An open access publication of the American Academy of Arts & Sciences
Summer 2021

Constraining Bureaucracy Beyond Judicial Review

Christopher J. Walker
View PDF

The modern regulatory state–and the field of administrative law that studies it–is in need of “deconstruction.” That does not mean that it should be dismantled entirely. This essay does not embrace the reformers’ fixation on courts as the bulwark against agency overreach. Rather, this essay develops the concept of bureaucracy beyond judicial review: not only agency actions that statute or judicial doctrine precludes from judicial review, but also agency actions that are technically subject to judicial review yet effectively insulated from it. Appreciating the phenomenon of bureaucracy beyond judicial review should encourage us to rethink theories and doctrines in administrative law. If judicial review provides no safeguard against potential abuses of power in most regulatory activities, we must turn to other mechanisms. All three branches of the federal government must play their roles, as should civil society and the agencies themselves.

Christopher J. Walker is the John W. Bricker Professor of Law at The Ohio State University Moritz College of Law and Chair of the American Bar Association’s Section of Administrative Law and Regulatory Practice. He previously clerked for Justice Anthony Kennedy in the U.S. Supreme Court and worked on the Civil Appellate Staff at the U.S. Department of Justice. He blogs regularly for the Yale Journal on Regulation and has published in such journals as the California Law Review, Georgetown Law Journal, Michigan Law Review, and Stanford Law Review.

The vast majority of federal lawmaking today takes place not in the halls of Congress, but in the bureaucratic trenches: by hundreds of thousands of political and career bureaucrats in Washington, D.C., and throughout the nation. As regulation rises and legislation declines, administrative law, too, grows in importance. Administrative law, after all, sets the ground rules for regulation. It dictates how federal agencies regulate and how the other federal government actors–the president, Congress, and the courts–supervise, review, influence, motivate, and constrain agency action. It also opens up space for public participation in the regulatory process, while attempting to close out undue outside influence and lobbying. When there is a change in presidential administration, administrative law enables law and policy change without legislative action. Indeed, with a Congress that has arguably lost much of its lawmaking ambition, change we can believe in must inevitably come from the administrative state. This ascendant vision of bureaucratic governance goes well beyond the “presidential administration” Elena Kagan articulated two decades ago.1

With this rise and rise–and further rise!–of the administrative state in federal lawmaking, it is no surprise that administrative law itself has become an ideological battleground.2 During the Obama administration, we began to see an upswing in scholars (largely conservative and libertarian) questioning the modern administrative state’s legitimacy in our constitutional order.3 In response, Gillian Metzger dedicated her foreword to the Harvard Law Review volume on the 2016 Supreme Court term to declare that the administrative state is “under siege” and to divide the legal academy into two camps: those who favor a robust administrative state and the “anti-administrativists.”4

More recently, legal scholar Jeffrey Pojanowski attempted to bring granularity to this us-versus-them dichotomy by disaggregating the field into three main camps.5 The “administrative supremacy” camp views the administrative state as constitutionally necessary for modern governance. Courts should not patrol agencies’ substantive actions or their choice of procedures, only review to encourage effective governance. “Administrative skepticism,” by contrast, is formalist in nature and finds much of the modern administrative state unconstitutional. Courts should review de novo administrative interpretations of law, utilize the nondelegation doctrine to strike down broad statutory delegations, and otherwise embrace judicial doctrines that constrain bureaucratic action.

“Administrative pragmatism,” which Pojanowski situates in between these two extremes, “seeks to reconcile the reality of administrative power, expertise, and political authority with broader constitutional and rule-of-law values.” In many respects, administrative pragmatism is the conventional view, reflected in current administrative law doctrine and regulatory practice. Pojanowski argues for a neoclassical alternative to administrative skepticism, in which courts would not defer to administrative interpretations of law but would defer to agency policy decisions. It would disarm the constitutional calls to deconstruct the modern regulatory state. Instead, it would encourage courts to faithfully interpret the Administrative Procedure Act and the agencies’ organic statutes to ensure agencies do not exceed their statutory authority.

However administrative law scholars are categorized, it is beyond serious dispute that the academic criticisms of the modern administrative state have risen over the last decade, and the academic rebuttals and defenses have followed.6 These academic criticisms have made their way from the ivory tower into the real world (and vice versa, perhaps). A growing number of federal judges and members of Congress (again, largely conservative and libertarian) have called for administrative law reform. For example, they have argued for eliminating judicial deference to administrative interpretations of law and for reinvigorating the nondelegation doctrine to strike down as unconstitutional broad statutory grants of lawmaking authority to federal agencies.7

Donald Trump’s election as president, moreover, ushered in a deregulatory agenda, one that perhaps went beyond a typical Republican presidential administration. Shortly after the 2016 election, President Trump’s chief strategist Stephen Bannon grabbed headlines by demanding a “deconstruction of the administrative state.”8 The Trump administration took many measures to curtail administrative governance, even in ways that inhibit the president’s power to make law and policy through the executive branch. Reforms to agency guidance, adjudication and enforcement policies, rulemaking processes, and the civil service come immediately to mind. Yet the Trump administration also leveraged the regulatory state to wield administrative power in unprecedented ways. One need look no further than its various sweeping immigration regulatory actions as well as its attempts to respond to the COVID-19 pandemic independent of Congress.

