abstract. In its recent decisions in Skilling, McDonnell, and Percoco, the Supreme Court has cut back on honest services fraud (18 U.S.C. § 1346)—a broad-reaching statute that federal prosecutors have used to convict corrupt governors, senators, and congressmen.

Despite its broad reach, the statutory text is remarkably thin, comprising just twenty-eight words: “For the purposes of this chapter, the term ‘scheme or artifice to defraud’ includes a scheme or artifice to deprive another of the intangible right of honest services.” The statute’s undefined terms and its perceived intrusions on federalism have invited critical scrutiny from federal judges, who have decried Section 1346 as “amorphous and open-ended,” “facially vague,” permitting a “standardless sweep” of prosecutions, and imposing an “extreme version of truth in politics.”

This Essay charts a path out of this doctrinal morass. The solution, I argue, is to interpret honest services fraud to require a predicate violation of state law. All fifty states have enacted detailed anticorruption statutes which often outstrip the reach of their federal analogues. Tethering federal honest services fraud to these state regimes closes the enforcement gap, while offering an intelligible standard that addresses both the due process and federalism concerns the Supreme Court raised in Percoco, McDonnell, and Skilling. Moreover, this approach finds support in a constitutional provision that gives Congress authority to protect against state and local corruption by criminalizing what we call “honest services fraud”: the Guarantee Clause of Article IV, Section 4.

Part I of this Essay examines the development of the honest services statute, from the Fifth Circuit’s early judicial gloss on the mail fraud statute in 1941, to the Supreme Court’s McNally decision in 1987 followed by Congress’s enactment the next year of 18 U.S.C. § 1346, and finally to the Supreme Court’s recent attempts to curtail the reach of honest services fraud. Part II introduces the Constitution’s Guarantee Clause, uncovering the Clause’s anticorruption roots and the Clause’s invocation by President Franklin Delano Roosevelt, President Lyndon B. Johnson, and then-Senator Biden to counter political corruption. Part III argues that honest services fraud should be limited by the Fifth Circuit’s pre-Skilling rule set forth in Brumley v. United States and addresses the federalism and due process concerns raised in Percoco, McDonnell, and Skilling. Part IV considers counterarguments, and the Essay then concludes.

Introduction

In a trifecta of decisions—Skilling v. United States, McDonnell v. United States, and Percoco v. United States1—the Supreme Court curtailed federal prosecutors’ ability to use the honest services fraud statute (18 U.S.C. § 1346), a once broad-reaching statute historically used to target state and local corruption by convicting corrupt governors, senators, and congressmen.2 Prosecutors have used the statute to bring charges against Illinois Governor Rod Blagojevich for conspiring to sell Barack Obama’s Senate seat;3 Virginia Governor Bob McDonnell for accepting an engraved Rolex and borrowing a Ferrari from a constituent;4 and most recently, Senator Bob Menendez for accepting gold bars and other bribes from donors.5

Despite its broad reach, Section 1346’s statutory text is remarkably thin, consisting of just twenty-eight words. The statute’s undefined terms and perceived intrusions on federalism have invited critical scrutiny from federal judges, who have decried Section 1346 as “amorphous and open-ended,”6 “facially vague,”7 permitting a “standardless sweep” of prosecutions,8 and imposing an “extreme version of truth in politics.”9 Justice Scalia called for the statute to be abrogated entirely, criticizing its coverage of a “staggeringly broad swath of behavior.”10 More recently, Justice Gorsuch remarked that “[h]onest-services fraud and this Court’s vagueness jurisprudence are old friends,” criticizing Congress’s decision to “give the Judiciary uncut marble with instructions to chip away all that does not resemble David.”11

Yet Congress has repeatedly declined invitations to clarify or abrogate the statute, leaving federal prosecutors, defendants, and courts in the lurch. With an imperfect tool for policing state and local corruption, the enforcement gap has only grown wider over time. Between 2003 and 2022, convictions of federal, state, and local officials in cases brought by federal prosecutors decreased by forty-nine percent, despite no indication that political corruption has become any less common.12 Nor is it likely that state-level prosecutions have expanded to fill the gap.13

This Essay charts a path out of this doctrinal morass. The solution, I argue, is for federal courts to interpret honest services fraud as requiring a predicate violation of state law. All fifty states have enacted comprehensive anticorruption statutes, many of which outstrip the reach of their federal analogues.14 Yet, state prosecution of state public-corruption offenses remains anemic.15 Tethering federal honest services fraud to these state regimes would close the enforcement gap, while offering a clear standard that addresses the due process and federalism concerns that the Supreme Court raised in Skilling, McDonnell, and Percoco.16

The argument for this model is grounded in two sources of authority. First, the Fifth Circuit in the pre-Skilling case of United States v. Brumley created a state-law predicate rule that was adopted by a small minority of circuits.17 Though a majority of circuits rejected Brumley as lacking constitutional grounding, this Essay argues that those circuits failed to consider an alternative font of authority: the Guarantee Clause of the U.S. Constitution.18 A little-understood constitutional provision once described by Senator Charles Sumner as a “sleeping giant,”19 the Guarantee Clause played a major role in the legislative history of the honest services statute. Though federal courts have traditionally viewed the Guarantee Clause as nonjusticiable20—a shield for the states, rather than a sword for the federal government—this Essay uncovers the original understanding of the Clause to argue that it conferred broad powers on the federal government to enforce anticorruption laws.21 On this view, the Guarantee Clause is both a sword for the federal government and a shield for the states against corrupting influences from within.

