Calling the Shots: Multistate Challenges to Federal Vaccine Mandates

Multistate litigation brought by state attorneys general (“AGs”) frustrated the Biden administration’s efforts to combat the COVID-19 pandemic by vaccinating workers. Nationwide injunctions played an important role in halting the implementation of vaccine mandates in multistate actions. State challenges to vaccine mandates are consistent with AG lawsuit trends against recent presidential administrations. These challenges also reveal new emerging patterns that shed light on the future of multistate litigation and nationwide injunctions. State vaccine mandate lawsuits have continued to raise criticisms of nationwide injunctions and, at the same time, provide insights on pathways forward for reform.

Download

Self-Defense Exceptionalism and the Immunization of Private Violence

After the high-profile trial of Kyle Rittenhouse, the parameters of lawful self-defense are a subject of intense public and scholarly attention. In recent years, most commentary about self-defense has focused on “Stand Your Ground” policies that remove the duty to retreat before using lethal force. But the reaction to Rittenhouse’s case reflects a different, more extreme way that the law governing defensive force is changing. In particular, advocates and legislators say that private citizens like Rittenhouse who exercise self-defense should be entitled to immunity—an exemption from prosecution—giving them an extraordinary procedural benefit not attaching to other defenses that are adjudicated at trial. As this Article reveals, this effort to transform self-defense into something exceptional within criminal law began more than a decade ago in the shadows of Stand Your Ground. One-quarter of U.S. states have already enacted laws providing for self-defense immunity.

This Article examines this fundamental yet understudied shift in self-defense law. It shows how the concept of immunizing defensive force is foreign to the Anglo-American legal tradition as well as settled principles of modern criminal law and procedure, including the exceedingly narrow role of immunities. It tells the story of how self-defense immunity arose not as part of the broader criminal justice reform movement, but rather at the behest of the movement to insulate defensive gun use from liability. And it demonstrates the costs of treating self-defense as an immunity, such as increasing violence, diminishing the institution of the jury, delegitimizing criminal law outcomes, and undermining judicial economy. After exposing the unreasoned rise and inevitable costs of self-defense immunity, this Article concludes that self-defense should remain an affirmative defense to criminal charges rather than immunize a defendant from being prosecuted at all. Self-defense reform should move in lockstep with other criminal law defenses so as to avoid the societal harms that result from immunizing defensive violence.

INTRODUCTION

On August 25, 2020, seventeen-year-old Kyle Rittenhouse traveled to Kenosha, Wisconsin, with an illegally obtained AR-15–style rifle in the wake of the shooting of Jacob Blake by a police officer. Rittenhouse said he went heavily armed to provide medical aid and protect property, albeit strangers’ property, during racial justice protests and unrest following yet another police shooting of a Black man. Instead, he shot three men during altercations, killing two of them. Rittenhouse was charged with crimes including murder, and in his defense he asserted self-defense: he feared that the men would disarm him and use his own rifle against him unless he shot them first.

Rittenhouse’s case was closely watched and controversial, splitting the nation into diametrically opposed camps regarding the appropriateness of his conduct. It also raised difficult factual and legal questions, including whether he provoked the confrontations and thereby negated the lawfulness of his defensive force. At the end of a two-week trial at which dozens of witnesses testified, a jury deliberated for three days and returned a verdict of not guilty. The outcome should have pleased those who supported Rittenhouse’s conduct that summer night. Instead, a common reaction was, as former President Donald Trump put it, that Rittenhouse “shouldn’t have been prosecuted in the first place.”

If that sentiment were simply a feature of modern political rhetoric, it might be undeserving of close scrutiny. Indeed, the politics of self-defense shone brightly after the Rittenhouse trial. U.S. Representative Marjorie Taylor Greene even introduced a bill to award Rittenhouse a civilian’s highest congressional tribute, a Congressional Gold Medal, for his “courageous actions.” Several Republican politicians invited Rittenhouse to intern in their offices. Just days after the verdict, he was welcomed at Trump’s Mar-a-Lago Club in Florida.