One would think that the Trump administration’s regulatory actions would cause even “administrative supremacists” to become concerned about bureaucratic sprawl and overreach–perhaps even more so as the field of administrative law took a critical race theory turn during the summer of 2020.9 Administrative skeptics certainly have not changed their tune about the need to rein in the regulatory state. The vast majority of administrative law scholars, however, are not what Pojanowski labels administrative supremacists. Nor, of course, are they administrative skeptics. Instead, they are administrative pragmatists who view the modern administrative state as imperfect yet necessary. These pragmatists recognize the importance of both enabling administrative discretion and constraining that exercise of discretion to avoid arbitrary and capricious agency action. In shaping administrative law, they promote values such as agency expertise, reasoned decision-making, due process, fairness, consistency, transparency, and public accountability in administrative governance.

In other words, the vast majority of administrative law scholars have always been concerned with constraining bureaucratic power. And many of us–particularly administrative skeptics but also many administrative pragmatists–are growing increasingly concerned about the shift from legislation to regulation to make major policy decisions at the federal level and what that means for the future of administrative law. Yet our focus has been myopically court-centric. Administrative law, as a field, has long fixated on the role of federal courts in reviewing and constraining agency action. Each year hundreds of law review articles are published on administrative law’s judicial deference doctrines and other standards of judicial review. Indeed, since its birth in 1984, the Supreme Court’s landmark judicial deference decision in Chevron v. Natural Resources Defense Council has been cited on Westlaw more than ninety thousand times, including in more than twenty thousand law review articles and other secondary materials. In the last year alone, the Chevron decision has appeared in more than fifteen hundred secondary materials. As legal scholars Kevin Stack and Peter Strauss have argued, the history of our approach to teach administrative law has no doubt also contributed to the field’s emphasis on courts.10

This judicial focal point should come as no surprise. Federal courts serve as a critical safeguard in the modern administrative state. But it is a mistake to focus just on courts. Much of administrative law happens without courts. Put differently, federal agencies regulate us in many meaningful, and sometimes frightening, ways that either evade judicial review entirely or are at least substantially insulated from such review. I am not the first to make this observation in the field of administrative law. Among others, Jerry Mashaw has been examining this phenomenon for decades, including in his seminal book Bureaucratic Justice.11 No doubt sparked by Mashaw’s work, internal administrative law has become a trending subfield in administrative law.12

To be sure, scholars of public administration have spent decades developing theories about internal bureaucratic organization and control.13 In the field of administrative law, however, a more comprehensive and sustained inquiry is needed, especially for those of us intent on “deconstructing” the modern administrative state by strengthening safeguards against bureaucratic overreach. This essay focuses on the state of the administrative law field, but more interaction with these other fields is sorely needed. To help move this work forward, this essay sketches out a research agenda for a more systemic investigation into this phenomenon, which I will call bureaucracy beyond judicial review. I have two main goals.

First, in the field of administrative law, the concept of bureaucracy beyond judicial review is undertheorized. The conventional account focuses on one under-inclusive category of agency action: where judicial review is expressly precluded by statute or judicial doctrine. If our goal is to constrain bureaucracy beyond judicial review, at least three additional categories deserve attention. On the one hand, judicial review is technically available for many agency actions, yet for a variety of reasons they never make it to federal court. On the other, even agency actions that make it to court are often subject to deferential standards of review that create an administrative policy-making space insulated from judicial review.

This agency policy-making space is further complicated by the fact that federal agencies play a substantial, judicially unreviewable role in drafting the statutes (and presidential budgets and executive orders) that govern them. In other words, federal agencies have the potential to essentially self-delegate the bureaucratic power that is insulated from judicial review. In theorizing bureaucracy beyond judicial review in the first part of this essay, I draw on recent examples from both the Obama and Trump administrations.

Second, understanding bureaucracy beyond judicial review should encourage us to rethink existing theories and doctrines in administrative law. So much scholarly attention has focused on refining judicial deference doctrines and standards of review to strike the right balance of allowing agencies to reasonably exercise their expertise yet rein in arbitrary exercises of agency discretion. But because judicial review provides no adequate safeguard against potential abuses with respect to these regulatory activities, we must turn to other actors and actions. We must develop a theory of administrative law that better incorporates the various other actors who can help monitor, constrain, and protect against agency abuse in regulatory activities insulated from judicial review.

That does not mean we give up on judicial review. When reframed in light of bureaucracy beyond judicial review, administrative law’s theory of judicial review would focus not just on the individual cases that make it to court but also on how courts can have a more systemic effect on those agency actions that never reach them. The second part of this essay explores how courts, Congress, and the agencies themselves can help counteract the dangers of bureaucracy beyond judicial review.

The conventional account of bureaucracy beyond judicial review focuses on agency actions that statute or judicial doctrine expressly excludes from the courts’ purview. The founders of the Administrative Procedure Act (APA) of 1946 envisioned that some agency actions would be precluded from judicial review. Indeed, in Section 701(a) of the APA, Congress makes clear that the APA does not apply when “(1) statutes preclude judicial review; or (2) agency action is committed to agency discretion by law.” These two categories arguably make up the standard view of bureaucracy beyond judicial review. Each category merits some further elaboration.

