Presidential Administration Via Criminal Enforcement

Cristina Isabel Ceballos[*]

Abstract

When the president seeks to influence administrative agencies, he can choose from a “toolkit” of different techniques for presidential control. The president might intervene into an agency’s funding, for example, by conditioning the release of agency grants. Alternatively, the president might intervene into an agency’s rulemaking, for example, by directing the agency to rescind a prior rule. Or the president might intervene into an agency’s enforcement, for example, by announcing a high-profile enforcement crackdown against certain types of offenses. By utilizing a broad toolkit of techniques, the president can influence agencies in a process known as “presidential administration,” a term coined by then-Professor Elena Kagan in 2001. Presidential administration has become the dominant scholarly theory used to describe the president’s influence over administrative agencies, and in the decades since Kagan’s article, many scholars have expanded upon Kagan’s theory by identifying new techniques in the toolkit for presidential control.

Fewer scholars, however, have examined criminal enforcement as a tool for presidential administration. Administrative agencies typically pursue civil enforcement (which involves civil penalties such as fines or injunctions), but agencies may also pursue criminal enforcement (which involves criminal penalties such as imprisonment). The president has the power to direct both civil and criminal enforcement, and many recent presidents have turned to criminal enforcement. President Biden, for example, oversaw crackdowns against environmental and corporate crime. President Trump used criminal enforcement against immigrants to create the Family Separation Policy. President Obama encouraged criminal prosecution of workplace safety violations. And President Bush used civil and criminal enforcement in his attempt to outlaw physician-assisted suicide in Oregon. In these examples, criminal enforcement has functioned as a tool for presidential administration, allowing the president to influence areas like immigration law, environmental law, and workplace safety.

I. Introduction

In recent years, the president has intervened in criminal enforcement across a wide range of policy areas. President Trump combined criminal and immigration enforcement to create the Family Separation Policy.¹ President Biden oversaw a crackdown on environmental and corporate crimes.² President Obama encouraged criminal prosecution of workplace safety violations³ and announced new guidelines for mandatory minimum sentences.⁴ And President Bush invoked both civil and criminal enforcement in his attempt to outlaw physician-assisted suicide in Oregon.⁵ Across both Democratic and Republican administrations, the president has used criminal enforcement for policymaking in immigration law, environmental law, workplace safety, securities regulation, and other regulatory areas that include criminal penalties.⁶

In many of the above examples, the president enlisted the resources of both traditional administrative agencies, such as the Environmental Protection Agency, and the Justice Department, which is the sole agency authorized to bring federal criminal charges.⁷ This approach is common for presidents who use criminal enforcement: many presidential policies will cut across both administrative and criminal law.⁸ So in order to fully address the president’s use of criminal enforcement, this Article adopts a cross-cutting approach that draws on both administrative and criminal law.

This Article makes two contributions. First, it recognizes that criminal enforcement is a powerful and largely overlooked tool for presidential administration. Coined by then-Professor Elena Kagan, “presidential administration” is the leading theory that scholars have used to describe the president’s control over administrative agencies.⁹ Yet scholars of presidential administration have devoted relatively little attention to criminal enforcement. In fact, Kagan did not include the word “criminal” in the main text of her seminal article on the theory.¹⁰ Many subsequent articles about presidential administration have followed Kagan’s approach and have largely overlooked criminal enforcement.¹¹ This Article advances the presidential administration literature by identifying criminal enforcement as a powerful tool for the president.¹²

Second, this Article proposes a mens rea-based constraint on the president. This constraint is based on the president’s epistemic disadvantage: the president simply does not know as much as prosecutors and line-level agency officials. If the president seeks directive control over prosecutors, he may pressure prosecutors into bringing charges even if there is weak evidence of mens rea.¹³ When this happens, the lower courts can use mens rea to constrain presidentially directed criminal prosecutions. For example, in New Mexico, a magistrate judge dismissed criminal charges in dozens of cases against immigrants in the Trump administration’s new militarized border zones.¹⁴ The judge first applied a presumption in favor of scienter, which reads in an implied mens rea requirement even in statutes that do not expressly require mens rea, and then he held that prosecutors had failed to allege sufficient facts about mens rea.¹⁵ This Article urges the lower courts to adopt the New Mexico court’s more protective approach to mens rea. In addition, the Article urges the Supreme Court to signal to prosecutors that they should resist presidential pressure to charge cases with weak evidence of mens rea.¹⁶

This Article proceeds in five parts. Part II situates criminal enforcement within Kagan’s theory of presidential administration. Kagan’s traditional toolkit described rulemaking, civil enforcement, Office of Information and Regulatory Affairs (“OIRA”) review, and other tools for presidential control. Part III describes a relatively new tool for presidential policy control—criminal enforcement—which recent presidents have used across many areas of administrative law. Criminal penalties are embedded within many otherwise civil regulatory regimes,¹⁷ so the president has a strong incentive to use criminal enforcement. Part IV evaluates Kagan’s traditional normative argument in favor of presidential administration. This Part argues that Kagan did not fully address concerns about “presidential lawlessness,” which occurs when the president adopts aggressive policies that push agencies beyond their statutory mandates.¹⁸ Kagan believed that presidential lawlessness could be managed through judicial review, yet her reasoning applies most squarely to agency rulemaking (where judicial review is presumptively available) and less squarely to agency enforcement¹⁹ (where judicial review is harder to obtain).²⁰ If the president opts for criminal enforcement rather than rulemaking, then he can avoid the presumption of reviewability under the Administrative Procedure Act (“APA”).

Finally, Part V argues for a mens rea-based constraint on the president. Under a more protective conception of mens rea, the president would be free to set general enforcement priorities, but he would be prevented from issuing binding or quasi-binding directives to lower-level officials, because binding directives squeeze out lower-level officials’ ability to consider mens rea.

This more protective conception of mens rea would operate in two places: in the lower courts and the Supreme Court. In the lower courts, judges would apply a more protective conception of mens rea during criminal prosecutions, as illustrated by recent cases in New Mexico.²¹ In the Supreme Court, a ruling on mens rea would allow the Court to send a strong signal to prosecutors and to the president, urging the president to exercise restraint in his use of criminal enforcement. Many Justices on the Court are already deeply committed to mens rea,²² and they might be more willing to use mens rea, as opposed to cabining the Vesting Clause, to rein in the president’s use of criminal enforcement.²³

This Article draws on a wide range of scholarship from both criminal and administrative law. It was enriched by prior work on the growth of the criminal code,²⁴ the federalization of criminal law,²⁵ and the empirics of environmental crimes.²⁶ Other scholarship discussed enforcement crackdowns by agencies,²⁷ separation of powers in the criminal law,²⁸ and the moral dimensions of regulatory crimes.²⁹ Recent scholarship has chronicled the Court’s preoccupation with criminal delegations³⁰ and its adoption of a strong unitary executive theory³¹ based on Justice Scalia’s interpretation of the Vesting Clause. This Article is deeply indebted to this array of excellent prior scholarship.

