Reviewing Presidential Orders

Introduction

In early 2017, a newly inaugurated President Donald J. Trump tried to force policy change through a flurry of written orders. While some opponents took to the streets to protest,1 others identified a different forum for resistance: the federal courts.2 Lawyers, mobilizing at a breakneck pace, sued the President in name to enjoin the implementation of several of his signature orders, including his first travel ban and an executive order involving sanctuary cities.3 Within weeks, plaintiffs succeeded in securing preliminary relief from the courts.4 Resort to the judicial branch thereby allowed litigants, in dramatic fashion, to thwart several of Trump’s earliest policymaking efforts.

As Trump’s presidency progressed and additional orders flowed from his desk, more lawsuits arose, including challenges to orders affecting transgender troops, land conservation, tax policy, regulatory rollbacks, and more.5 These separate lines of litigation already have generated extensive commentary. Yet observers generally have overlooked what these lawsuits represent as a whole: a new and particularly forceful form of litigation aimed at the president.

Prior to Trump’s entrance into the White House, litigants only occasionally brought lawsuits directly and immediately against the initiatives contained in a president’s written orders.6 Instead, litigants tended to wait for an agency to act in response to a president’s order and then challenged the agency’s action, rather than the president’s order, using now well-established administrative law principles.7 This approach remained constant even as presidents—from Ronald Reagan forward—began to deploy executive orders and other written directives in increasingly heavy-handed ways to control executive-branch policymaking.8 For example, early in his tenure, President Reagan issued a highly controversial order, Executive Order No 12291, that directed agencies to consider costs and benefits when engaging in rulemaking.9 Despite the opposition it garnered, adversaries did not immediately file suit challenging the order’s legality. Instead, they waited for agencies to act in response to Executive Order No 12291 and then filed lawsuits challenging specific agency actions.10 Similarly, when President Barack Obama issued a written directive to the Department of Labor (DOL) ordering it to “modernize and streamline” rules governing overtime pay, litigants did not challenge Obama’s instruction itself.11 Instead, they waited for DOL to issue a final rule on the subject. They then challenged DOL’s rule, not the President’s written directive.12 These two examples illustrate the usual chronology surrounding challenges that implicate written presidential orders: after a president issues an order that provides the bureaucracy with instructions, opponents of the order wait for the order to be implemented by an executive-branch official and then file suit against the executive-branch actor, not against the president himself.

Not so in the Trump era. Trump’s adversaries have filed a multitude of lawsuits directly against the President, challenging a wide range of presidential orders almost immediately after the orders were made public.13 For example, when Trump issued an order—his so-called “one-in, two-out” order—directing executive agencies to rescind two regulations each time the agency finalizes a new regulation,14 opponents did not wait for agencies to act. Instead, just one week after Trump signed the executive order, opponents filed a lawsuit naming the President himself as one of the defendants.15 Many of the lawsuits filed in opposition to Trump-era policies fit this new model—namely, they involve direct and immediate challenges to the legality of presidential orders, and they name the President himself as a defendant.16

This Article identifies the novelty of this new form of litigation. It also demonstrates just how far these lawsuits fall outside of the well-developed and extensively theorized legal framework that governs challenges to agency action. A partial explanation lies in the Administrative Procedure Act17 (APA), a statute that has provided the scaffolding for more than seventy years’ worth of judicial and scholarly discussion about judicial review of agency action. In 1992, the Supreme Court concluded that the APA reaches only agencies, not the president.18 As a result, the APA simply does not apply to claims brought against the president, including claims that directly challenge the legality of a presidential order.19

Nor do existing judicial precedents provide anything close to a well-developed or coherent legal framework for courts to follow when reviewing presidential orders. In 2018, the Ninth Circuit acknowledged as much when it noted that “[i]n contrast to the many established principles for interpreting legislation, there appear to be few such principles to apply in interpreting executive orders.”20 Indeed, only a smattering of preexisting precedents even speak to judicial review of presidential orders,21 and the few precedents that do exist tend to proceed in a highly case-specific manner that offers little guidance going forward.22 Some of these existing precedents are based on doctrines that predate the APA—doctrines that have been largely ignored since its enactment and that therefore suffer from decades of judicial neglect.23 Meanwhile, presidents over the past several decades have become increasingly bold in their attempts to control agency policymaking.24 The net result is that existing judicial precedents involving presidential orders tend not to reflect the present-day reality of presidential involvement in federal policymaking.

Until recently, this absence of a well-developed framework to guide judicial review of presidential orders did not prove particularly problematic. This is because litigants tended to wait to challenge agency action rather than the presidential orders themselves.25 However, the recent flurry of lawsuits aimed at various Trump-era orders has forced courts and litigants to grapple with a host of difficult and unsettled legal issues relating to, among other things, justiciability, deference, remedies, and more. Often, judges have been called upon to resolve these complex issues in an expedited posture and under intense public scrutiny.

Although this new burst of litigation began with Trump, it is not likely to recede when he leaves office. Instead, the recent rise in litigation aimed at the President may well suggest an enduring change in the way litigants challenge executive-branch policies—one that reflects not only the controversial nature of Trump’s tenure but also, on a broader level, the increasingly aggressive attempts by presidents, over decades, to control executive-branch regulatory activity.26 The convergence of these trends, and the likely permanence of this new style of litigation, confirms the need for a legal framework to help courts navigate this particularly thorny form of judicial review, both now and in the future. Such a framework cannot easily be developed through case-by-case rulings issued in highly charged and often fast-moving cases. Rather, a cohesive framework to guide judicial review of presidential orders would best be formed through deliberative discussion among scholars, judges, litigants, and members of Congress.

In the hopes of initiating such a discussion, this Article sketches out what a framework for judicial review of presidential orders might look like, as well as what specific doctrines should be included.27 In so doing, this Article uses the term “presidential orders” as an inclusive shorthand, intending it to cover various forms of unilateral written directives publicly issued by the President, regardless of whether a given directive is formally labeled as an “executive order,” a “proclamation,” or a “presidential memorandum.” The labels generally have no bearing on the substance or the legal effect of presidential orders, and presidents tend not to use these labels in a consistent fashion.28

In tackling the subject of judicial review of presidential orders, this Article proceeds in four parts. Part I describes the robust and now well-developed administrative law framework that courts use when assessing the legality of agency action, including presidentially influenced agency action. Part II, in contrast, describes the relative dearth of precedents that speak to judicial review of presidential orders. As Part II describes, prior to the Trump presidency, the courts had little occasion to develop a cohesive framework for reviewing presidential orders. This was because direct challenges to presidential orders remained infrequent and sporadic, even as presidents—from Reagan forward—began deploying written directives in increasingly heavy-handed ways to influence agencies’ policymaking work. Part III turns to the Trump era, describing the emergence of a new and forceful form of litigation aimed directly at presidential orders. It then describes how this new kind of legal challenge to executive-branch action—combined with prior presidents’ increasingly forceful efforts to direct agencies’ policymaking activities—confirms the need for the development of a coherent legal framework to guide judicial review of presidential orders not only now but also in the future.

Finally, Part IV sets forth what such a framework might look like.29 It begins by identifying several high-level differences between presidential action and agency action. With these differences in mind, Part IV identifies and analyzes a number of specific legal doctrines that should form part of any coherent approach to judicial review of presidential orders. These legal doctrines include those relating to: threshold reviewability doctrines, such as standing, ripeness, and cause of action requirements; standards of review; and the availability and appropriateness of judicial relief, as well as questions of severability.

I. Judicial Review of Presidentially Influenced Agency Action

The United States’ legal system has undergone various transformations throughout its history. Most notably, in the second half of the twentieth century, the nation’s legal system shifted from one governed primarily by congressional statutes to one governed largely by regulatory law created by administrative agencies within the executive branch.30 As a result of this shift, Congress no longer plays the starring role in setting major policies that govern everything from air quality standards to mortgage disclosure requirements.31 Instead, executive-branch agencies have taken center stage.

Given the massive amount of regulatory power that Congress chose to hand over to executive-branch agencies, it was only a matter of time before presidents would start to try to direct agencies’ regulatory work. Indeed, this is precisely what happened beginning in the 1970s and 1980s.32 As we will describe in more detail in Part II, presidents’ mounting focus on agencies’ regulatory activities eventually led to the entrenchment of what then-Professor and now-Justice Elena Kagan has called an era of “presidential administration”—one in which presidents routinely exert significant control over agencies’ regulatory actions, often via written directives.33 For now, however, we focus on a different—but also likely inevitable—consequence of the ascendance of the modern administrative state: the development of various administrative law doctrines that, although sometimes murky around the edges, provide familiar guidance to courts engaged in judicial review of agency action.

Among administrative law’s many central doctrines, three are particularly relevant to this Article. They are: (1) arbitrariness review, sometimes referred to as “hard look” review; (2) procedural review; and (3) Chevron deference. All three doctrines have particular salience to this Article’s study of presidential orders because, as we discuss, they are regularly used by the courts to review agency action, including agency action that has been directed or influenced by the president. Yet, in all the cases that we discuss in this Part, the courts’ review is limited to the lawfulness of agency action. None of the doctrines was designed to facilitate judicial review of challenges brought directly against a president’s own actions.

A. Arbitrariness Review

One of the most important and well-established doctrines governing judicial review of agency action involves what is known as arbitrary and capricious review.34 This form of judicial review has its roots in § 706(2)(A) of the APA, which directs courts to “hold unlawful and set aside agency action” that is found to be “arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with the law.”35 Although § 706(2)(A) originally was understood to call for only minimal judicial review akin to mere rationality review,36 things changed in the 1960s and 1970s.37 At that time, agencies’ regulatory powers were surging, and concerns were mounting that agencies were being captured by private interests.38 In response, a number of federal judges crafted a ramped-up version of § 706(2)(A)’s arbitrariness review, which came to be called “hard look” review.39 Using this more aggressive take on arbitrariness review, the courts carefully scrutinized agency action to ensure that agencies took “a ‘hard look’ at the salient problems,”40 adequately considered all relevant factors, and provided thoroughly reasoned explanations to support their actions.41

Though it was difficult to reconcile with traditional understandings of § 706(2)(A), the Supreme Court eventually came to embrace hard look review. This endorsement came in 1983 in a case involving agency action that had been influenced by the president: Motor Vehicle Manufacturers Association v State Farm Mutual Auto Insurance Co.42 In this case, an agency rescinded a Carter-era safety standard soon after President Reagan swept into the White House on a deregulatory platform.43 In response to a challenge to the agency’s decision, the Supreme Court rejected the highly deferential form of review advanced by the government. The Court instead carefully parsed the fine-grained details of the agency’s explanation and, in the end, concluded that it had failed to adequately justify its rescission of the Carter-era rule.44

In a separate opinion, then–Associate Justice William Rehnquist brought up the politics behind the agency action, noting that the “agency’s changed view of the [safety] standard seems to be related to the election of a new President of a different political party.”45 Yet nothing in the majority opinion expressly acknowledged the political backstory.46 Instead, the majority opinion focused on the adequacy of the agency’s technocratic justifications.47 As a result, State Farm generally has been read to require that agencies explain their decisions in expert-driven rather than political terms,48 even when presidential pressure may have played an important part in the rulemaking. Indeed, courts now routinely use hard look review to assess the legality of agency action, including agency rules that have been influenced by the president or others within the White House.49

One more recent example of hard look review, as applied to presidentially influenced agency action, emerged from Department of Homeland Security’s (DHS) 2012 policy of Deferred Action for Childhood Arrivals (DACA). The DACA policy, which DHS put into place during Obama’s presidency, enabled certain undocumented immigrants who had been brought to America as children (often referred to as “Dreamers”) to seek work authorization, as well as the postponement of deportation.50 Although President Obama publicly and unambiguously announced his support for DACA, he never signed an executive order or other directive embracing the policy or setting forth its details.51 Rather, DACA was put into place through an agency memorandum—one signed by Janet Napolitano, the Secretary of Homeland Security.52

After President Trump entered the White House in 2017, DHS decided to rescind DACA. It did so by issuing another memorandum—this one signed by Elaine Duke, the then–Acting Secretary of Homeland Security.53 Nevertheless, Trump made his involvement clear, tweeting on the day DHS announced the rescission that if Congress failed to find a solution for the Dreamers, he would “revisit the issue.”54

DHS’s 2017 decision to end DACA triggered challenges,55 and the first judge to weigh in on one of these challenges ruled at the preliminary-injunction stage that DHS’s rescission of DACA was likely arbitrary.56 Specifically, the judge reasoned in January 2018 that DHS had terminated DACA based on a mistaken point of law,57 which rendered its decision arbitrary and capricious.58 This judge’s ruling was preliminary, and it was quickly appealed.59 Nevertheless, the opinion highlights the powerful role that § 706(2)(A) arbitrariness review can play in judicial review of agency actions, including those directly influenced and supported by the president.

By contrast, as we discuss later, the legal landscape is quite different for litigants challenging a president’s order directly. This is because § 706(2)(A) of the APA does not apply to actions of the president.60 As a result, it is not at all clear what standard of review—arbitrariness or otherwise—might apply when courts are asked to review the legality of presidential orders.61

B. Procedural Review

Another well-established type of judicial review of agency action involves review of the procedures that agencies use when they make regulations. Although different procedural requirements apply in different contexts,62 one of the most important sets of procedural constraints emanates from § 553 of the APA.63 This section speaks to the specific procedures that agencies often must use when making rules. In particular, the text of § 553 mandates that agencies: (1) issue some kind of a notice of the proposed rulemaking (NPRM), which could merely include a “description of the subjects and issues involved”; (2) accept written comments from interested persons; and (3) issue a “concise general statement” explaining the final rule’s “basis and purpose,” referred to as a statement of basis and purpose (SOBP).64

Despite the seeming simplicity of these procedural requirements, the courts have layered a variety of additional requirements on top of § 553’s text. These judicial glosses emerged in the 1960s and 1970s as the importance of rulemaking was surging, and they persist to this day.65 For example, the courts have interpreted § 553’s notice requirement to mean that agencies must disclose technical data and studies on which the agencies relied in formulating their rules.66 In addition, the courts have declared that agencies must respond to every “significant” comment made by interested parties who participate in the rulemaking.67

Despite the immense importance of these sorts of procedural requirements, there is a catch: the procedural glosses that courts have layered on top of § 553 apply only when agencies enact certain kinds of rules, not others.68 In particular, § 553’s notice-and-comment requirements do not apply when agencies issue policy statements—meaning agency pronouncements that merely advise the public of how the agency might exercise its discretion moving forward and that neither impose any legal rights or obligations nor overly restrict an agency’s discretion.69 By contrast, § 553’s notice-and-comment requirements do apply when agencies enact legislative rules—meaning rules that carry the force and effect of law, or rules that effectively create binding norms because they so severely constrict an agency’s discretion.70 As we discuss later, this distinction potentially provides significant insight into how courts might address challenges to presidential orders.71

Further fueling the significance of this divide between policy statements and legislative rules is the fact that the APA enables judicial review only of “final” agency action.72 The Supreme Court has stated that agency action is final (and hence subject to judicial review) only if it: (1) marks the “consummation” of the agency’s decision-making process and (2) has determined “rights or obligations” or imposed “legal consequences.”73 Legislative rules tend to meet these criteria. By contrast, policy statements generally do not impose legal consequences or determine rights or obligations. As a result, policy statements often do not constitute “final” agency action and are therefore not subject to immediate judicial review.74

In light of these doctrines, determining whether an agency’s rule is a policy statement or a legislative rule can affect not only the procedural requirements that an agency must satisfy when formulating a rule, but also whether the rule is subject to immediate judicial review. Debates over a rule’s characterization therefore can produce contentious, high-stakes fights in the courts—including fights over administrative rules that reflect the president’s own influences or directions.

As just one example, consider a politically charged lawsuit filed in 2014.75 Led by Texas, a coalition of states challenged a DHS memo that, building on the Obama administration’s 2012 DACA policy involving undocumented children discussed above,76 outlined a policy of deferred action for millions of undocumented parents. DHS called its new program Deferred Action for Parents of Americans and Lawful Permanent Residents (DAPA). Jeh Johnson, the Secretary of DHS, signed the memo that put DAPA into place on the same day that Obama went before the cameras and delivered a national address on immigration reform.77 As a result, even though DAPA was formally put into place via an agency memo signed by Secretary Johnson, rather than via a presidential order, DAPA had the President’s fingerprints all over it. Tellingly, some media outlets reported—mistakenly—that DAPA had been announced via an executive order signed by Obama.78

When the State of Texas filed the suit challenging DAPA, it argued, among other things, that DAPA was unlawful because the agency had issued it without complying with § 553’s notice-and-comment requirements. In support of this argument, Texas argued that DAPA created binding criteria that curtailed the agency’s discretion and that effectively announced a substantive change in immigration policy.79 In response, DHS argued that the memo announcing DAPA merely set forth a policy for the exercise of case-by-case prosecutorial discretion—one that did not create any substantive rights and that was therefore exempt from § 553’s procedural requirements.80

The district court judge who first ruled in the case sided with the plaintiffs. On the threshold question of finality, the judge concluded that DHS’s enforcement policy constituted final agency action (and thus was judicially reviewable) because it gave DAPA beneficiaries “the right to stay in the country” and also clearly imposed “obligations” on DHS.81 Then, as to whether DAPA set forth a new legislative rule that needed to go through the APA’s notice-and-comment requirements, the judge concluded that—for similar reasons—it likely did.82

Over the dissent of one judge, the Fifth Circuit agreed that DAPA was likely unlawful due to its failure to undergo notice-and-comment procedures.83 The case eventually came before the Supreme Court, where a tie resulted among the Court’s eight members (shorthanded due to Justice Antonin Scalia’s death).84 The tie prevented the Court from providing clarification as to whether DAPA needed to go through notice-and-comment procedures—or, relatedly, whether it was judicially reviewable. The tie also meant that the district court’s preliminary injunction of DAPA remained in effect. As a result, the Obama administration never was able to implement the policy. Instead, when Trump entered the White House in 2017, DAPA was still on hold.

The Texas case, accordingly, highlights the powerful role that procedural review can play in constraining agency action, including agency action like DAPA that forms a central part of a president’s desired policy agenda. Exactly when and how the APA’s procedural requirements apply to constrain agency action can be controversial and messy around the margins. Indeed, the courts’ rulings in the DAPA case generated a great deal of controversy. Nonetheless, with respect to its existence and general application, procedural review is well established, and it can have a significant effect on a president’s attempts to put a policy like DAPA into place via agency action.

In contrast, as we will discuss in Part IV, when it comes to the president’s issuance of a presidential order, the APA’s notice-and-comment procedures pose no obstacle at all. This is because, as we have explained, the APA’s procedural requirements do not apply to the issuance of presidential orders.85 Unlike agency rules, presidential orders can be issued with almost no process whatsoever (other than any process that the president chooses to impose on herself),86 and the courts, accordingly, tend not to engage in procedural review of presidential orders. Still, as we explore in more detail in Part IV, there are reasons to think that the procedural notice-and-comment constraints that govern agency rulemaking should be considered when crafting a framework for judicial review of presidential orders.87

C. Review of an Agency’s Statutory Interpretations

A third significant legal doctrine that courts often invoke when reviewing agency action involves Chevron deference. Named after the 1984 decision of the Supreme Court, Chevron U.S.A. Inc v Natural Resources Defense Council, Inc,88 Chevron deference calls on the courts to defer to agencies’ reasonable interpretations of statutory ambiguities.89

Chevron deference rests in large part on the notion that presidential preferences should be allowed to influence an agency’s interpretive decisions about the meaning of statutory ambiguity.90 Still, this strong form of deference does not give the President unchecked authority to direct agencies to interpret statutory ambiguities in any way the president pleases. Rather, the agency interpretation claiming eligibility for Chevron deference cannot run contrary to Congress’s clear intent, and it must be reasonable. In addition, due to decisions like Christensen v Harris County91 and United States v Mead Corp,92 the agency interpretation must be set forth in an appropriate format, such as a notice-and-comment rule, that indicates Congress would have intended the agency’s views to carry the force and effect of law and to effectively bind the courts.93 Chevron’s applicability, in other words, depends on whether Congress intended the agency to be able to speak with the force and effect of law in the format that it did. Only when these limits are met will Chevron demand that the courts defer to an agency’s policy choices.

Food and Drug Administration v Brown & Williamson Tobacco Corp,94 a politically charged case involving a president’s efforts to influence rulemaking, provides a concrete example of Chevron at work.95 As President Bill Clinton was preparing for his 1996 reelection campaign, he publicly directed the Food and Drug Administration (FDA) to rely on its preexisting statutory authority under the Food Drug and Cosmetic Act96 to develop regulations that would “restrict sharply the advertising, promotion, distribution, and marketing of cigarettes to teenagers.”97 Consistent with Clinton’s instructions, the FDA launched a lengthy notice-and-comment rulemaking process. Throughout the regulatory process, Clinton threw his weight behind the FDA’s efforts, including by issuing campaign statements that emphasized his efforts to protect children from smoking.98

When the FDA finally issued a rule on the subject, Clinton stood in the Rose Garden next to children wearing red “Tobacco-Free Kids” T-shirts and announced: “Today, we are taking direct action to protect our children from tobacco and especially the advertising that hooks children on a product.”99 Yet despite Clinton’s efforts to personally claim ownership of the rulemaking, the tobacco rule as a legal matter belonged to the FDA. It was the FDA that issued the final rule and the lengthy accompanying statement of basis and purpose, and it was the FDA that claimed that Congress had given it the authority to regulate tobacco. 100 As a result, when various tobacco companies filed suit in federal court, they challenged the agency’s rule, not the President’s actions, and they argued that the FDA had exceeded its statutorily granted powers.

The Supreme Court ultimately sided with the tobacco industry, dealing a major blow to the Clinton administration.101 The Court did so after concluding that Congress had “clearly precluded the FDA from asserting jurisdiction to regulate tobacco products.”102 As a result, the FDA’s new rule was not entitled to Chevron deference, and Clinton’s efforts to regulate tobacco were wiped away.

The Chevron framework that was at work in Brown & Williamson has generated—and continues to generate—lots of debate. Nonetheless, despite some calls to rethink Chevron, it currently stands as a fixture of administrative law. By contrast, as we discuss in Part IV, it is entirely unsettled whether the courts should apply Chevron-like deference in the context of challenges to presidential orders, and in particular to challenges involving a president’s interpretation of statutory ambiguity.103 Ultimately, Chevron deference—much like other well-established administrative law doctrines, including arbitrariness review and procedural review—was designed to enable judicial review of the lawfulness of agency action. It was not designed to facilitate judicial review of direct challenges to presidential orders.

