The Yale Law Journal

February 2022

Agency Control and Internally Binding Norms

Administrative Law

abstract. Lower courts have consistently held that agencies may not issue guidance that purports to bind the public. In their parlance, guidance cannot create a “binding norm” on regulated parties. The courts have been far less clear, however, on the extent to which guidance can appropriately bind the issuing agency or its staff. Courts have approached this issue in widely divergent ways, and some have held that guidance cannot even bind low-level agency officials.

In the shadow of this uncertainty, agencies continue to use guidance as an important tool for internal administration. Guidance facilitates bureaucratic supervision of frontline officials, enables agency administrators to exercise the agency’s discretion transparently, and communicates the agency’s interpretations of the law to both internal and external actors. To serve these functions, guidance must impose some binding norms on agency staff. Despite the importance of guidance to the internal operations of the administrative state, little empirical work exists to shed light on how agencies design and deploy these policies and whether their practices comport with the assumptions of the binding-norm doctrine.

To fill this gap, this Note conducts a comparative assessment of the guidance practices at three agencies: the Equal Employment Opportunity Commission, the Occupational Safety and Health Administration, and U.S. Citizenship and Immigration Services. Drawing on employee manuals, briefings in response to litigation, and interviews with agency insiders, I evaluate how these agencies use guidance to manage the discretion and work of agency staff. I find that officials at each agency believe that guidance must necessarily be capable of binding internal agency actors, particularly frontline officials, to effectuate consistent and transparent internal administration. These findings reveal a disconnect between the actual practices at these agencies and recent judicial decisions invalidating agency guidance. To resolve this discrepancy, I propose a new contextual approach to judicial review of guidance that encourages courts to distinguish between the internal and external binding effects of guidance. This approach weighs the underlying authority of the agency to act absent the guidance, the power of the agency generally to create indirect binding effects on regulated parties, the agency’s internal procedures for contesting guidance, and the audience for the guidance to determine whether the guidance is appropriate or simply a substitute for a legislative rule.

author. Yale Law School, J.D. 2021; University of California, Los Angeles, B.A. 2017. Special thanks to Nicholas Parrillo for supervising this project and for his invaluable support and guidance throughout its development. I am also indebted to Michael Beechert, Monica Bell, Nathaniel Donahue, Blake Emerson, Lucas Guttentag, James Linnell, Christopher Otmar, Cristina Rodríguez, Simone Seiver, and Melody Wang for their insightful comments and support, and to the participants in the Spring 2021 Yale Law Journal Student Scholarship Workshop for their helpful suggestions. I am grateful to Jackson Busch, Joe Linfield, Rachel Sommers, and the editorial staff of the Yale Law Journal for their excellent suggestions throughout the publication process. Finally, I thank my mother, Skarlete Nabavi-Noori, for her endless support in all my endeavors. All errors are my own.


Administrative agencies carry out their missions, delegated to them by Congress, in part by issuing legislative rules. Agency legislative rules (also referred to as “regulations”) vastly outpace the legislative output of Congress and, together with ordinary statutes, create a web of requirements that regulated parties must adhere to. Still, these legislative rules and statutes often leave many unanswered questions about what the law requires and consequently leave agencies with significant discretion in enforcing their mandates. As a result, agencies also generate significant amounts of “guidance”: general statements of policy or interpretations of existing law that advise the public—and agency staff—on how the agency intends to exercise this discretion.

Scholars, practitioners, and courts have long sought to develop a reliable means of distinguishing agency legislative rules from agency guidance.1 Though a seemingly mundane distinction, the stakes involved in setting boundaries around guidance documents are high. Although guidance documents, unlike legislative rules, cannot formally bind regulated parties, they comprise the vast majority of the regulatory output of the administrative state and collectively play a central role in structuring agency action and discretion.2 These documents can determine, for example, the procedures that workplace-safety inspectors dutifully follow in carrying out their on-site safety inspections, or whether a frontline immigration officer will favorably exercise discretion to provide an individual with relief from removal.3

The fundamental distinction between agency guidance documents and legislative rules is grounded in the Administrative Procedure Act of 1946 (APA). To promulgate new legislative rules, the APA ordinarily requires that agencies undergo a set of time-consuming and resource-intensive procedures, including notice and comment, which allow regulated parties and interested members of the public to play a role in shaping eventual regulations.4 These legislative rules carry the force of law and can create new rights for or obligations on regulated parties. By contrast, agencies can issue guidance documents with few procedural requirements at all thanks to the APA’s exemptions for “general statements of policy” and “interpretative rules.”5 Consequently, agencies can issue guidance documents in far greater volume, at a much faster pace, and with far less accountability than they can promulgate legislative rules. Agencies therefore use guidance to rapidly respond to their experience with regulations in the real world and to ease the burden of implementing large, complex policy initiatives.6