But this Article shows how the notion that people “should not fear exposure to criminal prosecution when they use firearms to defend themselves and their homes” is more than rhetoric. Rather, it is the foundation for an effort to grant an exemption from prosecution to those who, like Rittenhouse, claim self-defense in defending against criminal charges. After Rittenhouse’s acquittal, one advocate penned “Kyle’s Law” to cement the exalted status of self-defense. The proposed statute would alter the law in various ways, including effectively immunizing lawful defensive force from prosecution altogether. As it turns out, more than one-fourth of U.S. states have already done just that, and the trend is likely to continue.

In the past decade, legal scholarship has explored “Stand Your Ground,” or the removal of the common law duty to retreat before using lethal defensive force in public. That literature shows how Stand Your Ground interacts with an expansion of gun rights in a way that can lead to more violence and exacerbate existing patterns of discrimination in the criminal justice system. Articles have likewise explored additional features of the intersection of criminal law, self-defense, and gun rights. And legal scholars are starting to explore whether self-defense law might be bolstered in light of changed circumstances—especially the proliferation of gun carry—to limit the unnecessary loss of life.

Yet the notion that self-defense is exceptional and “deserves” to be immunized, as one legislative witness put it, has evaded close scrutiny. Articles about Stand Your Ground have acknowledged what Cynthia Ward termed the “curious beast” of self-defense immunity as well as the “confusion” it invites. However, self-defense immunity warrants a sustained analysis in terms of how it began as an adjunct to the gun rights movement and how it fits within the criminal justice system today. That, in turn, calls for an examination of a more general topic that similarly has received little attention: the procedural treatment of criminal law defenses and why prosecutorial immunities are so few in number. To exempt a category of defendants from the ordinary criminal process is profound, bestowing “a far greater right than any encompassed by an affirmative defense, which may be asserted during trial but cannot stop a trial altogether.” Examining why the criminal law is generally opposed to granting an exemption from prosecution is an important, understudied part of the inquiry.

This Article proceeds in three parts. Part I shows how justifications for otherwise criminal conduct, like self-defense, have traditionally been adjudicated: as affirmative defenses to criminal charges. Some have argued that immunizing self-defense is simply a return to past protections that have been lost in recent times. But those engaging in private violence have always been exposed to criminal prosecution and trial. The argument that self-defense exceptionalism is rooted in tradition is unsupported.

Part I also shows how modern pretrial criminal procedure is consistent with the historical antecedents. The formal process is overwhelmingly structured to bring cases forward to trial, even if few cases get that far. Pretrial screening is largely geared toward questioning the basis for the charged offense, not adjudicating potential defenses. The criminal law makes exceptions for a narrow set of pretrial matters—narrower than in the civil context. The scant prosecutorial immunities and their narrow justifications can be linked to the criminal law’s aims and distinctive character, which are especially protective of public prosecutions. The exceptions that receive prosecutorial immunity tend to be fundamentally different than self-defense in both their scope and purpose. In particular, other criminal law immunities benefit narrow classes of defendants and must be addressed ahead of trial to protect distinctive public interests like maintaining foreign relations or preserving the balance of powers. Self-defense, in contrast, can be invoked by any defendant and, like a multitude of other defenses, can be adjudicated at trial without undermining its role as justifying otherwise unlawful conduct. Moreover, interests served by self-defense law—like maintaining the legitimacy of the legal order—are actually undermined by immunity.

Part II then turns to the next logical question: Why are states now diverging from American legal tradition and standard practices to treat self-defense as something exceptional? The Article traces self-defense immunity from a barely debated and misunderstood change to Colorado law in the 1980s to a primary ambition of gun rights advocates in the 2000s. The resulting legal changes are often characterized as “Stand Your Ground laws,” but that understates the transformation that is afoot. Stand Your Ground relates to just one of many ways that legislators are remaking the law governing defensive force. Indeed, one possible reason why self-defense immunity has escaped close scrutiny is that the typical focus is on the substantive elements establishing what lawful self-defense is, and especially the duty to retreat, while glossing over changes to how self-defense is adjudicated.