Conditioning honest services fraud on a predicate violation of state law—which I term the Guarantee Clause model—balances the need for robust federal anticorruption enforcement with the Supreme Court’s concerns about due process and federalism. This model discards Skilling’s effort to provide a “uniform national standard”22 in favor of a pluralism of standards, enabling states to experiment and capture local norms about political corruption through legislation. And it safeguards that pluralism with a federal guarantee, combining the statutory flexibility of states with federal investigative resources and detachment from local politics. The core promise of the Guarantee Clause model, in short, is that it restores power to the governed to define the standards they expect of those who govern.

This Essay proceeds in four Parts. Part I examines the development of the honest services statute—from Shushan v. United States’s judicial gloss on the mail fraud statute, to the enactment of the current honest services statute at 18 U.S.C. § 1346, to the Supreme Court’s recent attempts in Skilling, McDonnell, and Percocoto curtail the statute. Part II introduces the Guarantee Clause, uncovering the Clause’s anticorruption roots and its invocation by President Franklin Delano Roosevelt, President Lyndon B. Johnson, and then-Senator Joseph Biden to counter political corruption. Part III argues that honest services fraud should be limited by the Brumley rule, whose authority should be derived from the Guarantee Clause. Part III then demonstrates how the due process and federalism concerns raised in Percoco, McDonnell, and Skilling are addressed by the Guarantee Clause model. Finally, Part IV addresses counterarguments.

I. honest services fraud from shushan to percoco

A. Pre-McNally Intangible Services Fraud

The 1941 case Shushan v. United States, in which federal prosecutors charged members of Louisiana Senator Huey Long’s infamous political machine for their involvement in a bribery scheme, marked the first federal honest services fraud prosecution.23 The use of the mail fraud statute to bring these charges was novel: since that statute’s enactment in 1872, no court had held that mail fraud extended to anything beyond traditional pecuniary fraud.24 But, in a groundbreaking decision, the Fifth Circuit upheld the charges, holding that the mail fraud statute extended to schemes to defraud the public of the intangible right to good government.25

By 1987, every federal circuit court adopted Shushan’s gloss on the mail fraud statute.26 The warm judicial reception to the expansive reading of mail fraud was perhaps a reaction to the political atmosphere. Rocked by Watergate, the Vietnam War, and civil-rights protests, public trust in government in the 1970s fell to historic lows.27 Other branches of the federal government were eager to reverse that trend. In 1976, the Justice Department created the Public Integrity Section, its first-ever dedicated public-corruption unit.28 At the same time, Congress enacted powerful new statutory authorities, including the Racketeer Influenced Corrupt Organizations (RICO) Act in 1970,29 the Foreign Corrupt Practices Act in 1977,30 and the federal-program bribery statute in 1984.31

The Shushan theory of mail fraud proved to be tremendously popular among prosecutors, despite the availability of other public-corruption statutes.32 Unlike RICO (18 U.S.C. § 1962), honest services fraud did not require proving multiple predicate acts, and, unlike federal bribery (18 U.S.C. § 201), honest services fraud extended to state and local officials.33 Deprivation of the “intangible right to honest services” was a capacious concept, one that the Justice Department was eager to begin pursuing. By 1987, the Justice Department had secured honest services convictions of two governors, and many more state judges, party chairmen, and local aldermen.34

B. McNally and the Congressional Response

Then in 1987, the Supreme Court decided McNally v. United States. Overturning the interpretation of every circuit court, McNally held that the mail fraud statute covered only “money or property,” not intangible rights to honest government.35 The decision landed like a lead balloon. In fiery congressional hearings, the Justice Department reported that McNally would shutter 186 convictions and 98 ongoing investigations.36

The Justice Department’s admonition spurred Congress to engineer a legislative fix. The task fell to Representative John Conyers, then chairman of the House Subcommittee for Criminal Justice. Conyers, a progressive civil-rights figure, was an unlikely proponent for expanding criminal prohibitions.37 But he quickly became one of the Justice Department’s most ardent backers, introducing one of the first bills to overturn McNally just five weeks after the decision was announced.38

On the House floor, Conyers was the first to introduce the idea of grounding the honest services fraud statute in the Guarantee Clause of the Constitution.39 The novel proposal soon caught fire. John Keeney, the acting head of the Justice Department’s criminal division, testified in its support.40 Conyers’s proposal also had the support of Keeney’s predecessor, Bill Weld—a particularly credible voice on the subject, as Weld had just resigned from the Justice Department in protest of the Attorney General’s alleged ethical violations.41 Several months later, then-Senator Joe Biden took to the Senate floor to back Conyers’s proposal, repeating Conyers’s argument regarding the Guarantee Clause.42 The Senate enacted Biden’s version of the bill on October 14, 1988, and sent the bill to the House Judiciary Committee five days later, on October 19.43 Two days later, on the final day of the legislative session, Congress enacted the Omnibus Anti-Drug Act with Biden’s honest services statute attached as a rider.44