II. Judicial Review of Presidential Orders

Given the exceedingly important role that administrative agencies play in setting national policies via regulations, it makes sense that a wide range of actors—including lawmakers, courts, and scholars—have worked hard to develop robust legal principles aimed at facilitating judicial review of agency action. Yet agencies do not go about their policymaking role in a vacuum, and certainly they are not the only important policymakers within the executive branch. Rather, as we have indicated, the head of the executive branch—the President of the United States—also plays a central role in the regulatory sphere, often by deploying unilateral written directives either to announce significant policies on her own or to direct government actors to help further her policy goals.

This Part begins by describing the important function that presidential orders serve in the policymaking arena. It then describes how, despite the importance of presidential orders, lawsuits directly challenging the legality of presidential orders were relatively rare as a historical matter—at least prior to President Trump’s entrance into the White House. As a result, there is no coherent or well-theorized legal framework analogous to the APA to guide courts’ review of presidential orders.

A. Presidents’ Reliance on Presidential Orders to Shape Agency Policy

Executive orders—along with other kinds of unilateral written directives, such as those contained in presidential memoranda and proclamations—serve as an extraordinarily important tool in the president’s toolkit.104 Indeed, presidents have relied on executive orders and other written directives throughout our nation’s history to make “momentous policy choices”105 —policy choices that eventually led to the desegregation of the nation’s military106 and the internment of thousands of Japanese Americans during World War II.107

Yet, despite their importance, presidential orders do not enable the president simply to announce by fiat whatever rules she might like. Instead, presidential orders must be authorized by some source of law.108 To demonstrate that such authorization exists, the president normally must be able to point to some legal authority either in the Constitution or emanating from Congress that gives her the power to take the action in question.109 In addition, when issuing an order, the president must take care not to run afoul of any applicable statutory or constitutional limits, such as the First or Fifth Amendments.

As a historical matter, presidents—seizing on their constitutional and statutory powers—have issued many different kinds of presidential orders to try to achieve many different aims.110 Some of these orders have been labeled as memoranda, others as proclamations, and still others as executive orders.111 Indeed, precisely because presidential orders can take so many different forms, it can be difficult to describe them in a general manner. At the risk of oversimplification, however, we believe that the discrete instructions contained within presidential orders can be grouped loosely into two largely distinct (though at times overlapping) categories.

The first category involves specific instructions found in presidential orders that carry the force and effect of law—much like how legislative rules issued by agencies carry the force and effect of law.112 We refer to these throughout this Article as “legally binding orders.” Often these sorts of orders directly regulate private actors outside of the executive branch and alter legal rights or obligations. As an example, in 1934, President Franklin D. Roosevelt issued a proclamation that made it unlawful for private actors in the United States to sell arms to Bolivia or Paraguay.113 The federal government later criminally indicted a company for violating this presidentially imposed prohibition.114 Another example can be found in President Richard Nixon’s 1971 order attempting to stabilize the economy by, among other things, freezing prices and wages, including those of private businesses, across the country.115 This order made its legal effects clear, announcing that certain violations of the order could result in fines of up to $5,000 per violation.116 Presidents have issued these sorts of legally binding orders throughout the nation’s history, generally by relying on some combination of their constitutional and statutory powers.117

The second category involves instructions embedded within orders that do not themselves alter legal rights or obligations, even though they may well prompt subsequent executive branch action that does have legal effect. We refer to these throughout this Article as “nonlegally binding orders.” Orders that fall into this second category often operate as a presidential communication tool, enabling the chief executive to tell executive-branch officers what to do, or not to do, as they carry out their duties and administer the nation’s laws. For example, if a president wishes to raise the wages of low-paid workers, he lacks the power simply to increase the nation’s statutorily set minimum wage. But he might seize on powers already granted to him by Congress and the Constitution to issue an order seeking to achieve a similar, albeit more limited end. For example, he might issue a written order, directed at federal agencies entering into new government contracts, that requires each such contract to include provisions ensuring contractors will pay their workers an increased minimum wage.

President Obama issued just such an order in 2014 when he announced $10.10 as the minimum wage for workers of employers who contracted with the federal government.118 Importantly, however, Obama’s order did not itself have a legally binding effect. Instead, Obama’s order instructed various agencies to ensure that new contracts they entered into on behalf of the federal government complied with this minimum wage, and it directed the Secretary of Labor to enact regulations to implement the requirement. Obama’s order, in other words, told agencies to take steps that, once implemented, would raise the minimum wage paid by contractors. What the order did not do was raise wages itself. Nor did Executive Order No 13658 give workers (or anyone else) a right, enforceable in court, to force compliance with its dictates.119

At this stage, it is important to note three things about the categories of orders that we have just identified here—legally binding orders and nonlegally binding orders. First, these two categories can blur together around the margins. Indeed, as we explore when we turn to various challenges to Trump-era orders,120 it is not always clear whether a given order is legally binding or not. As a result, some of the most contentious legal battles unfolding under Trump’s presidency have concerned, in essence, whether the challenged order in question is legally binding or instead has no legal effect.121 To further complicate matters, the very same presidential document often will include multiple instructions, some legally binding and some not, and court challenges may be directed at one, some, or all of them.122

Second, even orders that are most properly characterized as nonlegally binding orders can have a very real effect. Although such orders do not immediately alter the legal landscape, they operate to make the president’s wishes clear to affected agency heads. To this end, if an agency head fails to comply with a nonlegally binding order, she risks significant political blowback, including, potentially, termination via the president’s power of removal.123 The effect of such an order, in other words, is significant, even if the effect is largely political instead of legal.

Third, presidents now routinely deploy nonlegally binding orders as a means of controlling agencies’ regulatory work. This, however, was not always the case. To the contrary, prior to President Reagan, presidents generally “shunned direct intervention” in agency rulemaking and tried to avoid meddling in agencies’ regulatory work.124 Reagan, however, set the stage for greater presidential involvement in the regulatory arena in 1981 when he issued Executive Order No 12291, a nonlegally binding order directed at the heads of executive agencies.125 It ordered agencies to submit drafts of their proposed rules, along with regulatory impact analyses, to the Office of Management and Budget (OMB), an entity within the Executive Office of the President,126 for prepublication review.127 The order also required executive agencies, to the extent permitted by law, to take cost-benefit principles into account when making regulations.128 Reagan’s order made clear that it did not create any legally enforceable rights or obligations and that it was intended “only to improve the internal management of the Federal government.”129 Nonetheless, Reagan directed the order at the heads of executive agencies whom he easily could remove if they failed to heed his directions. The order, as a result, had bite even though it was not legally binding.

Reagan’s order triggered significant controversy and debate.130 Yet, as subsequent presidents entered the White House, Reagan’s ambitious program of White House review did not recede. To the contrary, the use of nonlegally binding orders to direct agencies’ regulatory activities became increasingly entrenched. President Clinton, in particular, built on Reagan’s foundation. When he entered the White House, Clinton replaced Reagan’s order with a new nonlegally binding order, Executive Order No 12866,131 that embraced an even “stronger view than [Reagan’s orders] had of the President’s authority over the administrative state.”132 For example, Clinton’s order gave the president (or the vice president acting at the request of the president) the ultimate authority to resolve any disagreements that arose between executive agencies and OMB during the regulatory review process.133

As then-Professor Kagan noted in her foundational article on the relationship between the president and the regulatory state, this and other Clinton-era orders helped to make “presidential intervention in regulatory matters ever more routine and agency acceptance of this intervention ever more ready.”134 This, in turn, paved the path for future presidents to deploy nonlegally binding presidential orders in increasingly aggressive ways, ultimately leading to a new era of presidential administration.135

Consider, to this end, Obama. Throughout his time in the White House, Obama relied extensively on nonlegally binding orders to instruct agencies to take action on all sorts of matters.136 In 2014, for instance, Obama issued a memorandum involving overtime pay.137 Obama’s memorandum ordered DOL, in no uncertain terms, to use the power it had under the Fair Labor Standards Act138 (FLSA) to update regulations relating to overtime-pay exemptions.139 Not surprisingly, DOL, which is headed by a political appointee who works at the pleasure of the president, promptly complied with Obama’s instructions, even though they were not legally binding.140

In short, presidents have deployed presidential orders—both legally binding and nonlegally binding—throughout the nation’s history. Legally binding orders, like Roosevelt’s arms-sales proclamation, alter legal rights and obligations, and they can be enforced in court. In contrast, nonlegally binding orders, like Obama’s overtime directive aimed at DOL and Reagan and Clinton’s regulatory review orders, contain instructions—akin to marching orders—directed at executive branch officers, but they do not alter legal rights and obligations. Yet even orders within the latter set have real effect, due to the likelihood that agency officials will comply with them. As a result, modern-day presidents have increasingly deployed nonlegally binding orders as a means of controlling agencies’ regulatory activities.

B. The Unsettled Nature of Judicial Review of Presidential Orders

Despite the significance of presidential orders in the contemporary policymaking arena, legal challenges to these orders have only occasionally worked their way into the federal courts—at least prior to Trump’s presidency. Later, in Part III, we describe what has unfolded during Trump’s time in office and, in particular, the unprecedented onslaught of legal challenges that litigants have brought against orders he has issued.141 For now, however, we turn to the legal landscape predating Trump’s entrance into office—one in which challenges to presidential orders were infrequent, yielding only a thin collection of precedents that provide judges with little meaningful guidance.142

1. Infrequent legal challenges.

Historically, litigants only rarely raised direct challenges to presidential orders.143 Presidents, accordingly, generally did not expect that their orders would be subject to routine judicial challenges.144 Moreover, courts failed to gain much experience adjudicating these sorts of lawsuits.145

To understand this history and some of its nuances, it is helpful to return to the distinction between presidential orders that we identified above: legally binding orders versus nonlegally binding orders. Relatively few instructions contained within presidential orders have fallen into the first category—although, when they have, it has not been particularly unusual for them to trigger a direct and occasionally swift legal challenge.146 Take Roosevelt’s 1934 proclamation, discussed above. This presidential proclamation made it a crime, enforceable by criminal sanctions, to sell arms to certain countries.147 By regulating the conduct of private actors, this order effected an immediate change in the law—and it soon triggered a legal challenge. Responding to a criminal indictment, defendants argued that the order violated the Constitution and therefore that it should be set aside. The Supreme Court eventually rejected this claim on the merits in United States v Curtiss-Wright Export Corp.148

A more recent illustration of a legally binding order challenged in the courts comes in a set of presidential proclamations issued by Clinton late in his presidency that designated portions of land as national monuments.149 These proclamations themselves effected an immediate change in the law by setting aside land and prohibiting certain private uses of that land.150 They were, as a result, legally binding. These proclamations quickly elicited challenges to their legality in a lawsuit brought directly against the orders and against the President in name. The federal courts also rejected these challenges on the merits.151

Contrasted against this first set of orders are nonlegally binding orders—orders that do not themselves alter legal rights or obligations. Two landmark orders we discuss above (Executive Orders Nos 12291 and 12866, issued by Reagan and Clinton, respectively) provide illustrations. These orders inserted the White House into the regulatory process by, among other things, requiring certain agencies to submit proposed rules to an office within the Executive Office of the President. What these orders did not do was effect a binding change in the law. Instead, they instructed executive-branch officials to act in certain ways when using their own powers to effect binding changes in the law.

Nonlegally binding orders of this sort are commonplace.152 Yet they have triggered very few direct and immediate legal challenges—at least prior to Trump entering the White House. Consider again Executive Order No 12291, which was issued by Reagan. The courts appear not to have grappled in any concerted way with the legality of this order, despite the political and legal controversy it generated. While plenty of plaintiffs brought related claims—for example, claims challenging regulations that later were promulgated under Executive Order No 12291’s dictates—no lawsuits appear to have directly challenged the lawfulness of the executive order itself.153

There are at least four possible overlapping explanations for why, historically, litigants have refrained from directly and immediately challenging nonlegally binding orders in the federal courts. The first is that such an aggressive challenge may not have seemed necessary. Opponents of a president’s order—committed to impeding its implementation—may have felt satisfied coming at the problem indirectly, by suing to enjoin the president’s subordinates from taking disfavored actions consistent with the president’s directives. An illustrative example came in the early 1980s, when the Environmental Protection Agency (EPA) issued a controversial decision and acknowledged that Reagan’s newly issued Executive Order No 12291 had been the reason behind it.154 In response, the National Resources Defense Council (NRDC) sued—but this lawsuit neither named the President as a defendant nor centered its challenge around the lawfulness of the President’s order.155 Instead, the NRDC sued the EPA itself, challenging the lawfulness of the agency’s decision. On this theory, the NRDC won.156 In another illustrative case, plaintiffs tried to enjoin provisions of an order, Executive Order No 13202, which President George W. Bush had issued in an effort to alter the way that federal agencies set requirements in construction contracts.157 In this line of litigation, the plaintiffs did challenge the lawfulness of the executive order itself. Still, rather than sue the President directly, the plaintiffs sued the federal agencies tasked with implementing it.158 In both cases, litigants enlisted the courts in their efforts to resist agencies’ attempts to effectuate a president’s orders. However, in neither case did they bring a direct and immediate challenge to the executive order against the president in name.

A second, related explanation for the infrequency of direct challenges to presidential orders involves Congress’s 1946 enactment of the APA, coupled with the judiciary’s development of the now-entrenched administrative law doctrines that guide judicial review of agency action. We describe above several important principles characterizing this body of law.159 By providing litigants with a straightforward and well-established way of challenging executive-branch activity, the APA may have taken the pressure off more direct challenges to presidential actions.160 The development of APA-related case law also may have had the effect of setting norms surrounding the review of executive-branch action, including with respect to the types of executive actions that seem to be appropriate, and inappropriate, to challenge in court. Indeed, as we explain in Part I, even when a president clearly has influenced an agency’s actions, litigants historically have drawn on the APA (or other administrative law principles) to challenge the actions of the agency, rather than focus on any underlying conduct of the president.161

A third explanation involves prevailing understandings of various justiciability doctrines, including ripeness and standing doctrines. Given that nonlegally binding orders simply provide marching orders to members of the executive branch and do not themselves effect a change in the law, lawyers have tended to assume that these orders do not “engender a ripe judicial controversy.”162 Indeed, writing in 1982, Professor Harold Bruff took it as a given that orders that represent nothing more than directives to the bureaucracy “are shielded from immediate judicial review by such doctrines as standing and ripeness.”163 As Bruff stated, it is only “when such orders are implemented in a fashion that affects the public that judicial review occurs, if at all; moreover, review then focuses on the implementary decisions themselves.”164

A fourth possible explanation involves Congress’s decision, over the course of most of the twentieth century, to delegate more and more powers to agencies, rather than to the president in name.165 This approach marked a shift; in the nation’s early years, most of Congress’s delegations of regulatory power ran straight to the president. In light of Congress’s decision to hand away authority to agencies,166 rather than to the president, it makes sense that most challenges to executive-branch activity also would run against agencies, rather than the president.

Taken together, these factors appear to provide some explanation for why direct challenges to presidential orders have been infrequent as a historical matter. Regardless of the cause, however, the effect on the courts has been clear: an extended period in which litigants generally refrained from bringing direct and immediate challenges to presidential orders, especially to nonlegally binding orders. As we now explain, few judicial precedents therefore exist to guide the courts when they are asked to review the lawfulness of presidential orders.

2. Limited legal precedents.

The historical infrequency of presidential-order challenges has produced an anemic set of judicial precedents—one that then–Associate Justice Rehnquist described in 1981 as “rare, episodic, and afford[ing] little precedential value for subsequent cases.”167 More than three decades later, a legal scholar, Erica Newland, reached a similar conclusion after reviewing nearly three hundred opinions in which the courts meaningfully engaged with doctrinal questions presented by presidential orders.168 Specifically, Newland concluded that this area of law is “born of disorder,”169 with courts generally failing to “acknowledge, in a particularly theorized way, the special challenges and demands of the executive order as a form of lawmaking.”170 Likewise, according to Newland, courts have tended not “to recognize the common jurisprudential questions that executive orders raise about sources of lawmaking and interpretive authority.”171

Although Newland’s dataset confirms that the courts have, on occasion, adjudicated challenges to presidential orders,172 it also reveals how the courts have done so largely without engaging with the overarching doctrinal questions presented by this form of judicial review. Important issues are left unaddressed and even unacknowledged, with cases decided narrowly or in subject-specific silos. As a result, the ad hoc smattering of judicial opinions that do exist in this area fail to provide anything close to a coherent or well-developed legal framework—and certainly nothing akin to the APA-centered regime that governs judicial review of agency action.173

A case like Youngstown Sheet & Tube Co v Sawyer,174 which is among the most prominent of judicial decisions testing the limits of presidential power, is illustrative. Decided on an expedited basis, the case involved whether President Harry S. Truman had acted constitutionally when he issued Executive Order No 10340, which instructed the Secretary of Commerce to take possession of and operate a number of privately owned steel mills. In a brief opinion, a majority of the Supreme Court concluded the President lacked statutory or constitutional authority for the order and, as a result, that its enforcement should be enjoined.175 In what proved to be a highly influential concurrence, Justice Robert H. Jackson concluded that the President had acted contrary to the will of Congress and without independent constitutional authorization, and therefore that the order could not stand.176

Youngstown certainly provides some insight into the substantive limits of presidential power. It also confirms that the courts will not blindly accept a president’s legal conclusions regarding his own authority. The many issues Youngstown fails to address, however, range from threshold matters (relating, for example, to what cause of action supported the companies’ challenge)177 to those arising at the back end of a case (relating, for example, to how severability analysis should proceed).178 Nowhere, moreover, did the Court squarely grapple with fundamental questions of interpretation, deference, or reviewability, among others.179 In its failure to engage meaningfully with these sorts of issues, Youngstown typifies the approach that courts historically have taken in addressing challenges to presidential orders.

The Supreme Court provided similarly scant guidance when it resolved Dames & Moore v Regan180 nearly three decades later. Dames & Moore presented the question whether Presidents Jimmy Carter and Reagan acted constitutionally when they issued a series of executive orders implementing the terms of an executive agreement between the United States and Iran.181 The Court concluded they had.182 Lest the reader make too much of this pronouncement, however, the Court explicitly insisted that its opinion should provide little guidance in future cases.183 Curtiss-Wright and Korematsu v United States184 likewise provided little meaningful insight into the overarching legal framework that should guide review of presidential orders. Indeed, much of the discussion in Korematsu addressed a civilian exclusion order issued by the President’s subordinates, not by the President himself.185

Despite the relative lack of guidance contained in these pre-cedents, the cases do collectively provide several insights into judicial review of presidential-order challenges. First, existing pre-cedents confirm that challenges to presidential orders can involve a range of legal claims. Sometimes the central point of contention between the parties is whether the Constitution itself provides a basis for the president’s order.186 Other times, the question is whether the order violates some applicable legal restriction, such as the First Amendment.187 In still other cases, the focal point becomes whether Congress has provided a legal basis for the president’s order.188 And in others, litigation turns on whether the relevant congressional authorization is itself constitutional.189 Thus, existing precedents confirm that presidential-order challenges can involve a variety of legal claims—some rooted in the Constitution and others in statutes.

Second, existing judicial precedents confirm that the APA cannot supply a framework for courts’ review of presidential orders. As discussed above, the Supreme Court foreclosed this possibility in 1992 when it decided Franklin v Massachusetts,190 the case in which it held that, as a matter of statutory interpretation, the APA’s references to “agency” do not include the president.191 In light of this holding, the APA simply does not apply to claims brought against the president in name. For those claims, the APA provides no cause of action, no waiver of sovereign immunity, no standards for review, and no rules for relief.192

A plaintiff seeking to contest the lawfulness of a presidential order nevertheless might activate some of the APA’s provisions by bringing an indirect challenge. These indirect challenges may be brought against a subordinate of the president,193 challenging some final agency action that the subordinate took on the authority or instruction of an allegedly unlawful presidential order. For these lawsuits, some of the APA’s provisions may well apply.194 The APA may provide a cause of action, for example, as well as a waiver of sovereign immunity and rules for relief. In addition, the APA’s standards of review still may govern the court’s review of the actions of the agency. Importantly, however, the APA’s standards of review still do not apply to the actions or orders of the president. As a result, questions central to this class of challenges continue to fall outside the APA’s dictates.195

A third defining feature of the current state of this jurisprudence relates to courts’ willingness, on occasion, to respond to the unsettled nature of this area of the law by reaching out to more familiar areas, such as those constituting the fields of administrative law and legislation. In Minnesota v Mille Lacs Band of Chippewa Indians,196 for example, the Supreme Court considered a dispute over usufructuary rights that the plaintiffs argued they retained pursuant to an 1837 treaty with the United States. A pivotal question was whether an 1850 executive order, signed by President Zachary Taylor, could be severed. Acknowledging the novelty of the inquiry, the Court observed that it never had addressed whether executive orders “can be severed into valid and invalid parts, and if so, what standard should govern the inquiry.”197 Rather than definitively resolve these issues, however, the Court decided to “assume, arguendo, that the severability standard for statutes also applies to Executive Orders.”198 Under this standard, the Court concluded that the order was not severable.199 Tellingly, the Court reached this conclusion without setting a definitive standard going forward—or even engaging meaningfully with the question whether transferring standards from the field of legislation was appropriate.

A federal court also looked to other areas of the law for inspiration in attempting to resolve the dispute in Chamber of Commerce of the United States v Reich,200 a 1996 decision out of the Court of Appeals for the DC Circuit. Reich has taken on near-canonical status due to its willingness to do what so few cases in this area are willing to do: engage in a meaningful way with broad doctrinal questions affecting presidential orders. In Reich, employer associations challenged Executive Order No 12954, which had been issued by Clinton. This order prohibited the federal government in certain circumstances from contracting with employers who hire permanent replacements during a labor strike. Plaintiffs argued that this prohibition violated the National Labor Relations Act (NLRA) and the Procurement Act.