Yet while Stand Your Ground has garnered the most attention, advocates—and especially gun rights advocates—have pursued a deeper goal: insulating defensive gun use from legal oversight to the greatest extent possible. It is hard to overstate the degree to which the quick rise of self-defense immunity is due to lobbying by advocates for one deadly weapon (the gun) that is used in a minuscule percentage of self-defense confrontations. The loudest voices advocating for immunizing self-defense tend not to be those seeking criminal justice reform generally but rather those seeking to expand gun rights. A National Rifle Association (“NRA”) lobbyist, for example, drafted and led the campaign to institute self-defense immunity in Florida, which then became a model for states across the nation. The playbook for transforming self-defense into an immunity mirrors the one used to expand gun rights. The overlap between gun rights and self-defense rights advocacy begs the question of whether any principle other than bestowing a benefit on gun users is guiding self-defense’s transformation from an affirmative defense into an immunity. Part II raises several possibilities, but it finds each too thin to justify such an immense procedural departure.

Part III then explores functional and institutional costs of immunizing private violence. Self-defense immunity sends a signal that people can judge for themselves when to deploy violence in the name of self-protection without exposure to prosecution, thereby encouraging unnecessary violence. Meanwhile, by preventing the community, through the jury, from evaluating the lawfulness of defensive force, immunity jettisons the institution best suited for adjudicating self-defense. In addition, immunizing self-defense creates an inefficient process by which courts consider the same witnesses and arguments that will be presented at trial during a separate pretrial hearing, setting up the sort of mini-trial that criminal procedure generally disfavors.

Trials like Rittenhouse’s spark intense disagreement and debate. But such trials are a feature—not a bug—of the American justice system. The Article concludes that policymakers should keep self-defense in its traditional place as an ordinary affirmative defense to criminal charges. Criminal justice reform is desperately needed, but treating private violence as privileged at the behest of gun rights advocates is a perilous path.

I.  SELF-DEFENSE AND PRETRIAL CRIMINAL PROCEDURE

       As Carl Sagan famously put it: “You have to know the past to understand the present.” That maxim applies equally well for modern criminal law. This Part thus explores how self-defense was historically implemented in criminal procedure. It shows how the criminal justice system that the United States adopted from England was “trial-centered, in the sense that the legal system sought to resolve most criminal business at trial,” including claims of self-defense. This Part then shows how that treatment continued in modern times until the recent effort to grant pretrial prosecutorial immunity for self-defense. The effort to recharacterize self-defense as an immunity invites a question about how immunities fit within the criminal justice system. This Part closes by addressing that question, showing how and why prosecutorial immunities are few in number and narrowly construed, and how and why their typical rationale does not apply to self-defense.

A.  Historical Procedure

In 1841, in People v. McLeod, a New York court considered a habeas corpus petition for a defendant charged with murder. The defendant sought his “unqualified discharge” on the basis of pretrial evidence that, among other things, he acted in lawful self-defense. The court emphatically rejected the “extraordinary” request, noting the “absurdity of such a proposition in practice, and its consequent repudiation by the English criminal courts” whose law and procedure the United States inherited. Among other things, granting the defendant’s request “would be to trench on the office of the jury.” As the court explained, “[a]n innocent man may be, and sometimes unfortunately is[,] imprisoned. Yet his imprisonment is no less lawful than if he were guilty. He must await his trial before a jury.” That early American understanding of the appropriate time—and the appropriate entity—to adjudicate self-defense was firmly rooted in the English common law tradition.

During the seventeenth and eighteenth centuries in England, after a felony was charged, judges lacked authority to discharge defendants “without further trial.” This was true regardless of whether the defendant was believed to be justified in engaging in the alleged offense conduct. In the 1700s, judges began conducting a “pretrial inquiry” that “increasingly took on the trappings of a public hearing, which would ultimately come to be known as the preliminary hearing.” At such hearings, however, the defense attorney was limited to challenging the prosecution’s case and was not entitled to present the defense’s case.

Classic common law treatises demonstrate how self-defense was just like other defenses in that it was a trial issue, not a pretrial issue. For example, Michael Foster, a judge on the King’s Bench and the author of a widely read treatise published in 1762, observed that the defendant raising self-defense “standeth upon just the same foot that every other Defendant doth: the Matters tending to Justify, Excuse, or Alleviate, must appear in Evidence before He can avail himself of them.” And the opportunity to introduce that evidence was not until trial: “[W]hether the Facts alledged by way of Justification, Excuse, or Alleviation are True, is the proper and only Province of the Jury.”