It is not uncommon for Congress to statutorily exclude judicial review for certain agency actions. Immigration law is a prime example. In Department of Homeland Security v. Thuraissigiam (2020), the Supreme Court confronted the constitutionality of the lack of judicial review for one such agency action: expedited removal of noncitizens at or near the border.14 Expedited removal is one form of what immigration scholars have coined “shadow removals” or “speedy deportations”: where Congress has generally precluded not only Article III (of the U.S.  Constitution) judicial review but even administrative review in an Article II immigration court.15 The Thuraissigiam Court rejected constitutional challenges to expedited removal under both the Due Process Clause and the Suspension Clause. In dissent, Justice Sonia Sotomayor declared that the “decision handcuffs the Judiciary’s ability to perform its constitutional duty to safeguard individual liberty and dismantles a critical component of the separation of powers.”

The breadth of shadow removals is staggering. In 2018, immigration judges, who are agency adjudicators within the Justice Department’s Executive Office for Immigration Review, received roughly three hundred thousand cases and concluded more than two hundred thousand cases.16 Those cases receive administrative review in the immigration courts. If the noncitizens are ordered removed at the conclusion of the administrative proceedings, they generally can seek further review in an Article III federal court. But, as immigration law scholar Jennifer Koh has documented, the vast majority of removal orders today never make it to immigration court. They are issued through shadow removals “by front-line immigration officers acting as investigator, prosecutor, and judge, thus bypassing the immigration courts entirely.”17 Indeed, in 2018, more than four out of five removals were shadow removals, conducted without a formal administrative hearing or Article III judicial review.

Many agency actions are not judicially reviewable because they are “committed to agency discretion by law.” Agency enforcement discretion is the quintessential example. As the Supreme Court held in Heckler v. Chaney (1985), agencies enjoy a form of prosecutorial discretion for enforcement decisions: a “presumption that agency decisions not to institute enforcement proceedings are unreviewable.”18 In Department of Homeland Security v. Regents of the University of California (2020), the Court confronted this category of discretionary agency action in the context of the Trump administration’s decision to rescind the Obama administration’s Deferred Action for Childhood Arrivals program (DACA). There, again, the Court reaffirmed that agency enforcement decisions are generally unreviewable as committed to agency discretion. Yet the Court disagreed that DACA is just a nonenforcement policy, as DACA also grants certain benefits.19

This nonreviewable agency discretion extends not just to under-enforcement but also to over-enforcement. Or, as legal scholar Mila Sohoni calls it, “crackdowns.” A crackdown is “an executive decision to intensify the severity of enforcement of existing regulations or laws as to a selected class of offenders or a selected set of offenses.”20 Consider the Trump administration’s immigration enforcement crackdown in San Francisco and surrounding cities. In 2018, reports swirled that Immigration and Customs Enforcement (ICE) sought to arrest more than 1,500 noncitizens and that the crackdown was motivated in part by California’s decision to become a sanctuary state and thus not fully cooperate with the federal government to enforce federal immigration law. Indeed, ICE’s acting director publicly declared: “California better hold on tight”; if state and local officials “don’t want to protect their communities, then ICE will.”21

Deciding when and where to dedicate enforcement resources is a powerful regulatory tool. Agency decisions to refrain from enforcement benefit the potential enforcees. And they harm the beneficiaries of the potential enforcement action: the consumers, competitors, investors, employees, and so forth, whose rights and interests go unprotected by the regulators’ decision not to enforce the laws. Conversely, when agencies decide to crack down, the subjects of the crackdown suffer, whereas similarly situated regulated parties do not, for reasons beyond the control of the regulated. That, too, can create arbitrary advantages and disadvantages for similarly situated regulated parties, in addition to the accompanying externalities for third parties. Yet courts generally cannot patrol agency decisions on when and how to wield their enforcement authority.

The concept of bureaucracy beyond judicial review should also include agency actions for which judicial review is technically available, yet for a variety of reasons never make it to federal court. High-volume agency adjudication is a classic example. As I have explored elsewhere, the Article III federal judiciary receives outsized attention compared with the attention paid to the federal administrative judiciary.22 After all, more than twelve thousand agency adjudicators hold hearings and decide cases, compared with fewer than nine hundred Article III judges. Much has been made of the Trump administration’s appointment of some two hundred Article III judges. Yet its hiring of nearly two hundred and fifty Article II immigration judges has hardly been noticed (outside of immigration law circles).

In the realm of formal agency adjudication, one perhaps would not anticipate discovering bureaucracy beyond judicial review. After all, formal adjudication involves trial-like agency proceedings before an administrative law judge or some other agency adjudicator, where the parties have the statutory right to seek judicial review of the agency’s final decision. But even formal agency adjudication can be insulated from judicial review. This is particularly true for mass agency adjudication–such as immigration, Social Security, and veterans’ adjudications–in which only a fraction of cases ever reach federal courts.

