative Law

Presidential Administration and the Traditions of Administrative Law

American administrative law has long been characterized by two distinct traditions: the positivist and the process traditions. The positivist tradition emphasizes that administrative bodies are created by law and must act in accordance with the requirements of the law. The process tradition emphasizes that agencies must act in accordance with norms of reasoned decisionmaking, which emphasize that all relevant interests must be given an opportunity to express their views and agencies must explain their decisions in a public and articulate fashion. In the twentieth century, American administrative law achieved a grand synthesis of these two traditions, with the result that deficiencies from the positivist perspective—such as very broad delegations of discretion to agencies—were acceptable, as long as process norms were vigorously enforced. Professor Peter Strauss and other architects of this synthesis never envisioned that the process tradition could completely displace positive law. In recent years, however, commentators have begun to argue that the process tradition can take on a life of its own and can function as a complete substitute for the positivist tradition. This can be seen in a variety of contexts where traditions of legislative supremacy are weak, such as multinational treaty regimes and various forms of “presidential administration.” This Essay offers some grounds for skepticism about the long-term prospects of an administrative law based solely on the process tradition. When acting in the positivist tradition, courts function as agents of sovereignty. Their judgments, assuming they are perceived as being faithful to the law, are backed by the sovereign power of the state, which means they are likely to be obeyed. The process tradition rests on norms of reasonableness, as to which reasonable people may disagree. Especially where judicial review is weak or nonexistent, internal review institutions are unlikely to have enough institutional capital to impose their judgments about reasonableness on other government actors. Enforcement of administrative law norms may come to be seen as merely a matter of contestable opinion. Instead of acting as a check on admin­istrative abuse, administrative law may devolve into a ration­alization for the exercise of raw power.





Administrative law has always been concerned with constraints on executive action. Historically, administrative law scholars have focused primarily on constraints imposed by courts through judicial review of administrative agency action. 1 The classic works are John Dickinson, Administrative Justice and the Supremacy of Law in the United States (1927), and Louis L. Jaffe, Judicial Control of Administrative Action (1965). Modern casebooks, including the one edited by Peter Strauss, are dominated by issues that arise in the course of judicial review of agency actions. Peter L. Strauss et al., Gellhorn and Byse’s Administrative Law: Cases and Comments chs. 7–9 (11th ed. 2011). This Essay sketches two broad traditions that have played a critical role in the evolution of judicial review of agency action—the positivist tradition and the process tradition. Where judicial review is available, both traditions continue to play an important role today. 2 See infra Part I (discussing “Grand Synthesis” of positivist and process traditions and its continuing relevance today). In contrast, where judicial review is not available and the primary constraint on administrative action comes from internal review within the executive itself, there are signs—necessarily tentative and inconclusive—that the process tradition has greater appeal. 3 See infra Part II (discussing pure process approaches in both European Union and American governments). The question this Essay raises is whether an expanded sphere of “presidential administration” 4 See generally Elena Kagan, Presidential Administration, 114 Harv. L. Rev. 2245, 2246 (2001) (defining presidential administration as form of governance in which regulatory activity of executive branch becomes extension of President’s policy and political agenda). that operates free of the constraints of judicial review can be meaningfully constrained by precepts drawn from the process tradition.

What are the key features of the traditions developed in the context of judicial review? Under the positivist tradition, the critical question is whether the government agency has legal authority for the action it is taking. 5 See, e.g., Henry P. Monaghan, The Protective Power of the Presidency, 93 Colum. L. Rev. 1, 5 n.27 (1993) (noting requirement of legal authority is “so deeply ingrained in our constitutional tradition that it is seldom articulated”); Daphna Renan, Pooling Powers, 115 Colum. L. Rev. 211, 220 (2015) (“As a legal matter, an agency requires affirmative authority to undertake any type of action. The source of that legal authority is generally statutes, though in rarer instances it might be constitutional authority delegated by the President.” (footnote omitted)). Administrators must justify their actions in terms of some higher law, such as the Constitution or a statute. Professor Richard Stewart famously described this as the “transmission belt” model of admin­istrative law. 6 Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1667, 1675 (1975). In democratic regimes, power is said to flow from the people, acting to ratify the Constitution, to the elected legislature, to administrative bodies established by the legislature. 7 Id. at 1671–76 (describing traditional model of administrative law). Authority to act with the force of law moves along a series of delegations, running from the people, to the legislature, to administrators. The positivist tradition developed under the assumption that some other institution—typically assumed to be the courts—stands ready to block the actions of government agents when those actions exceed the authority conferred by law. 8 See, e.g., Jerry L. Mashaw, Recovering American Administrative Law: Federalist Foundations, 1787–1801, 115 Yale L.J. 1256, 1258 (2006) (noting American administrative law has long assumed courts serve as checks on overreaching administrative agencies). At least in theory, however, the requirement that an agency’s action must conform to law can also be enforced by institutions internal to the executive branch, such as the Office of Legal Counsel (OLC) in the Justice Department.

Under the process tradition, the critical question is whether agency action comports with reasoned decisionmaking. 9 See, e.g., Indus. Union Dep’t, AFL-CIO v. Hodgson, 499 F.2d 467, 475 (D.C. Cir. 1974) (demanding “careful identification” by the Administrator “of the reasons why he chooses to follow one course rather than another”). Process, as I use the term, is distinct from procedure. See Gary Lawson, Outcome, Procedure and Process: Agency Duties of Explanation for Legal Conclusions, 48 Rutgers L. Rev. 313, 318 (1996) (noting even if agency complies with all applicable procedural rules, “chain of reasoning employed by the agency to reach its conclusion” must “satisfy a minimum standard of rationality” whereby agency explains why it “reached the conclusion that it did,” which Lawson calls process review). “Reasoned” here does not refer to whether the decision conforms to higher law, but rather to the manner in which the decision was reached. 10 See, e.g., Michigan v. EPA, 135 S. Ct. 2699, 2706 (2015) (“‘Not only must an agency’s decreed result be within the scope of its lawful authority, but the process by which it reaches that result must be logical and rational.’” (quoting Allentown Mack Sales & Serv., Inc. v. NLRB, 522 U.S. 359, 374 (1998))). The process tradition emphasizes that all relevant interests should be given an opportunity to express their views, that these views must be fully considered, and that agencies must explain their decisions in a public and logical fashion. 11 See, e.g., Int’l Harvester Co. v. Ruckelshaus, 478 F.2d 615, 652 (D.C. Cir. 1973) (Bazelon, C.J., concurring) (arguing for public participation and transparency in agency decisionmaking); Lisa Schultz Bressman, Procedures as Politics in Administrative Law, 107 Colum. L. Rev. 1749, 1762 (2007) (noting administrative law allows “broad public partici­pation”); Ernest Gellhorn, Public Participation in Administrative Proceedings, 81 Yale L.J. 359, 380–81 (1972) (discussing importance of public participation in agency decisionmaking). This conception of reasoned decisionmaking can plausibly be seen as having been derived from the model of judicial decisionmaking. 12 Thomas W. Merrill, Article III, Agency Adjudication, and the Origins of the Appellate Review Model of Administrative Law, 111 Colum L. Rev. 939, 942 (2011) [hereinafter Merrill, Article III]. But again it is not strictly dependent on the existence of an external enforce­ment agent like the courts. It can also be invoked as a norm that can be internalized by government agents and which serves to legitimize their decisions, without regard to whether those decisions are subject to review by an external institution like a court. European scholars have gener­alized the process tradition in terms of norms of “transparency” and “accountability,” and this terminology has spread to the United States, the birthplace of the process ideal. 13 For the European version, see, e.g., Benedict Kingsbury, Nico Krisch & Richard B. Stewart, The Emergence of Global Administrative Law, Law & Contemp. Probs., Summer/Autumn 2005, at 15, 17 (defining “global administrative law” to include “social understandings” that promote “accountability of global administrative bodies, in particular by ensuring they meet adequate standards of transparency, participation, reasoned decision, and legality, and by providing effective review of the rules and decisions they make”). For the American splashback, see, e.g., Mark Fenster, The Opacity of Transparency, 91 Iowa L. Rev. 885, 900 (2006) (“[Transparency] enables the free flow of information among public agencies and private individuals, allowing input, review, and criticism of government action . . . .”); Jennifer Shkabatur, Transparency With(out) Accountability: Open Government in the United States, 31 Yale L. & Pol’y Rev. 79, 82–84 (2012) (arguing for transparency and accountability in agency actions as preconditions to democratic governance).

In the United States, contemporary administrative law as applied by the courts consists of a synthesis or integration of the positivist and process traditions. Courts enforce “clear” commands found in the law that delegate discretionary authority to agencies. 14 Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 842–43 (1984) (“If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.”). They also enforce norms of reasoned decisionmaking, the content of which has been developed over time through a process of common law elaboration. 15 See Gillian E. Metzger, Foreword: Embracing Administrative Common Law, 80 Geo. Wash. L. Rev. 1293, 1310 (2012) (arguing much of administrative law, including norms of reasonable process, is form of judge-made common law). When pressed, however, courts justify such process norms as resting on interpretations of positive law, most notably the Administrative Procedure Act (APA). 16 See United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 244 (2d Cir. 1977) (grounding process norms in APA). Courts disclaim any inherent power to develop process norms not grounded in positive law. 17 For example, courts will not review agency action on process grounds when the action is unreviewable. See Webster v. Doe, 486 U.S. 592, 601 (1988) (declining to review claim that employee’s dismissal from Central Intelligence Agency was arbitrary and capricious because National Security Act precluded review of such claim).

In the nineteenth century, administrative law was rooted exclusively in the positivist tradition. Enforcement of legal constraints was spotty, since there was no general means of securing judicial review of admin­istrative action. 18 See Jerry L. Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law 301–08 (2012) [hereinafter Mashaw, Creating the Administrative Constitution] (concluding review generally depended on availability of common law action for damages or prerogative writ, but when it was available courts decided matter de novo). It was necessary to find a common law writ like mandamus in order to challenge the action or inaction of a government agent. 19 See id. at 65–78 (summarizing common law actions). If review was available, however, the courts would determine whether the action was authorized by law. 20 Merrill, Article III, supra note 12, at 951 (noting nineteenth-century administrative law featured “little rhetoric of deference, and even less evidence of it in practice”). Little attention was given to the process used by the government in reaching its decisions.