Several years after Foster’s publication, William Blackstone completed “the preeminent authority on English law for the founding generation,” in which he explained that “it is incumbent upon the prisoner to make out, to the satisfaction of the criminal court and jury,” any “circumstances of justification, excuse, or alleviation.” The jury, Blackstone wrote, is “to decide whether the circumstances alleged [regarding self-defense or other affirmative defenses] be proved to have actually existed”; the judge then decides “how far [the proved circumstances] extend to take away or mitigate the guilt.”

Edward Hyde East, in his influential 1803 treatise, built on Blackstone’s and Foster’s accounts and elaborated on the lack of a pretrial process for asserting self-defense. He wrote that “the jury alone [is] to decide” on “the truth” of the defendant’s allegations of “justification, excuse, or alleviation,” though the judge could consider such defenses when deciding on bail. The McLeod case demonstrates that this current continued in the United States into the nineteenth century. In his 1872 Commentaries on the Law of Criminal Procedure, Joel Prentiss Bishop described how a defendant entering a plea of not guilty at arraignment formally “puts himself upon the country,” or submits to a trial by jury. The jury therefore remained the primary entity to decide disputed fact issues in criminal cases, including regarding self-defense.

Pretrial processes, like the preliminary hearing and the grand jury, generally did not provide a defendant an opportunity to introduce evidence of any particular defense. As the 1918 edition of Francis Wharton’s treatise on criminal procedure observed, “the better opinion is that on a preliminary hearing the magistrate is to hold the defendant for trial” when “there is made out a probable case of guilt.” Similarly, in a proceeding before the grand jury, “it is not the usage to introduce, in matters of confession and avoidance, witnesses for the defense, unless their testimony becomes incidentally necessary to the prosecution.”

The notion that self-defense could be adjudicated by a judge before trial thus has no basis in the common law tradition imported from England and implemented in America. The next Section shows how that basic understanding carried forward to modern times.

B.  Modern Procedure

In 1971, Indiana passed a statute providing that “[n]o person . . . shall be placed in legal jeopardy of any kind whatsoever” after exercising lawful self-defense. Armed with that broad statutory language, one defendant sought a pretrial determination of the lawfulness of his claimed self-defense. In Loza v. State, Indiana’s highest court recognized the novelty of the proposition before reacting much like the New York court did more than a century earlier in McLeod. In particular, in order “to prevent absurdity,” the court held that the new law “neither creates a new remedy nor does it alter our procedure in any respect.” In other words, self-defense remained a trial issue. The Loza court’s understanding was consistent with modern pretrial procedure.

Modern criminal procedure is heavily constitutional, and an overview of the minimalist pretrial constitutional requirements for defenses (like self-defense) is therefore instructive. Under the Fourth Amendment, police officers must have probable cause before making an arrest, and an impartial magistrate must review whether probable cause exists if the arrestee is to remain in custody. The Supreme Court has described probable cause as “a fluid concept” that “requires only a probability or substantial chance of criminal activity, not an actual showing of such activity.” Admittedly, probable cause is “not a high bar.”

Importantly, moreover, probable cause does not require robust consideration of self-defense, if it requires any at all. The Third Circuit has held that “affirmative legal defenses”—like self-defense—“are not a relevant consideration in [a police] officer’s determination of probable cause.” In contrast, the Second Circuit has held that “a police officer’s awareness of the facts supporting a defense can eliminate probable cause.” That said, such evidence must be “conclusive” or first-hand, and once an officer has probable cause to make an arrest, the officer does not constitutionally have “to investigate exculpatory defenses offered by the person being arrested or to assess the credibility of unverified claims of justification.” Self-defense is not singled out for special treatment, but rather is treated like any other defense.

Subsequently, once a prosecutor makes a charging decision, there is “no federal constitutional right to any review” of that decision before trial “apart from the grand jury clause of the Fifth Amendment.” The grand jury, meanwhile, is also guided by the standard of whether there is “probable cause necessary to initiate a prosecution for a serious crime.” In United States v. Williams, the Supreme Court held that, notwithstanding the constitutional obligation to disclose material exculpatory evidence to a defendant before trial, the Constitution does not require prosecutors to disclose substantial exculpatory evidence to the grand jury, including regarding a potential claim of self-defense. Looking back to the common law history, the Court explained that the grand jury is “an accusatory [body],” not “an adjudicatory body,” and its task is “to assess whether there is adequate basis for bringing a criminal charge.” Historically, “it has always been thought sufficient for the grand jury to hear only the prosecutor’s side.”