II. The Traditional Toolkit for Presidential Administration

When Kagan introduced her theory of presidential administration, she described a broad toolkit of techniques for presidential control.³² President Reagan, for example, used OIRA, which allowed him to delay or return agency rulemakings.³³ President Clinton was fond of press conferences, which he used to announce new rules and to personally claim credit for agency action.³⁴ President Bush inherited an “arsenal” of techniques from his predecessors, and he continued to expand the toolkit for presidential control.³⁵ As Kagan noted, each new president had a strong incentive to preserve or expand the powers handed down by his predecessors.³⁶

A. Kagan’s Original Toolkit

1. Rulemaking

In Kagan’s article, rulemaking was the most frequently cited tool for presidential administration.³⁷ Rulemaking allows an agency to promulgate a new rule which carries the “force and effect of law.”³⁸ Rules are typically formulated via the notice-and-comment process, which requires a notice of proposed rulemaking, a period for members of the public to submit comments, and publication in the Federal Register.³⁹ Most pre-publication rules are reviewed by OIRA, which has the power to delay or return agency rulemakings.⁴⁰ If a rule survives OIRA review, then the newly promulgated rule is considered “binding” upon the agency and the public.⁴¹

President Clinton often used rulemaking as a substitute for legislation, because his legislation was frequently blocked in Congress. During President Clinton’s two terms, Congress was mostly controlled by Republicans who proved “hostile” to President Clinton’s progressive agenda.⁴² President Clinton failed when he tried to enact legislation for universal healthcare,⁴³ failed when he attempted to legislate tobacco restrictions,⁴⁴ and failed again when he pursued other legislative goals.⁴⁵ Rulemaking allowed President Clinton to accomplish what he had been unable to accomplish through legislation.

As Kagan noted, President Clinton inserted himself at every stage of the rulemaking process. He announced new rules before the agency initiated the rulemaking process, he weighed in on rules as they made their way through OIRA review, and he personally claimed credit for new rules after they were finalized.⁴⁶ In some instances, President Clinton even dictated the specific contents of some rules: for example, when President Clinton announced a new anti-tobacco rule, he described six specific proposals (including a ban on “Joe Camel” advertising for children and a prohibition on “cigarette vending machines” in schools).⁴⁷ After the agency completed its rulemaking process, President Clinton’s six proposals were included in the agency’s final rule, essentially unchanged.⁴⁸

2. Civil Enforcement

Civil enforcement was a secondary tool in President Clinton’s toolkit, one which he used “less frequently” than rulemaking.⁴⁹ Kagan described two significant civil enforcement efforts from President Clinton, against tobacco companies⁵⁰ and gun manufacturers.⁵¹ Oftentimes, President Clinton’s civil enforcement efforts were accompanied by rulemakings: his tobacco enforcement announcement, for example, followed his push for an anti-tobacco rulemaking.⁵²

When Kagan discussed civil enforcement by President Clinton, she sounded a rare note of caution, even disapproval.⁵³ She argued that specific enforcement actions might tempt the president to engage in “the crassest form of politics,” such as the pursuit of “personal favors and vendettas.”⁵⁴ Presidential interventions should be discouraged in enforcement cases,⁵⁵ she said, and she even proposed a “prohibition” on presidential administration in specific enforcement actions.⁵⁶

Kagan’s prohibition was inspired by the risks of abuse. Enforcement decisions center on “particular individuals and firms,”⁵⁷ Kagan noted, which created a “danger” of politicized and unprofessional decisions.⁵⁸ If the president involved himself in selecting specific enforcement targets, he might target his political rivals or pursue “vendettas” or “favors.”⁵⁹ To avoid these risks, Kagan favored a “prohibition” on presidential administration in specific enforcement cases.⁶⁰ Notably, Kagan’s prohibition applied only to specific enforcement actions but not to general enforcement policy.⁶¹ As Kagan clarified:

I do not mean this prohibition to include the articulation of broad enforcement policy or strategy, of the kind President Clinton sometimes made part of his program of administrative governance . . . . I do mean to raise serious questions about presidential direction of decisions to file suit against discrete entities . . . .⁶²

As Kagan herself acknowledged, her “prohibition” cut against the grain of traditional theories, particularly unitary executive theory.⁶³ Unitary executive theory views enforcement as “the heart” of presidential power.⁶⁴ Part V will discuss unitary executive theory in more detail.

3. OIRA Review, Press Conferences, and Other Tools

In addition to discussing rulemaking and civil enforcement, Kagan also mentioned additional tools for presidential administration, including OIRA review, press conferences, and executive orders. According to Kagan, different presidents favored different “methods of control,” depending on their personalities and the tools available to them.⁶⁵ For example, President Reagan preferred to use the OIRA review process, which gave him greater control over rulemaking.⁶⁶ Via OIRA, the White House could delay, dilute, or even defeat agency rulemakings.⁶⁷

President Clinton perfected the art of the press conference, which he used to announce high-profile executive orders. In Kagan’s count, President Clinton issued 107 executive orders, compared to only nine by President Reagan.⁶⁸ According to President Clinton’s chief speechwriter, “The White House settled into a rat-a-tat-tat of announcements and statements. A ‘message event’ of the day. Each one with an element of news—issuing an executive order, announcing the results of a study, setting a regulation, passing out grants.”⁶⁹ Another Clinton aide joked that “we expect that we’ll be producing executive orders until the morning he leaves office.”⁷⁰

After President Clinton left office, he handed down an expanded toolkit to President Bush, who inherited an “arsenal” from his predecessors.⁷¹ President Bush took office a few months before Kagan published her article,⁷² and Kagan (correctly) predicted that President Bush would be an enthusiastic adopter of presidential administration, despite some early predictions to the contrary.⁷³ According to Kagan, President Bush was “unlikely” to voluntarily “cede” the presidential powers which had been built up by Presidents Reagan and Clinton.⁷⁴ Kagan noted that President Bush had taken an “aggressive posture” in the OIRA process,⁷⁵ and she predicted that President Bush would use his powers to advance a deregulatory agenda, as President Reagan had.⁷⁶

B. Expanding the Toolkit

In the decades since Kagan’s article, scholars have identified numerous additional tools for presidential control. This section briefly summarizes the expanded presidential toolkit, which includes tools like agency funding, agency personnel, and agency nonenforcement.

1. Agency Funding

Agency funding has gained renewed attention since 2025, when President Trump began actively intervening into areas like foreign aid spending and federal research grants.⁷⁷ These interventions have been called “appropriations presidentialism,” a term coined by Eloise Pasachoff, Zachary Price, and Matthew Lawrence.⁷⁸ In their recent article, they argue that President Trump’s actions represent a “dramatic, destabilizing expansion of a preexisting trend.”⁷⁹ They follow the preexisting trend to President Obama, who used his powers to implement the Affordable Care Act’s cost-sharing reductions.⁸⁰ However, they argue that President Trump has aggressively expanded the president’s powers in ways not undertaken by previous presidents. For example, the Trump White House accessed the Treasury Department’s payment system, which historically has not been used for policymaking,⁸¹ cut off funds without following agencies’ own internal procedures,⁸² and clawed back funds already disbursed.⁸³ Pasachoff, Price, and Lawrence critique President Trump’s “dramatic escalation” of appropriations presidentialism,⁸⁴ and they urge Congress to reassert its power over the purse.⁸⁵

2. Agency Personnel & Capacity

Presidents can influence agency capacity by changing an agency’s high-level leadership or lower-level personnel. In recent years, Anne Joseph O’Connell has documented the president’s use of “actings,” or temporary agency leadership.⁸⁶ By appointing “actings,” the president can bypass the usual Senate confirmation process and exert greater control over agency leadership.⁸⁷ More recently, Nicholas Bednar has shed new light on agency “capacity,” which includes know-how by lower-level staff.⁸⁸ As Bednar writes, “Some agencies are overworked, under-resourced, and poorly managed,” and these capacity issues can derail the president’s agenda even in agencies that are broadly aligned with the president’s policies.⁸⁹ Agency capacity has also been discussed by Jody Freeman and Sharon Jacobs, who have warned that the president can “hollow out” agencies by deliberately pushing out “key staff.”⁹⁰