The twentieth century witnessed the growth of the administrative state, which posed an enormous challenge to the positivist tradition. Administrative agencies became more numerous and were delegated large discretionary powers. 21 For an overview of the range of administrative functions, see Peter L. Strauss, Administrative Justice in the United States 152–86 (2d ed. 2002); see also Agency List, Fed. Reg., [] (last visited Aug. 8, 2015) (listing 438 current administrative agencies). The expansion of the administrative state created a demand for wider availability of judicial review in order to police against abuses by administrators. Yet it made no sense to delegate authority to administrative agencies and then have courts decide everything all over again. 22 Dickinson, supra note 1, at 201–02 (arguing “double process” of administrative judgment followed by judicial review only “reduplicates the uncertainty”). Even if one had more faith in the courts than in administrators, courts did not have the capacity or the expertise to oversee everything agencies decided. 23 See Peter L. Strauss, One Hundred Fifty Cases Per Year: Some Implications of the Supreme Court’s Limited Resources for Judicial Review of Agency Action, 87 Colum. L. Rev. 1093, 1095 (1987) [hereinafter Strauss, One Hundred Fifty Cases] (noting systemic consequences of courts’ limited capabilities for reviewing agency action). The solution, which was worked out with great ingenuity over time, was to supplement the positivist tradition with a new ideal—the process tradition. Courts would review agency decisions to assure not only that they were consistent with law, but also that they were reached in a reasoned fashion. 24 See infra Part I (discussing this synthesis). This melding of the two traditions began to take shape in the first half of the twentieth century and was fully worked out in the second half of the twentieth century. Professor Strauss, whom we honor with this Symposium, was one of the foremost architects of this grand synthesis, and he remains unsur­passed as an expositor of its many implications. 25 Among his works that have had the most influence with me, I would include Peter L. Strauss, Formal and Functional Approaches to Separation-of-Powers Questions—A Foolish Inconsistency?, 72 Cornell L. Rev. 488 (1987); Peter L. Strauss, On Resegregating the Worlds of Statute and Common Law, 1994 Sup. Ct. Rev. 429; Strauss, One Hundred Fifty Cases, supra note 23; Peter L. Strauss, The Place of Agencies in Government: Separation of Powers and the Fourth Branch, 84 Colum. L. Rev. 573 (1984); Peter L. Strauss, Publication Rules in the Rulemaking Spectrum: Assuring Proper Respect for an Essential Element, 53 Admin. L. Rev. 803 (2001); Peter L. Strauss, The Rulemaking Continuum, 41 Duke L.J. 1463 (1992); see also infra notes 158, 160 & 177.

In the twenty-first century, we may or may not be on the threshold of a new era in administrative law, in which the positivist tradition is signifi­cantly displaced by a dominant process tradition. It is too early to tell, but there are signs that such a further evolution may be in the offing. The driving force behind such a development is that admin­istrative gover­nance is increasingly outrunning legislative authorization. 26 See infra Part II (describing examples of such potential overreach). The clearest example is found in treaty-based regimes like the European Union, where an elaborate administrative apparatus has been established with­out the benefit of a delegation of power from a sovereign legislature exercising equivalent power. 27 See id. (discussing European Union specifically). It is not surprising in such a context that administrative law scholars would seek to justify administrative edicts exclusively in terms of the process tradition (articulated in terms of the norms of transparency and accountability), since the positivist tradition would suggest that such an exercise of governmental power is problem­atic. In the United States, the rise in power of the President and relative decline of Congress has begun to generate analogous examples. 28 E.g., Memorandum from Jeh Charles Johnson, Sec’y of Homeland Sec., to Leon Rodriguez, Dir. of U.S. Citizenship and Immigration Servs. et al., Exercising Prosecutorial Discretion with Respect to Individuals Who Came to the United States as Children and with Respect to Certain Individuals Who Are the Parents of U.S. Citizens or Permanent Residents (Nov. 20, 2014), publications/14_1120_memo_deferred_action.pdf [] [here­inafter Johnson Memorandum] (implementing presidential action regarding immi­gration policy without explicit statutory authority); see also Exec. Order No. 13,693, Planning for Federal Sustainability in the Next Decade, 80 Fed. Reg. 15,871 (Mar. 25, 2015) (directing reduction in federal government’s greenhouse gas emissions). The War on Terror and the Great Recession produced major expansions of executive power, often justified by the need to respond to unprecedented crises. 29 Eric A. Posner & Adrian Vermeule, The Executive Unbound: After the Madisonian Republic 34–41 (2010). More recently, aggressive waivers of statutory requirements under the Affordable Care Act (ACA) 30 See, e.g., Scott Gottlieb, Decision to Waive Obamacare Rules on Small Business Health Plans Comes with Costs, Forbes (Mar. 5, 2014), [] (discussing Obama Administration waiver for small busi­nesses); Jon Healey, The Truth Behind All Those Obamacare Waivers, L.A. Times (Oct. 3, 2013), [] (discussing IRS waiving requirement that companies with fifty or more employees offer certain minimum benefits). and President Barack Obama’s effort to reform immigration law by administrative action 31 Johnson Memorandum, supra note 28, at 1–5; Memorandum from Janet Napolitano, Sec’y of Homeland Sec., to David V. Aguilar, Acting Comm’r, U.S. Customs and Border Prot. et al., Exercising Prosecutorial Discretion with Respect to Individuals Who Came to the United States as Children 1–3 (June 15, 2012), [] [hereinafter Napolitano Memorandum]. present other exam­ples. Not surprisingly, since the authority for these executive initia­tives in enacted law is often weak or nonexistent, supporters of these executive acts have stressed the “transparent” manner in which they have been developed, 32 See, e.g., Anil Kalhan, Deferred Action, Supervised Enforcement Discretion, and the Rule of Law Basis for Executive Action on Immigration, 63 UCLA L. Rev. Discourse 58, 65 (2015) (listing “transparency” as rule-of-law value ignored by opposition to President Obama’s immigration actions). and have cited the involvement of the President, who is elected by all the people, as a sign of that these initiatives suffer from no deficit of “accountability.” 33 See, e.g., id. at 85 (listing “accountability” as rule-of-law value ignored by opposition to President Obama’s immigration actions). These arguments are designed to confer legitimacy on these efforts without regard to whether they have been authorized by positive law.

The principal question posed by this Essay is whether administrative law can continue to provide a meaningful source of constraint on admin­istrative agencies if it is based solely or even primarily on the process tradition. The synthesis of the twentieth century was a success, certainly in the sense that it allowed a new form of government to develop without expressly amending the Constitution or abandoning traditional ideals associated with the protection of federalism, separation of powers, or individual rights. That success was dependent, to a significant degree, on a sense of continuity between the positivist tradition and the process tradition, and the appearance—most closely associated with the enactment of the APA—that the process ideal was itself required by positive law. 34 Court decisions—including several famous APA decisions—cite public participation, which is itself a form of both transparency and accountability, as a value the APA encourages. E.g., United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 251–53 (2d Cir. 1977).

The long-term prospects of an administrative law based solely on process norms are cause for concern. Unless process norms are themselves embodied in and enforced as positive law, the authority of any internal reviewing institution to insist on compliance with those norms is weak. Those with decisional responsibility will advance arguments favoring confi­dentiality rather than openness and expedition rather than participation, and the internal review body cannot respond that these preferences are contrary to law. Also, the norms associated with the process tradition are elusive, even more so than those embodied in enacted law. Under pressure to accomplish discrete policy goals, these norms are likely to give way, or to morph into novel and more attenuated forms. Perhaps most troubling, the positivist tradition has been the primary vehicle for preserving an archi­tecture of government that features checks against runaway govern­ment power. It is through the enforcement of enacted law that constitu­tional norms associated with federalism, separation of powers, and individual rights are enforced. It is through interpretation and enforce­ment of statutes that Congress’s assignment of different functions to different government offices or to the private sector is maintained. Abandonment of the primary means of preserving the architecture of government would have far reaching consequences that are difficult to foresee. Even if every government action is “transparent,” and every government actor is in some theoretical sense “accountable,” individual freedom and local autonomy as we have come to know them could be irretrievably lost.

The Essay proceeds as follows. Part I describes how administrative law in the United States evolved to reflect both the original positivist tradition and a newer process tradition. Part II surveys examples of executive policy­making that moves beyond authority delegated by democratically elected legislatures—most prominently in the United States the emergence of presidential administration—and the invocation of the process tradition, commonly generalized in terms of the norms of transparency and account­ability, in an effort to confer legitimacy on these efforts. Part III raises questions about whether these efforts to legitimize aggressive executive policymaking in terms of process norms will be successful.

I. American Judicial Review and the Synthesis of Two Traditions

In order to assess the prospect of an administrative law grounded solely in the process tradition, it is important to consider how admin­istrative law in the twentieth century came to rest on the twin pillars of positive law and process review. The story is complex and filled with many conflicting developments, and it is impossible to present anything like a complete account in a short essay. It will be necessary to trace only some broad themes.

In terms of constitutional law, two interpretations of the founding document were critical in laying the foundation for the administrative state that emerged in the twentieth century. Both paved the way for a melding of the older positivist tradition with a newer emphasis on administrative process.