In some jurisdictions, by either law or internal policy, prosecutors are held to a higher standard than the federal constitutional baseline with respect to grand juries. However, most such departures only require presenting “evidence that is clearly exculpatory” or “that would exonerate the accused or lead the grand jury to refuse to indict.” Given the low bar for indictment—again, probable cause—even these jurisdictions stop far short of adjudicating self-defense before trial.

The Federal Rules of Criminal Procedure, which “almost always reflect the basic position adopted in a substantial number of states,” provide other pretrial procedural steps apart from the grand jury, most notably a preliminary hearing. Yet the preliminary hearing—consistent with historical practices—focuses on the prosecution’s evidence for the charged offense, and not evidence of self-defense or any other affirmative defense. Again, the standard is probable cause: the prosecutor need only show “probable cause to believe an offense has been committed and the defendant committed it.” Moreover, the prosecutor gets to decide whether to have a preliminary hearing at all: if the prosecutor secures an indictment before a grand jury, then the defendant has no right to demand a pretrial hearing.

It thus has remained true under conventional criminal procedure that “[i]f a defendant claims innocence or has a defense,” including self-defense, “the proper body to decide the issue is the petit jury.” Recent reform efforts, however, characterize self-defense not as a “defense” but as an “immunity,” calling to mind exceptions to the general rule—a category of traditional immunities and other matters that are adjudicated pretrial. The next Section addresses such pretrial issues in relation to self-defense.

C.  Immunities from Prosecution

Recent legislation declaring that self-defense is an immunity from prosecution has led judges and commentators to treat self-defense as a “true immunity” comparable to others. This classification invites questions about how other prosecutorial immunities operate, why they exist, and whether they share anything in common with self-defense.

Common immunities from prosecution include diplomatic immunity, judicial immunity, legislative immunity, executive immunity, immunity after compelled testimony, and immunity bestowed on the basis of a plea agreement. These are “defenses” in the sense that they are asserted by a defendant as a way to avoid a conviction. But their essence goes beyond ordinary defenses because immunities operate to exempt a person from the mandate of the criminal law, not to justify otherwise criminal conduct because of the circumstances surrounding that conduct. Black’s Law Dictionary cross-references “impunity” in its definition of “immunity,” which similarly denotes an “[e]xemption from punishment.” The example that Black’s uses to describe impunity relates to diplomatic immunity: “because she was a foreign diplomat, she was able to park illegally with impunity.” Immunity gets asserted early in the criminal process to head off the prosecution of someone possessing such an exemption.

As such, prosecutorial immunities are a remarkable departure from the ordinary criminal process described above; moreover, they are in tension with a basic, distinctive function of criminal law. Criminal law is traditionally viewed as a means to declare “a formal and solemn pronouncement of the moral condemnation of the community.” The community’s role in implementing the criminal law—through a public prosecution and jury trial—is intertwined with that function. It is no coincidence that the prosecutor in a criminal case is called “The People” in many jurisdictions.

Prosecutorial immunity dilutes the formal power of the public in assessing an alleged crime, and it thus raises special concerns in criminal law that might exist only to a lesser extent in the civil context, where immunity is sometimes granted, for example, primarily to avoid costs. In the criminal context, immunities tend to be justified by a narrower, more compelling rationale. As a general matter, only when avoiding the criminal justice process is a defense’s entire raison d’être is it exempted from prosecution as an “immunity.” Put differently, the public policies underlying the above-mentioned criminal law immunities necessarily require the avoidance of prosecution and trial.