3. Nonenforcement

Finally, many scholars have written about nonenforcement, which gained national attention during President Obama’s policy on Deferred Action for Childhood Arrivals (“DACA”). DACA hinged on nonenforcement: immigration agencies were told to cease enforcement against immigrants who had been brought to the United States as children, who had earned a high school diploma, and who met other requirements.⁹¹ DACA also allowed immigrants to apply for employment authorization documents.⁹²

DACA kicked off a flurry of nonenforcement scholarship. Kate Andrias, for example, found that President Obama was not the only president who used nonenforcement: President Bush, for example, had dialed back enforcement of environmental regulations.⁹³ Nonenforcement also attracted the attention of constitutional scholars, including John Yoo, Robert Delahunty, Zachary Price, Mila Sohoni, and many others.⁹⁴ Some scholars argued that the Take Care Clause was flexible enough to accommodate policies like DACA, while others argued against a broad nonenforcement policy like DACA.⁹⁵

* * *

The toolkit for presidential administration was already large when Kagan described it in 2001, and it has grown even larger since. In the interest of space, this Article provides only a brief discussion, since this is already well-trodden ground. Next, this Article turns to a relatively new tool for presidential administration: criminal enforcement.

III. Criminal Enforcement as a New Tool for Presidential Control

The president has strong incentives to utilize criminal enforcement for regulatory control, because criminal enforcement is widely available and is more strongly deterrent than civil enforcement. Part III.A describes incentives for the president, who can use criminal enforcement across the many regulatory areas where criminal penalties are embedded alongside civil penalties. Part III.B describes examples from the Bush, Obama, Biden, and Trump administrations.

A. Incentives for the President to Use Criminal Enforcement

For the president, criminal enforcement may present an attractive option. First, criminal enforcement is widely available, because many regulatory regimes contain both civil and criminal penalties. Second, criminal enforcement carries a stronger deterrent effect, because criminal penalties are typically heavier than civil penalties. Third, the president can amplify the uptake of criminal referrals, because he has the power to direct agencies (which investigate regulatory violations) as well as the Justice Department (which receives criminal referrals from agencies).

1. Widely Available Criminal Penalties

Regulatory statutes often include both civil and criminal penalties. For example, criminal penalties exist in environmental regulations,⁹⁶ food and drug law,⁹⁷ securities law,⁹⁸ antitrust law,⁹⁹ tax law,¹⁰⁰ immigration law,¹⁰¹ and other areas. According to Lawrence Solan, “[s]tatutes that contain both civil remedies and criminal penalties are typical in the modern regulatory state.”¹⁰²

The number of criminal penalties has grown “explosively” since the 1970s.¹⁰³ Several authors have attempted to count the number of criminal penalties embedded in the United States Code; while the authors do not agree on an exact number, all agree that the number has grown significantly.¹⁰⁴ In 1998, for example, the American Bar Association found that “over forty percent” of federal crimes “have been created since 1970.”¹⁰⁵ In 2004 and 2008, two studies from John Baker found several hundred new federal crimes.¹⁰⁶ A more recent January 2022 study found 1,510 sections of the Code that created at least one federal crime,¹⁰⁷ and a subsequent study found additional sections that had been overlooked by the 2022 study.¹⁰⁸

It has been difficult to quantify the number of criminal penalties in the Code because criminal penalties are scattered throughout many different parts of the Code and they are often embedded in otherwise civil regulatory regimes. For example, the National Wildlife Refuge System Administration Act regulates wilderness conservation. The Act is usually civilly administered by the Fish and Wildlife Service,¹⁰⁹ yet it also includes criminal penalties, such as the following:

Any person who knowingly violates or fails to comply with any of the provisions of this Act or any regulations issued thereunder shall be fined under [T]itle 18 or imprisoned for not more than 1 year, or both.¹¹⁰

The statute uses the word “imprisoned,” which denotes a criminal, not civil, penalty.¹¹¹ Throughout the Code, keywords like “imprisonment,” “shall be punished,” and “shall be guilty of”¹¹² are used to create criminal penalties, which are often embedded into otherwise civil regulatory regimes.¹¹³

In recent years, many scholars have attempted to explain the “explosive growth” of criminal penalties.¹¹⁴ Some scholars blame overcriminalization and the “tough on crime” movement.¹¹⁵ Alternatively, William Stuntz has argued that legislative incentives are the primary driver of federal criminal legislation.¹¹⁶ Whatever the reason, the past fifty years have seen a substantial increase in federal criminal provisions.¹¹⁷ Criminal penalties exist in many modern regulatory regimes, thus creating more opportunity for criminal enforcement direction by the president.

2. Strong Deterrent Effect

Criminal penalties are often more severe than civil penalties, and many theorists have argued that criminal penalties have a stronger deterrent effect.¹¹⁸ For example, William Stuntz, who wrote about the civil-criminal distinction, noted that:

Criminal punishment often means prison, and prison is both different from and worse than money damages. It is also, by definition, worse than nonmonetary civil sanctions: Injunctions take away the freedom to do certain things, while prison takes away the freedom to do just about anything that prison officials don’t want you to do.¹¹⁹

Stuntz, like many criminal theorists, acknowledged that deterrence is not the only goal of criminal punishment. Such punishment may also aim to rehabilitate, to incapacitate the wrongdoer, to restore the social standing of the victim, or to accomplish a wide variety of other goals.¹²⁰ As Henry Hart, Jr. put it, deterrence need not be “the overriding and ultimate purpose of the criminal law, important though it is.”¹²¹

Yet despite competing theories, many theorists nevertheless agree that criminal penalties have a stronger deterrent effect than civil penalties. This intuition is borne out in the empirical literature, including a recent article by Nicholas Parrillo, who studied enforcement against large companies and healthcare providers.¹²² According to one attorney interviewed by Parrillo, the fear of criminal enforcement “was the main reason people in the industry followed [agency] guidance.”¹²³ Another interviewee noted that announcements about criminal indictments could trigger a “built-in level of hysteria” in the industry.¹²⁴

3. Amplified Uptake of Criminal Referrals

Finally, the president can amplify uptake of criminal referrals, because he can direct the Justice Department to accept criminal referrals from agencies.¹²⁵ When agencies refer their cases for criminal prosecutions, their referrals are sometimes declined by the Justice Department.¹²⁶ But the president can increase uptake of criminal referrals by simultaneously directing (a) agency investigations of regulatory violations and (b) Justice Department prosecutions.

a. Agency Investigations of Regulatory Violations

Administrative agencies typically investigate regulatory violations in their areas. According to Daniel Richman, a former federal prosecutor, agencies have “the expertise, the manpower, the technical resources, and, perhaps most importantly, the informational networks that no U.S. Attorney’s Office possesses.”¹²⁷ In many cases, federal prosecutors will “not even know that a crime has been committed” until an agency refers a criminal case to the Justice Department.¹²⁸

Mens rea, also known as scienter or “guilty mind,” typically divides criminal offenses from civil offenses. Some tax statutes, for example, allow civil penalties for good-faith mistakes, yet tax errors can become criminally punishable if the offender acted willfully or with some other culpable mental state.¹²⁹ Similarly, environmental violations are usually sanctioned with civil fines, yet violators may become criminally liable if the agency uncovers evidence of willfulness, intent to deceive, or other aggravating factors.¹³⁰