Perhaps the most important constitutional development was the relaxation of (or more accurately, the continued unwillingness to en­force) the proposition that Congress may not delegate the power to legislate. 35 See, e.g., Whitman v. Am. Trucking Ass’ns, 531 U.S. 457, 474–75 (2001) (observing Court has “‘almost never felt qualified to second-guess Congress regarding the permissible degree of policy judgment that can be left to those executing or applying the law’” (quoting Mistretta v. United States, 488 U.S. 361, 416 (1989))); Thomas W. Merrill, Rethinking Article I, Section 1: From Nondelegation to Exclusive Delegation, 104 Colum. L. Rev. 2097, 2103–09 (2004) [hereinafter Merrill, Rethinking Article I] (discussing history of nondelegation doctrine). The Supreme Court held early in the century that there was no violation of the nondelegation doctrine as long as the legislature laid down an “intelligible principle” for the government agent to follow. 36 J.W. Hampton, Jr., & Co. v. United States, 276 U.S. 394, 409 (1928). In the ensuing decades, this evolved into a “boundless standard” satisfied by even the most vaguely worded or incompletely specified delegations. 37 See Dep’t of Transp. v. Ass’n of Am. R.R.s, 135 S. Ct. 1225, 1246 (2015) (Thomas, J., concurring in the judgment) (“Although the Court may never have intended the boundless standard the ‘intelligible principle’ test has become, it is evident that it does not adequately reinforce the Constitution’s allocation of legislative power.” (quoting Whitman, 531 U.S. at 472)); see also Merrill, Rethinking Article I, supra note 35, at 2109 (concluding nondelegation doctrine “imposes no effective constraint on congressional legislation”). By refusing to enforce the nondelegation doctrine, courts gave admini­strative agencies breathing space to make policy. Significantly, however, courts also frequently said that one reason such broad delega­tions were permissible was because the legislature had made judicial review available to ensure that the resulting exercises of administrative discretion were reasonable. 38 See Cynthia R. Farina, Statutory Interpretation and the Balance of Power in the Administrative State, 89 Colum. L. Rev. 452, 487 (1989) (“‘Congress has been willing to delegate its legislative powers broadly—and the courts have upheld such delegation—because there is court review to assure that the agency exercises the delegated power within statutory limits.’” (quoting Ethyl Corp. v. EPA, 541 F.2d 1, 68 (D.C. Cir. 1976) (en banc) (Leventhal, J., concurring) (footnote omitted))); Thomas W. Merrill, Delegation and Judicial Review, 33 Harv. J.L. & Pub. Pol’y 73, 73 (2010) (“Broad delegations of power to executive actors are constitutionally permissible, the Court has suggested, in significant part because courts stand ready to assure citizens that the executive will discharge its discretion in a . . . fashion that otherwise satisfies the requirements of reasoned decision making.”). One can readily discern here a partial substitution of process for positivism. By declining to require the legis­lature to spell out in any detail the policies that administrators were to follow, the courts weakened the positivist tradition. At the same time, the newly developed requirement of reasoned decisionmaking was advanced as a substitute for guidance from higher lawmaking authority. The resulting synthesis combined both traditions—a watered-down positivist tradition and an emergent process tradition.

The second important constitutional interpretation allowed agencies rather than courts to exercise primary authority in adjudicating disputes between individuals and the government. In the nineteenth century, due process was understood to mean the right to a hearing in a common law court, subject to narrow exceptions, before one could be deprived of life, liberty, or property. 39 E.g., Murray’s Lessee v. Hoboken Land & Improvement Co., 59 U.S. (18 How.) 272, 284 (1855). It was also widely presumed that Article III of the Constitution required that disputes between individuals and the government be resolved by the independent federal courts established by Congress. 40 See, e.g., id. (“To avoid misconstruction upon so grave a subject, we think it proper to state that we do not consider congress can . . . withdraw from judicial cogni­zance any matter which, from its nature, is the subject of a suit at the common law, or in equity, or admiralty . . . .”). The Supreme Court revised these understandings in the early decades of the twentieth century. 41 Crowell v. Benson, 285 U.S. 22, 62–63 (1932), is generally regarded as the watershed decision. The key determination was that due process was satisfied if a hearing was held before an administrative body that emulated the features of a common law court, such as a right to present evidence and a decision based on the evidence presented to the hearing officer. 42 See Daniel R. Ernst, Toqueville’s Nightmare: The Administrative State Emerges in America, 1900–1940, at 5 (2014) (“By 1940, the rule of law no longer required that individuals subject to economic regulation receive a ‘day in court’ as long as administrators had given them a ‘day in commission.’”). Excep­tions remained, primarily for matters of private right that would have been heard by a court at common law. 43 See N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 458 U.S. 50, 89–92 (1982) (Rehnquist, J., concurring in the judgment) (finding broad grant of power to bankruptcy courts unconstitutional as bankruptcy was part of “stuff of the traditional actions at common law tried by the courts at Westminster in 1789”); see also Wellness Int’l Network, Ltd. v. Sharif, 135 S. Ct. 1932, 1950–60 (2015) (Roberts, C.J., dissenting) (summarizing jurisprudence). But the role of the courts in administrative matters was increasingly defined as monitoring the process deployed by administrative adjudicators, rather than resolving disputes themselves. 44 Ernst, supra note 42, at 137. The APA ratified this transformation by directing courts to uphold findings of fact by administrators if they were supported by “substantial evidence.” 45 5 U.S.C. § 706(e) (2012). Here again we see the substitution of process review for the enforcement of commands originally thought to be required by the Constitution—that is, the positivist tradition.

In terms of subconstitutional law, the process tradition emerged in full flower in the 1970s. The lower courts, led by the D.C. Circuit, openly espoused the idea that courts should review the process followed by an agency in developing policy when determining whether a regulatory initiative was permissible. 46 See Antonin Scalia, Vermont Yankee: The APA, the D.C. Circuit, and the Supreme Court, 1978 Sup. Ct. Rev. 345, 348–52 (summarizing D.C. Circuit development of process review). An acceptable process required full disclosure of the studies and factual assumptions underlying a proposed regulatory initiative, an opportunity for the public to comment on the studies and assumptions as well as the policy embodied in a proposed regulation, and a cogent response by the agency if it rejected material objections raised by public commenters. 47 See United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 248–53 (2d Cir. 1977) (requiring those procedures for notice-and-comment rulemakings to be adequate under APA). Some judges went so far as to say that courts should confine themselves to reviewing the agency’s process and eschew substantive review altogether. 48 See David L. Bazelon, Coping with Technology Through the Legal Process, 62 Cornell L. Rev. 817, 823 (1977) (arguing courts should focus on “strengthening administrative procedures”).

The idea that the full-blown process tradition might be regarded as a substitute for the positivist tradition was explicated and rationalized in an important article by Professor Stewart written in 1975. 49 Stewart, supra note 6. Professor Stewart suggested that the emerging requirements of the process tradition—which he characterized as a general duty of administrators “to consider ade­quately all participating interests in decisions on agency policy” 50 Id. at 1756. —could be viewed as an alternative form of democratic legitimacy. Democracy was originally conceived as legitimating government because individuals were allowed to vote for representatives who would then deliberate about what types of rules backed by coercive force would be made binding on the citizenry—the “transmission belt” theory of democracy. 51 Id. at 1675. Professor Stewart argued that the emerging norm of interest representation could be regarded as a different mode of democratic legitimacy. The modern administrative agency, by providing an opportunity to all affected interests to participate in agency decisionmaking and requiring that agencies demonstrate that they have given adequate consideration to all interests, “gives citizens a sense of involvement in the process of government, and increases confidence in the fairness of government decisions.” 52 Id. at 1761. It should be noted that Stewart himself was skeptical that interest representation before agencies could be regarded as an acceptable substitute for more robust forms of democratic legitimacy. See id. at 1802.

Notwithstanding widespread academic endorsement of the process tradition, judicial decisions continued to reflect both the new process ideal and the older tradition of positivism. One interesting manifestation of this has involved the understanding of the APA, itself of course a form of positive law. One line of decisions has interpreted broad language in the APA as consistent with what I have called the process tradition. Thus, for example, the requirement that courts set aside agency action that is “arbitrary and capricious,” originally understood to mean something lacking even a minimally rational basis, 53 Pac. States Box & Basket Co. v. White, 296 U.S. 176, 182 (1935) (equating arbitrary and capricious with mere rationality standard for review of legislation). has been used to justify an insis­tence that agencies take a “hard look” at critical issues if their actions are to be upheld. 54 Greater Bos. Television Corp. v. FCC, 444 F.2d 841, 851 (D.C. Cir. 1970). The APA’s most general standard of review has thus been transformed into a demand for reasoned decisionmaking, the hall­mark of the process tradition. 55 Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402 (1971), and Motor Vehicle Manufacturers Ass’n v. State Farm Mutual Auto. Insurance Co., 463 U.S. 29 (1983), can be regarded as illustrations here. Similarly, courts interpreted the bare-bones proce­dures spelled out by the APA for informal rulemaking as requiring elaborate disclosure of the factual and policy rationale for proposed rules, extensive rights of public comment, and reasoned responses to material comments as part of the “concise general statement of basis and purpose” accompanying the final rule. 56 United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 248–53 (2d Cir. 1977); see also J. Skelly Wright, Courts and the Rulemaking Process: The Limits of Judicial Review, 59 Cornell L. Rev. 375, 380–81 (1974) (describing duty of court under APA § 553 as “satisfy[ing] itself that [genuine dialogue between agency experts and concerned members of the public] occurred and that it was not a sham”). These expansive interpretations of the APA served to legitimize the emergence of the process tradition. They suggested that process review was itself required by positive law.

Periodically, however, the Supreme Court has treated the APA like a set of binding instructions from which no deviation is permitted. 57 Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, Inc., 435 U.S. 519, 524 (1978), is of course the primary example. Last Term’s decision in Perez v. Mortgage Bankers Ass’n, 135 S. Ct. 1199, 1201 (2015), holding that agencies do not need to use notice-and-comment procedures in amending interpretative rules, is a more recent illustration of this line of decisions. In these cases, the positivist tradition has reasserted itself in its original form, with the APA interpreted as imposing a ceiling on procedural requirements rather than a floor. It is unclear exactly what triggers these episodic reversions to a purely positivist mode of analysis in explicating the APA. What they reveal is that the positivist and proceduralist tradi­tions continue to coexist, however uneasily.