Consider diplomatic immunity. A key reason why we immunize conduct by foreign diplomats in the United States is to protect American diplomats outside the United States from exposure to foreign court systems. There is no way to satisfy that goal through an affirmative defense at trial. Consistent with the purpose of diplomatic immunity, it also does not protect diplomats from sanction upon return to their home countries. Judicial, legislative, and executive immunities are similarly geared to specific policy rationales necessitating avoidance of a trial. Each protects “governmental officials from personal liability arising from their official duties” because of the strong interest in facilitating their ability to serve the public. The Supreme Court has explained how legislative immunity enables “representatives to execute the functions of their office without fear of prosecutions.” An added component of legislative and judicial immunity is to preserve the balance of power between the three branches of government by insulating legislative and judicial officers from prosecutions by the executive branch. Again, interests that these governmental immunities serve cannot be furthered—and indeed would be undermined—if they were treated as defenses to be proved at trial. The remarkable benefit of immunity is thus granted because of strong public policy arguments that inherently entail a bar to prosecution.

How does self-defense relate to immunities? Self-defense is not about trial avoidance but exculpation. Like other justification defenses and unlike immunities, it can be adjudicated in the traditional way—through trial—without undermining its rationale. Moreover, unlike typical immunities, self-defense furthers interests that are in fact undermined by short-circuiting a prosecution and trial.

T. Markus Funk has identified seven values served by self-defense law: protecting the state’s monopoly on force, protecting the individual attacker’s right to life, maintaining the equal standing between people, protecting the defender’s autonomy, ensuring the primacy of the legal process, maintaining the legitimacy of the legal order, and deterring attackers. Immunity arguably advances the interests in protecting a defender’s autonomy or deterring attackers. But it runs roughshod over other values, especially self-defense law’s dual roles of ensuring the primacy of the legal process and maintaining the legitimacy of the legal order. Both roles underlie the idea that “the authority to punish and condemn” remain with “the liberal state,” not with individual citizens. In his discussion of ensuring the primacy of the legal process, Funk notes that “[t]o the extent possible, . . . the justice system must promote the resolution of disputes in the courts.” Immunity, however, dilutes the state’s oversight of defensive violence and, perhaps worse still, undermines the community’s role through the jury to assess the lawfulness of violence—a point addressed in greater depth in Part III. In other words, in contrast to typical immunities, whose purposes areoverall advanced by providing an exemption from prosecution, key values underlying self-defense law are undercut by providing such an exemption.

Immunities, of course, are not the only matters that receive pretrial resolution. Some defenses—like those based on statutes of limitations, double jeopardy, and speedy trial requirements—are also adjudicated in advance of trial. Other issues, like competency to stand trial, also receive pretrial determination. In the effort to implement self-defense immunity, some have analogized self-defense to those other pretrial issues even though they are not technically “immunities.” Yet these issues, like traditional immunities, protect interests that necessarily call for avoiding trial and thus are dissimilar to self-defense. Statutes of limitations affirm the belief that “[a]fter a period of time, a person ought to be allowed to live without fear of prosecution.” Double jeopardy protections are “designed to protect an individual from being subjected to the hazards of trial and possible conviction more than once for an alleged offense.” Speedy trial guarantees mandate “the Government [to] move with the dispatch that is appropriate to assure [the defendant] an early and proper disposition of the charges against him.” And resolving competency questions must also happen before a trial since the entire point is to determine the defendant’s “ability to participate meaningfully in the trial.”

In connection with competency hearings, one exception to the general rule of limiting pretrial criminal matters to those that inherently require pretrial determination involves the insanity defense. Courts tend to draw a clear line between the question of competency to stand trial, which is adjudicated in advance of trial, and insanity at the time of the offense, which is a trial issue. As a general matter, therefore, an insanity defense is submitted to the fact finder at trial and is not decided at a pretrial hearing. However, such bifurcation is not universally followed. Pennsylvania law, for example, grants a judge the discretion to “hear evidence on whether the person was criminally responsible for the commission of the crime charged” so long as the judge is already conducting a competency hearing.

In that context, judicial economy might weigh in favor of considering evidence of both competency and insanity at a pretrial hearing. At least one other state—North Carolina—gives courts discretion to hold a pretrial insanity hearing so long as the state consents. That exception is highly limited in that courts and prosecutors can override a defendant’s request for a hearing, making it quite different from self-defense immunity. And in Washington, a defendant may request a pretrial insanity determination, but the statute notes that any acquittal under the statute cannot be used to contest mental health detention—a possibility that distinguishes insanity and self-defense.