Agencies that conduct criminal investigations must also “convince a prosecutor to take the case.”¹³¹ Per Robert Rabin, agencies that acquire a “reputation for poor quality preparatory work” may find that prosecutors are reluctant to accept their referrals.¹³² If the Justice Department does accept a referral, agency investigators may assist with the prosecution and “will generally be responsible for the evidentiary aspects” of the case.¹³³

When an agency refers a case to the Justice Department, it will sometimes send the criminal referral to “Main Justice” in Washington, D.C. At Main Justice, some units specialize in certain types of regulatory crimes: for example, Main Justice’s Tax Division will often handle referrals from the Internal Revenue Service,¹³⁴ and Main Justice’s Environmental Crimes Section will often work with the Environmental Protection Agency.¹³⁵ But agencies need not always work with Main Justice in D.C., because agencies may also refer cases to local U.S. Attorneys’ Offices throughout the country. The Securities and Exchange Commission, for example, often works with federal prosecutors in the Southern District of New York.¹³⁶ The Environmental Protection Agency, too, can refer cases to local U.S. Attorneys’ Offices, and these local U.S. Attorneys’ Offices can sometimes file charges without seeking approval from Main Justice.¹³⁷

b. Criminal Prosecutions by the Justice Department

The Justice Department’s control over criminal litigation derives from the Judiciary Act of 1789, which authorized federal prosecutors to represent the United States in civil and criminal litigation.¹³⁸ Over time, civil litigation authority was split between different agencies: for example, the Securities and Exchange Commission was created in 1934 and was granted its own civil litigation authority independent of the Justice Department.¹³⁹ Yet even though some agencies acquired civil litigating authority, criminal authority remained firmly ensconced within the Justice Department.¹⁴⁰ According to Kirti Datla and Richard Revesz:

The default rule is that the Attorney General is to direct all litigation to which the United States is a party ‘[e]xcept as otherwise authorized by law.’ Congress has carved out numerous exceptions to the centralized control of litigation in the DOJ. These exceptions—all related to civil, not criminal, litigation—’have been anything but systematic.’¹⁴¹

In the centralized system described by Datla and Revesz, some agencies can pursue civil litigation in federal court, but all agencies must refer cases to the Justice Department for federal criminal prosecution.¹⁴² Agencies are thus “dependent” upon the Justice Department for criminal prosecutions,¹⁴³ and Justice Department prosecutors are the “exclusive gatekeepers” for federal criminal enforcement.¹⁴⁴

The president can amplify uptake of criminal referrals in given policy areas because he can direct the Justice Department to prioritize certain types of referrals. During the Family Separation Policy, for example, the Trump administration directed the Justice Department to accept referrals from immigration agencies.¹⁴⁵ During President Biden’s term, prosecutors were instructed to prioritize corporate and white-collar crime.¹⁴⁶

B. Criminal Enforcement as Used by Recent Presidents

Many recent presidents have used criminal enforcement, including Presidents Trump, Biden, Obama, and Bush. However, President Trump’s policies have been unusually binding upon prosecutors, who have traditionally exercised a large degree of independence from the White House. This section provides five in-depth case studies of criminal enforcement as used by recent presidents.

1. New Mexico Military Zone (Trump Administration)

As part of its effort to reduce immigration, the Trump administration created new military zones in border states like New Mexico.¹⁴⁷ The military zones allowed the Army to “take a more direct role in securing our southern border,” and they were announced by President Trump, who directed the Department of the Interior to transfer jurisdiction over public lands to the Department of Defense.¹⁴⁸ President Trump noted that the transfer would allow the Army to patrol the newly created military zone and to “exclude persons” from entering the zones.¹⁴⁹ A few weeks later, in May 2025, Defense Secretary Pete Hegseth and the U.S. Attorney for the District of New Mexico announced that they would “work hand in glove” to criminally prosecute immigrants.¹⁵⁰

The military zones allowed prosecutors to bring more severe national security-related charges against immigrants. In previous administrations, immigrants might have faced deportation (a civil penalty)¹⁵¹ or charges of entry without inspection (carrying a criminal penalty of up to six months of imprisonment).¹⁵² Under the Trump administration’s new policy, immigrants were charged under multiple criminal statutes, including national security-related statutes.¹⁵³ These national security-related charges were often more severe, especially if the charges were stacked together: for example, immigrants were charged under 50 U.S.C. § 797 for violating a national security regulation (punishable by imprisonment up to one year), plus 18 U.S.C. § 1382 for entering military property (punishable by imprisonment up to six months), plus 8 U.S.C. § 1325 for entry without inspection (punishable by imprisonment up to six months).¹⁵⁴

By May 2025, prosecutors were bringing stacked criminal charges against immigrants detained in the New Mexico military zone.¹⁵⁵ Federal prosecutors filed hundreds of nearly identical charges, using a “cut-and-paste approach” that repeated the same generic allegations.¹⁵⁶ Public defenders contested the national security-related charges, arguing that they were unwarranted because immigrants did not even know they were entering a military zone.¹⁵⁷ Indeed, many local residents and local officials seemed confused about the boundaries of the New Mexico military zone, which contained large swaths of private land (such as a church-owned pilgrimage site)¹⁵⁸ and land frequently used by ranchers for livestock grazing.¹⁵⁹ In some instances, the U.S. Attorney’s Office erroneously charged immigrants who had not, in fact, been found within the boundaries of the military zone.¹⁶⁰

In May 2025, Magistrate Judge Gregory Wormuth dismissed the national security-related charges in dozens of cases.¹⁶¹ He held that prosecutors had failed to allege sufficient facts about mens rea.¹⁶² Prosecutors wrote that “signs” had been posted to alert passersby about the military zone, but the judge held that this generalized allegation was insufficient. Simply stating that “some ‘signs’ were posted,” he argued, “provides no basis on which to conclude that the Defendant could have seen, let alone did see, the signs.”¹⁶³ In dismissing the criminal charges, Judge Wormuth drew upon a more robust and far-reaching conception of mens rea, which this Article will discuss in detail in Part V.

2. Family Separation Policy (Trump Administration)

In April 2018, President Trump announced a crackdown against immigrant families at the border.¹⁶⁴ The crackdown began with an executive order from President Trump,¹⁶⁵ which was quickly followed by a memorandum from Attorney General Jeff Sessions, who announced a “zero-tolerance” policy for immigration cases referred to federal prosecutors.¹⁶⁶ Another memorandum, from Secretary of Homeland Security Kirstjen Nielsen, directed immigration officials to criminally refer cases to the Justice Department, even though the criminal referrals would separate parents from children.¹⁶⁷ Both memoranda were published in April 2018, and shortly thereafter, President Trump’s cabinet members approved the Family Separation Policy by a show-of-hands vote in the White House.¹⁶⁸

The Family Separation Policy used criminal referrals to redirect the flow of immigration cases. Prior to the Family Separation Policy, immigrant families typically faced civil immigration proceedings.¹⁶⁹ While more serious cases sometimes had been referred to the Justice Department (for example, immigration cases that involved human trafficking or sex offenders), many routine immigration cases were resolved without a criminal referral, in part due to concerns about family separation.¹⁷⁰