Notwithstanding the continuing tension, the larger picture in the twentieth century reveals a broad evolution toward a synthesis of the positive and process traditions. The emergence of the Chevron doctrine 58 Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837 (1984). as the dominant form of judicial review in the last two decades of the century is especially revealing in this regard. Chevron’s familiar two-step process can be seen as incorporating the positivist tradition at step one, where courts are instructed to exercise independent judgment in determining whether an agency has violated a statutory command. 59 Id. at 842–43. But if the court finds that the statute is ambiguous or that it does not address the precise question at issue, then courts are instructed at step two to uphold the agency’s interpretation if it is reasonable. 60 Id. at 843–45. Several commen­tators and some lower courts have urged that “reasonable” in this context should mean reasoned decisionmaking as defined by the process tradition. 61 See generally U.S. Telecom Ass’n v. FCC, 227 F.3d 450, 460 (D.C. Cir. 2000) (endorsing “reasoned decisionmaking” approach); Gary S. Lawson, Reconceptualizing Chevron and Discretion: A Comment on Levin and Rubin, 72 Chi.-Kent L. Rev. 1377 (1997) (reviewing literature). The Supreme Court has not explicitly endorsed this under­standing of step two. But the large and growing body of decisions applying the Chevron framework reveals a steady oscillation between measuring agency initiatives against the language of the authorizing statute (positivism) 62 See, e.g., MCI Telecomms. Corp. v. AT&T Co., 512 U.S. 218, 225–31 (1994) (invalidating agency reading of statute as too broad a construction of word “modify”). and accepting agency interpretations that are compatible with statutory language and are devel­oped in standard modes of administrative process (process). 63 See, e.g., Mayo Found. for Med. Educ. and Research v. United States, 562 U.S. 44, 58–60 (2011) (applying Chevron step-two deference to Treasury regulation adopted through notice-and-comment procedure).

Even more strikingly, the Court has held that Chevron-style review applies only when an agency interpretation has the force of law (such as an interpretation advanced in a binding rule); otherwise, agency inter­pretations should be considered under what has been called Skidmore review. 64 United States v. Mead Corp., 533 U.S. 218, 234–35 (2001). This alternative standard of review looks at multiple factors, including, critically, the persuasiveness of the agency’s explanation for its interpretation. 65 Id. at 228 (citing Skidmore v. Swift & Co., 323 U.S. 134, 139–40 (1944)). Although the fit is not perfect, this can be seen as a variant type of process review. Relatively formal agency process triggers strong deference; less formal agency process elicits closer judicial scruti­ny into the reasonableness of interpretation. The basic point is that the Chevron doctrine can be seen as a blending of positivism and process review, which is the key feature of the grand synthesis achieved by twentieth century administrative law.

I have emphasized the distinctive character of the positivist and process traditions, and the transformative nature of the emergence of process review. But change is often controversial, and consequently it is not surprising that those who favor new institutional arrangements frequently insist that they are continuous with established under­standings. So it has been in administrative law. A key formulation here is the use of the concept “rule of law.” Although originally associ­ated with the positivist tradition, the term “rule of law” is sufficiently ambiguous that proponents of the process tradition have been able to claim that administrative edicts adopted in a manner consistent with the reasoned decisionmaking ideal also partake of the “rule of law,” without regard to whether such edicts are securely grounded in any delegation of power. 66 Ernst, supra note 42, at 2–8 (arguing “rule of law” in America means right to appeal to common law court and modeling agency process on judicial process therefore satisfies rule of law). The term “rule of law” in this context thus serves as a device for emphasizing (or exaggerating) the continuity in administrative law.

An alternative way of describing the history, which puts more emphasis on the discontinuity, would be to say that by allowing the administrative process to satisfy the constitutional requirement of due process, the Court was taking the first steps toward supplementing the positivist tradition with the process tradition. A reasonable, that is, a court-like process, was deemed to be equivalent to a hearing by a real court. 67 See Crowell v. Benson, 285 U.S. 22, 53 (1932) (permitting administrative agencies to determine questions of fact subject to deferential review by courts). For antecedents, see Merrill, Article III, supra note 12, at 955–72 (describing historical development of deferential review of agency factfinding). Whether this comports with the rule of law depends on how one defines the rule of law. 68 For a recent attempt, see Dep’t of Transp. v. Ass’n of Am. R.R.s, 135 S. Ct. 1225, 1242 (2015) (Thomas J., concurring in the judgment) (suggesting rule of law requires that “ruler must be subject to the law in exercising his power and may not govern by will alone”). For one source of Justice Thomas’s inspiration, which places more emphasis on the need for independent judicial review, see Philip Hamburger, Is Administrative Law Unlawful? 143–48 (2014) (arguing judicial independence was central to development of modern rule of law). It clearly represents a substitution, or at least a supple­mentation, of the process tradition for the positivist tradition, which is the central feature of the grand synthesis achieved by the twentieth century.

II. The Rise of Presidential Administration and the Prospect of Pure Process

In the early years of the twenty-first century, it is possible to discern a further turn away from the positivist tradition and in the direction of the process tradition. It would be misleading to suggest that there has been anything like a complete displacement of the positivism by process. But there are increasing signs of movement in this direction.

The movement toward pure process has proceeded furthest in the European context. The motivating development has been the emergence of treaty-based systems of multinational regulation, including the General Agreement on Tariffs and Trade and World Trade Organization , but most prominently the European Union. The problem posed by these regimes is that there is no direct delegation of authority from the electorate to the bodies that exercise primary authority in promulgating directives having the force of law. 69 See Nat. Res. Def. Council v. EPA, 464 F.3d 1, 7–10 (D.C. Cir. 2006) (holding that action by Environmental Protection Agency (EPA) cannot be reviewed for compliance with norms established by international treaty organization because such norms are not “law” within meaning of U.S. domestic law). In the context of the European Union, legislative proposals are initiated by a Commission, whose members are appointed by the heads of the governments that participate in the regime. 70 David Edward & Robert Lane, Edward and Lane on European Union Law 96, 100–01 (2013). Most legislation must gain the assent of the European Parliament, which is directly elected, and the Council of Ministers, which consists of representatives from the member governments, but the Commission is the driving force. 71 Id. at 105–06. The Commission also oversees an extensive bureaucracy, which promulgates secondary legislation (regulations) and engages in enforcement activity. From a positivist perspective, the Commission and its bureaucracy are themselves agents appointed pursuant to treaty provisions ratified by the member governments; as such, they represent a further extension of the admin­istrative model, rather than a consolidation of the democratic sovereignty model. To borrow Professor Stewart’s metaphor, the transmission belt has been stretched, not widened. 72 Stewart, supra note 6, at 1675. All of which has given rise to widespread concern that the E.U. regime suffers from a “democracy deficit.” 73 See, e.g., Paul Craig, Integration, Democracy, and Legitimacy 28–31, in The Evolution of EU Law (Paul Craig & Grainne De Burca eds., 2d ed. 2011) (summarizing debate over Europe’s democracy deficit).

In this context, scholars sympathetic to a strong European Union have responded creatively by invoking what I have called the process tradition as an alternative source of legitimacy for the directives of the European Commission. They have not borrowed American terms, like “hard look” 74 See Thomas J. Miles & Cass R. Sunstein, The Real World of Arbitrariness Review, 75 U. Chi. L. Rev. 761, 761–63 (2008) (discussing development of hard-look doctrine). or “reasoned decisionmaking,” 75 See Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 52 (1983) (disallowing agency action as not being product of “reasoned decisionmaking”). which emerged from the efforts of federal judges to meld the process tradition to the language of the APA. Instead, they have developed their own vocabulary, invoking the concepts of “transparency” and “accountability” (sometimes also “participation”) to describe the features of administrative regulation that partake of enhanced legitimacy. 76 See, e.g., E. Madalina Busuioc, European Agencies: Law and Practices of Accountability 9 (2013) (offering systematic survey of “accountability regimes of European agencies”); Paul Craig & Grainne De Burca, EU Law: Text, Cases, and Materials 541–49 (5th ed. 2011) (surveying origins and development of transparency as general principle of E.U. law). These concepts are slippery (no less than “hard look” or “reasoned decisionmaking”), but in most applications they track the features of the process tradition that developed in the United States and reached its full form in the 1970s. A regulation is “transparent” if its terms are fully spelled out in a publicly accessible text and if the regulatory body provides a comprehensive rationale for the provisions of the regulation. 77 See, e.g., P. Nikiforos Diamandouros, European Ombudsman, Transparency, Accountability, and Democracy in the EU, Lecture at the School of Advanced International Studies of Johns Hopkins University, Bologna, Italy (Oct. 17, 2006), [] (positing public accessibility to information is necessary for transparency); Transparency Portal, European Comm’n, [] (last updated July 9, 2015) (explaining European citizens have “right” to access E.U. rules and decisions). The regulators are “accountable” if mem­bers of the public are allowed to participate in some fashion in the development of the regulations and the regulatory body responds to concerns raised by these participants in some meaningful fashion, or if the regulators are subject to review or oversight by some institution such as a national court that has a stronger claim to democratic legitimacy. 78 See, e.g., Elizabeth Fisher, The European Union in the Age of Accountability, 24 Oxford J. Legal Stud. 495, 514 (2004) (book review) (noting “promotion of accountability in the EU cannot be disentangled from debates about the legitimacy of European governance and in particular what the role of democratic processes and principles should be”); Diamandouros, supra note 77 (defining accountability as having “to explain and justify one’s actions in terms of appropriate criteria and in sufficient detail”). In urging regulators to make their regulations “transparent” and to hold themselves “accountable,” these scholars are in effect replicating what Stewart called the “interest representation” model of democratic legitimacy in the United States in the 1970s—that is, the process tradition. 79 See Charles F. Sabel & William H. Simon, Epilogue to Law and New Governance in the EU and the US 402 (Grainne de Burca & Joanne Scott eds., 2006) (noting legitimacy of peer review processes of legislation promulgated by European Union “will depend on their transparency and more ambitiously, on their openness to directly deliberative participation by affected stakeholders”). The difference being that in the multinational regime context, the process tradition is the only game in town. 80 Or at least has been perceived to the only game in by many students of the European Union. As Professor Peter Lindseth has argued, the fact that the application of E.U. directives in particular cases is often reviewable in national courts may provide a critical link between European edicts and conventional conceptions of sovereignty that accounts in significant part for national acceptance (up to a point) of the European enterprise. Peter L. Lindseth, Power and Legitimacy: Reconciling Europe and the Nation-State 4–57 (2010).