This Section has set out the limited nature of criminal law immunities and other pretrial matters and offered a normative explanation, rooted in the criminal law’s distinctive role, for that narrow scope. Below, the Article considers additional arguments for and against expanding immunities to include self-defense. First, however, the Article turns to the story of how self-defense immunity arose in the first place.

II.  THE PUSH TO MAKE SELF-DEFENSE EXCEPTIONAL

In light of the American criminal law tradition of adjudicating self-defense at trial, how did self-defense immunity arise? This Part shows how self-defense immunity emerged out of Colorado in 1986, laid dormant for almost two decades, and then became a central component of gun rights advocacy in the 2000s. The Part then analyzes the thin rationales put forward for treating self-defense as deserving of exceptional treatment through prosecutorial immunity.

A.  Inauspicious Beginning in Colorado

Accounts of recent self-defense reforms tend to begin with Florida’s 2005 Stand Your Ground legislation. Indeed, Florida’s law served as a model that influenced legal changes across the country. But the first example of a self-defense immunity statute was not Florida’s but rather a last-minute compromise bill from Colorado twenty years earlier.

The Colorado law did not, at first, provide for prosecutorial immunity. Rather, the bill initially added a legal presumption to self-defense law to enhance the scope of lawful self-defense against home intruders. To be sure, homeowners already had an expanded right to self-defense through the “Castle Doctrine,” which generally removed a person’s duty to retreat before using lethal defensive force in the home. However, Colorado policymakers wanted to do more, so they borrowed from a California statute that a person confronting a home intruder is legally “presumed” to fear for their life. That presumption would satisfy one requirement of lethal defensive force—that the defender reasonably perceives a threat of death or serious bodily injury—thereby relieving the defendant of the need to produce evidence of such heightened fear.

Prosecutors objected because they “believed that it would be very difficult, if not impossible, to rebut the presumption in favor of the homeowner.” There was little public debate regarding the subsequent compromise that became the nation’s first law providing immunity from prosecution for self-defense. Yet the law appears to have imported a civil immunity provision enacted in Colorado in 1982 into the criminal law.

By way of background, in 1981, a Colorado jury awarded a plaintiff more than $300,000 in damages from a defendant for gunshot injuries incurred while the plaintiff was burglarizing the defendant’s shop. The public outcry was swift and the shop owner’s lawyer helped to draft a bill immunizing people like his client from civil damages. The resulting law barred payouts for personal injuries “sustained during the commission of or during immediate flight from” a felony if the person inflicting the injury reasonably believed that physical force was “reasonable and appropriate” to prevent both injury and the commission of the felony. The wisdom of such civil immunity is beyond the scope of this Article; more important for present purposes is that it did not address immunity from criminal liability. As discussed above, criminal liability is geared toward vindicating public harms in a way that civil liability is not. Nonetheless, the criminal immunity bill that later passed in Colorado in 1986 mirrored the earlier civil immunity law. The law provided that a person “shall be immune from criminal prosecution” if the person used defensive force and four conditions were met relating to an unlawful home intrusion.

The 1986 law’s legislative sponsors and the negotiating prosecutors appeared to have different beliefs about what the new law actually accomplished. The sponsors appreciated that they had achieved “greater protection [for defendants] than a presumption for the homeowner as part of an affirmative defense at trial.” The negotiating prosecutors, in contrast, believed that they gave up nothing. Denver’s district attorney, for example, publicly commented that the “compromise is just a clarification of existing law.”

In that vein, some prosecutors tried to argue in subsequent litigation that the new provision could not possibly grant true immunity for self-defense. Among other things, they pointed out that the provision appears alongside other affirmative defenses in Colorado’s criminal code. When the issue reached the Colorado Supreme Court, however, the justices rejected the prosecutors’ interpretation that self-defense remained an ordinary defense to be proved at trial, noting that “[i]t must be presumed that the legislature has knowledge of the legal import of the words it uses.” The plain meaning of “shall be immune from criminal prosecution” in the statute, they concluded, was “to bar criminal proceedings against a person for the use of force under the circumstances set forth” in the law. In the course of reaching that holding, the justices acknowledged what went unsaid during the legislative hearings: that “the immunity created by [the law] is an extraordinary protection which, so far as we know, has no analogue in Colorado statutory or decisional law.” In fact, immunity for self-defense in criminal cases does not appear to have existed anywhere else in the country.