Let us return to immigration adjudication. As noted above, immigration courts decide several hundred thousand cases per year. According to one 2015 study, roughly two in five immigrants in removal proceedings in immigration court had legal representation, and less than half of those represented had representation for all of their agency hearings.23 Unsurprisingly, immigrants represented by counsel are more likely to prevail: that same study found that represented immigrants won tenfold (21 percent) more often than unrepresented immigrants (2 percent). That is in part because unrepresented immigrants were fifteen times less likely to even seek relief from removal. The lack of legal representation no doubt plays a significant role in creating the stark disparities in the immigration adjudication system, and in preventing many potentially successful claims from reaching an Article III court. There’s a reason why a seminal empirical study on immigration adjudication labels the system “refugee roulette.”24

So what does this mean for the phenomenon of bureaucracy beyond judicial review? Because noncitizens often navigate agency adjudication without legal representation, it is much more likely that individuals will not seek judicial review of erroneous agency decisions. Either they lack the knowledge or resources to navigate the process, or they have otherwise procedurally defaulted meritorious claims in the administrative process. Thus, courts never have the opportunity to directly help these individuals. Courts’ ability to correct agency errors is limited to the subset of cases in which individuals have the wherewithal to seek judicial review.

Subregulatory guidance is another context in which agency action is substantially insulated from judicial review. The conventional understanding is that agency guidance does not have the force of law, and thus is not judicially reviewable absent the agency’s application of that guidance in enforcement or adjudication. Whether agency guidance is actually nonbinding on regulated parties is subject to considerable debate. For instance, last year, the Justice Department issued an interim final rule that sets forth a number of requirements and procedures for creating agency guidance documents, including that “guidance documents may not be used as a substitute for regulation and may not be used to impose new standards of conduct on persons outside the Executive Branch.”25

Regardless of whether agency guidance can be formally binding yet escape judicial review, it often functionally binds regulated parties in ways insulated from judicial review. As legal scholar Nicholas Parrillo has documented, even when agency guidance is not legally binding, regulated parties often have strong incentives to comply due to significant risks of agency enforcement, certain agency preapproval requirements, the need to maintain a good relationship with the agency, or “intra-firm constituencies for compliance beyond legal requirements.”26

Indeed, in the context of the Obama administration’s “dear colleague letter” to universities on Title IX sexual harassment claims procedures, some scholars observed that, “terrified, administrators not only complied; they over-complied.”27 To be sure, the universities could have sought judicial review. They could have refused to comply, and then challenged in court the agency’s enforcement decision or the federal government’s withdrawal of all federal funding. But the stakes (losing all federal funding) were obviously too high. And it certainly does not encourage regulated parties to seek judicial review when, under the Auer deference doctrine, the court must defer to the agency’s regulatory interpretation advanced in agency guidance “unless plainly erroneous or inconsistent with the regulation.”28 I have previously called this effect regulation by compliance.29

In discussing the potential dangers of agency guidance, I do not mean to suggest we should abandon it. Agency guidance serves important purposes. Its role in the modern regulatory state is critical. My point is that it is greatly insulated from judicial review. And as Parrillo observes, administrative law scholarship on guidance “misses much about the everyday workings of guidance that pervade the administrative state, for it focuses on the tiny fraction of guidance documents that get challenged in litigation, and only on the kinds of facts about guidance that reach the courts.”30

Bureaucracy beyond judicial review should also encompass the administrative policy-making space that administrative law’s judicial deference doctrines create. Chevron deference is perhaps the prime example. As the Supreme Court explained in the Chevron decision itself, the reviewing court “need not conclude that the agency construction was the only one it permissibly could have adopted to uphold the construction, or even the reading the court would have reached if the question initially had arisen in a judicial proceeding.”31 Agencies thus enjoy Chevron policy-making “space” to regulate in ways subject to judicial review only for their reasonableness.32

This Chevron policy-making space is real and substantial. In reviewing every published federal court of appeals decision from 2003 through 2013 that refers to Chevron deference, administrative law scholar Kent Barnett and I found a difference of nearly twenty-five percentage points in agency-win rates when judges decide to apply the Chevron deference framework.33 And once the circuit courts got to Chevron’s second step, agencies prevailed 93.8 percent of the time.

It is also clear that federal agencies are keenly aware of this Chevron space. In a survey of 128 federal agency rule drafters, Chevron deference was the most-known interpretive tool by name (94 percent) and most reported as playing a role in agency rule drafting (90 percent) among twenty-two interpretive tools included in the survey.34 Nearly nine out of ten rule drafters agreed or strongly agreed that they think about subsequent judicial review when drafting statutes, and two out of every five rule drafters surveyed agreed or strongly agreed–with another two in five somewhat agreeing–that a federal agency is more “aggressive” in its interpretive efforts if it is confident that Chevron deference applies (as opposed to some less-deferential standard).

In other words, agencies know they have policy-making space under Chevron; not surprisingly, they act differently when they believe the threat of judicial invalidation is low.

How Chevron’s policy-making space enhances bureaucracy beyond judicial review is further complicated by the fact that federal agencies play a substantial role in drafting the laws that delegate them that space in the first place. As I have documented elsewhere, federal agencies are substantially involved in the legislative process. They propose substantive legislation to Congress and provide confidential technical drafting assistance on nearly all legislation drafted by congressional staffers that affect the agency.35 Legal scholars Eloise Pasachoff and Tara Leigh Grove have similarly documented the substantial role federal agencies play in the drafting of presidential budget and executive orders, respectively.36

Courts, of course, review enacted statutes to determine their meaning and constitutionality. But courts do not review how agencies participate in statutory drafting (or in the drafting of presidential directives). They do not assess if agencies self-delegate lawmaking authority by leaving statutory mandates broad and ambiguous, much less the role agencies may play in drafting statutes that eliminate judicial review of agency action altogether. This judicially insulated legislative role is remarkable in and of itself. But it may well also compound the problematic lack of judicial review for the categories of agency action discussed above. After all, all of these agency actions are at least in part creatures of statutes–statutes the agencies themselves helped create.