Yet, despite its prevalence throughout the legislative history, the Guarantee Clause basis was never mentioned in the final conference report.45 Nor was the Guarantee Clause ever mentioned in the enacted version of the honest services statute, as it had been in earlier iterations of Conyers’s bill.46 One potential reason for this omission, both in the text of the statute and in the final legislative report, was legislative delay. By the time the House received Biden’s bill, Congress was already two weeks past a scheduled adjournment.47 Many were eager to wrap up the legislative session, and over a hundred members had already left for home.48 Facing the loss of a quorum if there were further delays, the House version of the Omnibus Act eventually passed by voice vote, after only two days of consideration.49

For his part, Conyers never had a change of heart. On the eve of the Omnibus Bill’s enactment, Conyers took to the House floor once more to repeat his belief in the Guarantee Clause as the constitutional basis for Congress’s authority to enact the honest services prohibition.50

C. Circuit Splits and the Supreme Court’s Response in Skilling, McDonnell, and Percoco

Intentional or not, the failure of Congress to invoke the Guarantee Clause proved consequential as it deprived the lower courts of a guiding principle for interpreting the meaning of “honest services fraud.” After being repudiated by Congress, the Supreme Court handed the reins back to the lower courts to interpret the metes and bounds of 18 U.S.C. § 1346. These courts struggled to decipher Congress’s intent from the barebones statute, and the doctrinal confusion resulted in a kaleidoscope of differing approaches. One set of circuits imposed a “foreseeable harm” test51 that another set of circuits rejected.52 The Seventh Circuit required a showing of private gain to the defendant, while its sister circuits did not.53

One early split emerged between the Fifth Circuit and the Ninth Circuit on whether honest services fraud requires a predicate violation of a duty arising under state law. In United States v. Brumley, the en banc Fifth Circuit answered the question affirmatively and justified its limiting rule solely on federalism grounds.54 The Brumley court expressed doubt that Congress wanted to “garner to the federal government the right to impose upon states a federal vision of appropriate services . . . in other words, an ethical regime for state employees.”55 The Third Circuit later concurred with the Fifth’s approach in United States v. Murphy, sharing similar concerns about federalism and fair notice.56

The Ninth Circuit in United States v. Weyhrauch agreed in principle that Brumley’s approach would satisfy many of the concerns raised about federalism, fair notice, and potential for partisan abuse.57 But the Weyhrauch court reasoned that it could not find “any basis in the text or legislative history of § 1346” for the Brumley rule and that no pre-McNally case law had adopted the Brumley approach.58 The Ninth Circuit was eventually joined by four other circuits—the First, Sixth, Seventh, and Eleventh.59

By 2009, the mess in the lower courts spurred some Justices to advocate for the Supreme Court to reenter the fray. Dissenting from the denial of certiorari in Sorich v. United States, Justice Scalia decried the “staggeringly broad swath” of the honest services statute, arguing that the statute seemingly encompassed everything from “a mayor’s attempt to use the prestige of his office to obtain a restaurant table without a reservation” to an “employee’s phoning in sick to go to a ball game.”60 Judge Jacobs of the Second Circuit asked: “How can the public be expected to know what the statute means when the judges and prosecutors themselves do not know, or must make it up as they go along?”61

The public would not have to wait long for an answer. The next year, in Skilling v. United States—a case arising from the implosion of Enron—the Court unanimously agreed to limit honest services fraud to the pre-McNally “core” cases of bribery and kickbacks.62 Skilling foreclosed a larger body of cases involving undisclosed self-dealing and conflicts of interest, which, as the U.S. Attorney for the Northern District of Georgia described it, cast a “chilling effect” on honest services prosecutions.63 To many, the Court’s decision to “construe, not condemn” the statute departed from the usual rules of statutory interpretation.64 As Scalia observed in his concurrence, “Among all the pre-McNally smorgasbord-offerings of varieties of honest services fraud, not one is limited to bribery and kickbacks. That is a dish the Court has cooked up all on its own.”65

Despite Skilling’s intent to “establish a uniform national standard” that “define[d] honest services with clarity,”66 the decision failed to address the preexisting circuit splits, including the Brumley-Weyhrauch split.67 Skilling also failed to clarify from where the definition for “bribery” or “kickbacks” would derive—federal or state law, or some version of common law.68 Justice Scalia’s concerns would be vindicated, as the Supreme Court was forced to intervene again six years later.