American administrative law has not yet begun to approach the situ­ation of the European Union, but there are intimations that it is headed in that direction. As originally conceived, and throughout most of the nineteenth century, Congress was the dominant institution in the American system of government. 81 Woodrow Wilson, Congressional Government 11 (1885), (on file with the Columbia Law Review). By the end of the twentieth century and accelerating at the beginning of the twenty-first, the President and the vastly expanded executive branch have become the most powerful engine of government. 82 See, e.g., Posner & Vermeule, supra note 29, at 5–7 (noting modern expansion and size of executive branch and its power over agency policymaking). Presidential influence over policy has gone far beyond proposing and vetoing legislation. 83 See, e.g., id. at 11 (describing executive control over policy agenda); Johnson Memorandum, supra note 28 (laying out comprehensive federal nondeportation policy for certain types of illegal immigrants). Presidents have worked assiduously to increase their control over the executive branch and inde­pendent agencies, and have used this control to engage in what has been called “presidential administration.” 84 E.g., Kagan, supra note 4, at 2246 (“We live today in an era of presidential administration.”). Under this conception, the President and his staff develop a comprehensive policy agenda designed to maintain popular support and ensure reelection. 85 Id. at 2345 (noting presidential administration drives “broad domestic policy agenda”). The President’s agenda is then promoted by issuing “directives” to administrative agencies to implement items on the agenda, all the while seeking to rally public opinion with presidential speeches and press conferences, weekly radio addresses, and photo opportunities. 86 Id. at 2290–99 (discussing Clinton White House’s usage of directives). Congress, meanwhile, stymied by ponderous procedures and afflicted with partisan gridlock, is relegated to a largely reactive role, holding oversight hearings, occasion­ally ratifying presidential initiatives with legislation, and periodically trim­ming presidential sails with appropriations riders. 87 Id. at 2256 (noting Congress’s failure to “exercise any effective control over administrative policymaking” and instead only using weak methods of administrative control); see also Michael S. Greve & Ashley C. Parrish, Administrative Law Without Congress, 22 Geo. Mason L. Rev. 501, 502–03 (2015) (attributing “powerful shift in the direction of executive government” to “debilities of the United States Congress”).

The most extreme analysis of the emergence of presidential admin­istration, by Professors Eric Posner and Adrian Vermeule, argues that the President and the administration are no longer meaningfully constrained by law. 88 Posner & Vermeule, supra note 29, at 31–37. This is almost surely an overstatement. If Presidents are uncon­strained by law, it is unclear why they always seek to justify their actions as being consistent with law, threaten to veto legislation they do not like, and obey judgments of courts based on judicial interpretations of the law. 89 See Saikrishna B. Prakash & Michael D. Ramsey, The Goldilocks Executive, 90 Tex. L. Rev. 973, 988–92 (2012) (reviewing Eric A. Posner & Adrian Vermeule, The Executive Unbound: After the Madisonian Republic (2010)) (asserting President is not “unbound from the laws of Congress” or court judgments). Given their analysis, however, it is unsurprising that Professors Posner and Vermeule put no stock in the positivist tradition in admin­istrative law. They have no interest in the process tradition either, viewing the APA and related process restraints as filled with “black holes” and “gray holes” that allow administrative agencies to dispense with proce­dural requirements whenever they become inconvenient. 90 Posner & Vermeule, supra note 29, at 92–101. Although Professors Posner and Vermeule have no truck with process review, it is interesting that they recommend that Presidents “commit to transparency” about the actions they take in order to enhance their credibility with the public, id. at 145, an echo of the European version of the process tradition. Instead, they argue that Presidents are constrained only by politics and public opinion. In particular, Presidents need “to maintain popularity and credibility” in order to govern effectively. 91 Id. at 13. Since this is the only truly meaningful constraint on presidential action, administrative law is an irrelevancy that can be dispensed with.

Few, if any, of the other partisans of presidential administration would go so far. Instead, mainstream lawyers and scholars tend to defend presidential administration in terms of the process tradition. Three examples can be cited in support of this proposition, although others could be cited as well.

Perhaps the most significant manifestation of the turn toward presidential administration is the emergence of systematic White House review of major agency regulations. The instrument for this review is an entity known as the Office of Information and Regulatory Affairs (OIRA), located in the Office of Management and Budget (OMB), which is part of the Executive Office of the President (EOP). 92 Office of Mgmt. & Budget, Office of Info. & Regulatory Affairs, White House, [] (last visited Aug. 8, 2015). Although there were precursors during the administrations of Presidents Gerald Ford and Jimmy Carter, the major impetus for systematic review of regulations came during the Reagan Administration. 93 See, e.g., Richard H. Pildes & Cass R. Sunstein, Reinventing the Regulatory State, 62 U. Chi. L. Rev. 1, 3 (1995) (noting most important development came from President Ronald Reagan’s executive order). Based on an executive order, OIRA was charged with determining, “to the extent permitted by law,” whether regulations issued by executive departments would deliver social benefits in excess of their costs. 94 Exec. Order No. 12,291, 3 C.F.R. 127, 128 (1982), reprinted in 5 U.S.C. § 601 (1988), revoked by Exec. Order No. 12,866, 3 C.F.R. 638 (1993), reprinted as amended in 5 U.S.C. § 691 (2012). The regulatory hook for such review was the asserted authority of OMB to act as gatekeeper in permitting the publication of regulations in the Federal Register. 95 Id. § 3(f)(2), at 130. Because the regu­latory review process was located in the EOP, its determinations were assumed to be immune from judicial review under the APA. 96 See, e.g., Peter Ketcham-Colwill, Presidential Influence over Agency Rulemaking Through Regulatory Review, 82 Geo. Wash. L. Rev. 1622, 1635 (2014) (noting OIRA’s immunity from APA); Michael A. Livermore & Richard L. Revesz, Regulatory Review, Capture, and Agency Inaction, 101 Geo. L.J. 1337, 1339 (2013) (emphasizing OIRA actions are not subject to judicial review).

The OIRA review process has no clear statutory foundation and thus sits on shaky ground from the perspective of the positivist tradi­tion. 97 The only legal authority cited in the original Reagan executive order was “the authority vested in me as President by the Constitution and laws of the United States of America.” Exec. Order No. 12,291, supra note 94. It also is conducted largely in secret, with no opportunity for formal participation by affected interests, 98 See Pildes & Sunstein, supra note 93, at 5 (discussing criticism of OIRA review as being too secretive). and therefore runs counter to the norms associated with the judicially developed process tradition. Regulatory review was lambasted by critics on both grounds during its early deregulatory orientation in the Reagan Administration. 99 See, e.g., Alan B. Morrison, OMB Interference with Agency Rulemaking: The Wrong Way to Write a Regulation, 99 Harv. L. Rev. 1059, 1067–68 (1986) (criticizing OIRA review as unaccountable and secretive); Erik D. Olson, The Quiet Shift of Power: Office of Management & Budget Supervision of Environmental Protection Agency Rulemaking Under Executive Order 12,291, 4 Va. J. Nat. Resources L. 1, 3 (1984) (mentioning “simmering controversy over whether, and under what conditions, the President has the [legal] authority to supervise executive agency rulemaking”). During the Bush I Administration, legislative hostility to regulatory review forced a transfer of its function to the Office of the Vice President. With the election of President Bill Clinton, a Democrat more sympathetic to activist regulation, many assumed that the office would be abolished. 100 See Bruce Ackerman, The Decline and Fall of the American Republic 35 (2013) (noting with Democrats’ return to power in 1992 “one might have expected them to call upon President Clinton to abolish OIRA”). Instead, President Clinton issued a new executive order, reestablishing the office and relocating it in OMB. 101 Exec. Order No. 12,866, 3 C.F.R. 638 (1994), reprinted as amended in 5 U.S.C. § 601 (2012) (supplemented by Exec. Order No. 13,563, 3 C.F.R. 215 (2012), reprinted in 5 U.S.C. § 601 (2012)). The Clinton version of regulatory oversight was less explicitly deregulatory. It also included procedural reforms designed to limit ex parte contacts during the review process and to provide for more disclosure of communications between OIRA and the agencies whose regulations were under review. 102 Kagan, supra note 4, at 2286–87. This newer, softer version of regulatory review encountered little opposition from Congress or from legal academics. 103 See Nicholas Bagley & Richard L. Revesz, Centralized Oversight of the Regulatory State, 106 Colum. L. Rev. 1260, 1267 (2006) (claiming appropriateness of OIRA review is not seriously challenged). Perpetuation of the office by President George W. Bush 104 Exec. Order No. 13,422, 3 C.F.R. 191 (2008) (modifying and amending Executive Order 12,866); Exec. Order No. 13,258, 3 C.F.R. § 204 (2003) (same). and President Obama 105 Exec. Order No. 13,497, 3 C.F.R. § 218 (2010) (revoking Executive Orders 13,258 and 13,422). appears to have cemented its existence as a permanent fixture of the regulatory state.

Notwithstanding bipartisan acceptance of OIRA review, it represents a highly discordant feature within the American administrative process. The review process rests on a series of executive orders, not on legislation enacted by Congress delegating authority to the President to engage in such review. 106 See supra notes 101 and 104–105 (noting executive orders creating OIRA). And it proceeds largely behind closed doors, lacking any of the features of public participation or judicial review. 107 See Pildes & Sunstein, supra note 93, at 5 (discussing secretive nature of OIRA review).

In order to legitimize the office, it was necessary to draw on the process tradition in a newly creative way. Then-Professor Elena Kagan, who served as an advisor in the Clinton White House and is now a Supreme Court Justice, provided the principal justification. In a major article in the Harvard Law Review, she argued that presidential administration “enhances transparency,” because the high visibility of presidential pronouncements and press releases about regulatory affairs enables the public “to comprehend more accurately the sources and nature of bureau­cratic power.” 108 Kagan, supra note 4, at 2331–32. And presidential directives “promote[] account­ability,” because the President is the only official “elected by a national constit­uency in votes focused on general, rather than local, policy issues.” 109 Id. at 2331, 2334. When directed and reviewed by the President, bureaucratic action “thus turns out to have a democratic pedigree purer even than Congress’s in our system of government.” 110 Id. at 2334. Then-Professor Kagan credited Professor Jerry Mashaw for developing the argument that regulations adopted by administrative agencies have a superior claim to democratic legitimacy because of oversight by the nationally elected and accountable President. See Jerry L. Mashaw, Prodelegation: Why Administrators Should Make Political Decisions, 1 J.L. Econ. & Org. 81, 95 (1985).