In light of the vast, underexplored terrain of bureaucracy beyond judicial review, how should administrative law theory and doctrine adjust? As I noted at the outset, administrative law as a field must exit the courtrooms and enter into the expansive world of external and internal laws, doctrines, and practices that assist the various actors who monitor, constrain, and protect against agency abuse in regulatory activities that are insulated from the courts. Here, I focus on the three branches of the federal government. But states and civil society obviously also play important constraining roles.

The Judicial Branch. Federal courts must view their role in the modern administrative state as one of more than mere error correction. Much ink has been spilled arguing for shrinking or eliminating the Chevron policy-making space. Others have argued to make certain actions more judicially reviewable, such as enforcement decisions, agency guidance documents, and agency actions currently precluded from judicial review by statute. Many of these proposals would likely require congressional action, or at least a judicial philosophy that disregards stare decisis (law by precedent) and the Bickelian “passive virtues” I generally embrace.

In light of bureaucracy beyond judicial review, however, courts could more fully embrace one substantial shift in mindset: courts should view their role in the administrative state not only as reviewing the agency actions that reach them but also as engaging in a dialogue with the political branches. This vision reorientation may be particularly important in the context of high-volume agency adjudication, where many individuals have meritorious claims but lack the wherewithal to seek judicial review. As I have documented elsewhere, federal courts possess a toolbox of dialogue-enhancing tools that they can employ when remanding flawed agency adjudications back to the agency.37

Where courts are skeptical of the agency getting it right on remand, concerned about undue delay, or worried about the petitioner getting lost on remand, some courts require the agency to provide notice of its final determination, retain panel jurisdiction over the matter, or set deadlines for an agency response to the remand. Others suggest (or order) that administrative judges be replaced on remand, certify issues for decision on remand, or set forth hypothetical answers in dicta or concurring opinions. Some courts, moreover, obtain concessions from the government at argument to narrow the potential grounds for denial of relief on remand. And courts through their published opinions can set off fire alarms for Congress, the president, and the public to draw attention to potential systemic issues in a regulatory process.

These tools help courts play a more active role in improving equity, efficiency, and consistency in the agency adjudication system generally, rather than just the limited number of cases that make it to a federal court. Yet the tools still respect the proper separation of powers by using mere words instead of orders that may exceed their statutory (or, in some cases, perhaps constitutional) authority. Using this toolbox is one example of how judicial review in administrative law should be enhanced to address the present-day realities of mass agency adjudication and other bureaucratic actions that otherwise evade judicial review.

The Executive Branch. The executive branch itself can play a powerful role in constraining bureaucracy beyond judicial review. Here, I focus on the role of the agencies themselves, and leave for another day the role of the president and centralized regulatory review. The APA and the agencies’ organic statutes set the minimum procedural requirements for agency action. The Supreme Court has repeatedly reaffirmed that federal “agencies are free to grant additional procedural rights in the exercise of their discretion.”38 Agencies do so through the creation of internal administrative law, which encompasses a wide range of internal agency procedures, structures, practices, and guidance that seek to constrain their exercises of discretion. This Vermont Yankee “white space,” as legal scholars Emily Bremer and Sharon Jacobs have called it, has the potential to serve as a potent defense against agency overreach, especially in the context of bureaucracy beyond judicial review.39 The universe of internal law that could constrain bureaucracy is vast, and I have surveyed it elsewhere.40 But a few examples for each type of bureaucracy beyond judicial review can illustrate the constraining power of internal administrative law.

For agency actions where judicial review is precluded by statute or judicial doctrine, federal agencies can embrace a variety of internal procedures to protect individuals in those processes. On the shadow removals front, for example, the agency could establish internal review procedures and additional procedural protections. It could create what civil rights law scholar Margo Schlanger has termed an “office of goodness”: an internal ombuds office that looks out for the rights of noncitizens in the informal adjudicative process and ensures the agency complies with its external and internal laws.41

The Internal Revenue Service’s (IRS) Taxpayer Advocate Service provides a model that may be worth adapting in other agency contexts.42 The Taxpayer Advocate Service is an independent office within the IRS with two distinct main objectives. First, it has physical offices in every state, where individual taxpayers can get free help with their tax problems with the IRS. Second, leveraging these tens of thousands of annual individual interactions nationwide, the Service is required to report regularly to Congress to recommend systemic reforms to the federal tax system. Similar internal structures and procedures could be beneficial in the context of agency enforcement discretion.