Unlike legislative rules, however, guidance documents lack formal legal force. In the language of the courts, agencies may not use these documents to establish a “binding norm.”7 Instead, guidance must satisfy two criteria. First, it must “act[] prospectively,” proceeding carefully so as not to “impose any [new] rights and obligations” on a regulated party.8 Second, and more controversially, the guidance must “leave[] the agency and its decision-makers free to exercise discretion.”9 To enforce this test, the courts engage in a legal fiction, determining that agency guidance that inappropriately creates such a binding norm is not guidance at all, but instead a procedurally invalid legislative rule that should have undergone notice-and-comment rulemaking. So, to qualify for the APA’s guidance exemption from notice-and-comment procedures, guidance documents may only telegraph what an agency might do in a certain scenario, leaving room for alternative approaches proposed by regulated parties and allowing agency staff to exercise discretion.

The paradigmatic use case for permissible guidance under this test arises where a statute or regulation leaves businesses, individuals, or other regulated parties uncertain exactly how to achieve compliance, either because it does not anticipate a particular factual scenario or because the relevant requirements are vague, leaving room for competing interpretations. An agency might then choose to issue guidance to address these ambiguities. The guidance would offer regulated parties insight into how they might achieve compliance with regulatory mandates in the face of this uncertainty, while emphasizing that the agency’s view is tentative and remains open to alternatives. In other words, the guidance is not “binding” on the regulated parties; it is merely a suggestion.

Much of the existing scholarship on agency guidance and the binding-norm test is preoccupied with the external effects of guidance documents—that is, their effects on regulated parties. Scholars have, for example, long been skeptical that agencies use guidance as merely a tool to clarify existing regulatory obligations. Instead, these scholars argue that agencies regularly use guidance to create new obligations without having to undergo notice and comment. As a result, much has been written to address the question of just when agencies go too far over the line in writing guidance that is unlawfully “binding” on regulated parties—and whether agencies do so regularly and with malicious intent to subvert otherwise necessary procedures.10 Recently, however, a new body of empirical research has begun to challenge this simplistic view of guidance, elucidating the subtler practical realities that define the relationship between agencies and regulated parties. Most notably, a recent study by Nicholas R. Parrillo on behalf of the Administrative Conference of the United States (ACUS) identifies structural and institutional factors that help explain why regulated entities may choose to comply with guidance—and, in fact, even request that agencies produce more guidance to help them achieve compliance—rather than ignore it.11 Nevertheless, the existing doctrine and its emphasis on rooting out impermissibly “binding norms” are still rooted in the traditional skepticism of guidance’s external effects.

There is, however, another important function for agency guidance that has received far less scholarly attention in connection with the binding-norm test: how agencies use guidance internally as a tool for managing the work of their staffs. Like supervisors at any complex organization, agency administrators need to direct and control the work of lower-level employees. In addition to the important role guidance documents play in structuring the relationship between agencies and regulated parties, they also represent a crucial mechanism through which agency administrators communicate with agency staff about how to exercise the agency’s often considerable discretion. This communication often happens in the form of enforcement guidelines, field guidance, policy manuals, and a range of other documents that agency supervisors use to encourage the consistent and transparent exercise of the agency’s discretion by frontline officials.

On its face, the binding-norm test poses a serious challenge for the internal functioning of agency guidance. As often written and recited by the courts, the test emphasizes that guidance must leave the “agency and its decision-makers free to exercise discretion” without distinguishing between staff at different levels of the agency hierarchy or further elaborating on where this decision-making flexibility must be located.12 The test therefore suggests that an agency cannot compel its own frontline employees, who must routinely make decisions and exercise discretion on behalf of the agency, to follow the agency’s own guidance documents. At least some courts have adopted this position, insisting that even frontline officials must be free to exercise individual enforcement discretion unbound by guidance promulgated by the agency head13—a result that would significantly undermine agencies’ ability to engage in bureaucratic supervision of low-level staff. A more modest view would suggest that so long as the agency itself remains free to exercise discretion to depart from guidance, either through the agency head or through appeal to supervisors within the agency, then the guidance can avoid creating an impermissible binding norm.14 But the question of which of these views, if any, reflect the empirical realities of how agency administrators expect guidance to operate within the agency, or, for that matter, how frontline agency officials expect to carry out their roles has yet to be explored.