The DC Circuit began by reversing the district court’s ruling on ripeness—a recurring issue that looms large in this area201 —thereby allowing the case to go forward.202 Less than a year later, the case came back up on appeal, and the court of appeals again reversed, concluding (contrary to the district court) that the executive order was preempted by the NLRA.203 To reach this decision, the court of appeals explored a range of issues, including questions relating to what cause of action might be available to litigants not proceeding pursuant to the APA,204 the circumstances under which an order of the president may be reviewable by a court,205 and, at least in general terms, the degree of deference due to a president in his interpretation of statutes.206 In addressing this final issue, the court of appeals acknowledged—but rejected—the district court’s conclusion that Chevron-type deference should apply to the President’s interpretation of his own statutory authority.207

The court again considered the applicability of principles pulled from administrative law in American Federation of Government Employees, AFL-CIO v Reagan,208 a case out of the DC Circuit. Plaintiffs had challenged the lawfulness of a presidential order excluding various subdivisions from “coverage” under the Federal Service Labor-Management Relations Act.209 The challengers argued that the order was invalid because the statute required, as a predicate for the coverage exclusions, that the President first make particular determinations, and the President had not made those determinations (or, at least, had not done so explicitly).210 The court rejected the challengers’ argument. Citing cases across a range of administrative areas, the court noted the rebuttable presumption of regularity that normally attaches to agency action, observing that this presumption had been applied “in a variety of contexts.”211 The court then concluded that this presumption was “clearly applicable” to the issuance of an executive order.212

As we explore in more detail in Part IV, the judiciary’s willingness to analogize to administrative law often makes sense given that both sets of challenges (to presidential orders and agency actions) involve challenges to executive-branch action. Yet, given the relative infrequency with which the courts have reviewed presidential orders, such borrowing generally has been sporadic and poorly theorized, and it has failed to yield a carefully considered or a well-accepted legal framework for judicial review of presidential orders.213 The lack of an established framework for judicial review may well have been tolerable in the past, when challenges to presidential orders were few and far between. However, as the next Part of this Article describes, Trump’s 2017 entrance into the White House brought a flood of litigation aimed at the President. These Trump-era lawsuits—combined with prior presidents’ increasingly forceful uses of orders to control agency activity—reveal the need for a more coherent and purposeful framework to guide courts in reviewing presidential orders, not only now but also in the future.

III. The Emerging Need for a Legal Framework to Guide Judicial Review of Presidential Orders

As a historical matter, executive-order challenges have been relatively rare, and so the lack of a coherent framework to guide courts has not proven terribly problematic. The 2017 inauguration of President Trump marked a change. It corresponded with a shift in how litigants attack presidential orders, with legal claims against Trump’s orders coming more quickly, more directly, and more often than they had before.214 Indeed, Trump’s presidency appears to have ushered in a new era of litigation targeted directly at presidential orders.215

It is impossible to predict with certainty whether this new form of litigation will dissipate once a new president enters the White House. However, a confluence of events—relating both to how presidents over time have interacted with the administrative state and how the courts more recently have responded to legal challenges—suggests that these changes are unlikely to reverse themselves anytime soon. As such, courts and litigants, both now and in the future, would benefit from the development of a more coherent and purposeful framework to guide judicial review of presidential orders.

A. The Explosion of Trump-Era Litigation Aimed at the President

An unprecedented deluge of lawsuits hit President Trump after his inauguration in January 2017.216 Many took direct aim at his presidential orders.217 The wave of litigation was distinctive not only due to its volume, but also due to the nature of the challenges. Many Trump-era litigants have elected to challenge the legality of the President’s orders, rather than (or along with) the legality of subsequent agency action—a strategy that, as we have discussed, historically has been quite unusual. Moreover, many of these litigants have brought their challenges immediately, without waiting for final agency action to occur.218 This rush to the courthouse helps to explain why the President himself was named as a defendant in many of these Trump-era lawsuits, a phenomenon that also has relatively little precedent.219

The following table, organized by the issuance date of the order challenged, provides a summary of much of this litigation in Trump’s first year in office.

Table 1: A Partial List of Presidential Orders Challenged in Court During Trump’s First Year in Office

Document Label	Document No	Document Title	Date Issued	Challenge Filed on or Before	Notes
Executive Order	13768	“Executive Order: Enhancing Public Safety in the Interior of the United States”	1/25/2017	1/31/2017	Includes directives aimed at “sanctuary jurisdictions”
Executive Order	13769	“Executive Order Protecting the Nation from Foreign Terrorist Entry into the United States”	1/27/2017	1/28/2017	Often referred to as “first travel ban”
Executive Order	13771	“Presidential Executive Order on Reducing Regulation and Controlling Regulatory Costs”	1/30/2017	2/8/2017	Often referred to as “one-in, two-out” order
Executive Order	13780	“Executive Order Protecting the Nation from Foreign Terrorist Entry into the United States”	3/6/2017	3/8/2017	Often referred to as “second travel ban”
Executive Order	13798	“Presidential Executive Order Promoting Free Speech and Religious Liberty”	5/4/2017	5/4/2017	Addresses IRS enforcement of “Johnson Amendment”
Executive Order	13799	“Establishment of Presidential Advisory Commission on Election Integrity”	5/11/2017	7/18/2017	Intended to address, inter alia, “fraudulent voting”
Presidential Memo	N/A	“Presidential Memorandum for the Secretary of Defense and the Secretary of Homeland Security”	8/25/2017	8/9/2017220	Addresses “Military Service by Transgender Individuals”
Presidential Proclamation	9645	“Presidential Proclamation Enhancing Vetting Capabilities and Processes for Detecting Attempted Entry into the United States by Terrorists or Other Public-Safety Threats”	9/24/2017	9/29/2017	Often referred to as “third travel ban”
Presidential Proclamation	9681	“Presidential Proclamation Modifying the Bears Ears National Monument”	12/4/2017	12/4/2017	First of two monuments-related orders
Presidential Proclamation	9682	“Presidential Proclamation Modifying the Grand Staircase-Escalante National Monument”	12/4/2017	12/4/2017	Second of two monuments- related orders

Combined, these lawsuits appear to have ushered in a new era: one of direct and immediate challenges to presidential orders. This litigation includes challenges to presidential orders that clearly have immediate legal effect; it also includes challenges to orders that do not have any clear immediate legal effect—which executive-branch lawyers have attempted to characterize (in substance) as nonlegally binding. Take, for example, Executive Order No 13768. As relevant, this Order directs federal officials to take various actions in response to so-called sanctuary jurisdictions.221 A central dispute in litigation over Executive Order No 13768 is the extent to which the order actually changed existing law.222 Before the district court, the government argued that the order is “simply an example of the President’s use of the bully pulpit,” which does not change the law but rather merely directs federal officials to “enforce existing law.”223 In contrast, the challengers argued that the Order does change the law. The district court agreed, rejecting the government’s narrow reading and concluding that the order as written “change[d] the law.”224 On appeal, the Ninth Circuit agreed with the district court, suggesting that the motivation behind the government’s argument—that the order was “all bluster and no bite”—seemed to be driven by its desire to “avoid legal consequences” of the order’s issuance.225

This kind of debate—over the legal effect (or lack thereof) of a given presidential order—has recurred with some frequency in the Trump era.226 Indeed, some of the fiercest legal battles have concerned, in essence, whether the order in question is legally binding or instead has no legal effect. Tellingly, there is little case law providing courts with guidance on this central issue—a void that helps to further confirm the unprecedented nature of much of the litigation filed in response to Trump’s orders.

The forces leading to the dramatic uptick in litigation aimed at attacking presidential orders—including orders that the executive branch claims to be nonlegally binding—are complicated. However, a broad trend that appears to have played a significant role relates to the increasingly heavy hand that modern presidents have wielded over agency rulemaking.227 More specifically, it reflects the intensifying efforts by the nation’s chief executives, particularly since President Reagan, to blur the lines between presidential and agency action, all in an effort to push forward their policy goals.

From his very earliest days in office, Trump embraced this trend. Less than two weeks after his inauguration, for example, he issued Executive Order No 13771, which is analogous in some ways to Reagan’s Executive Order No 12291 and President Clinton’s Executive Order No 12866. Often referred to as Trump’s “one-in, two-out” order, the directive purports to require under certain circumstances that agencies identify two preexisting regulations to be repealed for every new regulation promulgated.228 It also purports to require that agencies limit or offset certain regulatory costs.229 Tellingly, in a signing ceremony for the order, Trump employed rhetoric suggesting little-to-no distinction between the President’s own desires and an agency’s rulemaking decisions—and even implied that he personally would play a role in the regulatory changes.230

This early example is but one of many, as Trump has acted aggressively throughout his presidency to blur the lines between the President and the agencies he oversees.231 This approach poses legal risks. When a president tries to inject herself as the ultimate decision-maker in the administrative process, this invites litigants to direct their legal challenges not at subsequent agency action, but rather at the predicate presidential decisions. Such an invitation may be particularly enticing to litigants who want legal relief fast, without waiting for what is often a lengthy agency process to unfold.

The invitation to litigate also may be particularly attractive when a president is willing to use presidential orders to advance politically and legally controversial policies. Such an approach certainly describes Trump’s presidency.232 This combination—the controversial nature of Trump’s policies, coupled with his enthusiastic embrace of the longstanding trend toward collapsing the distinction between presidential and agency action—seems, in a sense, to have lit a match. Litigants just needed a fuse—and the high-profile DAPA-related lawsuit discussed above may well have provided just such a fuse.233

As previously noted, the DAPA-related line of litigation, which was pursued at the end of Obama’s presidency, did not involve a presidential order. Rather, the memorandum announcing DAPA emanated from DHS. As a result, the APA could, and did, provide the framework for Texas’s legal challenge.234 Still, a few defining features of the DAPA proceedings seemed to blur the lines in a way that may have set the stage for a change in litigation strategies.

First, the DAPA order in some ways felt like an executive order. Even though President Obama was not the one formally to sign the order, he presented it almost as though it was his own, and that is how it popularly was perceived. In addition, the DAPA order sought to achieve change through an abrupt and unilateral set of directives, much like an executive order, rather than through a lengthy notice-and-comment process.235

Second, the plaintiffs challenged DAPA before it had even gone into effect, seeking to block its implementation based on a controversial theory of standing—namely, Texas’s assertion that it would have to issue driver’s licenses to DAPA beneficiaries and would thereby suffer a financial loss. The government warned that Texas’s theory of standing would open the floodgates for future challenges to federal policies brought by states.236

The courts ended up accepting Texas’s theory of standing, and they also approved of Texas’s decision to sue early in the process, before DAPA had gone into effect. Moreover, as discussed above, the courts accepted the state’s argument concerning the nature of the memorandum (that is, that it set forth a legislative rule and therefore could not, given the APA’s procedural requirements, be enforced).237 Texas accordingly received the relief it had requested. Thus, by bringing an early and direct challenge to a set of directives emanating from the executive branch, the plaintiffs had managed to derail a high-profile initiative, central to a president’s agenda, before it had even begun.

This precedent captured the attention of many politically minded observers, who began to wonder: if a state like Texas could directly and immediately challenge an agency memorandum under a novel theory of standing, and thereby invalidate a policy initiative central to the agenda of a Democratic president, why could a state like Washington not do the same in response to a signature policy initiative of a Republican president? And if such a lawsuit could work against an agency memorandum heavily influenced by a president, then why not against a presidential order itself?

This line of reasoning perhaps helps to explain the wave of lawsuits flooding Trump’s White House in just the first few weeks of his presidency. The President began his term by issuing a series of far-reaching orders in rapid secession. Litigants responded with a type of lawsuit that, as discussed above, rarely had been lodged before: an immediate and direct challenge to a presidential order itself, in a lawsuit brought against the president in name, without waiting for the regulatory process to unfold. The most vivid illustration of this line of litigation came in response to Executive Order No 13769, which contained Trump’s first travel ban. Within days of the order’s issuance, litigants filed a multitude of cases challenging its lawfulness.238 As this litigation unfolded, Trump’s other orders also began triggering challenges, with at least two additional orders challenged within his first month in office.239

While all these lawsuits received attention, the travel ban litigation was extraordinarily high profile. It involved the detention of individuals from all over the world; it elicited widespread protests and extensive debate; and it captured the attention of the national and global media. This early travel ban litigation was also, at least for the plaintiffs, extraordinarily successful—even when considered in light of later developments before the Supreme Court.240 In less than a week, a lawsuit initiated by the State of Washington produced a nationwide injunction prohibiting the federal government from enforcing several important provisions of Executive Order No 13769.241 On appeal, the court of appeals refused to stay this injunction, which ended up prohibiting the government from enforcing those provisions for the life of the order.242

In short, the State of Washington’s early success in the travel ban litigation provided litigants with a clear blueprint to follow in response to a disfavored presidential order: before the president’s policies can take hold, bring an immediate and direct challenge and seek a broad injunction to halt the order’s implementation. The stage was set for a surge in a new style of litigation.

And the surge came. Multiple jurisdictions filed suit, for example, in response to Executive Order No 13768, which sought to push back on sanctuary jurisdictions, just days after Trump had issued the order.243 These complaints did not challenge agency action. They could not; no agency had time to act.244 Likewise, litigants challenged Executive Order No 13771, often referred to as the “one-in, two-out” order, less than two weeks after its issuance, and therefore well before agencies had the opportunity to transform its dictates into final agency action.245

An even more striking example of such litigation came in response to the series of tweets that Trump sent on July 26, 2017, announcing that the “United States Government will not accept or allow . . . [t]ransgender individuals to serve in any capacity in the U.S. Military.”246 Two weeks later, plaintiffs filed suit challenging this apparent change of policy. This challenge truly was unprecedented. Not only did plaintiffs decline to wait for subsequent agency action; they refused to wait even for a formally issued presidential order, such as an executive order or a presidential memorandum. Rather, their lawsuit challenged a presidential decision that had been memorialized exclusively on Twitter.247

As these precedents confirm, Trump’s entry into office marked a dramatic uptick in the number and forcefulness of lawsuits directly challenging presidential orders. Since his inauguration, challenges to presidential orders have been arriving more quickly, more directly, and more often than before. That said, though Trump’s presidency appears to have ushered in this era, there is little reason to think this trend will reverse itself once he leaves office. Rather, the long-term trends motivating these legal developments—including those relating to the increased forcefulness of presidential control over administrative action—are well entrenched. And, of course, the horse is now out of the barn, with the precedents set by these high-profile cases offering both inspiration and a model for those seeking another lever of control over a president’s agenda. In short, this new form of legal challenge cannot easily be dismissed as a Trump-induced anomaly. More likely, it is the new normal.

B. The Long-Term Benefits of Developing a More Theorized Legal Framework

Given that challenges to presidential orders will likely persist long after Trump leaves office, the development of a more coherent and purposeful approach to reviewing presidential orders would benefit courts and litigants alike. In particular, the formulation of a more theorized approach to judicial review of presidential orders would serve at least three important long-term purposes. The first relates to uncertainty. By helping to clarify the role that courts can play in reviewing presidential orders, a more developed framework would provide helpful guidance not only to litigants and judges, but also to the president, her staff, agencies, and others affected by the president’s actions. Currently, all parties involved (including those charged with drafting the presidential orders in the first place) are forced to guess at the applicable legal framework, and this guesswork exacerbates uncertainty in an already very difficult and unsettled area of the law.

The value of consistency underlies the second purpose potentially served by more theorization. By articulating an analytical framework and core principles to follow, a more developed framework would help to increase consistency across cases and thereby promote the rule of law. It is true that the Supreme Court can promote consistency through case-by-case error correction. However, not all presidential-order cases reach the Supreme Court, despite their importance. In his first year in office, for example, Trump saw at least one of his presidential orders enjoined, effectively permanently, without any Supreme Court review.248 Moreover, even when the Supreme Court eventually is able to address inconsistency, the delay associated with that review can be substantial. It was not until June 2018, for example, that the Supreme Court handed down a ruling on the merits in Trump v Hawaii, a case involving Trump’s third travel ban—which the President had issued nearly a year earlier.249 And that case moved much more quickly than is typical through the lower federal courts.

The third purpose served by a more theorized approach relates to the quality of the framework that courts ultimately decide to use. Challenges to presidential orders present complicated issues relating to, among other things, the separation of powers, the role of procedure, and enforcement mechanisms. All these issues quickly translate into challenging questions relating to justiciability, deference, remedies, and more. Sporadic, case-by-case rulings in highly charged—and often fast-moving—cases tend not to provide an adequate forum for carefully considering how best to resolve these issues. Instead, these sorts of issues, including questions about which overarching analytical framework to apply, are likely to be best resolved in conjunction with concentrated attention and deliberative discussion among scholars, judges, litigants, members of Congress, and others. In the hopes of initiating just such a discussion, we turn now to the work of sketching out what such a framework for judicial review of presidential orders might look like, as well as the specific doctrines it might include.

IV. Toward a Framework for Reviewing Presidential Orders

This Part sets forth our thoughts about what a coherent framework for judicial review of presidential orders might look like. To this end, this Part addresses two separate but related issues. First, it considers whether administrative law principles should be transferred into the presidential-order context to provide an overarching, high-level analytical framework. Ultimately, it acknowledges the wisdom of looking to administrative law principles but cautions against borrowing from them too reflexively, given the significant differences between presidential action and agency action. These differences include those relating to procedures, enforcement mechanisms, and separation-of-powers concerns. Second, with these differences in mind, this Part turns to specific legal doctrines that frequently are implicated in challenges to presidential orders—doctrines that must form a part of any workable legal framework for judicial review. These doctrines include: threshold reviewability doctrines, such as standing, ripeness, and cause-of-action requirements; standards of review; and the availability and appropriateness of different forms of judicial relief, as well as severability.250

A. Choosing a High-Level Analytical Framework: The Imperfect Fit of Administrative Law Principles

As we have explained, courts occasionally borrow from administrative law when reviewing presidential orders.251 This borrowing makes a great deal of sense, given that administrative law principles were designed to help the courts police executive action. It also makes sense from an efficiency perspective: reinventing the wheel—and coming up with an entirely new framework to guide review of presidential orders—would be time consuming and difficult. Nevertheless, critical differences exist between presidential action and agency action, and these differences should be considered before transferring administrative law principles into the presidential-order context.

1. Differing procedural constraints.

One major difference between policies put into place via agency rules and those put into place via presidential orders involves the applicable procedural requirements. As we noted in Part I, when administrative agencies enact legally binding rules, those rules often must go through a lengthy notice-and-comment rulemaking process mandated by the APA.252 This procedurally cumbersome process provides the public with the opportunity to participate in the regulatory process and to comment on proposed rules.253 In addition, the notice-and-comment process helps create a record for the courts to review when agency action is challenged.254 Although the APA does not elaborate on what exactly should constitute the “record” for purposes of judicial review of notice-and-comment rulemakings,255 courts have held that the administrative record must contain materials that are considered in some manner by the agency.256

In contrast to agency rules, presidential orders can be issued quickly with few procedural checkpoints.257 This is because Congress has not imposed meaningful procedural requirements on the issuance of most forms of presidential orders—other than to require the publication of many, but not all, executive orders and proclamations in the Federal Register.258 Moreover, as we have explained, the APA does not apply to presidential action,259 and so its procedural requirements are inapplicable.

Indeed, the few procedural requirements that do constrain the issuance of presidential orders tend to come not from Congress but rather from the White House itself. An executive order originally issued by President John F. Kennedy, for example, requires that drafts of executive orders be submitted first to the director of OMB and then to the Attorney General for legal review.260 These sorts of “intra-executive” procedural requirements call for at least some deliberative processes within the executive branch.261 Yet precisely because these requirements emanate from the White House, not from Congress, they easily can be undone. Or, worse yet, they can simply be disregarded. This is what President Trump allegedly did, for example, when he hastily issued his first travel ban without subjecting it to intra-executive branch legal review.262

In short, significant procedural requirements do more than constrain agency rulemaking; they also ensure the production of an extensive record supporting the promulgation of agency rules. These features distinguish agency rulemaking from presidential orders in a manner that is highly relevant to courts’ review of the latter. At the outset, given that presidential orders are not generally supported by the same sort of clearly defined record that exists in the rulemaking context, it may be difficult for courts to determine what the “record” should include when they go about assessing the legality of a presidential order. Various Trump-era cases, for example, have raised questions about whether the record on review should include information like a president’s campaign statements and assertions made in media interviews.263 As a practical matter, moreover, the lack of procedural constraints governing the issuance of presidential orders—and, in particular, a president’s decision to dispense with intra-executive processes, such as internal legal review—may well encourage a court to review the substance of a presidential order with a more skeptical eye.264

In short, presidents, unlike agencies, encounter few binding procedural hurdles when issuing presidential orders. Any framework for judicial review should reflect this fundamental distinction.265

2. Differing enforcement mechanisms.

A second major difference between agency rules and presidential orders involves enforcement mechanisms. As we explain above, when agencies promulgate legislative rules (as opposed to interpretive rules or policy statements),266 those rules carry the force and effect of law.267 This means that agencies’ legislative rules can be enforced in administrative or judicial forums, and violators of these rules can be hit with significant monetary fines or even sent to prison.268

Some presidential orders—namely, legally binding orders such as President Roosevelt’s 1934 proclamation banning the sale of arms—operate in a similar way.269 They also carry the force and effect of law and, as such, can be legally enforced in court. Moreover, the mechanisms used to enforce legally binding orders, such as fines or prison sentences, can parallel those used to enforce legislative rules issued by agencies.

Many other presidential orders, by contrast, constitute nothing more than marching orders to government officials within the executive branch—orders that are generally enforced, if at all, not through legal mechanisms but rather through political means.270 As we explain above, these orders—nonlegally binding orders—might very well prompt agencies or other government officers to take subsequent actions that have legal effect. However, the orders themselves do not alter legal rights or obligations. One previously discussed example of such an order is Executive Order No 12291,271 President Reagan’s order, which helped to usher in our current era of presidential administration. The mechanisms used to enforce nonlegally binding orders, accordingly, do not parallel those used to enforce an agency’s legislative rules.

This distinction calls into question whether, for purposes of judicial review, all presidential orders should be treated similarly to agencies’ legislative rules—which, as we discussed above, tend to qualify for immediate review under the APA.272 In many respects, legally binding orders are, indeed, similar to legislative rules. As such, it may be appropriate to treat them in a similar fashion. Nonlegally binding orders, by contrast, arguably are more analogous to nonbinding agency policy statements—which, as we discuss above, receive a very different treatment under the APA,273 with immediate review generally not available.274

In short, there are important differences in how various presidential orders and agency rules are enforced. The nuances that emerge from this comparison should be considered in crafting a legal framework for judicial review of presidential orders.275

3. Differing separation-of-powers concerns.

A third difference between agency action and presidential action involves separation of powers. In the context of judicial review of presidential orders, three different separation-of-powers concerns have particular relevance.