In 2016, a unanimous Supreme Court in McDonnell v. United States overturned the honest services conviction of the former governor of Virginia.69 Governor McDonnell and his wife received $175,000 in gifts, including a Rolex and a loaned Ferrari, allegedly as bribes to induce the Governor to arrange government meetings to promote a company’s tobacco product.70 Invoking due process and federalism, McDonnell held that a public official’s “official acts” for bribery must be defined narrowly.71 McDonnell was an unusual honest services prosecution in that the case turned on the interpretation of another statute—the federal bribery statute, which ordinarily only applies to federal officials.72 However, the parties stipulated to importing § 201’s definition of bribery into § 1346.73 Hence, McDonnell’s holding is arguably limited to honest services prosecutions where the parties import § 201’s definition of bribery.74

Seven years later, the Court stepped in again in its decision in Percoco v. United States. Joe Percoco, a high-level aide to Governor Andrew Cuomo of New York, received $35,000 for intervening on behalf of a real-estate developer while he took an eight-month hiatus from government service to work on Cuomo’s reelection campaign.75 The Court unanimously overturned the conviction, holding that the Second Circuit’s test for when private individuals owe honest services to the public, which was first set forth in the 1981 Margiotta decision, was unconstitutionally vague.76 Though Percoco abrogated Margiotta, it left no alternative in its place, simply punting the question to the Second Circuit.77

Despite the rare consensus across three opinions on the same question in two decades, Skilling, McDonnell, and Percoco failed to craft a lasting solution to cure honest services fraud. Ideally, as Justice Gorsuch suggested, Congress would return to the drawing board and recraft the honest services statute with greater precision.78

But hope for a congressional intervention increasingly seems to be a pipe dream. As Congress’s overall productivity has declined, so too has the number of congressional overrides of Supreme Court statutory-interpretation decisions, which decreased significantly from 1988 to 2011.79 At its peak, Congress enacted over thirty-five overrides in the 104th Congress (1995-96). Between the 106th (1999-2000) and 112th Congresses (2011-2012), that number fell to below an average of five per year.80 If past is prologue, the Supreme Court will soon find itself entering the fray again, drawing another temporary and blurry line in the sand.

II. the anticorruption guarantee clause

This Part argues that the Court may look to a preexisting solution to the honest services fraud problem: the Fifth Circuit’s rule in Brumley, which addressed the fair-notice and federalism concerns first raised by McNally and subsequently discussed in Percoco, McDonnell, and Skilling. Though Brumley was rejected by other circuits as lacking a basis in the Constitution or in statutory text, this Part argues that Brumley’s rule can be grounded in the Guarantee Clause.

A. James Madison and the Guarantee Clause

The United States shall guarantee to every State in this Union a Republican Form of Government, and shall protect each of them against Invasion; and on Application of the Legislature, or of the Executive (when the Legislature cannot be convened) against domestic Violence.81

Once described by Senator Charles Sumner as a “sleeping giant in the Constitution,”82 the Guarantee Clause is perhaps among the least discussed provisions of the Constitution. Nevertheless, some leading constitutional theorists have attempted to decipher its meaning.83 This Essay does not attempt to quibble with any of those interpretations. As Professor Akhil Amar has remarked, “Republican Government . . . is indeed a spacious [concept], and many particular ideas can comfortably nestle under its big tent.”84 This Part merely proposes one additional idea to nest under the tent: the concept of an anticorruption Guarantee Clause.

The Constitution was influenced in large part by fears of political corruption. Corruption, bribery, and foreign influence were topics of discussion in over a quarter of the days of the convention.85 Many early public scandals—including an incident where Benjamin Franklin accepted a diamond-encrusted painting from the King of France—revealed the need for provisions like the Foreign Emoluments Clause.86 Many other provisions of the Constitution, from the prohibition of ennoblement, provisions for impeachment, and the residency requirements for election, take root in fears of foreign and private corruption.87

The Guarantee Clause reflected a similar distrust of corruption among state and local lawmakers.88 James Madison, the principal drafter of the Guarantee Clause, grew deeply disillusioned with the rank corruption of state legislators.89 As a Virginia legislator, Madison witnessed his fellow state legislatures enact “paper money” legislation to alleviate the private debts of indebted merchants, driving the states into financial ruin.90 In one particularly egregious post-Convention example, members of the Georgia legislature—nearly all of whom had been bribed—gifted away thirty-five million acres of land in present-day Mississippi and Alabama for two cents an acre.91

It was clear to Madison that if the nascent union were to survive, the states would need protection from malefactors within. As Professor Ralph Ketcham, a leading Madison historian, wrote: “The states left to themselves, Madison concluded, seemed invariably to trample on both private rights and the public good.”92 The laws passed by these state legislatures “bring[] . . . into question the fundamental principle of republican Government.”93

Madison proposed the Guarantee Clause at the Constitutional Convention.94 Explaining his reasoning in Federalist 43 and in a private memorandum, Vices of the Political System of the United States, Madison made three principal arguments.

First, federal intervention would be necessary to secure republicanism. Consequently, the Guarantee Clause vests the federal government with “authority to defend the system against aristocratic [] innovations” against the “ambition of enterpri[s]ing leaders, or [against] the intrigues and influence of foreign powers.”95

Second, “republicanism” would be defined not only by the right to elect one’s own representatives but also by the ability to exercise that right unencumbered by “aristocratic or monarchical innovations”—or what we now might call political corruption.96 Madison warned that a minority equipped with “superiority of pecuniary resources” and “secret succours [sic] from foreign powers” could destroy a republic as effectively as “an appeal to the sword.”97

Third, the federal guarantee must extend to “republican government” as defined by the state itself. Madison bemoaned the “[w]ant of Guaranty” under the Articles of Confederation “to the States of their Constitutions & laws against internal violence.”98 Despite his suspicion of state legislatures, Madison did not advocate for a uniform definition of republican government. Rather, the guarantee “supposes a pre-existing government of the form which is to be guaranteed.”99 States under the Guarantee Clause “may choose to substitute other republican forms,” backed by the federal guarantee, so long as the substitute form is not “antirepublican.”100

B. Awakening the “Sleeping Giant”: Twentieth-Century Invocations

While the anticorruption model of the Guarantee Clause lay dormant for most of its history, it enjoyed a brief revival in two instances. In 1935, President Roosevelt explored the prospect of invoking the Guarantee Clause in his quest to oust the populist senator and Roosevelt critic, Huey P. Long.