Indeed, the everyday internal workings of agency guidance documents are “rife with ‘unknown unknowns.’”15 As a result, important questions remain as to how agencies train and instruct frontline personnel to consistently apply the agency’s view of the law without impermissibly running afoul of the binding-norm test. How do agencies promote or discourage flexibility in implementing guidance at different levels of the agency’s decision-making hierarchy? How much rigidity do agencies expect from frontline staff in implementing agency guidance? If guidance is nominally binding at one level of the agency, are there easily accessible mechanisms through which regulated parties can request departure from higher-ups? And to what extent does the agency’s managerial hierarchy impact the force of guidance within the agency? These questions are crucial to understanding whether the application of the binding-norm test to internal guidance is consistent with how agencies practically administer internal guidance, and whether the benefits of the binding-norm test outweigh the potential costs to agencies’ ability to engage in legitimate internal administration.16

Answers to these questions have doctrinal, theoretical, and practical implications. Doctrinally, the binding-norm test has frequently been referred to as “enshrouded in considerable smog”17 and may be one of the “most frequently litigated and important issue[s]” of administrative law before the courts today.18 The test is often applied to guidance that purportedly binds regulated parties, but in recent years courts have increasingly invoked it in circumstances where the primary effect of the guidance is to create internal uniformity among agency staff in their exercise of the agency’s discretion.19 Although nearly every guidance document will have effects on both external regulated parties and internal agency actors, these recent applications of the binding-norm test threaten to erase the distinction between these two effects. Erasing this distinction has resulted in courts prohibiting agencies from binding their own decision makers in pursuit of policing the permissible effects of guidance on regulated parties, but without regard for the consequences that such decisions have on internal agency administration. Although scholars and courts are engaged in continuing efforts to clarify the distinction between guidance and legislative rules as they relate to regulated parties, the case law on whether and how guidance may purport to bind agency staff remains hotly contested and theoretically murky.20 The issue also remains live, attracting continuous litigation and producing an unresolved 4-4 split at the Supreme Court.21

Theoretically, understanding how agency administrators deploy guidance internally implicates important questions about the “meaning and boundaries of law itself in the administrative state.”22 From the perspective of traditional skeptics of guidance documents, even guidance that is only internally binding can create avenues for agencies to skirt procedural rules and engage in substantive regulation without undergoing the costly notice-and-comment process.23 From the agency’s perspective, however, these documents are part of the bread-and-butter toolset used to carry out the agency’s mission and can collectively contribute to the agency’s “internal law” of administration. They facilitate accountable bureaucratic supervision of agency staff by focusing and directing the agency’s discretion from the highest, most politically accountable levels; ensure uniform and fair application of the underlying regulatory or statutory requirements across agency staff; and communicate the agency’s internal views to the public.

Finally, the debate about the proper role of agency guidance—including within agencies as a tool of agency administration—has become a deeply political one with significant practical implications for the operation of the administrative state. Congress, the President, and the courts continue to lend credibility to the skeptical view of guidance as a potentially dangerous tool that enables runaway administrative overreach.24 In recent years, these concerns have grown louder in conservative circles, particularly after the Obama Administration’s issuance of controversial guidance documents that effected changes in education,25 immigration,26 and employment discrimination.27 Consequently, the Trump Administration was active in not only rescinding guidance documents from the Obama Administration,28 but also openly questioning the proper role of guidance within the administrative state generally29 and proposing new rules that would have significantly curbed the ability of agencies to produce any guidance at all.30 Many agency officials, by contrast, see limitations on issuing guidance that binds agency staff as a threat to their ability to carry out their regulatory mission.31

This Note responds to these debates and contributes to the empirical literature on agencies’ use of guidance documents. It turns inward to examine internal agency practices, with a special focus on how agencies of various designs use guidance to manage staff discretion in light of the binding-norm test. To this end, I have chosen three agencies—the U.S. Equal Employment Opportunity Commission (EEOC), the Occupational Safety and Health Administration (OSHA), and U.S. Citizenship and Immigration Services (USCIS)—spanning a core cross section of the administrative state for a comparative assessment of their practices.32 For each agency, I draw on employee manuals, document descriptions from the agency websites, briefings in response to litigation on guidance, and interviews with agency officials and industry attorneys to evaluate how these agencies deploy guidance to manage the discretion and work of agency staff.

This analysis reveals crucial differences between the internal and external binding effects of agency guidance that remain unrecognized by the current binding-norm test. Namely, agency officials depend on internal guidance as an important tool to cabin the discretion of frontline employees and ensure uniform application of the law—purposes that stand in tension with the binding-norm test as applied by the courts. These differences lead to a doctrinal mismatch: although the binding-norm test may make sense where courts are concerned about impermissible regulation of external parties without the benefit of notice-and-comment procedural protections, it does not reflect how or why agencies use guidance internally. A stringent application of the binding-norm rule against frontline agency decision makers without a deeper examination of agency practice and structure fails to account for the realities of internal agency administration.

The remainder of this Note is organized as follows. Part I discusses the role of guidance within agencies’ internal administration and the judicial controls on this practice. Next, Part II analyzes how three agencies with diverse organizations and missions formulate and implement guidance within their hierarchies. Finally, in Part III, the Note concludes by proposing a new contextual approach to judicial review of guidance that encourages courts to consider the internal and external binding effects of guidance separately. This approach calls for courts to weigh the underlying authority of the agency to act absent the guidance, the power of the agency generally to create indirect binding effects on regulated parties, the agency’s internal procedures for contesting guidance, and the audience for the guidance to determine whether the guidance is appropriate or simply a substitute for a legislative rule.