After the Family Separation Policy, however, immigration cases began flowing toward the Justice Department—specifically, toward U.S. Attorneys’ Offices along the Southwest border. Attorney General Jeff Sessions instructed these U.S. Attorneys to exercise “zero-tolerance” in immigration cases, even if the cases involved parents with children.¹⁷¹ By law, children can be held in immigration detention but not in criminal detention,¹⁷² so children were placed in the custody of the Department of Health and Human Services while their parents served criminal sentences.¹⁷³ Immigration officials failed to create a system that would allow them to reunify families after the parents’ criminal trials,¹⁷⁴ and in some instances, officials worked to prevent family reunifications.¹⁷⁵ One immigration official wrote that reunification “obviously undermines the entire effort” and that “the expectation is that we are NOT to reunite the families.”¹⁷⁶ An estimated 3,000 children were separated from their parents.¹⁷⁷

The Family Separation Policy was unusually binding upon federal prosecutors. Traditionally, U.S. Attorneys’ Offices have exercised a large degree of independence from Justice Department officials headquartered in Washington, D.C.¹⁷⁸ Yet Attorney General Sessions overrode the traditional independence of U.S. Attorneys’ Offices,¹⁷⁹ and he publicly claimed that his zero-tolerance policy should apply to “100 percent” of referrals.¹⁸⁰ Lower-level officials were allegedly pressured to comply.¹⁸¹

Was the Family Separation Policy truly binding upon prosecutors? Some prosecutors appear to have treated it as binding. The U.S. Attorney for the Western District of Texas, who was interviewed by the Justice Department’s Office of the Inspector General, stated:

Zero tolerance, per our instructions, it’s basically sapping prosecutorial discretion . . . . [I] still can determine whether it’s a readily provable offense . . . . But even with respect to age of the child, it was a categorical, ‘We’re prosecuting all.’ . . . [N]o one in this office, including me, had any discretion on whether to accept if we were under the zero tolerance requirement from the Attorney General.¹⁸²

Other AUSAs treated the policy as partially binding, attempting to “refine[]” the policy to find a middle ground.¹⁸³ For example, some AUSAs “declined cases with very young children, [one] through six year olds,” but applied the Family Separation Policy to children older than six.¹⁸⁴

Finally, some AUSAs resisted the Family Separation Policy, arguing that it was inhumane or that it diverted resources from more serious crimes. An AUSA in the Southern District of Texas, for example, warned higher-up officials that the family separation cases had “affected [the office’s] ability to prosecute other substantive crimes.”¹⁸⁵ In some instances, prosecutors were forced to deprioritize cases with sex offenders who were then released without criminal prosecution.¹⁸⁶

Yet the fact remained that despite pushback from some U.S. Attorneys, the Family Separation Policy was unusually binding upon prosecutors. This binding nature stood in stark contrast to earlier examples of criminal enforcement, discussed below.

3. Workplace Safety Memorandum (Obama Administration)

In other instances, the president has adopted less-binding policies that preserved decision-making power for prosecutors. Consider, for example, President Obama’s Workplace Safety Memorandum from 2015.¹⁸⁷ The Workplace Safety Memorandum was part of President Obama’s larger workers’ rights agenda,¹⁸⁸ and it encouraged criminal prosecutions for serious workplace violations—for example, an employer might face criminal charges if his “willful” violation caused an employee’s death.¹⁸⁹ President Obama’s efforts were prompted, in part, by several high-profile workplace disasters, including a deadly mining explosion and an oil spill.¹⁹⁰

The Workplace Safety Memorandum encouraged criminal referrals from the Department of Labor to the Justice Department.¹⁹¹ The memorandum listed the specific “points of contact” at each agency who would handle referrals,¹⁹² implemented training programs for agency staff,¹⁹³ and emphasized data-sharing between the two agencies.¹⁹⁴ Per the memorandum, the purpose of the agreement was to “aid both agencies in more effectively implementing our national workplace statutes.”¹⁹⁵

Nevertheless, the Workplace Safety Memorandum held back from binding or quasi-binding language. The memorandum avoided mandatory language and, in fact, expressly stated that it was nonbinding.¹⁹⁶ The nonbinding approach in the Workplace Safety Memorandum stands in sharp contrast to the language surrounding the Family Separation Policy, which stressed “zero-tolerance” and placed significant pressure on prosecutors.¹⁹⁷

As the Workplace Safety Memorandum illustrates, the president has a range of options for administering criminal enforcement: he can craft policies that are binding or quasi-binding upon prosecutors, or he can create less-binding policies that encourage criminal referrals without expressly binding lower-level officials.

4. Mandatory Minimum Sentences (Obama Administration)

President Obama also used criminal enforcement as a tool for leniency, not severity, by directing prosecutors to reduce their use of mandatory minimum sentences. As part of his larger reform agenda, President Obama highlighted racial disparities in policing¹⁹⁸ and proposed the “Smart on Crime” program, which aimed to reduce mandatory minimum sentences for nonviolent drug offenses.¹⁹⁹ According to President Obama, his policy sought to “[g]ive judges some discretion around nonviolent crimes” so that judges could “steer a young person who has made a mistake in a better direction.”²⁰⁰

President Obama’s policy was announced by Attorney General Eric Holder, who issued a memorandum that directed prosecutors to decline to bring charges that triggered mandatory minimum sentences for certain nonviolent drug offenses.²⁰¹ The Holder memorandum laid out four criteria, such as lack of significant criminal history and lack of significant ties to large-scale drug trafficking organizations, and directed prosecutors to avoid mandatory minimum sentences for defendants who met the four criteria.²⁰² The memorandum also urged prosecutors to reduce their use of 21 U.S.C. § 851 sentencing enhancements, which imposed longer sentences on offenders with prior convictions.²⁰³

The Holder memorandum represented a push toward leniency in criminal enforcement, rather than a push toward severity. While the empirical effects of the Holder memorandum are disputed, most scholars agree that the overall messaging of the Justice Department shifted toward leniency.²⁰⁴ As one professor put it:

The tone/attitude of DOJ ultimately matters even more than the particulars of the [Holder] memo . . . . Things got a lot more lenient during Obama’s second term in part because a signal was coming from everyone that federal prosecutors should be a lot more lenient, and the Holder memo was the most essential piece of this story for prosecutors.²⁰⁵

President Obama’s efforts show that presidentially directed criminal enforcement need not always tilt toward severity. Instead, the president can use his influence to push prosecutors towards leniency.

5. Gonzales v. Oregon (Bush Administration)

President Bush used a different variation on presidential administration when his administration challenged physician-assisted suicide in Gonzales v. Oregon.²⁰⁶ The memorandum at issue in Gonzales combined enforcement and interpretation of the criminal law, because it reinterpreted existing regulations in order to expand civil and criminal liability for Oregon doctors.²⁰⁷ If upheld, the Gonzales memorandum would have given the Bush administration the “power to criminalize” physician-assisted suicide in Oregon.²⁰⁸

The Gonzales memorandum directed the Drug Enforcement Administration (“DEA”) to intensify enforcement against Oregon doctors.²⁰⁹ According to the Gonzales memorandum, “assisting suicide” was not a “‘legitimate medical purpose’” under existing regulations for medical drug dispensing.²¹⁰ Under this interpretation, physicians who complied with the Oregon Death with Dignity Act could lose their federal medical registrations (a civil penalty) and face up to twenty years of imprisonment (a criminal penalty).²¹¹ The Gonzales memorandum also explained how Drug Enforcement Agency staff could obtain the dispensing records of Oregon doctors.²¹²

After several years of litigation, the Supreme Court enjoined the enforcement of the Gonzales memorandum in a 6–3 opinion that cited a welter of different arguments. The Court argued that the Gonzales memorandum violated the statute’s power-sharing scheme,²¹³ and it also cited an “antiparroting” canon criticized by Justice Scalia in his dissent.²¹⁴ In another part of its opinion, the Court critiqued the Bush administration’s attempts to create a new crime. According to the Court, the Gonzales memorandum “purports to declare that . . . physician-assisted suicide is a crime.”²¹⁵ If upheld, the Gonzales memorandum would have given the Attorney General a “power to criminalize” that was “unrestrained” by statutory limits.²¹⁶

Even though it was ultimately enjoined by the Court, the Gonzales memorandum revealed a powerful variant on presidential administration. By reinterpreting existing statutes, the Gonzales memorandum expanded criminal liability (in effect, allowing the president to create a new crime) while simultaneously intensifying enforcement. The Gonzales memorandum thus illustrates the potential breadth of the president’s criminal enforcement powers.