Long, a former governor and later senator from Louisiana, was a singular figure in American history. An economic progressive who pushed for wealth redistribution during the height of the Great Depression, Long was also widely reviled as a quasi-dictator. During his reign, Long abolished judicial review over voter registrars—enabling him to push his preferred cronies into office101—established a censorship board with the power to review and edit all movies published in the state,102 and declared martial law over the state.103

In response to Long’s abuses, the House of Representatives established a select committee to investigate whether Long’s “corrupt political machine” had caused Louisiana to lose its “republican form of government” under the meaning of Article IV.104 President Roosevelt expressed enough interest in the idea of invoking the Guarantee Clause to justify federal intervention, that he eventually commissioned a Justice Department memo to investigate the idea.105 In press conferences, President Roosevelt frequently raised the idea of invoking the Guarantee Clause in dialogue with journalists.106 The question was rendered moot, however, as Long was assassinated a week after the formation of the House committee.107

The second revival of the Guarantee Clause occurred three decades later. In 1966, the National Commission on Reform of Federal Criminal Laws drafted a proposal to enact a federal statute targeting state and local corruption, pursuant to the Guarantee Clause.108 The Commission, colloquially known as the “Brown Commission” after its chair, California Governor Pat Brown, was a leading body of jurists and law professors appointed by President Johnson to propose an overhaul to the federal criminal justice system. One of the Brown Commission’s proposals was to enact a federal public corruption statute which would have read, in relevant part, that:”[b]road federal jurisdiction in this area [of local bribery and corruption] might be rested on Article 4, Section 4 of the Constitution . . . as a power to preserve the states from any intrusion of nonpolitical pecuniary influences into government.”109

For unexplained reasons, the final Brown Commission report omitted any reference to the Guarantee Clause. But Representative Conyers would later credit the Commission’s draft as inspiration for his own proposed legislation that later became the honest services fraud statute.110

In sum, leading figures across all levels of government—Representative Conyers, Senator Biden, Governor Brown, and President Roosevelt—have all attempted to pull the Guarantee Clause’s sword from stone to prosecute political corruption. These efforts have all failed, in large part due to historical happenstance—such as Huey P. Long’s assassination and the legislative procrastination of the 100th Congress. As a result of these seeming historical accidents, the anticorruption Guarantee Clause has lain dormant, without occasion for federal courts to consider its scope.

These past failures should not hinder federal courts from considering the anticorruption Guarantee Clause today. The next Part proposes a new model of honest services corruption premised on the Guarantee Clause. This formulation of the Guarantee Clause rectifies the concerns regarding due process and federalism raised in Skilling, McDonnell, and Percoco, and it avoids the nonjusticiability concerns associated with the Guarantee Clause.

III. a new model of honest services fraud

A. The Anticorruption Guarantee Clause

Under the Guarantee Clause model, Article IV, Section 4 serves as a source of authority—independent of the Postal Clause, Commerce Clause, and Spending Power—for the federal government to prosecute public-corruption offenses.111 This model can be loosely analogized to the corporate derivative suit. Just as shareholders may sue on behalf of corporations whose directors breach their fiduciary duties, the federal government may intervene when state officers have breached their duties under their state constitutions and laws. The justification for intervention is the same in both cases: because internal actors cannot be trusted to discipline themselves, external intervention is necessary to restore faithful leadership.

The lynchpin of the Guarantee Clause, as discussed in Part III, is the federal “Guaranty to the States of their Constitution & laws against internal violence.”112 This vision of republican government is the strongest constitutional argument for adopting the Brumley rule. The practical effect is that when federal courts interpret § 1346, they should do so with the goal of “implementing” the constitutional norms of the Guarantee Clause in mind.113 The Brumley rule is the best expression of the constitutional norms of the Guarantee Clause, which extends the federal guarantee to republicanism as defined by the state itself.

This vision of the Guarantee Clause ameliorates the often-raised justiciability concerns with the Guarantee Clause. Courts have traditionally understood the Guarantee Clause to be nonjusticiable, ruling that determinations of whether a state has a “republican government” is a political determination lacking judicially manageable standards.114 But the anticorruption Guarantee Clause proposed here does not implicate these concerns because federal courts are merely enforcing the states’ own constitutional and statutory definitions and protections of republican government. In particular, state anticorruption prohibitions provide ample standards for federal courts to apply because state statutes are often far clearer and more specific than their federal equivalents.115

Another benefit of the Guarantee Clause model is that it would refocus honest services prosecutions on individuals with a nexus to public governance.116 Since Shushan, honest services doctrine has been divided between public- and private-fiduciary lines of cases, with the latter attracting much more judicial skepticism.117 The Guarantee Clause model would focus on the public line of cases, including private actors whose corruption touches on public government but excluding purely private deprivations of honest services.