Here we see the process tradition, as reformulated by European scholars in terms of the norms of accountability and transparency, being used to justify administrative action having a weak or nonexistent foundation when viewed from the perspective of positivism. The transfor­mation is startling, given the longstanding understanding, grounded in positivist theory, that the President is not a “lawmaker.” 111 E.g., Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 587 (1952). Not everyone agrees with then-Professor Kagan’s defense of OIRA review in terms of the superior transparency and accountability associated with the EOP. Others continue to fault the process for its lack of openness and public partic­ipation. 112 See, e.g., Nina A. Mendelson, Disclosing “Political” Oversight of Agency Decision Making, 108 Mich. L. Rev. 1127, 1165 (2010) (urging more disclosure of executive supervision of agencies); Peter M. Shane, Political Accountability in a System of Checks and Balances: The Case of Presidential Review of Rulemaking, 48 Ark. L. Rev. 161, 207–09 (1995) (noting criticism of process based on lack of transparency). It is a sign of the times, however, that criticism of OIRA is centered on whether it conforms to the norms of the process tradition—transparency and accountability. No one seems to care that it operates without any delegation of authority from Congress.

More recently, presidential administration has moved beyond regu­latory review to occupy new territory. The next example involves action, again directed by the EOP, which has been labeled “big waiver” in a recent article by Professors David Barron and Todd Rakoff. 113 David J. Barron & Todd D. Rakoff, In Defense of Big Waiver, 113 Colum. L. Rev. 265, 272 (2013). The authors note that divided government and legislative gridlock have led to increased invocation by the executive branch of waivers of requirements imposed by Congress in statutes such as the No Child Left Behind Act and the ACA. 114 Id. at 267–68. They do not claim that the executive has inherent authority to waive requirements imposed by law. A decision to make modifications in regulatory law through the use of waivers “depends on there being a distinct statutory waiver authority.” 115 Id. at 312. Nevertheless, the authors devote little attention to the ques­tion whether the waivers they deem “big” were actually authorized by the relevant legislation. To the contrary, they seem comfortable with aggres­sive invocations of broadly worded or ambiguous waiver authority in order to achieve an “effective, engaged, and democratically responsive administrative state” that is not “hemmed in by federal legislative baselines enacted decades ago.” 116 Id. at 310. In lieu of proposing that Congress spell out the terms of permissible waivers in greater detail, they suggest a heightened obligation on the part of agencies to explain their decision to waive a statute. This is prudent, they counsel, because exercises of big waiver will “generate[] headlines,” and “critics will seize upon weaknesses in the legal arguments.” 117 Id. at 319. Thus, “[t]he agency should explain why its big waiver is not just permissible but affirmatively desirable; it should explain, that is, why the purpose of the statute will, under existing circumstances, be better satisfied by departure from the specific rules of the statute.” 118 Id. at 332. By “forcing transparency” 119 Id. at 334. in this fashion, decisions to engage in big waiver, the authors argue, will have enhanced legitimacy in the eyes of the public.

The authors’ celebration of “big waiver,” technically exercised by agencies but directed by the White House, reflects a further subordi­nation of the positivist tradition, with a concomitant elevation of ele­ments of the process tradition in its place. The authors gloss over the fact that executive waivers of statutory requirements will rarely be subject to judicial review. 120 The authors allude to “the special standing issues that may arise as to some exercises of the waiver power,” id. at 319, but say these “warrant an article in their own right.” Id. at 319 n.201. As Professors Michael Greve and Ashley Parrish document, the D.C. Circuit has adopted an especially restrictive approach to standing in cases involving challenges to agency waivers. Greve & Parrish, supra note 87, at 539–43. This is because those most directly affected by a waiver will be relieved of a statutory burden and cannot claim to be adversely affected or aggrieved. Meanwhile, the beneficiaries of the waived provision presumably represent a diffuse general interest of the sort that typically does not support a claim of standing under the Court’s standing jurisprudence. 121 See Allen v. Wright, 468 U.S. 737, 755 (1984) (stating individuals have “no standing to complain simply that their Government is violating the law”); see also FEC v. Akins, 524 U.S. 11, 23–25 (1998) (summarizing decisional law holding that generalized grievances do not confer standing); United States v. Richardson, 418 U.S. 166, 173–74 (1974) (holding generalized grievance did not confer standing). Accordingly, even if the executive branch determines to waive statutory requirements in a “transparent” way, as urged by the authors, the accountability of the executive for such actions will ordi­narily rest on the factors invoked by then-Professor Kagan in support of other unreviewable presidential directives—general public support for the incumbent President and his or her prospects for reelection.

The final example concerns efforts by the Obama Administration to achieve major reform of the immigration system by unilateral executive action. In 2012, the Department of Homeland Security announced by public memorandum a program called Deferred Action for Childhood Arrivals (DACA). 122 Napolitano Memorandum, supra note 31, at 3. Under this program, unauthorized immigrants who had entered the United States before the age of sixteen and had been present continuously for five years were entitled to renewable two-year relief from removal (later expanded to three years), 123 Johnson Memorandum, supra note 28, at 3. as well as author­ization to work in the United States. 124 Napolitano Memorandum, supra note 31, at 3. In November 2014, shortly after mid-term elections in which the President’s party lost control of both Houses of Congress, President Obama announced a program called Deferred Action for Parents of Americans and Lawful Permanent Residents (DAPA). 125 Johnson Memorandum, supra note 28, at 4–5. Under this program, unauthorized immigrants who were parents of children born in the United States or otherwise lawfully present were authorized to apply for deferral of removal and work permits for three years, provided they could show that they were not “enforcement priorities” (i.e., subject to criminal prosecution or iden­tified as national security risks). 126 Id. at 3.

These executive actions were justified on the ground that they were exercises in prosecutorial discretion. 127 Id. at 1; Napolitano Memorandum, supra note 31, at 1. Exercises of prosecutorial discretion are generally unreviewable. Heckler v. Chaney, 470 U.S. 821, 831 (1985) (“This Court has recognized on several occasions . . . that an agency’s decision not to prosecute or enforce . . . is a decision generally committed to an agency’s absolute discretion.”). Traditionally, prosecutorial dis­cretion has been exercised by local prosecutors making highly contextual judgments based on multiple factors, such as the strength of the evidence, the culpability of the offender, and the available prosecutorial resources. 128 See Inmates of Attica Corr. Facility v. Rockefeller, 477 F.2d 375, 380 (2d. Cir. 1973) (noting general judicial reluctance to cabin prosecutorial discretion given complexity of deciding whether to prosecute). Advocates of the process tradition in administrative law have long urged that prosecutorial discretion should be cabined by regu­lations or written guidelines that would provide a publicly articulated rule to structure such decisions. 129 See, e.g., Kenneth Culp Davis, Discretionary Justice: A Preliminary Inquiry 223–25 (1969) (advocating promulgation of rules to constrain prosecutorial discretion). The Obama Administration’s immi­gration directives appear to reflect such a development, although the Administration also insists that individual immigration judges retain discretion to depart from the directives in individual cases. 130 The 2014 policy has been stayed by order of the federal district court in the Southern District of Texas, on the ground that it is in substance a legislative rule that must be promulgated by notice-and-comment procedures. Texas v. United States, No. B-14-254, 2015 WL 648579 (S.D. Tex. Feb. 16, 2015). The Fifth Circuit declined to lift the stay, 787 F.3d 733, 743 (5th Cir. 2015), and the matter is now on appeal. Key questions are whether the states challenging the policy have standing, and whether the guidelines are really a disguised legislative rule. General statements of policy, under the APA, do not have to be promulgated by notice-and-comment rulemaking. 5 U.S.C. § 553(b)(A) (2012).

A recent article by Professors Adam Cox and Cristina Rodriguez justifying these exercises in executive reform of immigration law again suggests a further evolution away from positivism toward the process tradition. 131 Adam B. Cox & Cristina M. Rodriguez, The President and Immigration Law Redux, 125 Yale L.J. (forthcoming 2015)  (manuscript at 4) (on file with the Columbia Law Review) [hereinafter Cox & Rodriguez, Redux]. The article builds on and updates an earlier piece by the same authors, entitled The President and Immigration Law, 119 Yale L.J. 458 (2009) [hereinafter Cox & Rodriguez, President and Immigration Law]. One part of their argument consists of the claim that the President enjoys a heightened degree of autonomous authority in the field of immigration. Based on a survey of the history of immigration law, they argue that executive power in this area rests not only on express delegations of power from Congress, as the positivist tradition would require, but also on what they call “de facto delegation.” 132 Cox & Rodriguez, Redux, supra note 131 (manuscript at 3) (“The intersection of the immigration code and on-the-ground enforcement realities has given rise, on a large scale, to what we have termed the ‘de facto delegation’ of immigration screening authority to the President” (quoting Cox & Rodriguez, President and Immigration Law, supra note 131, at 512)). These de facto delegations are based on a variety of unilateral actions by Presidents to permit entry of immigrants for humanitarian or foreign policy reasons that were then either ratified or acquiesced in by Congress. 133 Id. (manuscript at 13) (discussing programs used by former Presidents to extend relief for humanitarian or foreign policy reasons). A second part of the argument is that the executive reforms are a worthy innovation because they make “the exercise of discretion more rule-like, more centralized, and more transparent.” 134 Id. (manuscript at 3). In other words, executive revision of the immigration laws earns plaudits under the norms associated with the process tradition. Indeed, the only criticism of the executive reforms offered by the authors is that “the process that produced them was opaque.” 135 Id. (manuscript at 62). The policies “might have benefitted from more procedural formality” like the notice-and-comment provisions of the APA. 136 Id. “The public deliberation facilitated by the proceduralist APA can increase the accountability of the policymaking process while also bolstering public confidence in the measures ultimately adopted.” 137 Id. Thus, like then-Professor Kagan and Professors Barron and Rakoff before them, the authors envision an enhanced sphere of presidential auton­omy largely free of constraints grounded in positive law and justify the legitimacy of policymaking within this sphere based on executive adher­ence to process norms. 138 For the argument that the President lacks constitutional authority to make broad dispensations that prospectively excuse legal violations, see Zachary S. Price, Enforcement Discretion and Executive Duty, 67 Vand. L. Rev. 671, 688 (2014).