* * *

Part III provided five in-depth examples of criminal enforcement used as a tool for presidential administration. In addition, some brief examples are listed below:

The Biden administration announced a “Strategic Civil-Criminal Enforcement Policy” against environmental pollution.²¹⁷
The Obama administration used criminal enforcement to strategically crack down on national security leakers, as described by Mila Sohoni.²¹⁸
The Biden administration intensified criminal enforcement against corporate crimes and white-collar crimes.²¹⁹
The Bush administration rolled back criminal enforcement of workplace violations, and instead promoted a policy of “voluntary compliance,” as documented by Kate Andrias.²²⁰
The Obama administration created a “Financial Crimes” unit to help address the mortgage lending crisis and other financial crimes.²²¹

As Part III illustrates, criminal enforcement has become a popular tool for recent presidents from both parties. President Trump’s policies, however, have been unusual because of their bindingness upon prosecutors.

IV. Kagan’s Arguments in Favor of Presidential Control

When Kagan introduced her theory of presidential administration, she largely endorsed President Clinton’s shift toward greater presidential control. The president brought important virtues to administrative agencies, she argued, injecting them with energy, effectiveness, and democratic accountability.²²² While Kagan acknowledged some potential downsides to presidential administration, such as the threat of “presidential lawlessness,” she argued that these dangers could be managed through judicial review.²²³

Part IV critiques Kagan’s traditional normative arguments in favor of presidential administration. Kagan’s arguments apply most squarely to agency rulemaking (which is presumptively reviewable under the APA), but her reasoning applies less squarely to agency enforcement policy (which is largely exempt from judicial review).²²⁴ As a result, Kagan’s arguments are less applicable to the context of criminal enforcement policy, rather than rulemaking.

A. Kagan on the Presidential Virtues

1. Energy and Effectiveness

In her article, Kagan praised the energy and effectiveness of the president, which she contrasted with the “indecision and inefficiency” of agencies.²²⁵ Agencies could be slow and disorganized, she argued, because they were “collective entities” with multiple decision-makers and multiple layers of internal review.²²⁶ By contrast, the president was a single, “unitary” actor who was less reliant on multiple stakeholders.²²⁷ She argued that the president brought effectiveness and efficiency to the federal bureaucracy.²²⁸

Kagan acknowledged that the president’s effectiveness came with a dark side: it could be “threatening” or even dangerous when it served bad ends.²²⁹ As Kagan noted:

There remains, of course, a question whether to count this presidential capacity for leadership over administration as virtue or vice, as a promise or a danger. Some will say that it is neither—or to state the point another way, that it is contingently both. The desirability of such leadership depends on its content; energy is beneficial when placed in the service of meritorious policies, threatening when associated with the opposite.²³⁰

Yet despite acknowledging the potential downsides of presidential effectiveness, Kagan nevertheless defended her model of presidential control. In Kagan’s telling, the president “alone” had the ability to drive effective policymaking in administrative law.²³¹

2. Democratic Accountability

Kagan emphasized a second virtue of the president’s—democratic accountability—because she believed it brought much-needed legitimacy to the administrative state.²³² As Kagan noted, most agency officials are unelected and have no “electoral link” to the public.²³³ Due to their lack of electoral accountability, agencies might be less responsive to voters’ concerns.²³⁴ Kagan frequently cited agency capture theorists, who worried that agencies would tend to fall under the sway of narrow interest groups.²³⁵

To counter the dangers of agency capture, Kagan promoted strong oversight from the president. She argued that the president is less likely to be captured by local or “parochial” interests, because he is electorally beholden to a “national” constituency.²³⁶ During his first term, the president depends on voters for reelection, and during his second term, he retains “strong incentives” to appeal to a national audience.²³⁷ According to Kagan, the president is more responsive than agencies to the “preferences of the general public.”²³⁸

B. Objection: The Threat of “Presidential Lawlessness”

Even though she praised the president’s virtues, Kagan acknowledged that the president might also pose a threat of “lawlessness.”²³⁹ In her article, she discussed an objection from Professor Peter Strauss, who had critiqued President Clinton’s moves toward greater presidential control.²⁴⁰ As Strauss noted, President Clinton had taken a “proprietary interest” in rulemakings that went far beyond the actions of any prior presidents.²⁴¹ Strauss argued that President Clinton had politicized rulemaking and eroded the delicate balance of power between agencies, the president, and Congress.²⁴² Per Strauss, “[t]he [p]resident as lawmaker is more hazardous than the Environmental Protection Agency . . . as lawmaker, precisely because he is omnicompetent, remote from effective check by courts or even Congress.”²⁴³ In Strauss’ view, President Clinton had threatened important democratic values by claiming rulemakings as his own.²⁴⁴

When Kagan discussed the critique from Strauss, she acknowledged that the president might pose a threat to “rule of law” values.²⁴⁵ As she noted, the president had a tendency to “push the envelope” on statutory interpretations and to adopt more aggressive interpretations than an agency would adopt on its own.²⁴⁶ This tendency, she argued, could create a “threat of presidential lawlessness,” in which the president pushed agencies far beyond their statutory mandates.²⁴⁷ To address the threat, Kagan proposed a solution: judicial review.²⁴⁸

C. Kagan’s Response to the Objection: Judicial Review

Kagan argued that judicial review could assuage Strauss’s concerns, because the courts would limit the president’s most aggressive uses of agency rulemaking.²⁴⁹ As Kagan acknowledged, the president’s actions are not directly reviewable under the APA, because the president is not considered an “agency.”²⁵⁰ However, Kagan argued that the president’s actions are indirectly reviewable,²⁵¹ because the APA establishes a broad presumption of reviewability for agency action, including action undertaken at the president’s behest.²⁵² Judicial review will still apply, albeit indirectly, because the agency’s (subsequent) action is reviewable even if the president’s (initial) action is not.