First, a motivating force behind the judiciary’s creation of many administrative law principles has been political account-ability. As we previously described, agencies routinely enact rules that announce legislative-like policies. These govern everything from air quality standards to workplace safety.276 Yet agency heads, unlike members of Congress, are not elected by the people. Instead, they are only indirectly accountable via the president. Recognizing this, the courts have worked hard to craft administrative law principles that will further, rather than undermine, agencies’ political accountability. For example, many of the glosses that the judiciary has placed on top of § 553 of the APA, including requirements relating to public-participation mechanisms, reflect underlying concerns about the fact that Congress has been allowed to transfer legislative-like powers to unelected agencies.277

For judicial review of presidential orders, however, the calculus is quite different. This is because the president—unlike the head of any given federal agency—is elected by the people. As a result, when a new policy is announced via a presidential order, that policy flows from a politically accountable actor: the President of the United States.278

A second separation-of-powers concern that differentiates presidential orders from agency action reflects the fact that agencies are creatures of Congress. Agencies are both created by and empowered by statutes. As a result, the judiciary’s main goal in policing agency activity is to ensure that agencies stay within any relevant legal bounds set by Congress. Given that judges are quite comfortable interpreting statutes, the courts are well suited to this task.279

With respect to a president’s orders, however, the relationship to Congress is more complicated. While presidents base some of their orders primarily on powers delegated by Congress, they base others on the president’s own constitutional powers, such as the president’s commander-in-chief power. When it comes to policing the precise boundaries of these constitutional powers, the courts often feel ill equipped to weigh in, or they might prefer that the nation’s political branches work out the proper boundaries.280 As a result, various justiciability doctrines, such as the political-question doctrine, might loom larger in cases involving presidential orders than in cases involving agency action.

Finally, separation-of-powers concerns were what convinced the Court to decide in Franklin that the APA’s definition of agency does not include the president.281 Regardless of whether this determination was correct, the effect of this decision was to preclude the APA—a regime carefully crafted by Congress—from serving as a legal framework for reviewing presidential orders.282 In this sense, the courts resisted a congressionally designed framework in favor of a legal vacuum. In the absence of some newly enacted legislation, a vacuum of this sort requires the judicial branch to fill in the blanks. This responsibility very well may require courts to engage in more creative forms of analysis, as we discuss below.283 It might, for example, require courts to rely on principles—such as those relating to courts’ inherent equitable powers—that can feel unfamiliar or out of place to those accustomed to the statutorily driven APA-based regime.284

In short, many differences exist between agency action and presidential action. With these high-level differences in mind, we turn now to consider several specific legal doctrines that should form part of the courts’ framework for reviewing presidential orders.

B. Incorporating Specific Legal Doctrines

When litigants bring direct and immediate challenges to a president’s order, their challenges tend to implicate a variety of legal issues. These include, but are not limited to: (1) threshold issues, such as ripeness, finality, and cause-of-action requirements; (2) standards of review; and (3) issues arising after a court has deemed an order to be unlawful, such as those relating to availability of relief and severability.285 We turn now to address each of these three categories of issues. In doing so, we set out some preliminary ideas about how these issues might best be resolved, with the primary goal of kick-starting a more robust and ongoing conversation about these complex issues.

1. Threshold issues.

Many threshold issues lurk in the background of challenges to presidential orders. Among these issues are those relating to standing, ripeness and finality, and causes of action.

1. Standing, ripeness, and other timing doctrines.  It is well established that a litigant challenging agency action must overcome a series of threshold hurdles associated with exhaustion, finality, ripeness, and standing.286 These hurdles tend to be more onerous when litigants file lawsuits early in the administrative process. In Ticor Title Insurance Co v Federal Trade Commission,287 for example, the FTC filed an administrative complaint against six insurance companies.288 Rather than wait for the administrative proceedings to unfold, the companies filed a lawsuit in federal court challenging the mere initiation of the administrative proceedings, and the DC Circuit concluded that the challenge was premature. Although all three judges agreed with the basic logic underlying the outcome, each cited a separate doctrine in support: exhaustion, finality, and ripeness. In this way, Ticor confirms that these overlapping doctrines “serve the same general function”: to “avoid premature judicial involvement in the administrative decisionmaking process.”289

   A similar set of concerns relating to timing affects challenges to presidential orders. Indeed, questions of timing loom especially large in the new class of lawsuits this Article has identified: direct and immediate challenges to presidential orders. By design, these lawsuits are brought at the very beginning of the executive-branch process, thereby triggering many of the same concerns that the court grappled with in Ticor and related cases.290

   For cases like Ticor, which involve challenges to agency action, the case law on issues related to timing can be complicated and even convoluted, but judicial precedent nevertheless provides courts with significant guidance.291 By contrast, it is not at all clear how courts adjudicating presidential-order challenges should tackle these recurring questions. Principles of exhaustion and finality pose the most uncertainty. Arguably, these principles do not even apply in the context of challenges to presidential orders. The requirement of finality, for example, derives from statutes like the APA,292 and yet, as discussed, the APA does not apply to challenges to presidential orders.293 In addition, the requirement of exhaustion derives from a combination of statutory requirements and common law principles.294 With respect to the common law, the courts have not clarified whether—much less how—these principles might apply to presidential-order challenges.

   It nevertheless is clear that standing and ripeness—both of which are grounded in Article III of the Constitution295 —do apply to presidential-order challenges.296 Still, the courts struggle to figure out how to understand these doctrinal rules in this less familiar context. For purposes of illustration, consider the lawsuits, discussed previously, challenging Executive Order No 13768. These challenges, filed in 2017, required the courts to determine whether San Francisco and other jurisdictions have standing to challenge a vaguely worded executive order that, as relevant, targets so-called sanctuary jurisdictions.297 Among the many details that the order fails to clarify are what defines a sanctuary jurisdiction—and therefore which, if any, jurisdictions might actually be subject to the order’s directives.298 The order also fails to clearly indicate what steps the agency heads should or will take to fulfill the order’s commands.299 Despite such uncertainty, jurisdictions challenged the order less than a week after its issuance.

   Predictably, a central point of contention in the litigation over Executive Order No 13768 has involved the overlapping doctrines of standing and ripeness.300 The defendants argued that the order’s vagueness precluded the plaintiffs from seeking immediate review both because they could not establish injury in fact (which goes to standing) and because the legal issues were not yet fit for judicial decision (which goes to ripeness).301 Implicit in the defendants’ arguments was the understanding that this same set of objections would not necessarily bar the plaintiffs from challenging actions that agencies eventually take, in the future, pursuant to Executive Order No 13768. Rather, according to the defendants, the real trouble with the early crop of lawsuits was that they were premature. Rejecting these arguments, the Northern District of California concluded that Executive Order No 13768 caused immediate injury that may be considered by a federal court precisely because of the uncertainty the order already had caused.302 In August 2018, the Ninth Circuit largely affirmed the judgment of the district court.303

   For cases like these, courts and litigants may benefit from considering the distinction that we explored earlier between legally binding orders and nonlegally binding orders.304 This is because the distinction may help to resolve questions of ripeness and standing in the context of presidential-order challenges. As we explain above, presidential orders (which at times are issued separately and at times combined in a single document305 ) generally can be separated into two loosely defined categories.306 Legally binding orders carry the force and effect of law and can be legally enforced in court—in a manner analogous to legislative rules promulgated by agencies. By contrast, nonlegally binding orders do not themselves alter legal rights or obligations. In this sense, nonlegally binding orders are not at all analogous to legislative rules. Rather, they are much more analogous to nonbinding agency policy statements—statements that courts generally will not allow to be subjected to immediate judicial review.307

   In a 2017 opinion, a federal district court judge expressly recognized this analogy between policy statements and nonlegally binding presidential orders.308 The case involved various Trump-era executive orders addressing the federal civil service and collective bargaining rights. Citing a DC Circuit case involving agency policy statements, the district judge rejected the plaintiffs’ efforts to challenge specific executive order provisions that espoused only “abstract policy principles” and that failed to “dictate particular outcomes.”309 Furthermore, given that the challenged provisions of the executive orders lacked any “independent operative legal effect,” the judge questioned whether the plaintiffs had Article III standing to bring suit.310

   The trouble, of course, is that it is not always clear whether a challenged provision found in a given presidential order is or is not legally binding. Indeed, as we have already described, many legal challenges to Trump-era orders have involved intense debates over whether the president’s order did—or did not—have any legal effect.311 As a result, whether the distinction between legally binding orders and nonlegally binding orders can prove helpful in resolving questions of ripeness and standing in the context of a presidential-order challenge depends on whether the line is judicially ascertainable.

   Analysis of judicial treatment of analogous distinctions in the agency context suggests that ascertaining this distinction is difficult, but achievable. In the agency context, when courts must distinguish legislative rules from nonbinding agency policy statements, courts give weight to an agency’s characterizations of its own actions but refuse to accept those labels as determinative.312 In Appalachian Power Co v Environmental Protection Agency,313 for example, the petitioners sought review of an EPA “guidance document” that petitioners alleged improperly imposed requirements under the Clean Air Act. A threshold issue in the case was whether the guidance document was a mere “policy statement,” not subject to judicial review, or a “legislative rule,” potentially reviewable by the courts.314

   In determining where the guidance document fell within this framework, the court looked to whether the document had a “binding” effect—either on its own terms, or as a practical matter.315 Ultimately, the court concluded that the challenged guidance document did have a binding effect, despite boilerplate language to the contrary, and so it deemed the document reviewable and ultimately set it aside as unlawful.316 In this way, the Appalachian Power court set forth a roadmap for distinguishing between legislative rules and nonbinding agency policy statements.

   Transplanted from the administrative law context, the Appalachian Power line of cases could prove very helpful to courts in distinguishing between challenges to presidential orders that may go forward versus those that are premature. One effect of this transplantation would be to require many litigants to wait for subsequent agency action before challenging the policies set forth in nonlegally binding orders. A partially offsetting effect, however, would be to allow litigants (such as those in Appalachian Power) to push back on the executive branch’s characterization of a president’s orders. If a president or her subordinates were to treat a presidential order as binding—or if litigants otherwise could demonstrate that an order was binding as a practical matter—then the label assigned to that order would not be determinative. Rather, a court could conclude that the presidential order has a binding effect and that it can be subject to immediate judicial review.

2. Causes of action.  A claim brought in federal court must be based on a valid cause of action.317 For challenges to agency action, litigants tend to have little trouble satisfying this requirement. This is because the APA provides a sweeping, transsubstantive cause of action for parties “adversely affected or aggrieved by agency action” for which “there is no other adequate remedy in court.”318 Alternatively, litigants challenging agency action may be able to take advantage of special statutory proceedings,319 in which a separate statute provides plaintiffs with a more targeted cause of action.320

   By contrast, those challenging a president’s orders generally have no statutorily conferred cause of action to support their claim. This is largely due to two limitations contained in the APA. First, the APA provides a cause of action to challenge only “agency” action—which, as we have discussed, has been read not to include presidential action.321 Second, the APA’s cause of action extends only to challenges of “final” agency action.322 As a result of this second limitation, claims (particularly those brought early in the process) will at times fall outside the APA’s cause-of-action ambit, even when the lawsuits are brought against subordinate officials for agency actions taken pursuant to presidential orders. These interlocking limitations mean that immediate challenges to presidential orders generally cannot rely on the APA to provide a cause of action.323

   In the absence of a statutorily conferred cause of action, litigants challenging presidential orders must look elsewhere to satisfy the cause-of-action requirement. In light of the federal courts’ resistance to novel Bivens-style theories, the Constitution itself is unlikely to provide such a foundation.324 A separate source of authority nevertheless might provide the necessary support: an authority we refer to as the courts’ “inherent equitable powers.”325 (At times, lower courts and commentators refer to this phenomenon as “non-statutory review.”326 ) As Professor Jonathan Siegel has explained, this form of judicial review is not well known.327 Yet it arguably helps to explain how William Marbury was able to sue James Madison,328 how Youngstown Sheet & Tube Company was able to sue Charles Sawyer,329 and how, in a range of lawsuits, plaintiffs were able to sue defendants for injunctive relief without identifying any statutorily or constitutionally conferred cause of action.330

   Despite the long pedigree, the nature and source of these inherent equitable powers remain unclear. The Supreme Court nevertheless provided some context in a 2015 case, Armstrong v Exceptional Child Center,331 which addressed the ability of service providers to sue state officials for an alleged violation of the federal Medicaid Act. The Court concluded that the courts’ inherent equitable powers could not support the particular lawsuit the plaintiffs wished to bring.332 In an opinion written by Justice Scalia, the majority nevertheless acknowledged that the “[t]he ability to sue to enjoin unconstitutional actions by state and federal officers is the creation of courts of equity, and reflects a long history of judicial review of illegal executive action, tracing back to England.”333 Writing in dissent, Justice Sonia Sotomayor agreed that the underlying principle—“[t]hat parties may call upon the federal courts to enjoin unconstitutional government action”—was “not subject to serious dispute.”334 In support, she cited Ex parte Young,335 acknowledging that such lawsuits sometimes are referred to by that case’s name.336 Justice Sotomayor also cited Free Enterprise Fund v Public Company Accounting Oversight Board,337 a 2010 case in which the Supreme Court confirmed that review of this sort generally is available in response to a constitutional challenge.338

   Despite the uncertainties that surround this area of the law, it is the courts’ inherent equitable powers that appear to provide the basis for many recent challenges to presidential orders. A small handful of courts have reached this conclusion explicitly. In response to challenges to Proclamation No 9645 (Trump’s third travel ban), for example, the Ninth Circuit rejected the Government’s argument that the plaintiffs lacked a cause of action to sue for alleged violations of the Immigration and Nationality Act. The court initially concluded that the APA provided plaintiffs with a cause of action.339 Perhaps recognizing that this analysis would not necessarily hold up on review,340 the court then identified a second, independent way for the plaintiffs to meet the cause-of-action requirement: through what the court referred to as an “equitable cause of action.”341 The Supreme Court eventually did adjudicate the plaintiffs’ claim on the merits, though without acknowledging the cause-of-action requirement.342

   As the Supreme Court’s treatment of this issue helps to confirm, the Ninth Circuit’s discussion was unusual in the context of presidential-order challenges because it openly addressed the cause-of-action requirement. Often, courts and litigants avoid the subject altogether.343 Other times, litigants and judges respond to the requirement with apparent confusion. In one challenge to the President’s transgender-troop-related directives, for example, a sophisticated set of lawyers representing the plaintiffs cited a clearly inapposite statute—42 USC § 1983—as the relevant cause of action.344 Even those plaintiffs who more clearly try to invoke the courts’ inherent equitable powers often seem uncertain of what to call it.345

   These scattered approaches confirm the need for clarification. Among the issues that warrant examination are the scope and reach of the courts’ inherent equitable powers (an area whose unsettled nature is confirmed by Armstrong’s opaque holding),346 the nature of the review that accompanies suits brought under the courts’ equitable powers,347 and how the operation of inherent equitable powers changes, if at all, when the defendant is the president.348 These areas of uncertainty go to the heart of the courts’ equitable powers—which in turn go to the heart of many of the challenges to presidents’ unilateral orders.

   In deciding how to resolve these disputes, courts would benefit from focusing on the separation-of-powers concerns that contributed to the confusion in the first place. In Franklin v Massachusetts, the Supreme Court explained its consequential holding (that Congress did not mean for the president’s actions to be included within the ambit of the APA) not by citing the usual principles of statutory interpretation, but rather by invoking “the separation of powers and the unique constitutional position of the President.”349 The result of this approach did more than relieve the president from having to comply with the many requirements imposed by the APA; it also deprived litigants of a congressionally conferred cause of action that they otherwise could have used to help check the president’s actions.

   Particularly in light of the developments of the last century—which has seen a massive transfer of policymaking authority from the legislative branch to the executive branch, coupled with increasingly aggressive attempts by presidents to control that policymaking—separation-of-powers principles cut not only in the direction of protecting the president, as the Court recognized in Franklin. They also cut in the direction of checking the president, to help ensure that he remains within legal limits. With respect to the cause-of-action requirement, separation-of-powers principles therefore would seem to support (and perhaps even necessitate) a robust understanding of equitable powers in response to presidential-order challenges. To this end, courts adjudicating these challenges—whether the challenges assert a constitutional or a statutory violation—should consider implementing a variation of the presumption that Justice Sotomayor offered in dissent in Armstrong: a presumption that generally allows claims challenging the lawfulness of a presidential order to go forward under an equitable theory and that assumes, in the absence of strong evidence to the contrary, that Congress does not mean to disturb this arrangement.350 In other words, the presumption of the availability of judicial review should be particularly strong. Such a presumption would be consistent with Chief Justice John Marshall’s pronouncements made long ago in Marbury v Madison351 : “The very essence of civil liberty certainly consists in the right of every individual to claim the protection of the laws, whenever he receives an injury.”352 As Chief Justice Marshall explained, “the United States has been emphatically termed a government of laws, and not of men,” and it is the judiciary’s duty to protect the nation’s laws, even when this requires the courts to review the legality of executive-branch action.353

2. Standards of review.

Another important set of questions involves the standards of review that should govern courts’ adjudication of executive-order challenges. For example, should the courts give a president significant deference when reviewing her orders on the merits? Or should they engage in more searching review? The courts have provided little insight into this fundamental set of questions. Nonetheless, at least two competing answers have emerged in preliminary discussions among judges, litigants, and scholars.

The first approach encourages courts to borrow heavily from the complex web of judicial doctrines that Congress and the courts already have developed to govern judicial review of agency action. This web includes arbitrary and capricious review (also known as hard look review), which is used to assess the adequacy of agencies’ reasons.354 Some scholars have recently taken the position that the courts should borrow from these sorts of administrative law deference doctrines when reviewing presidential orders. Professor David Driesen, for example, has argued that courts should apply a form of arbitrary and capricious review to ensure that presidential orders are supported by facts, as well as by a rationale adequately connected to the source of legal authority that purportedly authorized the order.355 In a similar vein, Professor Kathryn Kovacs recently argued that the APA’s arbitrariness standard should be used to assess the legality of presidential action.356 Kovacs’ argument rests on the notion that “if the President acts like an agency, he should be treated like an agency.”357

An alternative approach encourages the courts to shun administrative law’s many deference doctrines and instead apply a more deferential, one-size-fits-all approach: rational basis review. Rational basis review, a highly deferential form of review that is generally used to assess the constitutionality of legislation, is what the Supreme Court applied in its June 2018 decision in Trump v Hawaii. That case involved an executive order that, according to the Court, implicates both immigration policy and “the area of national security”358 and therefore warrants particularly deferential review. Citing these considerations, the Court concluded that it would apply rational basis review and, in so doing, held that the government had set forth a sufficient justification in support of Trump’s third travel ban, at least at the preliminary stages of the case.359

Although these two different models might continue to provide some guidance, we believe that courts and litigants engaged in assessing the legality of presidential orders should avoid borrowing too reflexively from either of them. This is because both models were developed with very different kinds of governmental action in mind. The first was developed to review agency action, the other to review legislative action. Congressional legislation, in particular, receives significant deference (and is generally subjected only to mere rationality review) in large part because the Constitution vests legislative judgments with the people’s political representatives in Congress and because, in order to enact a law, the members of Congress must comply with cumbersome bicameralism and presentment procedures spelled out in the Constitution.360 By contrast, when the president issues a presidential order pursuant to a congressionally delegated power, he often can do so without adhering to any cumbersome procedural formalities, even though he is essentially filling Congress’s shoes and carrying out a lawmaking-like role pursuant to a congressional delegation.361 Thus, at least with respect to presidential orders issued pursuant to some kind of statutory authorization, mere rationality review may not adequately ensure that the president remains within the bounds set by Congress.

Similarly, there are various differences between agency action and presidential action that may well call into question the propriety of applying administrative law deference doctrines in the presidential order context. For example, as we have discussed, when agencies take action, they must create a sufficient record and provide detailed explanations to enable the courts, pursuant to hard look review, to assess the adequacy of the agency’s reasons for its action.362 And when the courts assess the adequacy of an agency’s reasons, the courts tend to demand technocratic justifications from agencies that are grounded in the law, facts, and evidence—not in political or policy-driven explanations.363 It would be difficult for the courts to apply a robust form of arbitrary and capricious review (akin to hard look review) to presidential orders without also effectively demanding more of the presidents who are issuing those orders: perhaps technocratic justifications, detailed records, or more. These de facto requirements could, in turn, raise serious separation-of-powers concerns; the judiciary might be seen as impermissibly micromanaging the president’s decision-making process or otherwise compromising the Constitution’s separation of powers.

These same considerations also suggest that there are good reasons for the courts to resist trying to transfer into the presidential-order context something akin to Chevron deference. As we discuss above, Chevron deference applies to agencies’ reasonable interpretations of statutory ambiguities.364 For an agency to receive Chevron deference, its interpretation must be set forth in a format that suggests that Congress intended the agency’s views to carry the force and effect of law.365 Normally, an agency cannot meet this requirement unless it subjected its interpretation to some kind of process, like the notice-and-comment rulemaking process, that is designed to ensure public participation and deliberation.366 This limitation helps preserve Chevron deference for agency interpretations that are the result of careful consideration and public input.367 In contrast to agency rules that are subject to notice-and-comment rulemaking or other deliberative processes, presidential orders are routinely issued “without any prior public procedure, and often without any accompanying explanation.”368 In light of this reality, presidential orders arguably should not be eligible for the same kind of strong deference that an agency enjoys.

As this discussion helps to confirm, we do not believe that it would be advisable for the courts to blindly transfer either administrative law’s many complex deference doctrines or the legislative arena’s highly deferential rationality review into the presidential-order context. Instead, we believe that an intermediate approach ultimately might make the most sense—one that neither rubber-stamps presidential orders nor intrudes too severely into the president’s policymaking sphere.

Specifically, with respect to reason-giving review, one reasonable approach would be for courts reviewing presidential orders to protect against arbitrariness—without going so far as to demand that the president set forth the same kinds of lengthy, detailed technocratic justifications that agencies must.369 Under this fairly deferential form of reason-giving review, the president would be required to set forth a nonarbitrary justification in the text of her orders themselves; justifications set forth in post-hoc litigating positions would not be sufficient.370 Then, if challenged in court, the judiciary would be required to recognize the adequacy of the president’s justifications, including any reliance on political and policy-based justifications, so long as the decisional factors that the president relied upon were not legally foreclosed by existing statutes or the Constitution,371 and so long as any factual justifications had adequate support.372 In other words, the focus of the judiciary’s reason-giving inquiry would be narrow, and the president (unlike agencies) would be given breathing room to rely explicitly on political or policy-based considerations, so long as neither Congress nor the Constitution foreclosed the president from considering those factors.373

With respect to questions of law, it does not seem to us that Chevron deference or any other form of strong deference will prove appropriate in the context of presidential-order challenges, for the reasons noted above. Instead, the courts should likely engage in de novo review when it comes to questions of law. De novo review of questions of law will help ensure that presidential orders remain within the confines of the president’s legal authority, including within the bounds of any congressionally imposed limits. It will also ensure that the courts—not the president—act as the ultimate deciders as to whether a given order does or does not have legal effect, thereby ensuring that the president cannot evade judicial review simply by interpreting a given order to lack legal consequences.374

3. Issues that arise after a court’s finding of unlawfulness.

Finally, in the context of challenges to presidential orders, many open questions relate to issues that arise after a court’s finding of unlawfulness. These include questions implicating the general availability of judicial relief, as well as the availability of particular doctrines such as severability.