There are several common themes among the three examples just surveyed. One is that they all represent an expansion of presidential power, rather than of traditional administrative agency authority. An executive department may be the instrument by which the policy initiative is implemented, but the decision to act comes from the White House. We are witnessing an aggrandizement of power by the Second Branch, not some expansion of authority by a mysterious Fourth Branch. A second is that the path of expansion follows various routes where action is likely to be immune from judicial review. Regulatory review escapes judicial scrutiny because OIRA is part of the EOP, which is not subject to the APA. 139 See supra note 96 and accompanying text (explaining regulatory review process’s location in EOP led to belief it was immune from judicial review under APA). Big waiver escapes review because the interests harmed by such action are diffuse general interests that lack standing to complain in court. 140 See supra notes 120–121 and accompanying text (noting executive waivers of statutory requirements are rarely subject to judicial review). And reform of immigration law by executive order escapes review (if traditional doctrine is followed) because decisions not to prosecute are not subject to judicial review. 141 Heckler v. Chaney, 470 U.S. 821, 831 (1985) (“This Court has recognized on several occasions . . . that an agency’s decision not to prosecute or enforce . . . is a decision generally committed to an agency’s absolute discretion.”). Like water flowing downhill in different channels, power expands where it meets no check from other sources of authority.

More fundamentally, we see in each of the examples a further evisceration of the positivist tradition. Regulatory review by OIRA has only the most gossamer foundation in enacted law—OMB’s asserted authority to control the timing of release of regulations for publication in the Federal Register. 142 See supra note 97 and accompanying text (noting lack of statutory foundation in OIRA review process makes it infirm from positivist perspective). Big waiver is said by its celebrants to require statutory waiver authority, 143 Barron & Rakoff, supra note 113, at 335 (“[T]o waive any, or at least major, substantive statutory provisions, there has to be explicit statutory authority.”). but in the absence of judicial review, there is little constraint on waivers of statutory requirements, and in some instances involving waivers of requirements under the ACA no such authority appears to exist. 144 See Nicholas Bagley, The Legality of Delaying Key Elements of the ACA, 370 New Eng. J. Med. 1967, 1969 (2014) (arguing postponing entire sections of ACA had no statutory justification and exceeded President’s authority to enforce law). Lastly, the reform of immi­gration law by executive order has emerged in a form that closely tracks legislation proposed in Congress that Congress has failed to enact. 145 See, e.g., Lauren Gilbert, Obama’s Ruby Slippers: Enforcement Discretion in the Absence of Immigration Reform, 116 W. Va. L. Rev. 255, 267–74 (2013) (noting similarities between DACA and DREAM Act). Cumulatively, these examples present the prospect of a revision of the constitutional order in which the President exercises autonomous policy­making authority without the need for any delegation of power from Congress, at least for the duration of the presidential administration.

Finally, we see in each of the examples an effort to legitimize the exercise of unilateral presidential power by invoking the norms of the process tradition. The most conventional move here is to emphasize the ways in which presidential policymaking has voluntarily adopted norms of openness, publication, and advance notification, i.e., “transparency,” and hence can be said to comport with the “rule of law” in the most minimal sense. 146 E.g., Kalhan, supra note 32, at 65. Commentators have urged the executive to offer more complete explanations for its initiatives, 147 See, e.g., Barron & Rakoff, supra note 113, at 327 (“A fundamental requirement of administrative law . . . is the agency’s duty to explain the decisions it makes.”). and there has been some effort along these lines, as in the Obama Administration’s release of the legal analysis of the OLC justifying the DAPA order. 148 Memorandum from Karl R. Thompson, Principal Deputy Assistant Att’y Gen., Office of Legal Counsel to the Sec’y of Homeland Sec. & the Counsel to the President, The Dep’t of Homeland Security’s Authority to Prioritize Removal of Certain Aliens Unlawfully Present in the United States and to Defer Removal of Others (Nov. 19, 2014), []. Entreaties to adopt some form of public participation, in the form of the APA’s notice-and-comment requirement, while popular with commentators, 149 See Cox & Rodriguez, Redux, supra note 131, at 62. have largely fallen on deaf ears within the administration. The most creative effort is then-Professor Kagan’s argument that presidential speeches and press releases satisfy the requirement of transparency, and presidential elections ensure accountability, and hence unilateral presidential policymaking is consistent with the process tradition, broadly conceived. 150 Kagan, supra note 4, at 2331–32. Whether future commen­tators will adopt these arguments to support initiatives like big waiver and reform by executive order remains to be seen.

III. Why Pure Process Review Will Not Work

The key question I wish to raise is whether an administrative law divorced from positivism and based solely on the process tradition will work in the long run. Can it meaningfully preserve the understanding that we live under a republican form of government subject to checks and balances? Can it preserve the values of stability, predictability, and equal treatment that we associate with the rule of law (however slippery that term may be)? We know that the positivist tradition can serve these ends. Whenever courts review governmental action for compliance with the Constitution or administrative action for compliance with a statute, they are reaffirming the supremacy of law ratified by the people or enacted by the people’s representatives, and they do so in a fashion designed to preserve stability of expectations about the meaning of these constraints. 151 See supra notes 5–8 and accompanying text (describing positivist tradition). We also know that the grand synthesis in administrative law developed in the twentieth century, which relied on a blending of positivism and process review, can serve these ends. Although courts supplemented a concern for fidelity to democratically enacted law with a concern for the process in which agencies acted, the process norms they developed were designed to facilitate public participation and under­standing, and courts endeavored to link these process norms to forms of enacted law. 152 See supra Part I (describing grand synthesis). The question is whether we can expect similar results from presidential administration, in which the sole form of admin­istrative law is an internally enforced commitment to the norms associated with the process tradition. 153 Kagan, supra note 4, at 2384 (answering question in affirmative).

It is difficult to disentangle questions about the prospect of pure process review from the availability of judicial review. At least in the American context, the intimations of pure process review surveyed in the last Part all arise in contexts where executive action is or is assumed to be immune from judicial review. 154 See supra notes 107, 127 and accompanying text (discussing immunity from judicial review). Where judicial review is available, something like the grand synthesis will persevere, at least for the fore­seeable future. Indeed, in the realm of judicial review, the positivist tradition—as manifested today most typically under step one of the Chevron doctrine—has become, if anything, more dominant than process review. 155 It is also reflected in a number of other judicial trends, such as the rise of textualism in statutory interpretation, the decline of federal common law, and the judicial hostility to implied private rights of action. But not all governmental policy is subject to judicial review. 156 See, e.g., supra note 107 and accompanying text (discussing OIRA immunity from judicial review). It is not difficult to imagine that future Presidents will continue to exploit the gaps where judicial review is not available and, building on these gaps, will seek to expand on presidential administration in ways perhaps not presently imaginable. 157 For one possible harbinger of things to come, consider the proposal to create an office within the White House that would set guidelines for discretionary enforcement authority across the entire administrative state. Kate Andrias, The President’s Enforcement Power, 88 N.Y.U. L. Rev. 1031, 1037–38 (2013). The central argument, predictably, is that centralization of enforcement power in this fashion would make the current system of prosecutorial discretion, which is “ad hoc, crisis driven, and frequently opaque,” much more transparent and accountable. Id. at 1031. And it is likely that efforts to square these initiatives with the rule of law will be expressed largely if not exclusively in terms of the process tradition.

The very idea of presidential administration is deeply problematic. Professor Strauss, for one, has long been skeptical of the idea that the President, without regard to any delegation from Congress of authority to perform such a role, is a “decider” rather than an “overseer” of the administrative state. 158 See Peter L. Strauss, Foreword: Overseer, or “The Decider”? The President in Administrative Law, 75 Geo. Wash. L. Rev. 696, 704–05 (2007) [hereinafter Strauss, Overseer] (“[W]here Congress has assigned a function to a named agency subject to its oversight and the discipline of judicial review, the President’s role—like that of the Congress and the courts—is that of overseer and not decider.”). Like Congress and the judiciary, he has observed, the President is a “they,” not an “it.” 159 See id. at 753–54 (discussing how presidential decisions are made by numerous, nonelected officials); see also id. at 715 (citing Lisa Schultz Bressman & Michael P. Vandenbergh, Inside the Administrative State: A Critical Look at the Practice of Presidential Control, 105 Mich. L. Rev. 47, 49–50, 68 (2006)) (noting study documenting that EPA received guidance from nineteen different White House offices which was often “conflicting” and “cacophonous”); cf. Kenneth A. Shepsle, Congress Is a They, Not an It: Legislative Intent as Oxymoron, 12 Int’l Rev. L. & Econ. 239, 248 (1992) (originating “they” versus “it” distinction in context of Congress). The EOP is itself a bureaucracy, superimposed on top of a much larger federal bureaucracy. Largely immune from judicial review, its functions are more political and much more weakly defined by legislation than those of the typical executive department or independent regulatory agency. 160 Cf. Peter L. Strauss, The President and the Constitution, 65 Case W. Res. L. Rev. 1151, 1163 (2015) (“The Executive Office of the Presidency has grown from a handful of officials tolerated by Congress . . . to hundreds of bureaucrats acting as intermediaries between President and agency, with ‘czars’ responsible for major policy concerns acting outside public administrative procedures and shielded by White House prerogatives from public view.”). As Professor Strauss has argued, it is appropriate that the EOP perform a supervisory and coordinating role—recommending budgetary appropriations, reminding agencies that they should exercise their discretion in ways that maximize aggregate social welfare, resolving policy disputes among agencies with overlapping authority, and acting as a constraint against excessive paper­work burdens on citizens. 161 See Strauss, Overseer, supra note 158, at 709 (noting President should only have “supervisory, not decisional, authority”). But to allow the EOP to displace the myriad agencies by becoming the “decider” would weaken legal constraints on administrative action, and deprive affected interests and individuals from having an effective voice in the implementation of regulatory policy. 162 See id. at 753–54 (arguing White House control of administrative decisionmaking would substitute White House employees with limited expertise and authority, “out of the reach of the APA and the Freedom of Information Act,” for politically accountable agency administrator acting with assistance from expert staff and operating with “enhanced transparency and procedural regularity”).