Refocusing honest services fraud on public rather than private corruption would likely cause minimal enforcement harm, as § 1346 is rarely the lead charge in private fraud cases. Of the eighteen honest services cases referred for prosecution in fiscal year 2024, only three related to purely private fraud.118 This stands to reason: when private fiduciaries commit fraud, prosecutors typically have more traditional criminal laws at their disposal (mail fraud, wire fraud, securities fraud, bank fraud, and federal program bribery). Moreover, civil law enforcement (e.g., action by the Securities and Exchange Commission or sanctions by licensing boards) remains a powerful deterrent against private fiduciaries, with a lower burden of proof.119

In sum, the Guarantee Clause model empowers the federal government to prosecute state and local officials who have deprived citizens of their honest services, as defined under state anticorruption laws. The following Section argues that this model would close the enforcement gaps created by Skilling, McDonnell, and Percoco, while addressing the Supreme Court’s concerns regarding due process and intrusions on federalism.

B. Addressing the Court’s Line of Honest Services Cases

Turning to contemporary cases, this Section applies the Guarantee Clause model to the Supreme Court’s three most recent honest services cases: Skilling, McDonnell, and Percoco. As will be shown, the Guarantee Clause model addresses the Court’s due process and federalism concerns, while closing the enforcement gaps left by those cases.

1. Percoco (Private Influence over Government Decisions)

Percoco dealt with the question of defining when former public individuals like Joe Percoco, a former top aide to Governor Andrew Cuomo, owe fiduciary duties the public under § 1346. The Court held that the Second Circuit’s Margiotta test relied upon by the Government—asking whether the private individual “dominates government” or exercised “substantial control over public officials”—failed to provide a clear standard with “sufficient definiteness that ordinary people can understand what conduct is prohibited.”120

Where the Margiotta test may have failed, state criminal codes can step in. Forty-four states and territories have enacted “revolving door” statutes restricting former public officials from representing private clients before local agencies.121 Most of these statutes specify the categories of public employees covered (elected or unelected employees; legislative, executive, or judicial), require cooling-off periods (ranging from one year to permanent bans), and bar some types of representations.

table 1. sample prohibitions on activities after government service (new york)122

Percoco himself would have fallen under the New York Public Officers Law § 73(8)(a)(iv), which prohibits employees who have served in the office of the governor from representing clients before any government agency for two years. In addition to these “revolving door” laws, fifty states and territories have enacted insider-trading statutes that prohibit current or former government employees from disclosing or trading on confidential information obtained from their public service.123

As described above, state laws are far more granular and detailed than federal honest services fraud in describing what conduct is prohibited. Tethering honest services to state law would address many of the concerns raised by Percoco about the need for ex ante notice and due process.

2. McDonnell (Official Acts)

In McDonnell, a unanimous Supreme Court ruled that a narrow conception of honest services bribery was necessary to exclude from criminal liability the activities of everyday politics (e.g., invitations to attend a ballgame or making a call on behalf of a constituent).124 “Federal [g]overnment . . . setting standards of good government for local and state officials” would cast a pall on democratic activities.125 The Court viewed it as “the prerogative [of the states] to regulate the permissible scope of interactions between state officials and their constituents.”126

One of the most conspicuous difficulties with the decision in McDonnell is that it defined bribery by reference to a federal statute that did not apply to state officials, while ignoring a bevy of state statutes that did.127 Fifty-four states and territories have statutes defining bribery of state and local officials. Of these states, only six follow the narrow federal “official acts” language. The remaining forty-eight largely apply broader language that extends to “exercises of discretion,”128 or enumerate specific types of actions that fall within the statute (e.g., interference in an investigation or decision to incarcerate).

table 2. state and territorial definitions of bribery129

As in Percoco, an advantage of the Guarantee Clause model in the McDonnell context is that the federal courts would be applying statutory language that state or territorial legislatures had already blessed. McDonnell expressed concern that federal judges, who rarely have experience in electoral politics, may begin to criminalize politics as usual.136 These concerns are readily addressed when federal prosecutors are holding state legislators to the standards that they have enacted themselves—which in forty-eight states exceed the federal standard.

3. Skilling (Undisclosed Self-Dealing)

Lastly, the Guarantee Clause model resolves one of the main enforcement gaps created by Skilling, which eliminated the use of § 1346 to target undisclosed self-dealing by public officials.137 Fifty-four states and territories have enacted statutes addressing undisclosed self-dealing by public officials.138 These statutes largely focus on nondisclosure from public officials, not private citizens. Recentering honest services fraud on the Guarantee Clause would also create a principled line to limit the scope of honest services fraud, rather than following the artificial “bribes and kickbacks” line drawn by Skilling.139

C. Toward a Democratic Theory of Honest Services Fraud

Ultimately, the core promise of the Guarantee Clause model is that it democratizes federal policing of state and local political corruption. Twenty million Americans are employed as civil servants across nearly twenty thousand municipalities across the country.140 Surely, there is consensus that corruption should be punishable by law. But reasonable differences of opinion exist as to when, whether, and how criminal punishment should be meted out.141 Should tribal employees be subject to the same antinepotism laws as state officials in Albany? Should postal workers be subject to the same conflicts-of-interest laws as the governor? When does a broken campaign promise—perhaps considered by many as mere political puffery—become honest services fraud?