1. Availability of relief.  The law surrounding remedies for unlawful agency action is relatively well understood and well settled.375 Under the APA, for example, a plaintiff can sue an agency, agency officials, or the United States, seeking different forms of relief, such as a judicial ruling that sets aside unlawful action agency.376 The rules governing judicial remedies, however, become much less clear once a litigant brings a lawsuit directly against the president. Complications arise, for example, in light of longstanding uncertainty over the degree to which the president is susceptible to legal process, including whether the courts can issue injunctive relief or a declaratory judgment against the president.377

   A 2017 lawsuit brought against Trump and two of his subordinates helps to illustrate.378 The plaintiffs sought an injunction requiring that the defendants allow the plaintiffs greater access to the President’s Twitter feed.379 Maintaining that the court lacked jurisdiction to grant the requested relief, the government argued categorically that “the constitutional separation of powers precludes [a federal court] from assuming control over the President’s discretionary official conduct in the form of an equitable injunction.”380 In response to prior decisions—such as United States v Nixon381 and Youngstown, which were arguably in tension with the government’s position—the government argued that those prior precedents were limited to three strict categories: (1) judicial relief pursued against the president for unofficial conduct; (2) judicial relief pursued against the president himself with respect to a subpoena duces tecum; and (3) judicial relief pursued against the president’s subordinates, rather than against the president himself.382 The plaintiffs, by contrast, interpreted the prior precedents in a more expansive manner—as confirming the ability of federal courts to issue the requested injunction against the president383 —with amici arguing broadly that “[f]ederal courts have equitable power to order the President of the United States to comply with the law and the Constitution.”384 Rejecting the categorical arguments advanced by the defendants, the district court nevertheless proceeded cautiously, announcing it would grant only declaratory relief vis-à-vis Trump. It nevertheless suggested that in the future it would address, if necessary to ensure compliance, whether injunctive relief against the president might also be available.385

   The district court’s guarded approach helps to confirm the uncertainty that surrounds whether and how courts can exercise direct control over a president’s actions. Given that Trump-era litigants so often bring lawsuits against the President in name, this area of the law likely will require clarification sooner rather than later. Still, in light of the acute separation-of-powers concerns presented when a judge directs a president to comply with a court order, courts should strive to avoid these questions—where they can without prejudice to the plaintiffs’ case—by issuing judicial orders against a president’s subordinates, rather than against the president himself. Given that a president so rarely executes his orders personally, almost any presidential action effectively can be enjoined through injunctions against the president’s subordinates.386 Recognizing this dynamic, courts adjudicating Trump-era lawsuits have repeatedly issued relief against subordinates, rather than the President himself, thereby punting on the thorny questions that arise when courts try to enjoin a president directly.387 This sort of judicial avoidance is prudent both because orders running against the president inevitably require a difficult form of case-by-case analysis that is resistant to bright-line rules,388 and also because a judicial order against a president could give rise to a constitutional crisis if the president refused to comply.

   Nevertheless, there eventually will arise a case in which full relief can be granted only if an order runs against the president directly.389 When this occurs, courts will be forced to grapple with competing separation-of-powers concerns. In that case, the better argument—based both on historical precedent and structural constitutional analysis—appears to be that the courts can, when necessary to accord relief, compel the president to comply with the law.390

2. Severability.  Judicial review—whether the review is of a statute, an agency regulation, or a presidential order—requires a court to consider whether the challenged rule is lawful. On occasion, a court will conclude it is not. This, in turn, presents the question how to construe that defective legal rule going forward. Faced with this dilemma, courts generally resort to severability.391

   Severability can be thought of as a framework for analysis—one that requires a court to disregard an unlawful rule either in whole or in part.392 This framework aims to achieve at least three interrelated purposes. First, it seeks to provide the courts with guidance in response to a difficult predicament: what to do with a rule that remains on the books but that is partially unlawful. Second, it tries to constrain the courts’ options to a mechanical, even bimodal, choice, rather than allow the court to engage in wholesale reconstruction of another branch’s legal rule. Third, it seeks to limit the disruptive effect of the court’s legal ruling.393

   In determining how to conduct the severability analysis with respect to statutes, the courts are expected to discern how the enacting Congress, had it known about the defect, would have preferred the statute to be treated.394 Partially in an effort to accurately identify this legislative intent, courts have adopted a presumption of severability—at least, so long as the courts “can identify a relatively surgically precise way of curing the defect.”395

   Courts also have experience engaging in severability analysis in the context of legally defective agency regulations.396 In this context, courts generally have treated the severability analysis as analogous to that applicable in the statutory context. Looking to the intent of the agency, the courts attempt to determine whether the agency would have promulgated the regulation without the offending portions.397 If so, the regulation is severable; if not, it is not severable.

   Severability precedents are far sparser in the context of presidential orders. The relatively few courts to have grappled with unlawful presidential orders have engaged in little open discussion regarding how the severability framework should apply in this distinct context. Instead, the courts generally have either engaged in severability analysis without acknowledging what they are doing or have adopted, largely without discussion, an approach analogous to that which governs in the legislative and regulatory contexts. In the latter set of precedents, accordingly, the courts have tackled questions of severability by attempting to discern the intent of the president.

   One example of courts engaging in implicit severability analysis came in response to Trump’s first travel ban order. This order included a number of proclamations and instructions, including those contained in § 3(c), which imposed a temporary ban on entry by citizens of seven countries, as well as § 5, which, among other things, directed the Secretary of State to suspend for a time the US Refugee Admissions Program.398 In preliminary stages of the litigation challenging the order, a district court in Seattle enjoined the federal government from enforcing both § 3(c) and parts of § 5.399 Yet other parts of § 5—such as a proclamation capping the number of refugees to be admitted in 2017—remained untouched. Implicitly, the district judge had conducted a severability analysis and concluded that the remainder of the executive order could remain in force. Yet notwithstanding the legal and practical importance of this severability-related decision, the court never expressly identified its reasoning.

   Reviewing this decision on an expedited posture, the Ninth Circuit refused to grant the government’s request for a stay pending appeal.400 In so doing, the majority did not call into question the district judge’s implicit severability analysis. Dissenting from the denial of reconsideration en banc, however, five judges insisted that the severability analysis should have proceeded differently. “The law of our circuit is that we consider the severability of an executive order just as we would consider the severability of a statute,” the dissent wrote, citing a case called Matter of Reyes.401 For the dissent, this translated into an “obligation to maintain as much of the order as is legal,” so long as it can remain “operational” without violating the Constitution.402 As applied to the case before them, the dissenting judges would have favored narrowing the injunction to protect only some of those individuals who were affected by § 3(c) and the relevant provisions of § 5.403

   Despite these differing views concerning precisely how much of the presidential order could remain operational, the judges adjudicating the travel ban challenges seemed to agree, largely without discussion, that the unconstitutional portions of the presidential order could and should be severed from the remainder. And they reached this conclusion without openly grappling with whether the legislative model’s presumption of severability should provide the proper model when it comes to presidential orders. However, there are reasons to question whether it should.

   As we discuss above, a president can issue a presidential order without the same cumbersome and often time-consuming processes required to enact a statute or a regulation.404 As a result, the president can reconsider, rewrite, and reissue partially unlawful orders with an ease that is unavailable to either Congress or administrative agencies. These procedural differences call into question whether it is appropriate to place a thumb on the scale in favor of severability when a court is faced with a single, presidentially issued document containing interlocking provisions. Indeed, in many circumstances, the three main purposes of severability might be better served by applying a presumption in favor of inseverability. As indicated above, severability’s first purpose seeks to provide the courts with guidance. Yet this goal is equally, if not better, served by a requirement that a partially unlawful presidential order simply be invalidated in full. Second, severability analysis aims to prevent the courts from engaging in wholesale reconstruction of another branch’s legal rule. In most cases, however, the complete invalidation of an executive order requires less reconstruction than a partial invalidation.405 Finally, the severability framework seeks to limit the disruptive effect of the court’s substantive legal ruling. It is true that complete invalidation of a presidential order may feel much more disruptive than a partial invalidation (and that, politically, it may be perceived that way). Yet ultimately the goal is to effectuate the executive’s intent. And given the relative ease with which a president can reissue the order in a legally valid form, there is a straightforward way by which to determine and effectuate the president’s preferences: just invalidate the order and wait for the White House to reissue it in the manner that the president feels is best.

   Were courts to employ a presumption of inseverability in the context of challenges to presidential orders, the approach would counteract at least some of the problems associated with trying to discern presidential intent. This task, difficult in any context, is particularly challenging when the courts do not have the benefit of legislative history or an APA-driven administrative record. Moreover, the courts could mitigate the disruptive effect of their severability rulings by delaying, where appropriate, the effect of an injunction.406

   Still, a presumption of inseverability should be just that: a presumption. If an order clearly indicates the president’s intent that the order be severable, fundamental principles of severability would seem to require that a court disregard the contrary presumption and effectuate the president’s intent.407 A president might attempt to express such an intent, for example, through use of a severability clause. In addition, there may be cases in which the retroactive effect of a presidential order is relevant. In those cases, it may not be possible for the president to reissue an order in a manner that captures this prior conduct, and, as a result, there may be a more significant cost to a court’s refusal to sever offending provisions. In such a circumstance, a presumption of inseverability may not be appropriate.

* * *

The specific doctrinal issues that we discuss are only a few of many that warrant attention in the context of challenges to presidential orders.408 Working through all these complex legal issues will take time and careful thought. Nonetheless, given the recent rise in presidential-order challenges and the likelihood that this new form of litigation aimed at the president will endure, those efforts are critical. It is also critical that those working to formulate a framework for judicial review take into account, as we have tried to do here, significant similarities and differences between presidential action and agency action.

Conclusion

The inauguration of President Trump ushered in a new type of legal challenge to executive-branch activity. Rather than wait for agency action, litigants now routinely sue in direct and immediate response to unilateral presidential orders. Several phenomena, including the decades-long trend toward increasingly forceful presidential control over the regulatory state, appear to have provoked this shift in litigation strategy. Historically, however, challenges against presidential orders have been rare. As a result, the courts lack a coherent or well-theorized framework for adjudication.

As this Article explains, presidents’ reliance on politically aggressive and controversial presidential orders did not begin with Trump, and they almost certainly will not end once he leaves office. Particularly in light of recent, high-profile victories by those enlisting the courts to resist the President’s efforts at policymaking, presidential orders inevitably will continue to trigger legal challenges. This confirms the need for a continuing dialogue—among courts, litigants, scholars, and, ideally, Congress, among others—over the proper framework for review. This Article sketches out what such a legal framework might look like. In so doing, it takes note of the relevance of administrative law principles and identifies critical differences between presidential action and agency action. With this in mind, it begins the work of unpacking how these differences should inform courts’ legal framework for reviewing presidential orders.

 