For agency actions that are technically subject to judicial review but often evade such review, federal agencies have and should continue to adopt internal laws to protect individuals subject to those regulatory processes. On the mass adjudication front, the Administrative Conference of the United States (an independent federal agency that studies administrative procedure) regularly recommends best practices, including public availability of practice rules, availability of adjudication materials on agency websites, establishment of recusal rules for adjudicators, best practices for assisting self-represented individuals, and a sweeping suite of procedural protections for agency hearings. Agencies have also adopted appellate review systems and other quality assurance programs by internal law.43 The Social Security Appeals Council is a prominent example: a creature of internal administrative law that now consists of nearly one hundred administrative appeals judges and officers and processes more than one hundred thousand appeals per year.

Finally, when it comes to an agency’s policy-making space created by judicial deference doctrines, the Administrative Conference has recommended a number of best practices agencies can adopt to increase public participation and accountability, including targeting and meeting with knowledgeable or affected parties for feedback, improving online access to rulemaking dockets and related materials, utilizing social media to improve public engagement and awareness of rulemaking activities, and drafting rules in plain language for better public comprehension, just to name a few. As the preceding paragraphs illustrate, the Administrative Conference plays an important role in assessing internal agency laws and practices and identifying best practices agencies can embrace internally to help address bureaucracy beyond judicial review.

Federal agencies can also bind themselves internally to seek only judicial deference if they follow certain procedures. As noted above, last year the Justice Department issued an interim final rule requiring the agency to follow certain procedures when creating guidance documents, with heightened procedures for “significant guidance documents.” The rule instructs the agency not to seek any Auer deference in litigation for a guidance document that does not “substantially comply” with these requirements. Along similar lines, immigration law scholar Shoba Wadhia and I have argued that the Justice Department and the Department of Homeland Security should create internal administrative law that shifts the default for major policy-making in the immigration context from agency adjudication to notice-and-comment rulemaking.44 Like the Justice Department’s new rule for agency guidance, this new immigration internal law should instruct the immigration agencies not to seek Chevron deference for agency statutory interpretations promulgated through agency adjudication (while preserving Chevron deference for rulemaking). We argue that shifting the default from adjudication to rulemaking for immigration policy-making is more consistent with Chevron’s theoretical foundations: to leverage agency expertise, to engage in a deliberative process, and to increase political accountability.

The Legislative Branch. Congress must play a more prominent role in constraining bureaucracy beyond judicial review. As legal scholar Josh Chafetz has documented, Congress possesses a suite of hard powers (power of the purse, personnel power, contempt authority) and soft powers (freedom of speech and debate, internal discipline, cameral rules) that it can employ to constrain the administrative state.45 Congress should utilize this toolbox to address agency actions that evade judicial review. And administrative law scholars should dedicate more attention to exploring how Congress can better wield these powers in this context; they should, in turn, also leverage the ample literature on the subject in other fields.

At the end of the day, though, increased congressional oversight is unlikely to be sufficient to effectively constrain bureaucracy beyond judicial review. The same is true for senatorial pressure during the confirmation process for the administration’s nominees to run the agencies. So, too, with using appropriations power to influence administrative policy change. Congress must also reinvigorate its ambition to legislate and revisit the often decades-old statutes that empower federal agencies.

To encourage Congress to return to passing laws on a regular basis, legal scholar Jonathan Adler and I have argued that Congress should embrace the practice of regular reauthorization of statutes that govern federal agencies.46 This engagement would include regular assessment of agency action and regular recalibration, if the agency’s regulatory activities are inconsistent with the current Congress’s policy objectives. In some regulatory contexts, it may require Congress to enact reauthorization incentives, such as sunset provisions designed to induce legislative engagement. In other contexts, Congress may decide that the costs of mandatory reauthorization outweigh the benefits. Nevertheless, Congress should more regularly use reauthorization to mitigate the democratic deficits that come with broad delegations of lawmaking authority to federal agencies. It goes without saying that, as with many proposals to reform Congress, ours would require a greater investment in congressional capacity–in terms of staffing and other resources.

A regular reauthorization process could have dramatic effects on constraining bureaucracy beyond judicial review. Congress would, for example, have to choose whether to continue to preclude judicial review by statute in certain circumstances. In the hearings leading up to reauthorization, it would have an opportunity to hear from the agency and those affected by agency enforcement decisions, and it could apply pressure for the agency to modify its enforcement policies or even legislate to constrain such discretion. For agency actions that are judicially reviewable but often evade review, Congress could similarly assess those systems through reauthorization hearings and could codify best practices for quality assurance, offices of goodness, and the like.

Regarding the agency policy-making space created by judicial deference doctrines, regular reauthorization could play an important role. For many of us, Chevron deference has become far more problematic in the current era of congressional inaction. Congress appears to have insufficient capacity or willpower to intervene when an agency has used statutory ambiguity to pursue a policy inconsistent with current congressional wishes, much less when an agency’s organic statute is so outdated as to not equip the agency with authority and direction to address new technologies, challenges, and circumstances. A regular reauthorization process would alleviate many of these concerns.

It is also possible that Congress would consider eliminating or narrowing judicial deference with respect to certain subject matters or administrative processes. Legal scholar Kent Barnett has explored how Congress did so in the Dodd-Frank Act with respect to the Office of the Comptroller of the Currency’s statutory interpretations that preempt state law.47 Similarly, as noted above, Wadhia and I have argued that Congress should eliminate Chevron deference in the immigration adjudication context, while preserving it for notice-and-comment rulemaking.