It is practically impossible for federal courts to decide these questions while remaining “faithful agents” to the twenty-eight words of the honest services statute. State legislatures are not so bound. Despite Madison’s antipathy toward state legislatures, state lawmakers have proven to be far more productive legislators than their federal counterparts. New York, for example, has enacted statutes punishing sixteen different forms of public bribery and seventeen forms of private bribery, including sports bribery, labor bribery, and horse-racing bribery.142

To be sure, courts have expressed concern that the Brumley rule may generate undesirable circuit splits.143 Why should one act be deemed criminal political corruption in the Fourth Circuit, punishable by up to twenty years in prison, but treated as a civil regulatory violation in the Fifth Circuit, where the maximum penalty is a fine? Skilling bemoaned such “intercircuit inconsistencies,”144 which seem to strike at the basic precept that like offenses should be treated alike.

One answer is that such divergence is already prevalent across federal criminal law. Three of the most frequently employed statutes—federal money laundering,145 the Travel Act,146 and RICO147—are partially predicated on state criminal law. I am aware of no evidence demonstrating that circuit splits over these statutes have hampered federal law enforcement. Another response is that state-by-state variations reflect the legitimate policy choices of the electorates. Just as the norms of a particular tribal government may differ from those of legislators in Albany, state legislatures should have free reign to craft divergent anticorruption laws that reflect local priorities.

In short, the core promise of the Guarantee Clause is that it restores power to the governed to define the standards they expect of those who govern. It discards Skilling’s “uniform national standard” in favor of a pluralism of opinions, enabling states to experiment and capture local sentiments. And it safeguards that pluralism with a federal guarantee, combining the statutory flexibility of states with federal investigative resources and detachment from local politics.

IV. defending the guarantee clause model

Given the Guarantee Clause’s long dormancy, this attempt to restore it will likely attract many critiques. Below are responses to four of the most likely objections.

A. The “Wolves in the Henhouse” Problem

One potential critique of the Guarantee Clause model is that it creates a “backsliding” problem. State legislators, aware that they are enacting statutes that may one day be cited in a federal indictment against them, have a perverse incentive to lower or abolish state anticorruption laws. In short, the Guarantee Clause arguably allows the wolves to guard the henhouse.

This critique is misplaced. As discussed in Part III of this Essay, existing state antibribery laws generally exceed federal standards.148 Repeals of anticorruption laws have proved to be deeply unpopular. In twenty-seven states, voters also have the right to place anticorruption measures directly on the ballot, and have exercised that right to enact anticorruption laws from Tallahassee, Florida, to San Francisco, California.149 Statewide referenda are not always successful—for example, in South Dakota, lawmakers successfully repealed a voter-backed referendum that would have imposed anti-revolving-door rules and limited lobbyist gifts.150 But statewide referenda have remained robust, despite efforts by lawmakers to curtail their availability.151

Lastly—though not addressed in depth in this Essay—the Guarantee Clause likely also imposes a floor of republicanism that states may not fall below. As Madison wrote in the Federalist Papers, states “shall not exchange republican for anti-republican Constitutions.”152 This limitation could be invoked to prevent the most egregious abuses of state government.

B. Federal Officials

Another critique meriting consideration is that, while the Guarantee Clause model may justify robust federal prosecution of state and local officials, it does not justify federal prosecution of federal officials. Although prosecutors have predominantly used honest services fraud to target state and local officials, federal officials have occasionally been prosecuted for honest services fraud.153 Under this critique, those prosecutions are unjustified under the Guarantee Clause, as nothing in the text or history of the Guarantee Clause indicates that the Framers intended for the Guarantee Clause to authorize the federal government to prosecute its own.

But there is no need for a separate source of authority for these types of prosecutions because federal prosecutions of federal officials do not implicate the concerns raised in Skilling, McDonnell, and Percoco. Grafting federal definitions of bribery onto honest services prosecutions of federal government defendants does not raise the same due process concerns implicated in McDonnell. Nor does prosecuting federal officials raise the same federalism concern as raised in Percoco.

Moreover, the Supreme Court has frequently reaffirmed the broad authority of the federal government to pass statutes that help “carry[] into Execution” its enumerated powers.154 In United States v. Comstock, which concerned the federal government’s power to impose civil confinement on mentally ill defendants past their term of incarceration, the Supreme Court recognized that the federal government has “broad authority” to enact criminal laws under the Necessary and Proper Clause.155 Though “Congress’ power to criminalize conduct [and] its power to imprison individuals” is nowhere mentioned in the Constitution, there is no dispute that the Constitution vests the federal government with that authority.156 Hence, many criminal statutes—like 18 U.S.C. § 1001 (prohibition of false statements)157 and 18 U.S.C. § 371 (so-called Klein conspiracies)—are given wide berth by the courts because the federal government must be able to protect itself from fraud and false statements in order to “carry[] into Execution” its enumerated powers.158

So too with honest services fraud. Just as Section 371 and Section 1001 protect the federal government from external threats to its powers, Section 1346 protects the federal government from internal threats. It would certainly be a strange scenario if the federal government lacked the authority to police corruption within, at least to the same extent that it is permitted to police corruption externally.