• 1See, for example, Karma Allen, Protests Erupt Nationwide Following Trump’s Transgender Military Ban Announcement (ABC News, July 27, 2017), archived at http://perma.cc/E297-G9JR. See also Niraj Chokshi and Nicholas Fandos, Demonstrators in Streets, and at Airports, Protest Immigration Order (NY Times, Jan 29, 2017), archived at http://perma.cc/U5J6-APBF.
• 2See, for example, Merrit Kennedy, 2 Lawsuits Challenge Trump’s Ban on Transgender Military Service (NPR, Aug 28, 2017), archived at http://perma.cc/A3RC-SU2M; Conrad Wilson, Portland Joins Lawsuit over Trump’s Sanctuary City Order (Oregon Public Broadcasting, June 30, 2017), archived at http://perma.cc/NW7N-PCQ3.
• 3See Lisa Manheim and Kathryn Watts, The Limits of Presidential Power: A Citizen’s Guide to the Law 1–3, 83–84 (Manheim & Watts, LLC 2018) (describing this early wave of litigation).
• 4See id.
• 5See Part III.A.
• 6See Part II.B.1.
• 7See id.
• 8See Elena Kagan, Presidential Administration, 114 Harv L Rev 2245, 2277–80 (2001) (documenting the dramatic rise in presidential control over agency activity that began with President Reagan). See also generally Kathryn A. Watts, Controlling Presidential Control, 114 Mich L Rev 683 (2016) (analyzing the rise of presidential control over the administrative state through executive orders and focusing particularly on the George W. Bush and Obama administrations).
• 9See Executive Order No 12291, 3 CFR 127 (1981).
• 10See, for example, Center for Science in the Public Interest v Department of the Treasury, 573 F Supp 1168, 1174 n 5 (DDC 1983).
• 11See Presidential Memorandum of March 13, 2014; Updating and Modernizing Overtime Regulations, 79 Fed Reg 18737 (2014).
• 12See Nevada v United States Department of Labor, 2017 WL 3780085, *1 (ED Tex 2017).
• 13See Part III.A.
• 14Executive Order No 13771, Reducing Regulation and Controlling Regulatory Costs, 82 Fed Reg 9339 (2017).
• 15Complaint for Declaratory and Injunctive Relief, Public Citizen, Inc v Trump, No 1:17-cv-00253 (DDC filed Feb 8, 2017).
• 16See Table 1 (listing lawsuits).
• 1760 Stat 237 (1946), codified as amended at 5 USC §§ 500 et seq.
• 18See Franklin v Massachusetts, 505 US 788, 800–01 (1992).
• 19See notes 190–92 and accompanying text. But see Kathryn E. Kovacs, Trump v. Hawaii: A Run of the Mill Administrative Law Case (Yale J Reg: Notice & Comment, May 3, 2018), archived at http://perma.cc/BZH8-S8S3 (arguing that the Supreme Court was wrong in Franklin and that presidential orders should be subject to APA review).
• 20City and County of San Francisco v Trump, 897 F3d 1225, 1238 (9th Cir 2018). See also Erica Newland, Note, Executive Orders in Court, 124 Yale L J 2026, 2035 (2015) (concluding that courts have not recognized, much less resolved, the “common jurisprudential questions” raised by executive orders); Matthew Chou, Agency Interpretations of Executive Orders, 71 Admin L Rev *3 (forthcoming 2019), archived at http://perma.cc/SY8U-8AQA (describing the “underdeveloped and outdated” nature of case law concerning judicial deference to agency interpretations of executive orders).
• 21See Part II.B.2.
• 22See Newland, Note, 124 Yale L J at 2035 (cited in note 20). See also Part II.B.2.
• 23See Kevin M. Stack, The Reviewability of the President’s Statutory Powers, 62 Vand L Rev 1171, 1176–77 (2009) (describing the lack of “doctrinal pressure” for courts to continue developing these preexisting doctrines).
• 24See generally Kagan, 114 Harv L Rev 2245 (cited in note 8); Watts, 114 Mich L Rev 683 (cited in note 8).
• 25See Part II.B.1.
• 26See id.
• 27Recently, scholars have begun tackling specific doctrinal issues implicated by judicial review of executive orders. See, for example, David M. Driesen, Judicial Review of Executive Orders’ Rationality, 98 BU L Rev 1013, 1015 (2018) (addressing the appropriate standard of review in the adjudication of executive-order challenges and noting that “[p]erhaps surprisingly, this question has generated almost no commentary and little case law”). By contrast, this Article takes a broader view, as it explores the overarching need for a more theorized and coherent approach to guide judicial review of executive orders in light of presidents’ increasingly forceful efforts to control agency policymaking, coupled with litigation strategies unfolding in the Trump era.
• 28See Tara L. Branum, President or King? The Use and Abuse of Executive Orders in Modern-Day America, 28 J Legis 1, 6–7 (2002) (explaining that while the president’s toolkit includes executive orders, proclamations, and memoranda, the difference between them all is “typically one of form, not substance”).
• 29This Article focuses its attention on judicial review of presidential orders used to set generally applicable policies and thus declines to engage directly with the separate legal considerations implicated by presidents’ use of unilateral written directives to target specific individuals for special treatment (for example, through the issuance of pardons). See, for example, Ralls Corp v Committee on Foreign Investment in the United States, 758 F3d 296, 325 (DC Cir 2014) (enjoining in part a presidential order relating to two individuals’ investment in a windfarm project). In other words, this Article focuses on presidential orders that look and operate in many respects like agency rules rather than agency adjudications.
• 30See, for example, Kevin Stack, Interpreting Regulations, 111 Mich L Rev 355, 420 (2012) (noting that the era of statutes has given way to an era of regulation). See also generally Richard J. Pierce Jr, Rulemaking and the Administrative Procedure Act, 32 Tulsa L J 185 (1996) (detailing the development of rulemaking).
• 31See Part II.A.
• 32See Harold H. Bruff, Presidential Power Meets Bureaucratic Expertise, 12 U Pa J Const L 461, 469 (2010) (“In the 1970s, Presidents began taking an active role in managing regulation.”). See also Part II.A.
• 33See Kagan, 114 Harv L Rev at 2277 (cited in note 8). See also Part II.A.
• 34See Kathryn A. Watts, Proposing a Place for Politics in Arbitrary and Capricious Review, 119 Yale L J 2, 5 & nn 1–4 (2009) (describing arbitrary and capricious review).
• 355 USC § 706(2)(A).
• 36See Louis J. Virelli III, Deconstructing Arbitrary and Capricious Review, 92 NC L Rev 721, 727 (2014) (“Around the time of the APA’s adoption, arbitrariness review mirrored the highly deferential rational basis review employed by pre-APA courts.”).
• 37See id. See also Matthew Warren, Active Judging: Judicial Philosophy and the Development of the Hard Look Doctrine in the D.C. Circuit, 90 Georgetown L J 2599, 2601–03 (2002).
• 38See Warren, 90 Georgetown L J at 2602 (cited in note 37) (explaining that “trust in agency experts evaporated” in the 1960s and 1970s); Virelli, 92 NC L Rev at 727 (cited in note 36) (“In the 1960s and 1970s, the interest group model and concerns about agency capture emerged as dominant features of administrative law.”).
• 39See Warren, 90 Georgetown L J at 2602–03 (cited in note 37) (describing the rise of hard look review); Thomas W. Merrill, Capture Theory and the Courts: 1967–1983, 72 Chi Kent L Rev 1039, 1039–44 (1997) (“[M]any federal judges became convinced that agencies were prone to capture and related defects and—more importantly—that they were in a position to do something about it.”).
• 40Greater Boston Television Corp v FCC, 444 F2d 841, 851 (DC Cir 1970).
• 41See, for example, id; Industrial Union Department, AFL-CIO v Hodgson, 499 F2d 467, 475 (DC Cir 1974); National Association of Food Chains, Inc v Interstate Commerce Commission, 535 F2d 1308, 1314 (DC Cir 1976).
• 42463 US 29 (1983).
• 43See Peter L. Strauss, ed, Administrative Law Stories 369 (Foundation 2006) (“High on Reagan’s list of immediate targets were regulatory requirements affecting the automobile industry.”).
• 44State Farm, 463 US at 46–48, 53–57.
• 45Id at 59 (Rehnquist concurring in part and dissenting in part).
• 46See Michael Herz, The Rehnquist Court and Administrative Law, 99 Nw U L Rev 297, 310–11 (2004) (“Justice White did not contradict Justice Rehnquist’s description of the political setting or conclude that it was outweighed by other factors. Rather, he ignored it altogether, implicitly deeming the politics of the rescission simply irrelevant.”).
• 47In so doing, the Court may have taken a cue from the litigants and the National Highway Traffic Safety Administration. See Stephen Williams, Book Review, The Roots of Deference, 100 Yale L J 1103, 1108 (1991) (arguing that the agency and the litigants identified “science” instead of “politics” as the decision-making paradigm in State Farm).
• 48See Watts, 119 Yale L J at 19 (cited in note 34) (noting that since State Farm, courts have “searched agency decisions to ensure they represent expert-driven, technocratic decisionmaking”).
• 49Hard look review, for example, has played an important role in cases involving everything from the judiciary’s review of the Environmental Protection Agency’s handling of greenhouse gases during the Bush administration to its review of the FDA’s handling of the emergency contraceptive “Plan B” during the George W. Bush and Obama administrations. See Massachusetts v Environmental Protection Agency, 549 US 497 (2007); Tummino v Torti, 603 F Supp 2d 519, 522 (EDNY 2009); Tummino v Hamburg, 936 F Supp 2d 162, 169 (EDNY 2013).
• 50See US Department of Homeland Security, Exercising Prosecutorial Discretion with Respect to Individuals Who Came to the United States as Children (June 15, 2012), archived at http://perma.cc/8U86-X7EF.
• 51Manheim and Watts, The Limits of Presidential Power at 51 (cited in note 3).
• 52See note 50.
• 53US Department of Homeland Security, Rescission of the June 15, 2012 Memorandum Entitled “Exercising Prosecutorial Discretion with Respect to Individuals Who Came to the United States as Children” (Sept 5, 2017), archived at http://perma.cc/GVQ5-BAEC.
• 54Donald J. Trump (@realDonaldTrump), (Twitter, Sept 5, 2017), archived at http://perma.cc/KNJ2-5SPY (“Congress now has 6 months to legalize DACA (something the Obama Administration was unable to do). If they can’t, I will revisit this issue!”). See also Regents of the University of California v United States Department of Homeland Security, 279 F Supp 3d 1011, 1026 (ND Cal 2018).
• 55See, for example, Complaint for Declaratory and Injunctive Relief, Trustees of Princeton University v United States, No 1:17-cv-02325 (DDC filed Nov 3, 2017); Complaint for Declaratory and Injunctive Relief, County of Santa Clara v Trump, No 3:17-cv-05813 (ND Cal filed Oct 10, 2017); Complaint for Declaratory and Injunctive Relief, Casa De Maryland v US Department of Homeland Security, No 8:17-cv-02942 (D Md filed Oct 5, 2017); Second Amended Complaint, Batalla Vidal v Nielsen, No 1:16-cv-04756 (EDNY filed Sept 19, 2017) (available on Westlaw at 2017 WL 10088243); Complaint, National Association for the Advancement of Colored People v Trump, No 1:17-cv-01907 (DDC filed Sept 18, 2017) (available on Westlaw at 2017 WL 4122438); Complaint for Declaratory and Injunctive Relief, Garcia v Trump, No 3:17-cv-05380 (ND Cal filed Sept 18, 2017) (available on Westlaw at 2017 WL 4157508); Complaint for (1) Violation of Fifth Amendment Equal Protection and (2) Violation of 5 USC §§ 553 & 706(2)(D), San Jose v Trump, No 3:17-cv-05329 (ND Cal filed Sept 14, 2017); Complaint for Declaratory and Injunctive Relief, Regents of the University of California v United States Department of Homeland Security, No 3:17-cv-05211 (ND Cal filed Sept 8, 2017); Complaint for Declaratory and Injunctive Relief, New York v Trump, No 1:17-cv-05228 (EDNY filed Sept 6, 2017) (available on Westlaw at 2017 WL 3883559).
• 56See Regents of the University of California, 279 F Supp 3d at 1037–46.
• 57Id at 1026 (“[T]he new administration didn’t terminate DACA on policy grounds. It terminated DACA over a point of law, a pithy conclusion that the agency had exceeded its statutory and constitutional authority.”).
• 58Id at 1037 (“When agency action is based on a flawed legal premise, it may be set aside as ‘arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.’”).
• 59See Regents of University of California v United States Department of Homeland Security, 908 F3d 476 (9th Cir 2018).
• 60See notes 190–92.
• 61See Part IV.B.2.
• 62Procedural requirements for formal rulemaking, for example, can differ from procedural requirements for informal rulemaking. Compare 5 USC § 553 (governing informal proceedings), with 5 USC §§ 556–57 (governing formal proceedings). This Article is focused on informal rulemaking. In addition, the APA provides only a default set of procedural requirements. As such, Congress can alter these requirements as they apply to specific categories or types of rulemaking. See, for example, Magnusson-Moss Warranty Federal Trade Commission (FTC) Improvement Act, Pub L No 93-637, 88 Stat 2183, codified as amended at 15 USC §§ 2301–12.
• 63See 5 USC § 553.
• 645 USC § 553.
• 65See American Radio Relay League, Inc v FCC, 524 F3d 227, 248 (DC Cir 2008) (Kavanaugh concurring in part and dissenting in part) (“Courts have incrementally expanded those APA procedural requirements well beyond what the text provides.”); Jack M. Beermann and Gary Lawson, Reprocessing Vermont Yankee, 75 Geo Wash L Rev 856, 857 (2007) (“[I]n the 1960s and 1970s, the lower federal courts essentially rewrote the APA’s notice-and-comment rulemaking provisions to require extensive procedural machinery, including elaborate notices of proposed rulemaking that disclose to the public all relevant evidence possessed by the agency.”).
• 66See Portland Cement Association v Ruckelshaus, 486 F2d 375, 393 (DC Cir 1973) (“It is not consonant with the purpose of a rule-making proceeding to promulgate rules on the basis of . . . data that, [to a] critical degree, is known only to the agency.”). See also Connecticut Light & Power Co v Nuclear Regulatory Commission, 673 F2d 525, 530–31 & n 6 (DC Cir 1982).
• 67See Reytblatt v United States Nuclear Regulatory Commission, 105 F3d 715, 722 (DC Cir 1997) (“An agency need not address every comment, but it must respond in a reasoned manner to those that raise significant problems.”).
• 68See 5 USC § 553(b)(3) (exempting interpretive rules, policy statements, and procedural rules).
• 69See, for example, Community Nutrition Institute v Young, 818 F2d 943, 949 (DC Cir 1987); Professionals and Patients for Customized Care v Shalala, 56 F3d 592, 601–02 (5th Cir 1995).
• 70See Broadgate Inc v United States Citizenship & Immigration Services, 730 F Supp 2d 240, 243 (DDC 2010) (“Notice and comment procedures are only required under APA § 553 for legislative rules with the force and effect of law.”). See also American Mining Congress v Mine Safety & Health Administration, 995 F2d 1106, 1111–12 (DC Cir 1993).
• 71See Part IV.B.1 (discussing the circumstances under which presidential orders should be immediately reviewable, and noting overlapping considerations with the distinctions in the administrative law context between policy statements and legislative rules).
• 72See 5 USC § 704 (“Agency action made reviewable by statute and final agency action for which there is no other adequate remedy in a court are subject to judicial review.”).
• 73Bennett v Spear, 520 US 154, 178 (1997).
• 74See generally Stephen Hylas, Note, Final Agency Action in the Administrative Procedure Act, 92 NYU L Rev 1644, 1667 (2017) (noting that the Supreme Court’s definition of finality overlaps with the tests used to draw lines between legislative rules and policy statements and thus “nonlegislative rules seem to be effectively immune from pre-enforcement judicial review”).
• 75See Amended Complaint for Declaratory and Injunctive Relief, Texas v United States, No 1:14-cv-254 (SD Tex filed Dec 9, 2014) (available on Westlaw at 2014 WL 7497780) (Texas Amended Complaint).
• 76See notes 50–52 and accompanying text.
• 77See Barack Obama, Remarks by the President in Address to the Nation on Immigration (White House, Nov 20, 2014), archived at http://perma.cc/S6XA-AU3C.
• 78See, for example, How Will Obama’s Executive Order Affect Cache Valley’s Immigrant Families, Students? (Herald Journal, Nov 29, 2014) (available on Westlaw at 2014 WLNR 33751435); President Takes Executive Action on Immigration; Implores Congress to Pass Bill (Law Firm Newswire, Dec 30, 2014) (available on Westlaw at 2014 WLNR 36927412). See also Jennifer Rubin, The GOP Should Look to the Heartland (Wash Post, Dec 30, 2014), archived at http://perma.cc/MX4T-VGUA; H. Roy Kaplan, Sorting Fears from Facts about Hispanic Migrants (Tampa Bay Times, Dec 27, 2014), archived at http://perma.cc/J8XL-4E6N.
• 79See Texas Amended Complaint at ¶¶ 54–55 (cited in note 75).
• 80See generally Defendants’ Memorandum of Points and Authorities in Opposition to Plaintiffs’ Motion for Preliminary Injunction, Texas v United States, No 1:14-cv-254 (SD Tex filed Dec 24, 2014).
• 81Texas v United States, 86 F Supp 3d 591, 648–49 (SD Tex 2015). The judge also noted that the government had “not specifically suggested” that DAPA is not final. Id.
• 82See id at 671. Among other things, the judge reasoned that DAPA “confers the right to be legally present in the United States and enables its beneficiaries to receive other benefits” and thus was not a mere policy statement. Id at 670.
• 83Texas v United States, 787 F3d 733, 764–66 (5th Cir 2015). The Fifth Circuit further reasoned that DHS’s enforcement policy likely ran contrary to the Immigration and Nationality Act. Id at 754.
• 84See United States v Texas, 136 S Ct 2271, 2272 (2016) (“The judgment is affirmed by an equally divided Court.”).
• 85See Part IV.A.1. See also notes 190–95 and accompanying text.
• 86See Tara Grove, Presidential Laws and the Missing Interpretative Theory, 168 U Pa L Rev *20–27 (forthcoming 2020), archived at http://perma.cc/V6XF-HU7K (noting that even though neither the Constitution nor statutes spell out the processes that must be used to create executive orders, presidents have chosen to impose various processes on the executive branch, such as intra-agency consultation requirements).
• 87See Part IV.B.1 (discussing the circumstances under which presidential orders should be immediately reviewable, and noting overlapping considerations with the distinctions in the administrative law context between policy statements and legislative rules); Part IV.B.2 (discussing different standards of review that might guide judicial review of presidential orders, and noting that the inapplicability and unavailability of procedural review might suggest that the courts should apply a robust standard of substantive review).
• 88467 US 837 (1984).
• 89Id at 863–66.
• 90See Lisa Schultz Bressman, Procedures as Politics in Administrative Law, 107 Colum L Rev 1749, 1764 (2007) (noting that Chevron is the “most prominent example” of how administrative law now reflects the presidential control model).
• 91529 US 576 (2000).
• 92533 US 218 (2001).
• 93See Mead, 533 US at 226–27 (noting that a delegation of authority to act with the force of law “may be shown in a variety of ways, as by an agency’s power to engage in adjudication or notice-and-comment rulemaking, or by some other indication of a comparable congressional intent”). See also Christensen, 529 US at 587.
• 94529 US 120 (2000).
• 95Id at 125–26.
• 9652 Stat 1040 (1938), codified as amended at 21 USC § 301 et seq.
• 97William J. Clinton, The President’s News Conference, 1995 Pub Papers 1237, 1237.
• 98See Manheim and Watts, The Limits of Presidential Power at 46 (cited in note 3).
• 99See id at 46–47. See also William J. Clinton, Remarks Announcing the Final Rule to Protect Youth from Tobacco, 1996 Pub Papers 1332, 1333.
• 100See Regulations Restricting the Sales and Distribution of Cigarettes and Smokeless Tobacco to Protect Children and Adolescents, 61 Fed Reg 44396 (1996).
• 101Brown & Williamson Tobacco, 529 US at 125–26.
• 102Id at 126.
• 103See Part IV.B.2. See also Part II.B.2.
• 104See Manheim and Watts, The Limits of Presidential Power at 38–42 (cited in note 3).
• 105Kenneth R. Mayer, With the Stroke of a Pen: Executive Orders and Presidential Power 4–5 (Princeton 2001) (describing how presidents have been able to use executive orders to effectively make law).
• 106See Executive Order No 9981, 13 Fed Reg 4313 (1948).
• 107See Executive Order No 9066, 7 Fed Reg 1407 (1942).
• 108See Youngstown Sheet & Tube Co v Sawyer, 343 US 579, 585 (1952).
• 109See id.
• 110See Phillip J. Cooper, By Order of the President: The Use and Abuse of Executive Direct Action 21 (Kansas 2002) (“Among the reasons executive orders are so difficult to master is that they are used in a variety of ways and for a plethora of reasons.”).
• 111See note 28 and accompanying text.
• 112See Kathryn A. Watts, Rulemaking as Legislating, 103 Georgetown L J 1003, 1015 (2015) (explaining that legislative rules “carry the force and effect of law”).
• 113See generally Franklin D. Roosevelt, Proclamation No 2087, Sale of Arms and Munitions of War to Bolivia and Paraguay, 48 Stat 1744 (May 28, 1934).
• 114See United States v Curtiss-Wright Export Corp, 299 US 304, 330–33 (1936) (upholding the legality of Congress’s delegation of power to the President to issue a proclamation that had the effect of criminalizing the sale of arms to Bolivia and Paraguay).
• 115See Executive Order No 11615, 36 Fed Reg 15727 (Aug 15, 1971). This order also instructed members of the public who were engaged in the business of selling or providing commodities or services to maintain records of their prices and rents for public inspection. See Amalgamated Meat Cutters & Butcher Workmen of North America, AFL-CIO v
  Connally, 337 F Supp 737, 763–64 (DDC 1971) (rejecting a challenge to the lawfulness of this executive order).
• 116See Executive Order No 11615, 36 Fed Reg at 15729 (cited in note 115).