Kagan’s reasoning works best when applied to rulemaking, for example, as depicted below:

In the diagram above, the president directs an agency to engage in a rulemaking.²⁵³ While the president’s initial order is nonreviewable under the APA,²⁵⁴ the subsequent agency actions are reviewable—thus subjecting the president’s actions to indirect judicial review. According to Kagan, this indirect judicial review helps assuage Strauss’s concern about overly aggressive presidential actions.²⁵⁵

D. Enforcement as an Exception to Judicial Review

1. The Chain of Enforcement Decisions

Kagan, however, largely overlooked enforcement. Her reasoning applies most squarely to rulemaking (which is reviewable), yet her reasoning does not apply to enforcement policy (which is largely unreviewable).²⁵⁶ As a result, courts typically cannot review presidentially directed criminal enforcement choices, as depicted below:

This diagram depicts the chain of enforcement decisions which precede a criminal prosecution. On the left, the president directs both the Justice Department and an agency. These presidential orders are nonreviewable under the APA.²⁵⁷ Next, on the bottom, the agency selects enforcement targets in accordance with the president’s order (again, the agency’s choice of enforcement targets is nonreviewable).²⁵⁸ In the next step, the agency opens an investigation into enforcement targets, then criminally refers cases to the Justice Department (again, these agency actions are nonreviewable because the actions are not final and because they involve the agency’s enforcement discretion).²⁵⁹ Finally, at the top, the Justice Department decides whether to bring criminal charges, a decision which is largely unreviewable as well.²⁶⁰

In this diagram, the president faces few judicial checks. While each individual criminal defendant will have the opportunity to go to trial (or to negotiate a favorable plea),²⁶¹ the president’s policy decision—to intensify criminal enforcement in the aggregate—is not subject to review.²⁶²

2. Nonreviewability of Agency Enforcement Discretion

In administrative law, enforcement is largely insulated from judicial review.²⁶³ Rachel Barkow has commented on the courts’ reluctance to second-guess agency enforcement decisions:

Most aspects of agency enforcement policy generally escape judicial review. A decision not to enforce is presumptively unreviewable under Heckler v. Chaney, as is an agency’s decision not to monitor those it is charged with regulating. Courts tend to steer clear of second-guessing an agency’s selection of which actors to target and which to ignore. The judiciary takes a similarly hands-off approach to reviewing an agency’s broader plans for how it will proceed with enforcement, changes in its nonbinding enforcement policies, or how it will allocate funds from a lump-sum appropriation for enforcement needs.²⁶⁴

In administrative law, then, agencies have great freedom to set their enforcement priorities. Agencies can decide whom to bring enforcement actions against, how severely to enforce, and whether to enforce criminally or civilly.²⁶⁵ And agencies can make these decisions largely unencumbered by judicial review.²⁶⁶

In criminal law, too, nonreviewability doctrines shield prosecutors from judicial review. Prosecutors have almost total discretion over charging decisions,²⁶⁷ as long as there is probable cause.²⁶⁸ Prosecutors have the power to choose the charge, to negotiate a plea deal, or to dismiss charges.²⁶⁹ Many of these prosecutorial decisions are nonreviewable or subject only to light judicial review.²⁷⁰ While courts may overturn prosecutions that are motivated by unconstitutional motives (for example, racist or sexist motives), these challenges are “exceedingly difficult” to prove.²⁷¹ In the words of Professor Sohoni: “Both criminal law and administrative law protect . . . executive discretion with doctrines that shield enforcement policy choices from judicial review.”²⁷²

* * *

Kagan’s argument is weaker when applied to criminal enforcement. Kagan had argued that judicial review would curb the threat of “presidential lawlessness.”²⁷³ Yet enforcement is often exempt from review. As such, the president can largely evade judicial review if he opts for criminal enforcement rather than rulemaking.

V. Constraining the President’s Use of Criminal Enforcement

Part V argues for a more robust conception of mens rea, which would serve to constrain the president’s use of criminal enforcement. This more protective conception of mens rea already exists in glimmers of caselaw like Morissette v. United States²⁷⁴342 U.S. 246, 251 (1952) (using mens rea to protect the defendant from a criminal conviction and holding that a crime requires the “concurrence of an evil-meaning mind with an evil-doing hand”). and in more recent cases like the New Mexico prosecutions.²⁷⁵ This Article urges the courts to adopt this more protective conception of mens rea which can act as a constraint on the president. Under this more robust conception of mens rea, the president would be free to set general enforcement priorities, but he would be prevented from issuing binding directives that squeeze out decision-making power from lower-level officials.

A. Mens Rea as a Constraint on the President

Mens rea is a basic concept in criminal law. As every first-year law student learns, a crime involves two components: mens rea and actus reus. Actus reus, or the “act” requirement, ensures that the government cannot punish you for thought alone—you can only be punished for some voluntary act.²⁷⁶ Mens rea, also known as scienter or “guilty mind,” helps determine your degree of culpability: you are less culpable when you negligently hit someone with your car, but more culpable if you intentionally mow them down in a fit of road rage. Mens rea is often divided into four levels of intent—negligence, recklessness, knowledge, and purposefulness.²⁷⁷

The mens rea determination involves a fact-specific inquiry that hinges on details about a defendant’s mental state. For example, law students are often taught to distinguish between premeditated murder versus murder that is committed in the heat of the moment. For a defendant who killed someone in a bar fight—did the defendant pull the gun out of his pocket? Or did he walk to his car to fetch a gun, forming a pre-meditated intent in the five minutes it took him to retrieve his weapon? These kinds of fact-specific analyses are essential to mens rea determinations, which seek to peer into the mind of criminal defendants.²⁷⁸

So how does mens rea relate to the president? The point is simple—the president is rarely in a position to assess the mens rea for any given defendant, because the president is too high-level to possess the necessary facts. My claim here is epistemic: namely, the president does not have access to the requisite facts about mens rea. And the epistemic claim also carries a normative pay-off: namely, the president should avoid issuing binding directives in criminal enforcement, out of respect for mens rea. If the president attempts to bind lower-level officials, then he risks infringing upon the mens rea requirement.

This mens rea constraint may play out in different forms, depending on the context. In the lower courts, mens rea can function as a hard constraint that halts criminal prosecutions, full stop. This recently occurred in New Mexico, where a federal magistrate judge dismissed the Trump administration’s criminal charges against immigrants for lack of sufficient proof of mens rea.²⁷⁹ The magistrate judge applied an expansive conception of mens rea that was strongly protective of defendants’ rights.²⁸⁰ In one part of his opinion, the judge critiqued prosecutors for providing generalized allegations that failed to establish the mens rea for any individual defendant: while prosecutors had alleged that “‘signs’ were posted,” they had alleged no individualized facts that might bear on mens rea.²⁸¹ As the judge put it:

Beyond the reference to signage, the United States provides no facts from which one could reasonably conclude that the Defendant knew he was entering the NMNDA [New Mexico National Defense Area]. Consequently, the Criminal Complaint fails to establish probable cause to believe that Defendant knew he/she was entering the NMNDA.²⁸²

In another part of his opinion, the judge applied the “presumption in favor of scienter,” which assumes an implicit mens rea requirement even for statutes that do not explicitly include a mens rea element.²⁸³ The presumption of scienter can be used to insert a mens rea requirement into statutes that would otherwise create strict liability crimes, and it assumes that Congress did not intend to criminalize “otherwise innocent conduct”²⁸⁴ without evidence of criminal intent.²⁸⁵

In the New Mexico scenario, and other scenarios with similar fact patterns, mens rea might function as a hard constraint on the president. The constraint applies at the final step of the process, after prosecutors have already carried out the president’s binding directives. In other instances, mens rea might also more softly constrain the president’s decisionmaking prudentially. When the president issues a directive about criminal enforcement, he might acknowledge his relative lack of knowledge about mens rea, at least relative to prosecutors and agency officials who are closer to the facts. Guided by this prudential concern, the president might choose to make his directives less binding rather than more binding.