Appreciating the phenomenon of bureaucracy beyond judicial review should encourage us to rethink theories and doctrines in administrative law, and to reconsider the direction of the field of administrative law. So much scholarly attention has focused on refining judicial deference doctrines and standards of review to strike the right balance of allowing agencies to employ their expertise to reasonably exercise their statutorily vested discretion while reining in arbitrary exercises of agency discretion. Administrative skeptics seem to have similarly fixated on courts, calling for the elimination of Auer and Chevron deference and the reinvigoration of an exacting nondelegation doctrine.

But if judicial review provides no safeguard against potential abuses of power in most regulatory activities, we must turn to other mechanisms. All three branches of the federal government must play their roles. As should civil society and the agencies themselves. (When it comes to the agencies, this also must include the role of a professionalized civil service.) This is the type of “deconstruction of the administrative state” that deserves more scholarly and real-world attention.

© 2021 by the American Academy of Arts & Sciences. Published under a CC BY-NC 4.0 license.

author’s note

This essay draws from the author’s address at the V International Congress on Institutional Theory at the Federal University of Rio de Janeiro, Brazil. It also weaves together a number of distinct lines in the author’s research agenda; the endnotes attribute such reliance. The themes presented in this essay are further developed in the author’s forthcoming book, Constraining Bureaucracy: Rethinking Administrative Law in a System without Courts (Cambridge University Press).