More fundamentally, presidential administration undermines the role of Congress in allocating power among governmental institutions. Only Congress, under the Supremacy Clause and the Necessary and Proper Clause, has authority “to arrang[e], order[], and distribut[e] power to act with the force of law among different institutions in society”—“to decide who decides.” 163 Thomas W. Merrill, The Disposing Power of the Legislature, 110 Colum. L. Rev. 452, 454 (2010). In contrast, “[t]he President has no inherent authority to make law, create institutions, set appropriation levels, or allocate enforce­ment authority among rival institutions.” 164 Id. at 473–74. The attempts by recent Presidents to occupy policy space not delegated to the White House by Congress are thus inconsistent with a fundamental design principle reflected in our evolved constitutional order.

To be sure, as Professors Posner and Vermuele point out, presi­dential administration is constrained by public opinion and by the need to maintain the President’s credibility with other political actors. 165 See Posner & Vermeule, supra note 29, at 12 (“[E]ven an imperial president is constrained by politics and public opinion.”). But these constraints operate primarily in the context of high-visibility presidential initiatives, such as those taken during the national security and economic crises that Professors Posner and Vermeule highlight in their book. 166 See id. at 12–15 (discussing increasing political constraints during perceived emergencies such as 9/11 and 2008 financial crisis). Yet as Professor Strauss rightly notes, “Given the over­whelming complexity and activity level of modern government, White House officials can attend to no more than a fraction of issues having to be decided.” 167 Strauss, Overseer, supra note 158, at 754. On a day-to-day basis, the regulatory state affects a vast array of interests that receive no media coverage and hence fail to register in opinion polls tracking the approval rating of the President. When the government sets safety standards for airplanes, 168 See 14 C.F.R. § 25 (2012) (codifying Federal Aviation Administration’s air­worthiness standards for passenger transport airplanes). regulates pipeline rates, 169 See 18 C.F.R. § 342 (2012) (codifying Federal Energy Regulatory Commission’s determination of oil pipeline rates and methodologies). or audits tax returns, 170 See I.R.C. § 7601 (2012) (codifying Secretary of Treasury’s authority to canvass districts for taxable persons or objects). the important constraints on the government are those found in the statutes that establish these functions and the process agencies follow in implementing and enforcing them. If one attends only to the statutory interpretation cases that reach the Supreme Court, or to the way the procedural requirements of the APA are implemented in national security emergencies, then it is possible to imagine that these constraints are infinitely plastic and manipulable. 171 See Posner & Vermeule, supra note 29, at 105 (“[I]t is inevitable, given the background conditions of the administrative state, that the norms governing judicial review of agency action will be embodied as loose standards and adjustable parameters.”); see also Adrian Vermeule, Our Schmittian Administrative Law, 122 Harv. L. Rev. 1095, 1098–105 (2009) (developing argument that modern administrative state cannot be constrained by rule of law). But this mistakenly generalizes from the extraordinary to the ordinary. To abandon the positivist tradition of administrative law, and allow large swathes of the administrative state to be taken over by a presidential administration subject only to the constraints of public opinion, would invite arbitrariness and oppression in a vast number of regulatory contexts that fly below the radar screen of media attention and public opinion.

Can presidential administration be rescued by calling on the process tradition, as argued by then-Professor Kagan and the other defenders of aggressive White House direction of the administrative state? Here, too, it is doubtful that the process tradition, as implemented by lawyers who are part of the executive branch, can serve the same legitimizing and constraining functions we associate with traditional forms of administrative law.

One problem concerns the authority of internal reviewing institu­tions to engage in process review absent some conferral of power to do so, either by the Constitution or a relevant statute like the APA. When acting in the positivist tradition, courts function as agents of sovereignty—either the sovereign people who have adopted the Constitution or the sover­eign legislature. Their judgments, assuming they are perceived as being faithful to the law, are backed by the sovereign power of the government. This means they are likely to be obeyed. An internal reviewing insti­tution, in contrast, is unlikely to have any statutory mandate to insist the administration adhere to judicially developed norms of reasonable process. Federal courts in the United States may have enough insti­tutional capital that they can insist that administrators adhere to norms of reasoned decisionmaking whether or not such norms are compelled by statute—at least for a time. But courts in other legal systems—not to mention internal review institutions in settings where judicial review does not exist—are unlikely to have enough institutional capital to impose their judgments about reasoned decisionmaking on other government actors. Enforcement of administrative law norms may come to be seen as merely a matter of contestable opinion. 172 See Mashaw, Creating the Administrative Constitution, supra note 18, at 304 (observing “reason lies in the eye of the beholder”). Instead of acting as a check on administrative abuse, administrative law could devolve into a rationalization for the exercise of raw power.

Another problem concerns the elusiveness of the elements of process review. Of course, interpretation of positive law is often sharply contested, especially where vague provisions (like due process or fair rates) are involved. 173 See, e.g., Obergefell v. Hodges, 135 S. Ct. 2584, 2596–99, 2604 (2015) (interpreting “due process of law” to mean same sex couples have right to marry, while acknowledging this possibility did not emerge until late in twentieth century). But often enacted law is quite clear, or has been determined over time to have a settled meaning, and in these contexts it imposes real constraints on government behavior—at least when it is understood that an independent institution like the courts stands ready to enforce these provisions. As we have seen, however, once process review slips beyond the confines of enacted law, it tends to fall back on gauzy generalizations like transparency and accountability that shift around from one context to the next and are hard to pin down. 174 See supra Part II (pointing out these issues with transparency and accountability). If policy announced at a presidential news conference is “transparent,” and directives from the White House are “accountable” if they enter in some small way into the approval ratings of the incumbent President, 175 See Kagan, supra note 4, at 2331–32 (“First, presidential leadership enhances transparency, enabling the public to comprehend more accurately the sources and nature of bureaucratic power. Second, presidential leadership establishes an electoral link between the public and the bureaucracy, increasing the latter’s responsiveness to the former.”). then the process tradition offers limitless possibilities for rationalizing uni­lateral policy initiatives taken at the direction of the President with or without any sanction in law.

Finally, an administrative law limited to the process tradition would have little or no capacity to enforce the evolved architecture of American government. The written Constitution has undergone considerable mutation over time through interpretation. But its basic postulates of separation of powers, federalism, and protection of individual rights continue to shape our political system. Maintenance of these postulates requires continued exercise of review in the positivist tradition. 176 See Jack M. Balkin, Living Originalism 35–58 (2011) (arguing written Constitution—especially rule-like provisions—continues to provide basic structure of government as we know it today). There is also the not-small matter of what Professor Strauss has called “Congress’s constitutional prerogatives in structuring government.” 177 Peter L. Strauss, Presidential Rulemaking, 72 Chi.-Kent L. Rev. 965, 985 (1997). How are the boundaries between different offices established by Congress going to be enforced if some external review agent does not enforce enacted law? It is true that non-judicial review, by institutions like the OLC in the Department of Justice, is possible, and these institutions can develop an internal culture that incorporates respect for enacted law. 178 See Trevor W. Morrison, Stare Decisis in the Office of Legal Counsel, 110 Colum. L. Rev. 1448, 1460–70 (2010) (discussing OLC’s structure and decisionmaking practices). Moreover, the interpretative norms employed by these institutions may mimic those developed by courts—especially if judicial review is a realistic prospect. 179 See id. at 1494 (“[A] rule of stare decisis similar to the one followed by courts has long been believed to inhere in the legal advisory function originally discharged by the Attorney General and later delegated to OLC.”). But experience has shown that internal review insti­tutions either bend to the political winds when they become imperative or are displaced by other “legal advisors” who are more overtly political in their orientation. 180 See Ackerman, supra note 100, at 109 (“We have seen that the entire setup at the OLC—its mode of recruitment, its relationship to the White House, its deference to ‘the views of the President who currently holds office’—propels its top lawyers toward presidential apologetics.”). The fundamental point is that process review, by asking whether individual government initiatives are “transparent” and their proponents are in some sense “accountable,” cannot address questions of government structure or individual rights. An administrative law de-linked from the positivist tradition will offer little resistance to power politics. This would be a tragedy for our ongoing experiment in democratic government.

Whether there is a solution to the challenge of presidential administration is beyond the scope of this Essay. The root of the problem is the inability of the Congress and the courts to expand their decisional capacities to match the demands of a rapidly changing and globalizing world, and the greater capacity of the White House to do so, at least in relative terms. To some extent, the capacity limitations of the legislature and judiciary are self-imposed, such as the Senate filibuster rules that require sixty votes (out of 100 Senators) to move a bill to the floor, 181 See U.S. Senate Comm. on Rules & Admin., Rules of the Senate, Rule XXII, [] (stipulating debate may only be closed if “three-fifths of the Senators duly chosen and sworn” vote to do so). and the Supreme Court’s restrictive rules of judicial standing. 182 See Lujan v. Defenders of Wildlife, 504 U.S. 555, 560–61 (1992) (plurality opinion) (setting out particular, narrow requirements for standing). But whether corrective mechanisms are available to head off the tide toward the “plebisticiary presidency” endorsed by Professors Posner and Vermeule 183 Posner & Vermeule, supra note 29, at 16. remains to be seen.


This Essay is about a growing phenomenon—executive or admin­istrative policymaking that exceeds the scope of authority delegated by democratically elected legislatures. Under conventional administrative law precepts, such unilateral exercises of power would be struck down under what this Essay calls the positivist tradition in administrative law. 184 See supra notes 5–8 and accompanying text (defining positivist tradition). But judicial review is incompletely available, and impatient and aggressive executives have increasingly exploited these gaps to engage in policymaking without delegated authority. 185 See supra Part II (describing rise of presidential administration and lack of judicial review). Defenders of these inno­vations have drawn upon a second tradition in administrative law, what this Essay calls the process tradition, to argue that unilateral executive policymaking can be reconciled with the rule of law as long as it is transparent and accountable. 186 See supra notes 32–34 and accompanying text (discussing defense of administrative actions as “transparent” and “accountable”).

The Essay has raised a number of questions about whether this justi­ficatory strategy is plausible. Lawyers working within the executive branch will have difficulty persuading their principals to adopt the precepts of the process tradition if they are not required to do so by law, and if the executive action is not subject to judicial review. Their task will be made more difficult by the elusive nature of the requirements of the process tradition. Even if they were to succeed, process review divorced from the enforcement of positive law would undermine the evolved structure of government, which is thoroughly dependent on enforce­ment of a complex body of statutory law, and rests on the fundamental precept that the legislature, not the executive, holds the power to allocate decisional power among the different institutions of our society.