• 117See Kevin M. Stack, The Statutory President, 90 Iowa L Rev 539, 548 (2005) (explaining that executive orders may “have the force and effect of law” if supported by “appropriate constitutional or statutory authorization”).
• 118See Executive Order No 13658, 79 Fed Reg 9851 (Feb 20, 2014).
• 119Indeed, Obama’s order expressly stated: “This order is not intended to, and does not, create any right or benefit, substantive or procedural, enforceable at law or in equity by any party against the United States, its departments, agencies, or entities, its officers, employees, or agents, or any other person.” Id at § 7(c).
• 120See notes 221–26 and accompanying text.
• 121See id.
• 122As an example, consider Trump’s Executive Order No 13769, Protecting the Nation from Foreign Terrorist Entry into the United States, 82 Fed Reg 8977 (2017), which included many discrete instructions, some legally binding and some not. Executive Order 13769’s legally binding orders included, for example, the instruction contained in § 5(c) in which the President suspended the entry of nationals of Syria as refugees. Its nonlegally binding orders included, for example, the instruction contained in § 3(a) in which the President instructed the Secretary of Homeland Security to take particular actions.
• 123See Manheim and Watts, The Limits of Presidential Power at 23–26 (cited in note 3) (describing the president’s removal power).
• 124See Kagan, 114 Harv L Rev at 2277 (cited in note 8), quoting Commission on Law and the Economy, Federal Regulation: Roads to Reform 73 (American Bar Association 1978).
• 125See Executive Order No 12291, 46 Fed Reg 13193 at § 3(c) (cited in note 9). See also Kagan, 114 Harv L Rev 2245 at 2277–78 (cited in note 8).
• 126See Peter L. Strauss, Todd D. Rakoff, Cynthia R. Farina, and Gillian E. Metzger, Gellhorn and Byse’s Administrative Law: Cases and Comments 211 (Foundation 11th ed 2011) (discussing the Office of Management and Budget).
• 127See Executive Order No 12291, 46 Fed Reg 13193 at § 3(c).
• 128Id at § 2(b).
• 129Id at § 9. See, for example, Michigan v Thomas, 805 F2d 176, 187 (6th Cir 1986) (“Given this clear and unequivocal intent that agency compliance with Executive Order 12,291 not be subject to judicial review, we hold that the Order provides no basis for rejecting the EPA’s final action.”).
• 130See generally, for example, Morton Rosenberg, Presidential Control of Agency Rulemaking: An Analysis of Constitutional Issues That May Be Raised by Executive Order 12,291, 23 Ariz L Rev 1199 (1981). See also Kagan, 114 Harv L Rev at 2279 (cited in note 8) (noting that Reagan’s oversight program “provoked sharp criticism, most of which related to perceptions of the scheme’s anti-regulatory bias”).
• 131See Executive Order No 12866, Regulatory Planning and Review, 58 Fed Reg 51735 (1994), reprinted in 5 USC § 601 app at 76 (1993).
• 132Kagan, 114 Harv L Rev at 2285 (cited in note 8). See also Watts, 114 Mich L Rev at 690 (cited in note 8), quoting Kagan, 114 Harv L Rev at 2285 (cited in note 8).
• 133See Executive Order No 12866, 58 Fed Reg at § 7 (cited in note 131).
• 134Kagan, 114 Harv L Rev at 2299 (cited in note 8).
• 135Id.
• 136See Watts, 114 Mich L Rev at 700–03 (cited in note 8).
• 137See Presidential Memorandum of March 13, 2014, 79 Fed Reg at 18737 (cited in note 11) (directing the Secretary of Labor “to propose revisions to modernize and streamline the existing overtime regulations”).
• 13852 Stat 1060 (1938), codified as amended at 29 USC § 201 et seq.
• 13929 USC § 207.
• 140See Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 80 Fed Reg 38516 (2015). See also Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 81 Fed Reg 32391, 32392 (2016) (“On March 13, 2014, President Obama signed a Presidential Memorandum directing the Department to update the regulations defining which white collar workers are protected by the FLSA’s minimum wage and overtime standards.”). Tellingly, at the outset of its rulemaking proceeding, DOL posted a whiteboard video to its blog featuring a sketch of Obama directing the agency to “update the rules!” See Heidi Shierholz, Everything You Need to Know About Overtime Pay (US Department of Labor Blog, July 8, 2015) (Wayback Machine 2019), archived at http://perma.cc/3KQD-AF6Q.
• 141See Part III.A (describing the flood of litigation aimed at Trump’s executive
  orders).
• 142See Part II.B.2.
• 143This conclusion is supported by research we conducted through multiple means, including through investigation of judicial decisions, case filings, and the existing academic literature. For illustrative scholarly works, see, for example, Newland, Note, 124 Yale L J at 2035 (cited in note 20); Mayer, With the Stroke of a Pen at 4–5 (cited in note 105). See also generally Graham G. Dodds, Take Up Your Pen: Unilateral Presidential Directives in American Politics (Pennsylvania 2013) (detailing the use by presidents of unilateral directives and the judicial and congressional permissiveness of such directives in the course of American history). The data presented in these academic works help to confirm what our own case-law-based research more directly revealed: historically, a very limited number of direct challenges to presidential orders (and almost no direct challenges to nonbinding legal orders), followed by a stark change in this pattern accompanying Trump’s entrance into office. See Part II.B.1; Part III.A. Our own research reflects not only targeted searches in databases such as Westlaw, but also a review of the citations and analysis contained in judicial decisions—as well as litigants’ filings—in recent cases. While a more quantitative study of these questions (including the number and nature of challenges to presidential orders over time) would be of value, it would be difficult to design given the complexity of the issues, the shifting use of forms and terms, and the absence of a robust set of studies already addressing these issues. See, for example, Newland, Note, 124 Yale L J at 2033 (cited in note 20) (“[Scholars] have not sought to define or divine a case law of executive orders.”); Mayer, With the Stroke of a Pen at 66 (cited in note 105) (describing how, historically, “it was often unclear which presidential actions, exactly, constituted an executive order”). See also id at 66–70 (describing related complications). The absence of quantitative data going to these questions is yet another indication that this area of the law is undertheorized.
• 144One measure of this expectation is the practice of including severability clauses in executive orders, which appears to be on the rise. Given that these clauses apply only if a court rules an order to be unconstitutional, an inclusion of a severability clause indicates an expectation by the President that an order may be challenged in court. According to one study, the use of severability clauses historically has been rare, with Clinton issuing only three orders with a severability clause, President George W. Bush issuing one, and Obama issuing ten. See Gregory Korte, Trump Tries Little-Known Legal Tactic to Protect Controversial Executive Orders from the Courts (USA Today, Dec 5, 2017), archived at http://perma.cc/7STD-QYRS. Now, by contrast, their inclusion appears to be routine. Id.
• 145See Part II.B.2.
• 146See generally, for example, Harold H. Bruff, Judicial Review and the President’s Statutory Powers, 68 Va L Rev 1 (1982) (noting that challenges to presidential orders that directly affected the public were increasing, and chronicling legal challenges against Carter-era orders).
• 147See Curtiss-Wright Export Corp, 299 US at 333.
• 148299 US 304, 333 (1936).
• 149See generally Mountain States Legal Foundation v Bush, 306 F3d 1132 (DC Cir 2002); Tulare County v Bush, 306 F3d 1138 (DC Cir 2002).
• 150See Proclamation No 7625, Establishment of the Grand Canyon-Parashant National Monument, 65 Fed Reg 2825 (2000).
• 151See note 150 and accompanying text. In the Trump era, legally binding orders include some of those contained in Executive Order No 13769, 82 Fed Reg (cited in note 122), Executive Order No 13780, Protecting the Nation from Foreign Terrorist Entry into the United States, 82 Fed Reg 13209 (2017), and Presidential Proclamation No 9645, Enhancing Vetting Capabilities and Processes for Detecting Attempted Entry into the United States by Terrorists or Other Public-Safety Threats, 82 Fed Reg 45161 (2017) (President Trump’s three “travel ban” orders), as well as Presidential Proclamations No 9681, Modifying the Bears Ears National Monument, 82 Fed Reg 58081, 58085 (2017) and No 9682, Modifying the Grand Staircase-Escalante National Monument, 82 Fed Reg 58089, 58093 (2017), which immediately “modified and reduced” the boundaries of two national monuments.
• 152See, for example, Bruff, 68 Va L Rev at 2 n 7 (cited in note 146) (noting that many presidential orders are simply “procedural directives to the bureaucracy”).
• 153See notes 9–10 and accompanying text.
• 154See Natural Resources Defense Council, Inc v Environmental Protection Agency, 683 F2d 752, 757 (3d Cir 1982) (discussing the EPA’s indefinite postponement of the effective date of amendments to certain regulations).
• 155See id at 754–56.
• 156Part of the plaintiff’s argument was that, as a legal matter, the directives of the Executive Order could not overcome the directives of the relevant statute (here, the APA). The court concluded that it was not necessary to resolve this indirect challenge to the Executive Order because, in this circumstance, the agency could have complied with both the statute and the President’s order. See id at 765 n 26.
• 157See Building and Construction Trades Department, AFL-CIO v Allbaugh, 295 F3d 28, 29 (DC Cir 2002).
• 158Id at 30.
• 159See Part I.B.
• 160See, for example, Stack, 62 Vand L Rev at 1176–77 (cited in note 23) (“Following the enactment of the APA, there had been little doctrinal pressure for evolution of [ ] preexisting doctrines, and so they persisted, apparently awaiting renewed application. But by resorting to these prior doctrines, courts fail to acknowledge the legal developments that undermine their continued application.”).
• 161See Part I.B (discussing challenges to agency action).
• 162Bruff, 68 Va L Rev at 2 (cited in note 146).
• 163Id at 2 n 7.
• 164Id.
• 165Counteracting this trend, of course, has been the increasing forcefulness with which presidents have attempted to control decision-making by the agencies themselves—a phenomenon that may, in turn, help to explain the more recent Trump-era spike in challenges to executive orders. See notes 227–47 and accompanying text. See also Part I.A.
• 166See Part I.A.
• 167Dames & Moore v Regan, 453 US 654, 661 (1981).
• 168Newland, Note, 124 Yale L J at 2035 (cited in note 20). Some of the opinions Newland reviewed involved challenges to the lawfulness of the executive orders, and therefore they addressed lawsuits falling within the purview of this Article. Other opinions she reviewed involved attempts to enforce executive orders against the president’s subordinates or other issues falling outside this Article’s purview.
• 169Id.
• 170Id.
• 171Id. See also Dodds, Take Up Your Pen at 55 (cited in note 143) (“The question of the constitutionality of unilateral presidential directives has not been a narrowly focused area of jurisprudence. Instead, it generally appears in the course of a variety of other areas of substantive concern, periodically cropping up at the margins and only rarely getting the direct attention of courts.”).
• 172But see note 168 (noting that many of the cases contained in Newland’s dataset, while involving executive orders, do not involve challenges to their lawfulness). Professor Graham Dodds also has published a helpful survey of early case law addressing the lawfulness of presidential directives; tellingly, it is quite sparse. See Dodds, Take Up Your Pen at 54–85 (cited in note 143) (describing these older precedents as falling into four main categories based on subject matter: maritime, habeas corpus, the Civil War, and public lands).
• 173See Newland, Note, 124 Yale L J at 2083 (cited in note 20) (noting that “jurisprudence of executive orders may not derive from any coherent doctrine of presidential exceptionalism but instead from an under-theorized understanding of the role of executive orders”). See also Bruff, 68 Va L Rev at 2 (cited in note 146) (noting in 1982 that “there exists no generally accepted method for judicial review” of presidential actions).
• 174343 US 579 (1952).
• 175See id at 585–89.
• 176See id at 634–55 (Jackson concurring).
• 177Although the challenge presumably could have proceeded through the cause of action provided by the APA, which had been enacted less than a decade earlier, the companies did not cite this statute in their complaints. The availability of causes of action in presidential-order challenges is discussed at greater length below. See Part IV.B.1.b.
• 178The treatment of severability in presidential-order challenges is discussed at greater length below. See Part IV.B.3.b.
• 179The treatment of some of these issues as they affect presidential-order challenges is discussed at greater length below. See Part IV.B.2.
• 180453 US 654 (1981).
• 181Id at 662–66.
• 182Id at 686–88.
• 183See id at 660–61 (“We attempt to lay down no general ‘guidelines’ covering other situations not involved here, and attempt to confine the opinion only to the very questions necessary to decision of the case.”).
• 184323 US 214 (1944).
• 185See id at 215–16.
• 186This was the central dispute requiring resolution, for example, in Youngstown, in which the Court rejected efforts to cite the Constitution as the source of the President’s authority in ordering the seizure of privately owned steel mills. See Youngstown, 343 US at 587–89.
• 187For example, at issue in Cornelius v NAACP Legal Defense and Educational Fund, Inc, 473 US 788, 790 (1985), was Executive Order No 12353, which limited participation in a charity drive aimed at federal employees. Plaintiffs-respondents challenged the order as violating the First Amendment, a challenge that the Supreme Court rejected on the merits. Id at 803–13.
• 188An important but complicated illustration of such a case is Dames & Moore, in which the Court alluded to Congress’s role in authorizing the challenged orders but deliberately remained vague as to the precise source of congressional authority. A more straightforward illustration of a case in this category is Matter of Reyes, 910 F2d 611, 612 (9th Cir 1990), in which the plaintiff argued, successfully, that a particular statute failed to give the President the power to issue a contested executive order.
• 189It was in response to such a challenge, for example, that the Supreme Court invalidated multiple executive orders in Panama Refining Co v Ryan, 293 US 388 (1935). See id at 430 (concluding that the authorizing statute constituted an unconstitutional delegation of legislative power). In a case decided nearly seventy years later, the Ninth Circuit considered but rejected a similar claim, as it refused to rely on a non-delegation theory to invalidate a presidential proclamation supported by the federal Antiquities Act. See Mountain States Legal Foundation, 306 F3d at 1136–38. See also, for example, A. L. A. Schechter Poultry Corp v United States, 295 US 495, 518 (1935); Amalgamated Meat Cutters, 337 F Supp at 763.
• 190505 US 788 (1992).
• 191See id at 796.
• 192See 5 USC §§ 702, 706.
• 193The APA also allows these challenges to be brought against an agency or the United States. See 5 USC § 703.
• 194See, for example, East Bay Sanctuary Covenant v Trump, 909 F3d 1219, 1246–47 (9th Cir 2018) (concluding that, under the APA, the court could review a presidential proclamation to the extent that it was incorporated by the agency rule under attack). In these sorts of lawsuits, however, disputes can arise over whether the “agency” action being challenged is more properly characterized as “presidential” action that is immune from review under the APA. Compare Sisseton-Wahpeton Ovate v United States Department of State, 659 F Supp 2d 1071, 1082 (D SD 2009) (“The court finds that the actions taken pursuant to [an executive order involving transboundary pipeline permits] are presidential in nature, and therefore, do not confer upon the plaintiffs a private right of action under the APA.”), with Indigenous Environmental Network v United States Department of State, 2017 WL 5632435, *6 (D Mont) (concluding that action taken by the State Department involving a transboundary presidential pipeline permit could be characterized as “agency action” that was reviewable under the APA).
• 195Professor Kevin Stack refers to these lawsuits as “APA hybrid[s].” Stack, 62 Vand L Rev at 1194 (cited in note 23). Still, even a challenge to a presidential order brought against a subordinate will not always qualify as an APA hybrid. This is because the APA only applies to claims seeking relief in response to “final” agency action. 5 USC § 704. As such, a litigant cannot attack an executive order pursuant to the APA, even indirectly, unless that executive order is sufficiently connected to some final agency action.
• 196526 US 172 (1999).
• 197Id at 191.
• 198Id.
• 199Id at 195.
• 20074 F3d 1322 (DC Cir 1996).
• 201The treatment of ripeness in presidential-order challenges is discussed at greater length below. See Part IV.B.1.a.
• 202See Reich, 74 F3d at 1325.
• 203Id at 1339.
• 204See id at 1327.
• 205See id at 1331.
• 206See, for example, Reich, 74 F3d at 1335–36. See also id at 1325. In passing, the court also discussed its reluctance to “question the President’s motivation” in analyzing the order. Id at 1336.
• 207See id at 1339. See also Chamber of Commerce v Reich, 897 F Supp 570, 577–78 (DDC 1995). See Part I.C (discussing Chevron deference).
• 208870 F2d 723 (DC Cir 1989).
• 209See id at 725.
• 210See id at 726.
• 211See id at 727.
• 212American Federation, 870 F2d at 727–28. In the wake of recent challenges to Trump-era orders, some have called for the presumption of regularity to be revisited. See, for example, Leah Litman, Revisiting the Presumption of Regularity (Take Care, Jan 28, 2019), archived at http://perma.cc/E6EM-RWGB.
• 213See Stack, 62 Vand L Rev at 1176–77 (cited in note 23) (confirming that “[f]ollowing the enactment of the APA, there had been little doctrinal pressure for evolution of [ ] preexisting doctrines” that apply when the APA does not).
• 214See notes 216–20 and accompanying text.
• 215See id. See also Part III.A.
• 216See, for example, Matt Viser, Trump Has Been Sued 134 Times in Federal Court since Inauguration (Boston Globe, May 5, 2017) (asserting that, approximately three months into his term, “Trump has been sued 134 times in federal court . . . , nearly three times the number of his three predecessors in their early months combined”). See also Glenn Fleishman, The People vs. Donald Trump: Every Major Lawsuit and Investigation the President Faces (Fortune, Dec 12, 2018) archived at http://perma.cc/58EG-NB63 (“The president has an unprecedented number of legal entanglements compared to even the most challenged previous president.”). Though these sources do not distinguish between lawsuits challenging presidential orders versus other claims, these sources help to confirm the unprecedented number of lawsuits lodged against Trump. It is difficult to compare these numbers on a more granular level over time (particularly as they relate specifically to challenges to presidential orders), for some of the reasons we have identified above, see note 143, though such an analysis would be of value.
• 217As discussed below, Trump had at least ten orders challenged within his first year in office. Many of these orders spawned a multitude of separate lawsuits.
• 218As discussed above, if a litigant brings an APA hybrid claim (in which the litigant challenges an executive order indirectly through a direct challenge to final agency action), the APA provides, among other things, a cause of action and a waiver of sovereign immunity. See note 195. It is in this sense that final agency action can provide a procedural hook for executive-order challenges.
• 219See Part II.B (describing the legal landscape prior to Trump’s inauguration).
• 220The apparent discrepancy in dates exists because the plaintiffs in this case initially did not even wait for the President to issue a formal order. Rather, the plaintiffs sued to challenge the policy apparently announced by the President’s tweets sent on July 26, 2017. These plaintiffs then filed an amended complaint, adding a challenge to the 8/25/2017 Presidential Memorandum, on August 31, 2017.
• 221Executive Order No 13768, Enhancing Public Safety in the Interior of the United States, 82 Fed Reg 8799 (2017).
• 222See, for example, County of Santa Clara v Trump, 250 F Supp 3d 497, 507–08 (ND Cal 2017).
• 223Id at 514, 517.
• 224Id at 517.
• 225City and County of San Francisco v Trump, 897 F3d 1225, 1238 (9th Cir 2018).
• 226See, for example, Complaint for Declaratory and Injunctive Relief, Washington v United States, No 2:18-cv-00939, ¶¶ 72–74 (WD Wash filed June 26, 2018) (available on Westlaw at 2018 WL 3139446) (arguing in a complaint challenging the Trump administration’s practice of separating families along the border that a presidential order issued by Trump did not actually “require an end to family separation” and that by its own terms, the order stated that it “[did] not confer any enforceable right or benefit on any person”).
• 227See Part II.A (detailing this trend over time).
• 228See Executive Order No 13771, 82 Fed Reg at 9339 (cited in note 14). See also Public Citizen, Inc v Trump, 297 F Supp 3d 6, 12 (DDC 2018).
• 229See Executive Order No 13771, 82 Fed Reg at 9339 (cited in note 14).
• 230Trump’s comments included the following:

  If you have a regulation you want, No. 1, we’re not gonna approve it because it’s already been approved probably in 17 different forms. . . . But if we do, the only way you have a chance is we have to knock out two regulations for every new regulation.

  Nolan D. McCaskill and Matthew Nussbaum, Trump Signs Executive Order Requiring that for Every One New Regulation, Two Must Be Revoked (Politico, Jan 30, 2017), archived at http://perma.cc/R2QC-DXY2.

• 231See, for example, Executive Order No 13768, 82 Fed Reg (cited in note 221) (directing federal officials, including the Secretary of Homeland Security and the Attorney General, to implement Trump’s desired policies); Donald J. Trump, Presidential Memorandum for the Secretary of Defense and the Secretary of Homeland Security (White House, Feb 20, 2018), archived at http://perma.cc/Y7QP-UNZC (“Today, I am directing the Department of Justice to dedicate all available resources to . . . , as expeditiously as possible, [ ] propose for notice and comment a rule banning all devices that turn legal weapons into machineguns.”).
• 232See, for example, note 1 and accompanying text.
• 233See notes 75–83 and accompanying text.
• 234See notes 79–83 and accompanying text.
• 235See notes 75–78 and accompanying text.
• 236See Brief for Petitioners, United States v Texas, No 15-674, *31–33 (US filed Mar 1, 2016).
• 237See notes 81–83 and accompanying text.
• 238See Civil Rights Litigation Clearinghouse, Civil Rights Challenges to Trump Refugee/Visa Order (University of Michigan Law School), archived at http://perma.cc/CN5B-XZMM (collecting cases).
• 239Two orders triggering legal challenges were Executive Order No 13768, 82 Fed Reg (cited in note 221), which included directives aimed at “sanctuary” jurisdictions, and Executive Order No 13771, 82 Fed Reg (cited in note 14), which often is referred to as the President’s “one-in, two-out” order.
• 240Although the Supreme Court eventually rejected the litigants’ challenges in Trump v Hawaii, 138 S Ct 2392 (2018), this later line of litigation addressed Trump’s third travel ban, not the earlier and broader ban that the State of Washington already had successfully enjoined.
• 241See Washington v Trump, 2017 WL 462040, *3 (WD Wash).
• 242Trump ended up revoking the order less than two months after it was issued. He did so on March 6, 2017, when he replaced Executive Order No 13769, 82 Fed Reg (cited in note 122), with Executive Order No 13780, 82 Fed Reg (cited in note 151). The latter constituted Trump’s second “travel ban,” and it too was successfully enjoined in court—notwithstanding a narrowing of that injunction by the Supreme Court, see Trump v International Refugee Assistance Project (IRAP), 137 S Ct 2080, 2088 (2017)—until the order expired by its own terms. See also note 240 (describing the fate of Trump’s third travel ban).
• 243See Table 1. These plaintiffs filed suit even before the success of Washington State’s challenge had become clear.
• 244Both the City of San Francisco and the County of Santa Clara later amended their complaints to challenge subsequent agency action. However, as discussed above, the inclusion of subsequent agency action does not eliminate the range of unsettled legal issues presented by a challenge to an executive order.
• 245See Table 1.
• 246Donald J. Trump (@realDonaldTrump), (Twitter, July 26, 2017), archived at http://perma.cc/9GUG-84R6; Donald J. Trump (@realDonaldTrump), (Twitter, July 26, 2017), archived at http://perma.cc/6UTW-T5YZ; Donald J. Trump (@realDonaldTrump), (Twitter, July 26, 2017), archived at http://perma.cc/5HBU-WKVU.
• 247See Table 1. In their complaint, the plaintiffs seemed at least implicitly to acknowledge the extraordinary nature of this claim by trying to identify some other directive also to challenge. They alleged, for example, that “[u]pon information and belief, the White House turned that decision into official guidance, approved by the White House counsel’s office, to be communicated to the Department of Defense.” Complaint for Declaratory and Injunctive Relief, Doe v Trump, No 17-cv-01597, ¶ 5 (DDC filed Aug 9, 2017) (available on Westlaw at 2017 WL 3421198).
• 248See Executive Order No 13769, 82 Fed Reg (cited in note 122). Another order suffered a similar fate. See Executive Order No 13780, 82 Fed Reg (cited in note 151). With respect to this latter order, the Supreme Court, without deciding the case on the merits, narrowed the injunction imposed by the lower courts. See IRAP, 137 S Ct at 2088.
• 249See Trump v Hawaii, 138 S Ct at 2423. See also, for example, Donald J. Trump, Presidential Memorandum for the Secretary of Defense and the Secretary of Homeland Security (White House, Aug 25, 2017), archived at http://perma.cc/WCQ6-DZV9 (purporting to require a change in military policy relating to transgender troops on a specific timetable that, due to court intervention, was not met).
• 250See Part IV.B.3.
• 251See Part II.B.2.
• 252See Part I.B.
• 253See, for example, 5 USC § 553(c) (allowing interested persons an opportunity to participate in rulemakings via the submission of “written data, views, or arguments”).
• 254See 5 USC § 706.
• 255See Leland E. Beck, Agency Practices and Judicial Review of Administrative Records in Informal Rulemaking *2 (Administrative Conference of the United States, May 14, 2013), archived at http://perma.cc/U5PA-TFXC (explaining that the “whole record” is now generally referred to as the “administrative record” and noting that the APA “provides little guidance on the creation and compilation of the ‘whole record’ or the ‘administrative record’ as it has come to be known”).
• 256See, for example, Thompson v United States Department of Labor, 885 F2d 551, 555 (9th Cir 1989) (“The ‘whole’ administrative record [ ] consists of all documents and materials directly or indirectly considered by agency decision-makers and includes evidence contrary to the agency’s position.”) (emphasis omitted), quoting Exxon Corp v Department of Energy, 91 FRD 26, 33 (ND Tex 1981); Tafas v Dudas, 530 F Supp 2d 786, 793–94 (ED Va 2008) (“The whole administrative record includes pertinent but unfavor-able information, and an agency may not exclude information on the ground that it did not ‘rely’ on that information in its final decision.”).
• 257See Cooper, By Order of the President at 17 (cited in note 110) (“There is no requirement for notice and public participation.”); Stack, 90 Iowa L Rev at 552–53 (cited in note 117) (“In contrast to legislation or agency regulation, there are almost no legally enforceable procedural requirements that the president must satisfy before issuing (or repealing) an executive order or other presidential directive. That, no doubt, is central to their appeal to presidents.”). But see Bruff, 68 Va L Rev at 14 (cited in note 146) (noting that, as a matter of historical practice, “[f]ew impending decisions reach the White House without previous, often extensive, analysis in one or more of the executive agencies,” and that as a result, “when the time comes for the President to exercise his discretion, an administrative record exists in the bureaucracy—a mass of raw data, analysis, and opinion from both within and without the government”).
• 258See 44 USC § 1505. Given Congress’s failure to impose meaningful procedural constraints on the president’s issuance of orders, a related question remains unresolved: Namely, to what extent does the Constitution limit Congress’s ability to impose such constraints?
• 259See notes 190–95 and accompanying text.
• 260See Executive Order No 11030, Preparation, Presentation, Filing, and Publication of Executive Orders and Proclamations, 27 Fed Reg 5847, 5847 (1962).
• 261See generally W. Neil Eggleston and Amanda Elbogen, The Trump Administration and the Breakdown of Intra-Executive Legal Process, 127 Yale L J F 825 (2018).
• 262See id at 829–31.
• 263See, for example, International Refugee Assistance Project (IRAP) v Trump, 857 F3d 554, 597 (4th Cir 2017), vacd and remd, 138 S Ct 353 (2017) (rejecting the government’s argument that the court should not consider “extrinsic evidence” or otherwise “look beyond [an executive order’s] text and operation” in determining its constitutionality) (quotation marks omitted). See also id at 592–600 (considering Trump’s campaign statements and other assertions in determining whether order was motivated by an impermissible purpose).
• 264See generally, Eggleston and Elbogen, 127 Yale L J F (cited in note 261).
• 265The lack of procedural formality that accompanies presidential orders could impact the relevant standards of review. See Part IV.B.2 (discussing different standards of review that might guide judicial review of executive orders and discussing the relevance of procedural checks to this analysis). The lack of procedural trappings surrounding the issuance of presidential orders also might convince a court to do away with the presumption of regularity that normally attaches to executive branch action. See note 212 (discussing the presumption of regularity). See also, for example, The Presumption of Regularity in the Executive Branch, 131 Harv L Rev 2431, 2452 (2018) (concluding that the presumption of regularity, which often attaches to executive branch action, rests in large part on an assumption of procedural regularity and fairness).
• 266See notes 68–74 and accompanying text.
• 267See id. See also Thomas W. Merrill and Kathryn Tongue Watts, Agency Rules with the Force of Law: The Original Convention, 116 Harv L Rev 467, 476–77 (2002).
• 268See Watts, 103 Georgetown L J at 1015 (cited in note 112).
• 269See note 113 and accompanying text.
• 270See note 123 and accompanying text.
• 271See notes 125–30 and accompanying text (discussing Executive Order No 12291). Another example is Executive Order No 13765, which Trump issued in an effort to minimize the burdens of Obamacare. See Executive Order No 13765, Minimizing the Economic Burden of the Patient Protection and Affordable Care Act Pending Repeal, 82 Fed Reg 8351 (2017).
• 272See notes 68–74 and accompanying text.
• 273Id.
• 274These nonbinding agency policy statements are generally not reviewable for multiple reasons. One stems from the doctrine initiated in Heckler v Chaney, 470 US 821 (1985), which held that the APA precludes review “if the statute [governing the agency’s actions] is drawn so that a court would have no meaningful standard against which to judge the agency’s exercise of discretion.” Id at 830. Another is because these nonbinding agency policy statements often lack legal consequences and therefore do not qualify as “final” agency action pursuant to 5 USC § 704.
• 275See Part IV.B (taking first steps in conducting this analysis).
• 276See Part I.A.
• 277See note 65 and accompanying text. See also Kathryn A. Watts, Constraining Certiorari Using Administrative Law Principles, 160 U Pa L Rev 1, 23 (2011) (noting that “many administrative law scholars have argued that Congress’s decisions to legislate only in very broad strokes and to leave the details of policymaking up to agencies threaten principles of accountability, transparency, and rationality”); Bruff, 68 Va L Rev at 22 (cited in note 146) (noting that concern about the limited accountability of agencies has manifested itself in various ways, such as in “Congress’s imposition of procedural constraints on agencies in order to ensure that the public can participate in and influence agency decisions”).
• 278See Bruff, 68 Va L Rev at 22 (cited in note 146) (noting that the president, unlike agency heads, is directly accountable “as the only elected official with a national constituency”).
• 279See, for example, Antonin Scalia, A Matter of Interpretation: Federal Courts and the Law 13–14 (Princeton 1997) (“By far the greatest part of what . . . all federal judges do is to interpret the meaning of federal statutes and federal agency regulations.”).
• 280See, for example, Bruff, 68 Va L Rev at 35 (cited in note 146) (describing Dames & Moore as a “a major contribution to separation of powers analysis because it enables courts to avoid rendering direct definitions of the President’s ‘inherent’ or implied powers when it is not necessary to do so”).
• 281Franklin, 505 US at 800–01 (“Out of respect for the separation of powers and the unique constitutional position of the President, we find that textual silence is not enough to subject the President to the provisions of the APA.”).
• 282See notes 190–95 and accompanying text.
• 283See Part IV.B (discussing how courts might modify preexisting legal doctrines to better fit executive-order challenges).
• 284See Part IV.B.1.b (discussing causes of action).
• 285Among the many implicated topics we do not address in this paper are those involving mootness, sovereign immunity, the doctrine of constitutional avoidance, and whether an Auer-type deference should apply to an interpretation of an executive order advanced by the president or some other executive branch official. See Auer v Robbins, 519 US 452, 456–59 (1997); Kisor v Wilkie, 139 S Ct 2400, 2418 (2019). See also City and County of San Francisco v Trump, 897 F3d 1225, 1241–43 (9th Cir 2018) (discussing the degree to which an agency memorandum purporting to implement an executive order should be entitled to deference); Amici Curiae Brief of Administrative Law Professors in Support of Plaintiffs-Appellees and Affirmance, City and County of San Francisco v Trump, No 17-17478, *17–20 (9th Cir filed Feb 12, 2018) (same); Udall v Tallman, 380 US 1, 4 (1965) (deferring to the Secretary of the Interior’s “reasonable” construction of an executive order); Chou, 71 Admin L Rev (cited in note 20) (discussing the question of judicial deference to agency interpretations of executive orders). Each of these topics, among others, warrants further analysis.
• 286See Richard J. Pierce, 2 Administrative Law Treatise § 15.1 at 965–67 (Aspen 4th ed 2002). See also id § 15.12 at 1052.
• 287814 F2d 731 (DC Cir 1987).
• 288Id at 732.
• 289Pierce, 2 Administrative Law Treatise § 15.17 at 1105 (cited in note 286).
• 290See id §§ 15.1–15.17 at 966–1106 (cited in note 286) (describing, in the administrative law context, precedents tackling questions of ripeness, exhaustion, and finality); Pierce, 3 Administrative Law Treatise §§ 16.1–16.15 at 1107–1225 (cited in note 286).
• 291See note 290.
• 292See id § 15.11 at 1036 (discussing finality requirement of APA § 704).
• 293See, for example, notes 190–91 and accompanying text (discussing Franklin).
• 294See Pierce, 2 Administrative Law Treatise § 15.2 at 967–82 (cited in note 286) (discussing common law duty to exhaust); id § 15.3 at 982–1000 (discussing statutorily imposed duty to exhaust).
• 295See Town of Chester, New York v Laroe Estates, Inc, 137 S Ct 1645, 1651 (2017) (“For all relief sought, there must be a litigant with standing.”). See also Pierce, 2 Administrative Law Treatise § 15.12 at 1052 (cited in note 286) (“At its core, the ripeness requirement is based on the jurisdictional limit in Article III.”).
• 296See, for example, National Treasury Employees Union v Reagan, 685 F Supp 1346, 1353 (ED La 1988) (finding challenge to executive order was not ripe for review when the Executive Order “simply sets guidelines and outer boundaries for agency heads to follow in determining which positions are so sensitive as to warrant [drug] testing” and when specific agencies still had to finalize the particulars of their own plans).
• 297See Executive Order No 13768, 82 Fed Reg (cited in note 221).
• 298See County of Santa Clara v Trump, 267 F Supp 3d 1201, 1217 (ND Cal 2017).
• 299See id at 1217–18.
• 300See, for example, id at 1207–08.
• 301Id at 1207.
• 302County of Santa Clara, 267 F Supp 3d at 1216–17.
• 303City and County of San Francisco, 897 F3d at 1243–45.
• 304See notes 112–19 and accompanying text.
• 305See note 122 and accompanying text (describing how both legally binding orders and nonlegally binding orders can appear in the same document).
• 306See notes 112–19 and accompanying text.
• 307See note 74 and accompanying text.
• 308American Federation of Government Employees, AFL-CIO v Trump, 318 F Supp 3d 370, 437–39 (DDC 2018).
• 309Id at 437–38, citing Sierra Club v Environmental Protection Agency, 873 F3d 946, 951 (DC Cir 2017) (“Policy statements are binding on neither the public nor the agency, and the agency retains the discretion and the authority to change its position.”) (quotation marks omitted).
• 310Sierra Club, 873 F3d at 437–38.
• 311See notes 221–27 and accompanying text.
• 312As we discussed above, the need to make these distinctions, in the context of agency action, often relates to whether an agency’s action has reached the point of being judicially reviewable. This is because legislative rules generally are understood to be final for purposes of judicial review, whereas nonbinding agency policy statements frequently are not. See notes 72–74 and accompanying text. See also Broadgate Inc v United States Citizenship & Immigration Services, 730 F Supp 2d 240, 243 (DDC 2010) (“Legislative or substantive rules are, by definition, final agency action, while interpretive rules and general policy statements are not.”).
• 313208 F3d 1015 (DC Cir 2000).
• 314See id at 1020–21.
• 315See id at 1020–23.
• 316Id at 1023. See also, for example, McLouth Steel Products Corp v Thomas, 838 F2d 1317, 1324–25 (DC Cir 1988) (concluding that the agency had treated its “policy statement” as binding, even though the agency claimed it was nonbinding); Broadgate Inc, 730 F Supp 2d at 245–47 (rejecting challengers’ attempt to characterize agency memorandum as a legislative rule, based on its allegedly binding effect, for purposes of obtaining judicial review).
• 317See, for example, Davis v Passman, 442 US 228, 236–41 (1979).
• 318See 5 USC §§ 702, 704.
• 319See 5 USC § 703. See also, for example, Leuthe v Office of Financial Institution Adjudication, 977 F Supp 357, 361 (ED Pa 1997).
• 320See, for example, 29 USC § 160(f) (providing a cause of action for litigants seeking review of a final order of the National Labor Relations Board).
• 321See 5 USC § 702. See also Franklin, 505 US at 800–01.
• 3225 USC § 704.
• 323There nevertheless are circumstances in which a plaintiff challenging a presidential order can rely on the APA to provide a cause of action. Such an opportunity might present itself, for example, if a plaintiff seeks to challenge an agency’s final action on the grounds that the presidential order purporting to authorize that agency action was itself unlawful. Even in this situation, however, the APA still would not “establish the standard or availability of judicial review for the President’s claim of statutory power.” Stack, 62 Vand L Rev at 1213, 1294 (cited in note 23) (discussing what he terms “APA hybrid” lawsuits).
• 324Compare Bivens v Six Unknown Named Agents, 403 US 388 (1971), with Ziglar v Abbasi, 137 S Ct 1843 (2017). See also generally Alexander v Sandoval, 532 US 275 (2001) (rejecting a lenient approach to courts’ recognition of implied rights of action in separate context); David Sloss, Constitutional Remedies for Statutory Violations, 89 Iowa L Rev 355, 440 (2004) (exploring tensions in the Supreme Court’s implied right of action jurisprudence in the preemption context, including as it relates to the Sandoval line of cases); Armstrong v Exceptional Child Center, 135 S Ct 1378 (2015).
• 325The unsettled nature of this authority is confirmed by the absence of a well-settled term to capture it. Referring to it as the courts’ “inherent equitable powers” most closely tracks the Supreme Court’s recent usage. See Armstrong, 135 S Ct at 1385 (2015) (referring to “our equitable powers”).
• 326See, for example, Reich, 74 F3d at 1327; Trudeau v Federal Trade Commission, 456 F3d 178, 189 (DC Cir 2006). But see Kathryn E. Kovacs, Scalia’s Bargain, 77 Ohio St L J 1155, 1175 (2016) (“Nonstatutory review allows a court to hear a claim against the government without a waiver of sovereign immunity; it does not provide a cause of action.”).
• 327See generally Jonathan R. Siegel, Suing the President: Nonstatutory Review Revisited, 97 Colum L Rev 1612, 1613 (1997).
• 328See Marbury v Madison, 5 US (1 Cranch) 137 (1803). See also Akhil Reed Amar, Marbury, Section 13, and the Original Jurisdiction of the Supreme Court, 56 U Chi L Rev 443, 447 (1989); Siegel, 97 Colum L Rev at 1630–31 (cited in note 327). But see Thomas W. Merrill, Marbury v. Madison as the First Great Administrative Law Decision, 37 John Marshall L Rev 481, 487 (2004) (“Marbury’s cause of action was statutory.”).
• 329See Youngstown, 343 US at 85–86. See also Siegel, 97 Colum L Rev at 1636 (cited in note 327) (referring to Youngstown as “perhaps the most notable nonstatutory review case ever”). It is not clear why the Youngstown plaintiffs declined to rely on the APA, which would seem to have provided a cause of action for their lawsuit against the Secretary of Commerce, but they appear not to have.
• 330See Siegel, 97 Colum L Rev at 1636–37 (cited in note 327) (discussing cases).
• 331135 S Ct 1378 (2015).
• 332Id at 1385 (“The sheer complexity associated with enforcing [the relevant provision of the Medicaid Act], coupled with the express provision of an administrative remedy [ ] shows that the Medicaid Act precludes private enforcement of [that provision] in the courts.”).
• 333Id at 1384. See also id (“It is a judge-made remedy.”).
• 334Id at 1390 (Sotomayor dissenting).
• 335209 US 123 (1908).
• 336See Armstrong, 135 S Ct at 1392–94 (Sotomayor dissenting).
• 337561 US 477 (2010).
• 338Id at 1391.
• 339See Hawaii v Trump, 878 F3d 662, 680–81 (9th Cir 2017), revd on other grounds, Trump v Hawaii, 138 S Ct 2392. See also International Refugee Assistance Project v Trump, 883 F3d 233, 283 (4th Cir 2017) (reaching similar conclusion in response to similar challenge).
• 340Among other things, the Ninth Circuit’s analysis did not indicate how the APA possibly could provide a cause of action against the President, who was one of the named parties.
• 341Hawaii v Trump, 878 F3d at 682. See also IRAP, 883 F3d at 287 (reaching similar conclusion in response to similar challenge and referring to the courts’ “inherent authority to review allegations that an executive action has exceeded the Constitution or a congressional grant of authority”).
• 342See Trump v Hawaii, 138 S Ct at 2392.
• 343See, for example, City and County of San Francisco, 897 F3d at 1238–39 (declining to identify the theory under which plaintiffs have a cause of action); First Amended Complaint for Declaratory and Injunctive Relief, City and County of San Francisco v Trump, No 3:17-cv-00485-WHO (ND Cal filed Feb 27, 2017) (same).
• 344See Complaint for Declaratory and Injunctive Relief, Doe v Trump, No 1:17-cv-01597 (DDC filed Aug 9, 2017) (available on Westlaw at 2017 WL 3421198). The citation was inapposite because 42 USC § 1983 provides a cause of action against state and local defendants, not federal defendants.
• 345See, for example, Complaint for Injunctive and Declaratory Relief, Hopi Tribe v Trump, No 1:17-cv-02590 at ¶¶ 14, 200–01 (DDC filed Dec 4, 2017) (referring to “non-statutory review” in one section and the court’s “inherent authority to issue equitable relief” in another).
• 346In Armstrong, the Court concluded that the federal courts’ inherent equitable powers had been displaced by Congress in light of (i) the existence of an alternative enforcement mechanism for the statute at issue and (ii) the “judicially unadministrable nature” of the relevant statutory standard. Armstrong, 135 S Ct at 1385. Among the questions that Armstrong raises, but does not answer, are the following: In what other contexts will courts conclude that statutes have foreclosed relief based on the courts’ equitable powers? Is the approach different if a plaintiff is suing the federal government, as opposed to a state government? In Armstrong, moreover, statutory claims and constitutional claims converged, as plaintiffs were suing pursuant to a Supremacy Clause issue. To what extent is its analysis transferrable to a purely statutory claim? A purely constitutional claim?
• 347There exists some precedent to suggest that these lawsuits must proceed pursuant to an “ultra vires” theory. See, for example, Dart v United States, 848 F2d 217, 224 (DC Cir 1988). However, what exactly this means—for example, how ultra vires review differs from the type of review a plaintiff receives pursuant to the APA—is unsettled. Moreover, there are numerous examples of courts adjudicating claims based on their inherent equitable powers without expressly resorting to ultra vires analysis.
• 348The Supreme Court punted on this question in Dalton v Specter, 511 US 462, 474 (1994), stating merely that the Court would “assume for the sake of argument that some claims that the President has violated a statutory mandate are judicially reviewable outside the framework of the APA.” See also, for example, Reich, 74 F3d at 1331 n 4 (interpreting this passage in Dalton as reflecting “the special status of the President”).
• 349Franklin, 505 US at 800.
• 350See Armstrong, 135 S Ct at 1392 (Sotomayor dissenting). This presumption need not conflict with the majority’s holding in Armstrong, whose scope is unclear and reasonably could be limited to judicially unmanageable preemption claims.
• 3515 US (1 Cranch) 137 (1803).
• 352Id at 163.
• 353Id.
• 354See at Part I.A (discussing arbitrariness review under § 706(2)(A) of the APA).
• 355See Driesen, 98 BU L Rev at 1044–50 (cited in note 27).
• 356See generally Kovacs, Trump v. Hawaii: A Run of the Mill Administrative Law Case (cited in note 19) (arguing that the APA’s arbitrariness standard of review should apply to assess the legality of Trump’s travel ban).
• 357Id.
• 358138 S Ct at 2419 (citation omitted).
• 359Id at 2420–23.
• 360See US Const Art I, § 7. See also, for example, National Federation of Independent Business v Sebelius, 567 US 519, 537–38 (2012) (discussing the role that Members of Congress play as “our Nation’s elected leaders” and the need for the courts to defer accordingly).
• 361See notes 257–58 and accompanying text. See also Watts, 103 Georgetown L J at 1005–06, 1013–18 (cited in note 112).
• 362See note 254 and accompanying text. See also generally Citizens to Preserve Overton Park, Inc v Volpe, 401 US 402 (1971).
• 363See Watts, 103 Georgetown L J at 1041–42 (cited in note 112).
• 364See Part I.C (discussing Chevron deference).
• 365See Mead, 533 US at 226–27. See also Christensen, 529 US at 586–88.
• 366See Mead, 533 US at 226–27.
• 367See, for example, Gonzales v Oregon, 546 US 243, 253, 258–59 (2006) (refusing to grant Chevron deference in a case where the Attorney General adopted a rule on physician-assisted suicide “without consulting [the affected State of Oregon] or apparently anyone outside his Department”).
• 368Bruff, 68 Va L Rev 17–18 (cited in note 146).
• 369See Watts, 103 Georgetown L J at 1043 (cited in note 112) (explaining that in the agency context, “hard look review is quite searching, requiring agencies to give detailed, lengthy justifications for their rules”).
• 370See, for example, Securities & Exchange Commission v Chenery Corp, 332 US 194, 196–97 (1947) (holding that an agency action may be upheld only on the grounds the agency expressly relied upon at the time it originally acted). See also Kevin M. Stack, The Constitutional Foundations of Chenery, 116 Yale L J 952, 1014–16 (2007) (arguing that Chenery’s rule arguably should apply to judicial review of presidential action, not just agency action).
• 371See American Federation of Government Employees, AFL-CIO v Trump, 318 F Supp 3d at 392 (“[W]hen assessing whether the President has acted beyond the bounds of his legal authority, a court may at times have to consider both the authority that congressional statutes have conferred upon him and the inherent authority that the Constitution assigns to the President.”).
• 372For more on presidential factfinding, see generally Shalev Roisman, Presidential Factfinding, 72 Vand L Rev 825 (2019).
• 373Even this kind of fairly narrow reason-giving review would open up a host of questions, such as when and how challengers could attempt to prove that the president’s proffered justifications were pre-textual and that the president actually had based his order on impermissible factors, such as racial or religious animus. See, for example, Trump v Hawaii, 138 S Ct at 2418–20 (noting that the plaintiffs asked the Court “to probe the sincerity of the stated justifications for the [President’s] policy by reference to extrinsic statements” and ultimately concluding that the Court would consider plaintiffs’ extrinsic evidence but would “uphold the [President’s] policy so long as it can reasonably be understood to result from a justification independent of unconstitutional grounds”).
• 374See, for example, City and County of San Francisco, 897 F3d at 1242 (refusing to defer to DOJ’s interpretation of an executive order when, among other things, “[t]he Administration’s claim that the Executive Order has meant nothing all along is belied by the Executive Order itself and the Administration’s public statements”).
• 375See generally Pierce, 3 Administrative Law Treatise §§ 18.1–18.7 at 1323–86 (cited in note 286) (addressing remedies). It is true that there is plenty of uncertainty around the edges. See, for example, Cissell Manufacturing Co v United States Department of Labor, 101 F3d 1132, 1140–42 (6th Cir 1996) (Rosen dissenting) (discussing when a departure from the “general rule” for remedies is appropriate); Nicholas R. Parrillo, The Endgame of Administrative Law: Governmental Disobedience and the Judicial Contempt Power, 131 Harv L Rev 685, 686–87, 789–94 (2018) (arguing that “the federal government’s compliance with court orders is imperfect and fraught”). Nevertheless, for more routine cases challenging agency action, questions of remedy do not pose particular difficulty.
• 376See 5 USC § 706(2).
• 377See, for example, Citizens for Responsibility and Ethics in Washington v Trump, 302 F Supp 3d 127, 138 (DDC 2018) (noting that whether a court can issue a declaratory judgment against the president is not “clearly settled”).
• 378See Knight First Amendment Institute at Columbia University v Trump, 302 F Supp 3d 541 (SDNY 2018). This line of litigation technically does not fall into the new category of cases we have identified because it does not involve a challenge to an executive order. It nevertheless involves a direct challenge to the President’s official conduct and therefore is instructive.
• 379See generally Complaint for Declaratory and Injunctive Relief, Knight First Amendment Institute at Columbia University v Trump, No 1:17-cv-05205 (SDNY filed July 11, 2017) (available on Westlaw at 2017 WL 2952634).
• 380Defendants’ Memorandum of Law in Opposition to Plaintiffs’ Cross-Motion and Reply Memorandum in Further Support of Defendants’ Motion for Summary Judgment, Knight First Amendment Institute at Columbia University v Trump, No 1:17-cv-05205-NRB, *3–6 (SDNY filed Nov 17, 17) (available on Westlaw at 2017 WL 9533190) (Defendants’ Memorandum), citing Mississippi v Johnson, 71 US 475 (1867), and Franklin, 505 US 788.
• 381418 US 683 (1974).
• 382See Defendants’ Memorandum at *4 (cited in note 380). See also Clinton v Jones, 520 US 681, 703–06 (1997); Nixon, 418 US at 687–90; Youngstown, 343 US at 584–86. Defendants conceded that the second category might also encompass other “ministerial” duties. Defendants’ Memorandum at *4 (cited in note 380).
• 383See Plaintiffs’ Reply Memorandum of Law in Support of Plaintiffs’ Cross-Motion for Summary Judgment, Knight First Amendment Institute at Columbia University v Trump, No 1:17-cv-05205, *9–13 (SDNY filed December 1, 2017) (available on Westlaw at 2017 WL 9533191).
• 384Brief for Federal Courts Scholars as Amici Curiae Supporting Plaintiffs, Knight First Amendment Institute at Columbia University v Trump, No 1:17-CV-05205, *1 (SDNY filed November 6, 2017) (available on Westlaw at 2017 WL 9533186).
• 385See Knight First Amendment Institute, 302 F Supp 3d at 549, 579–80, affd Knight First Amendment Institute at Columbia University v Trump, 928 F3d 226, 230 (2d Cir 2019).
• 386See, for example, Franklin, 505 US at 828 (Scalia concurring in part and concurring in the judgment) (“Review of the legality of Presidential action can ordinarily be obtained in a suit seeking to enjoin the officers who attempt to enforce the President’s directive.”).
• 387See, for example, IRAP, 857 F3d at 605–06 (vacating injunction to the extent it was directed at the President), vacd and remd on other grounds, 138 S Ct 353 (2017); County of Santa Clara v Trump, 275 F Supp 3d 1196, 1219 (ND Cal 2017) (imposing a “nationwide injunction” against defendants other than the President). At least one court, however, recently declined to enjoin agency officials, preferring instead to “vacate” the unlawful portion of the Executive Order at issue in the case. See League of Conservation Voters v Trump, 2019 WL 1431217, *13 (D Alaska 2019) (declining to issue an injunction against the Secretaries of the Interior and Commerce and choosing instead to “vacate” the unlawful section of the presidential order). The APA provides that courts shall “set aside” agency rules found to be unlawful. See 5 USC § 706(2). See also Sierra Club v Van Antwerp, 719 F Supp 2d 77, 78 (DDC 2010) (citing Black’s Law Dictionary for the proposition that “set aside,” as used in the APA, means “to annul or vacate”); Harmon v Thornburgh, 878 F2d 484, 495 n 21 (DC Cir 1989) (“When a reviewing court determines that agency regulations are unlawful, the ordinary result is that the rules are vacated.”).
• 388See, for example, Nixon, 418 US at 713 (concluding that the President could be ordered to comply with the particular subpoena at bar but suggesting a different outcome might adhere in another case, including one in which the President cited something other than a “generalized interest in confidentiality”).
• 389Arguably, the plaintiffs in Franklin required an injunction running against the President in order to ensure relief. See Franklin, 505 US at 824–26 (Scalia concurring in part and concurring in the judgment) (describing disagreement with the plurality over the adequacy of relief against the President’s subordinates). The plaintiffs in Knight First Amendment Institute face a similar predicament. See Knight First Amendment Institute, 302 F Supp 3d at 561–62 (implying that “complete” relief would require an order to be entered against the President himself). Tellingly, neither Franklin nor Knight First Amendment Institute involved traditional executive orders, which normally rely squarely on subordinates for enforcement.
• 390See generally Brief for Federal Courts Scholars (cited in note 384) (advancing arguments in favor of this position).
• 391See Lisa Marshall Manheim, Beyond Severability, 101 Iowa L Rev 1833, 1841–44 (2016).
• 392Id at 1839–40. See also id at 1872–85 (calling into question whether this understanding of severability is appropriate).
• 393Id at 1845–47.
• 394Id at 1839–40.
• 395Richard H. Fallon Jr, Fact and Fiction about Facial Challenges, 99 Cal L Rev 915, 955 (2011).
• 396See Charles W. Tyler and E. Donald Elliott, Administrative Severability Clauses, 124 Yale L J 2286, 2294–97 (2015). See also Severability in Agency Rulemaking (Administrative Conference of the United States, June 26, 2018), archived at http://perma.cc/RZN9-BERZ.
• 397See, for example, Davis County Solid Waste Management v Environmental Protection Agency, 108 F3d 1454, 1459 (DC Cir 1997) (“Whether an administrative agency’s order or regulation is severable, permitting a court to affirm it in part and reverse it in part, depends on the issuing agency’s intent.”) (citation omitted). See also K Mart Corp v Cartier, Inc, 486 US 281, 294 (1988) (“The severance and invalidation of [the trademark regulation at issue] will not impair the function of the statute as a whole, and there is no indication that the regulation would not have been passed but for its inclusion.”).
• 398See Executive Order No 13769, 82 Fed Reg at §§ 3(c), 5(a) (cited in note 122).
• 399See Washington v Trump, 2017 WL 462040, *2–3 (WD Wash).
• 400See Washington v Trump, 858 F3d 1168, 1168 (9th Cir 2017).
• 401Id at 1172 (Kozinski dissenting from the denial of reconsideration en banc), citing Matter of Reyes, 910 F2d 611, 613 (9th Cir 1990). Reyes was also cited in Mille Lacs Band of Chippewa Indians, in which the Supreme Court deemed an executive order inseverable after “assum[ing], arguendo, that the severability standard for statutes also applies to Executive Orders.” Mille Lacs Band of Chippewa Indians, 526 US at 191. But see id at 215–16 (Rehnquist dissenting) (arguing that severability should proceed differently in the context of executive orders). In Reyes, the Government conceded on appeal that the order, as relevant, was partially unlawful. The question then became: How should the court treat this partially unconstitutional executive order going forward? In response, the Government encouraged the court to “apply the same test to [the] Order that would be used where Congress has passed a law that is partially unconstitutional.” Reyes, 910 F2d at 613. The Ninth Circuit accepted this invitation without further discussion. Parsing the text of the order, including its lack of a severability clause, the court concluded that the President would prefer the order to be treated as inseverable. Id.
• 402Washington, 858 F3d at 1172 (Kozinski dissenting from the denial of reconsideration en banc).
• 403Id.
• 404Professor Tara Grove’s study of the process of issuing presidential directives somewhat complicates this comparison by confirming that presidents, working with their subordinates, nearly always adhere to a process of interagency consultation. See Tara Grove, Presidential Laws and the Missing Interpretative Theory, 168 U Pa L Rev *20–27 (forthcoming 2020), archived at http://perma.cc/GGX3-D2SZ. See also id at *27 (“Although it is easier to issue a presidential directive than to enact legislation, the process takes a good deal longer than one might expect—anywhere from several weeks to several months (or even years).”). Nevertheless, this process is less onerous than the process legally required of either Congress or agencies, and it remains possible for the White House to modify or even completely bypass these procedures when it desires to expedite the process, see id at *26–27 & nn 162–64.
• 405But see Manheim, 101 Iowa L Rev at 1872–85 (cited in note 391) (asking whether this sort of concern is based on a distinction without a difference).
• 406See, for example, Brianne J. Gorod, The Collateral Consequences of Ex Post Judicial Review, 88 Wash L Rev 903, 949 (2013) (describing the Court’s willingness, on occasion, to take this approach in response to constitutionally infirm statutes).
• 407See Manheim, 101 Iowa L Rev at 1840 (cited in note 391) (describing, in the context of legislation, the “centrality” of the drafter’s intent to severability analysis, and just how “deeply engrained” this principle has become).
• 408See note 285.