  • 1Elena Kagan, “Presidential Administration,” Harvard Law Review 114 (8) (2001): 2245.
  • 2Gary Lawson, “The Rise and Rise of the Administrative State,” Harvard Law Review 107 (6) (1994): 1231.
  • 3See, for example, Richard Epstein, The Dubious Morality of Administrative Law (Lanham, Md.: Roman and Littlefield, 2020); Philip Hamburger, Is Administrative Law Unlawful? (Chicago: University of Chicago Press, 2014); and Joseph Postell, Bureaucracy in America: The Administrative State’s Challenge to Constitutional Government (Columbia: University of Missouri, 2017).
  • 4Gillian E. Metzger, “1930s Redux: The Administrative State Under Siege,” Harvard Law ­Review 131 (1) (2017). Metzger included me among the anti-administrativists, and I share many of the confessions Aaron Nielson expressed in his response. See Aaron Nielson, “Confessions of an ‘Anti-Administrativist,’” Harvard Law Review 131 (1) (2017).
  • 5Jeffrey A. Pojanowski, “Neoclassical Administrative Law,” Harvard Law Review 133 (2019): 852.
  • 6See, for example, Jon D. Michaels, Constitutional Coup: Privatization’s Threat to the American Republic (Cambridge, Mass.: Harvard University Press, 2017); Cass R. Sunstein and Adrian Vermeule, Law and Leviathan: Redeeming the Administrative State (Cambridge, Mass.: Harvard University Press, 2020); and Paul R. Verkuil, Valuing Bureaucracy: The Case for Professional Government (Cambridge: Cambridge University Press, 2017).
  • 7For a survey of such criticisms, see Christopher J. Walker, “Attacking Auer and Chevron Deference: A Literature Review,” Georgetown Journal of Law and Public Policy 103 (1) (2018); and Jonathan H. Adler and Christopher J. Walker, “Delegation and Time,” Iowa Law Review 105 (5) (2020): 1931.
  • 8Philip Rucker and Robert Costa, “Bannon Vows a Daily Fight for ‘Deconstruction of the Administrative State,’” The Washington Post, February 23, 2017.
  • 9In summer 2020, in the wake of the killing of George Floyd and subsequent Black Lives Matter protests across the nation, the Yale Journal on Regulation Notice and Comment blog published more than two-dozen essays as part of its symposium on racism in administrative law. Those essays are collected here.
  • 10Kevin Stack, “Lessons from the Turn of the Twentieth Century for First-Year Courses on Legislation and Regulation,” Journal of Legal Education 65 (1) (2015): 28; and Peter Strauss, “Review Essay: Christopher Columbus Langdell and the Public Law Curriculum,” Journal of Legal Education 66 (1) (2016): 157.
  • 11Jerry L. Mashaw, Bureaucratic Justice: Managing Social Security Disability Claims (New Haven, Conn.: Yale University Press, 1983). See also Nicholas R. Parrillo, ed., Administrative Law from the Inside Out: Essays on Themes in the Work of Jerry L. Mashaw (Cambridge: Cambridge University Pres, 2017).
  • 12See, for example, Gillian E. Metzger and Kevin M. Stack, “Internal Administrative Law,” Michigan Law Review 115 (8) (2017): 1239.
  • 13See, for example, Donald P. Moynihan, “‘Our Usable Past’: A Historical Context Approach to Administrative Values,” Public Administration Review 69 (5) (2009): 813–821.
  • 14Department of Homeland Security v. Thuraissigiam, 140 S. Ct. 1959 (2020).
  • 15Shoba Sivaprasad Wadhia, Banned: Immigration Enforcement in the Time of Trump (New York: New York University Press, 2019), chap. 5; and Jennifer Lee Koh, “When Shadow Removals Collide: Searching for Solutions to the Legal Black Holes Created by Expedited Removal and Reinstatement,” Washington University Law Review 96 (2) (2018).
  • 16Executive Office for Immigration Review, Statistics Yearbook Fiscal Year 2018 (Washington, D.C.: U.S. Department of Justice, 2019), Tables 4 and 5.
  • 17Jennifer Lee Koh, “Removal in the Shadows of Immigration Court,” Southern California Law Review 90 (2) (2017): 181, 183.
  • 18Heckler v. Chaney, 470 U.S. 821, 837 (1985).
  • 19Department of Homeland Security v. Regents of the University of California, 140 S. Ct. 1891, 1906 (2020).
  • 20Mila Sohoni, “Crackdowns,” Virginia Law Review 103 (1) (2017): 31, 33.
  • 21Hamed Aleaziz, “Feds Planning Massive Northern California Immigration Sweep to Strike against Sanctuary Laws,” SFGATE, January 17, 2018.
  • 22Christopher J. Walker, “Charting the New Landscape of Administrative Adjudication,” Duke Law Journal 69 (8) (2020). See also Christopher J. Walker and Melissa F. Wasserman, “The New World of Agency Adjudication,” California Law Review 107 (2019): 141.
  • 23Ingrid V. Eagly and Steven Shafer, “A National Study of Access to Counsel in Immigration Court,” University of Pennsylvania Law Review 164 (1) (2015): 7–8.
  • 24Jaya Ramji-Nogales, Andrew I. Schoenholtz, and Philip G. Schrag, Refugee Roulette: Disparities in Asylum Adjudication and Proposals for Reform (New York: New York University Press, 2011).
  • 2528 C.F.R. § 50.27(C)(1)(I).
  • 26Nicholas R. Parrillo, Federal Agency Guidance: An Institutional Perspective (Washington, D.C.: Administrative Conference of the United States, 2017).
  • 27Elizabeth Bartholet, Nancy Gertner, Janet Halley, and Jeannie Suk Gersen, “Fairness for All Students Under Title IX,” Digital Access to Scholarship at Harvard, August 21, 2017.
  • 28Auer v. Robbins, 519 U.S. 452, 461 (1997).
  • 29Christopher J. Walker, “Administrative Law Without Courts,” UCLA Law Review 25 (2018): 1620.
  • 30Parrillo, Federal Agency Guidance, 5.
  • 31Chevron U.S.A. v. Natural Resource Defense Council, 467 U.S. 837, 843 N.11 (1984).
  • 32Peter L. Strauss, “‘Deference’ Is Too Confusing–Let’s Call Them ‘Chevron Space’ and ‘Skidmore Weight,’” Columbia Law Review 112 (5) (2012): 1143.
  • 33Kent Barnett and Christopher J. Walker, “Chevron in the Circuit Courts,” Michigan Law Review 116 (1) (2017): 5–6, 30 (Figure 1).
  • 34Christopher J. Walker, “Inside Agency Statutory Interpretation,” Stanford Law Review 67 (5) (2015): 999, 1019 (Figure 1), 1020 (Figure 2).
  • 35Christopher J. Walker, “Legislating in the Shadows,” University of Pennsylvania Law Review 165 (6) (2017): 1377.
  • 36Eloise Pasachoff, “The President’s Budget as a Source of Agency Policy Control,” Yale Law Journal 125 (8) (2016): 2182; and Tara Leigh Grove, “Presidential Laws and the Missing Interpretive Theory,” University of Pennsylvania Law Review 168 (2020): 877.
  • 37Christopher J. Walker, “The Ordinary Remand Rule and the Judicial Toolbox for Agency Dialogue,” George Washington Law Review 82 (5) (2014): 1553.
  • 38Vermont Yankee Nuclear Power Corporation v. National Resources Defense Council, 435 U.S. 519, 524 (1978).
  • 39Emily S. Bremer and Sharon B. Jacobs, “Agency Innovation in Vermont Yankee’s White Space,” Journal of Land Use and Environmental Law 32 (2017): 523.
  • 40Christopher J. Walker and Rebecca Turnbull, “Operationalizing Internal Administrative Law,” Hastings Law Journal 71 (5) (2020): 1225.
  • 41Margo Schlanger, “Offices of Goodness: Influence without Authority in Federal Agencies,” Cardozo Law Review 36 (1) (2014): 53.
  • 42See Nina E. Olson, “Taxpayer Rights, Customer Service, and Compliance: A Three-Legged Stool,” University of Kansas Law Review 51 (2003): 1239.
  • 43See, for example, David Ames, Cassandra Handan-Nader, Daniel E. Ho, and David Marcus, “Due Process and Mass Adjudication: Crisis and Reform,” Stanford Law Review 72 (1) (2020).
  • 44Shoba Sivaprasad Wadhia and Christopher J. Walker, “The Case Against Chevron Deference in Immigration Adjudication,” Duke Law Journal 70 (2021): 1197.
  • 45Josh Chafetz, Congress’s Constitution: Legislative Authority and Separation of Powers (New Haven, Conn.: Yale University Press, 2017). See also Christopher J. Walker, “Restoring Congress’s Role in the Modern Administrative State,” Michigan Law Review 116 (6) (2018): 1101 (reviewing Chafetz, Congress’s Constitution).
  • 46Adler and Walker, “Delegation and Time.”
  • 47Kent H. Barnett, “Codifying Chevmore,” New York University Law Review 90 (1) (2015).