C. Justiciability

Another likely critique is that the Guarantee Clause has traditionally been viewed as a nonjusticiable political question. On this view, the Guarantee Clause cannot serve as an independent jurisdictional basis for honest services fraud because courts are to avoid any effort to decide what constitutes republican government.

The genesis of this view traces back to the 1849 case of Luther v. Borden.159 There, the Supreme Court was tasked with deciding between two competing factions—the charter government of Rhode Island and the leaders of the Dorr Rebellion—both of whom claimed to be the legitimate government of Rhode Island. Rather than decide the case on the merits, the Court held that this was a “political question[],” reserved for the legislative and executive branches of the United States.160 On the Guarantee Clause, the Court held that the determination of what constitutes republican government is similarly a political question reserved for Congress.161

Many scholars have disputed the traditional account of Borden, arguing that Borden’s discussion of the Guarantee Clause was dicta162 or that the Guarantee Clause is justiciable for other reasons.163 But even taking the traditional view on its face, there is no reason why this Essay’s vision of honest services fraud must be deemed a nonjusticiable political question. The canonical test for what constitutes a political question comes from Baker v. Carr, where the Court set forth six factors to consider:

1. Whether there is a “textually demonstrable commitment of the issue to a coordinate political department”;

2. “[L]ack of judicially discoverable and manageable standards”;

3. “[I]mpossibility of deciding without an initial policy determination of a kind clearly for nonjudicial discretion”;

4. “[I]mpossibility of a court’s undertaking independent resolution without expressing lack of the respect due coordinate branches of government”;

5. “An unusual need for unquestioning adherence to a political decision already made”; and

6. “The potentiality of embarrassment from multifarious pronouncements by various departments on one question.”164

The Guarantee Clause model does not display any of these political-question hallmarks. On the first factor, unlike with Congress and the President’s shared war powers in Article I, Section 8, and Article II, Section 2, there is no commitment in constitutional text to having another branch of government exercise exclusive control over state prosecutions.

On factors two, three, and four: adopting this model does not require federal courts to make any “initial policy determination.” Instead, the model simply mirrors the standards of good government defined by the state itself, which supplies the “judicially discoverable and manageable standards” federal courts are to follow. Doing so ensures the “respect due coordinate branches of government.” The fifth factor—”unusual need for unquestioning adherence”—has no relevance here, having to do more with the finality of a decision like the declaration of cessation of hostilities.165

The last factor, the fear of “multifarious pronouncements,” favors justiciability as well.166 Multifarious announcements under the Guarantee Clause model could arise to the extent that federal courts differ from state courts in how they interpret state statutes. The district court in the Northern District of New York may interpret a New York criminal statute differently than the Court of Appeals for New York. But that risk of objectionable inconsistency is an inherent one in our federal system, and one that is easily mitigated by allowing state high courts to have the final word on the interpretation of their state statutes.

At bottom, nonjusticiability is about ensuring due respect for separate sovereigns that are entitled to political self-determination. This model guarantees that respect by ensuring that federal enforcement is defined by the states themselves.

D. Originalism and Federal Criminal Law

One final critique of the Guarantee Clause model presented here is that while it draws support from Founding Era sources, it is arguably an un-originalist reading of the Guarantee Clause. Though Federalist 43 makes the case that there would be federal enforcement of the guarantee, there is no indication that anyone at the Founding understood this to mean federal criminal enforcement.

That much is true. No evidence reveals that the Framers considered federal criminal law at all, save for a few enumerated offenses.167 But that same critique could be leveled at the entire apparatus of federal criminal law. Surely, there is no plausible argument that the original public meaning of the Commerce Clause in 1787 extended to intrastate growing of medical marijuana or foreign sex trafficking.168 The Guarantee Clause model works if one accepts the premise that the Constitution authorizes a robust federal criminal law enforcement. If so, the value of the Guarantee Clause is that it is a more constitutionally principled basis for local public-corruption prosecutions than the Commerce Clause, the Spending Clause, or the Postal Clause.

Conclusion

The Guarantee Clause model raises many more questions that remain unanswered. Which state’s law applies in cross-border crimes? Would prosecutors be permitted to charge multiple state-predicate violations?

This model is not a panacea for all that ails federal corruption prosecutions. Federal courts will still need to fill in the gaps, just as they do when interpreting even the most detailed statutes in other regulatory regimes. But the Guarantee Clause model saves federal courts from needing to “chip away” at “uncut marble,” as Justice Gorsuch put it.169 Rather, federal courts may build atop the scaffolding that states have already constructed.

J.D., Yale Law School, B.A., University of Pennsylvania. Many thanks to Professor Kate Stith for her generous feedback on this piece and for sparking my interest in federal criminal law. I am indebted to Arshan Barzani, Jim Huang, and Jandee Wallis for their comments on earlier drafts, to Rachel Troy Cameron for her excellent work as lead editor, and to the editors of the Yale Law Journal.