To illustrate how this prudential constraint might operate in practice, we can compare two presidential interventions: the Workplace Safety Memorandum (which was less binding), and the Family Separation Policy (which was more binding). In the Workplace Safety Memorandum, the president refrained from directing mandatory prosecutions.²⁸⁶ Instead, he allowed lower-level officials to choose civil or criminal enforcement for any given defendant.²⁸⁷ In the Family Separation Policy, by contrast, the president issued highly binding instructions that applied across the board.²⁸⁸ The Family Separation Policy was unusually binding upon prosecutors (as discussed in Criminal Enforcement as a New Tool for Presidential Control), and it arguably violated the prudential constraint of mens rea.²⁸⁹

* * *

As this section makes clear, mens rea may constrain the president in different ways, depending on the context. In some instances, mens rea may create a hard constraint, stopping criminal prosecutions in their tracks.²⁹⁰ In other instances, mens rea may exert a prudential pull on the president, nudging him toward less-binding directives that leave room for lower-level officials to exercise their discretion.²⁹¹ In both instances, mens rea acts as a constraint on the president.

B. Justice Scalia’s Dissent in Morrison v. Olson

The mens rea-based constraint is compatible with Justice Scalia’s dissent in Morrison v. Olson.²⁹² In Morrison, Justice Scalia argued that “all” executive power is vested in the president, including the prosecutorial power.²⁹³ Justice Scalia’s interpretation of the Vesting Clause creates a significant obstacle for courts that attempt to constrain the president’s use of criminal enforcement.²⁹⁴ Yet a mens rea-based constraint largely avoids the Vesting Clause obstacle, because mens rea is entirely conceptually distinct from the Vesting Clause. Even if the Justices hold a range of views about the Vesting Clause, they might still support a more protective version of mens rea.

Furthermore, Scalia’s dissent can be read narrowly—within its political context—in order to harmonize it with a mens rea-based constraint on the president. As Justice Scalia wrote, Morrison arose from a “bitter power dispute” between President Reagan and Congress, after Congress sought an independent counsel who would investigate Reagan officials.²⁹⁵ The Reagan administration challenged the Independent Counsel Act, arguing that the Act unconstitutionally diluted the president’s power over criminal enforcement.²⁹⁶ The Court ultimately upheld the Independent Counsel Act, ruling against the president in an 8–1 majority.²⁹⁷

Justice Scalia, the lone dissenter, wrote an opinion that became hugely influential in subsequent decades. He argued that the Independent Counsel Act violated the Constitution because it interfered with the president’s exercise of his “quintessentially” executive powers, including the prosecutorial power.²⁹⁸ In Justice Scalia’s view, the Constitution vested all executive powers in the president, including the power to direct criminal enforcement.²⁹⁹ Justice Scalia’s dissent—which read the Vesting Clause as granting exclusive control of prosecutions to the president—laid the intellectual foundations for modern-day unitary executive theory.³⁰⁰

Justice Scalia’s dissent repeatedly cited the political context of the separation of powers at issue in Morrison, including the incentives for Congress, the president, and the independent counsel. For example, Justice Scalia wrote that Congress might be tempted to avoid a formal impeachment if it could instead pursue an easier option via a criminal investigation.³⁰¹ As he put it: “How much easier it is for Congress, instead of accepting the political damage attendant to the commencement of impeachment proceedings against the President on trivial grounds . . . simply to trigger a debilitating criminal investigation of the Chief Executive under this law.”³⁰²

As Justice Scalia observed, the Independent Counsel Act created strong incentives for Congress. The Act allowed Congress to weaken the president without incurring the “political damage” of an impeachment.³⁰³ Just a few years later, Justice Scalia’s assessment of the incentives played out in practice, when President Clinton faced a protracted investigation from Independent Counsel Kenneth Starr.³⁰⁴

In other parts of his dissent, Justice Scalia once again appealed to the political context of Morrison. He noted that the independent counsel had only one mandate—to investigate the president—unlike most federal prosecutors who face a heavy caseload and “competing responsibilities.”³⁰⁵ Justice Scalia predicted that an independent counsel might lose “perspective” and begin to pursue relatively trivial or “technical” violations by the president and his staff.³⁰⁶ As Justice Scalia observed, the independent counsel had “nothing else to do but to investigate” the president, unlike other federal prosecutors.³⁰⁷

Justice Scalia’s dissent, then, drew heavily on the political context. He argued that Congress had a strong incentive to weaken the president without an impeachment, and that the independent counsel had a strong incentive to prolong the investigation.³⁰⁸ Crucially, Justice Scalia’s political analysis applies primarily to high-level interbranch disputes, where Congress, the president, and the independent counsel clash over investigations of the president and his staff.

So not only does a mens rea-based constraint avoid a direct confrontation with the Vesting Clause, but also this insight allows us to apply the now-popular conception of presidential control over prosecution described by Justice Scalia in Morrison more narrowly, to high-stakes disputes involving an independent counsel. If a case involves a high-stakes conflict where an independent counsel investigates the president or his staff, then Justice Scalia’s dissent weighs more heavily, because the political context is most salient. But if a case involves no independent counsel and no investigation of the president or his staff, then considerations like mens rea weigh more heavily. This approach allows the mens rea constraint to operate in its proper context and harmonizes the mens rea constraint with Justice Scalia’s Morrison dissent.

V. Conclusion

Criminal enforcement has become a powerful tool for presidential administration. Recent presidents, including Presidents Bush, Obama, Biden, and Trump, have used criminal enforcement across a wide range of policy areas.³⁰⁹ In some instances, the president has used criminal enforcement to set prosecutorial priorities, and in other instances, the president has issued binding or quasi-binding directives to prosecutors and lower-level officials.³¹⁰ This Article argues against the president’s use of binding directives, because binding directives can be constrained by the mens rea proof requirement. When the president issues binding directives, he squeezes out decision-making by prosecutors and lower-level officials who have greater knowledge about the facts, including those involving mens rea.

In response to the president’s use of binding directives, this Article proposes a more robust and protective conception of mens rea. In the lower courts, relying on the mens rea proof requirement would protect defendants during criminal prosecutions, as occurred in the recent New Mexico cases.³¹¹ At the Supreme Court, the Justices would emphasize mens rea as a signal to prosecutors and the president, urging the president to exercise restraint in his use of criminal enforcement.

Of course, it remains to be seen whether the Court might be willing to adopt a more robust conception of mens rea as a constraint on the president. On the one hand, the Court has defended the president’s “conclusive and preclusive” authority over criminal prosecutions.³¹² On the other hand, many conservative jurists are deeply committed to mens rea,³¹³ and Justice Gorsuch has spent over a decade promoting stronger mens rea protections for defendants.³¹⁴ My argument uses mens rea in a novel fashion—to protect defendants against the president’s use of criminal enforcement—and it might find traction with Justices who are attracted to mens rea and who harbor misgivings about the president’s growing use of criminal enforcement.

[*] Postdoctoral Scholar at the University of California, Berkeley at the Kavli Center for Ethics, Science, and the Public. J.D. and Ph.D., Stanford University, 2023. For helpful comments and conversations, thank you to Mariano-Florentino Cuéllar, David Engstrom, Stephen Galoob, Jon Gould, Craig Green, Dan Ho, Thomas Icard, Chris Kutz, Dorna Mohaghegh, Julian Nyarko, Anne Joseph O’Connell, Debra Satz, Bijal Shah, Glen Staszewski, Peter Strauss, Ekow Yankah, Diego Zambrano and participants at the Administrative Law Junior Scholars Roundtable and the AALS Administrative Law New Voices Workshop.

View PDF Version