Perhaps because of its unusualness, or because it was an eleventh-hour deal seemingly unrooted in any principle other than compromise, Colorado’s self-defense immunity law was not immediately enacted elsewhere. In 1987, for example, Oklahoma’s governor vetoed legislation similar to Colorado’s, which subsequently passed after the immunity provision was removed. Nonetheless, Colorado’s immunity provision was on the books, providing a template for future efforts.

B.  Auspicious Effort by Gun Rights Advocates

The Colorado self-defense immunity law was not instituted at the behest of gun rights advocates or other lobbyists, but rather, it arose as a compromise with prosecutors after a locally elected leader perceived a need for expanding self-defense protections against home intruders. In more recent times, however, gun rights advocates and the NRA in particular have led a campaign to expand not only the right to have and carry guns but also to brandish and shoot them when gun owners feel threatened. Most public attention to this campaign has centered around Stand Your Ground, but looking closely at testimony and commentary reveals a deeper ambition: immunizing defensive gun use from prosecution.

The parallels between the NRA’s lobbying for gun rights and its lobbying for self-defense immunity is striking. Gun rights advocates frequently claim that the right to keep and bear arms is being disrespected in the courts and therefore that the Second Amendment needs more protection. The claim with self-defense is similar: as one gun rights advocate put it, self-defenders are “victimized . . . in court.” The executive director of the NRA’s Institute for Legislative Action lamented that “people who defend themselves are more likely to be charged with crimes and, as the old sayings go, be forced to ‘tell it to the judge’ and ‘let the jury sort it out.’ ” That creates a problem, he explained, because “a murder trial puts the defendant at risk of a long prison sentence—or worse.” The NRA lobbyist most directly involved with Florida’s landmark Stand Your Ground bill in 2005 was likewise moved by this notion. A basic problem, in her view, was that people were “being arrested” and “prosecuted . . . for exercising self-defense that was lawful.”

An answer to that feeling of disregard for self-defense was to transform it from an affirmative defense to an immunity. The NRA devised a self-defense immunity law and found legislative sponsors in Florida who agreed with the complaint that, as one put it, “law-abiding citizens” who “protect themselves [are] in a posture that they have to defend themselves from their own government.” The measure passed in 2005 and went even further than Colorado’s, extending prosecutorial immunity to all self-defense—not just self-defense in the context of home invasions. In particular, the law provided that someone using lawful self-defense is “immune from criminal prosecution,” with “criminal prosecution” defined to “include[] arresting, detaining in custody, and charging or prosecuting the defendant.”

After some Florida judges placed the burden on the defendant to prove self-defense at a pretrial hearing, legislators stepped in to strengthen the immunity provision by clarifying that the burden of proof is on the prosecutor to disprove self-defense before trial by clear and convincing evidence. That standard is much higher than the probable cause standard that prosecutors must satisfy to indict, which, as discussed above, is the primary focus of traditional and modern pretrial screening. And there have been efforts to increase the burden even more, such as by requiring the prosecutor to disprove self-defense beyond a reasonable doubt—the same burden borne by the prosecutor at trial.

Unlike Colorado’s law, which failed to attract buy-in elsewhere, Florida’s law was aggressively promoted by the NRA and the conservative American Legislative Exchange Council (“ALEC”), which described the need to “[p]rotect[] citizens from prosecution or liability if they use a firearm in self defense [sic] inside or outside their homes.” Similar laws were introduced in states across the country, and the NRA-promoted sentiment that civilians asserting self-defense should have a path to immunity was frequently invoked. When legislators debated Iowa’s self-defense law, one objected that a person must “spend eternity in prison trying to defend themselves” after being put “in that untenable situation where they have to make that snap decision and defend themselves or another from an aggressor.” In Ohio, a legislative witness inveighed that “[t]he mere fact of acting justly in self-defense should not result in dragging folks who used defensive force in accordance with Ohio law through the mud, costing them valuable time and resources.” In South Carolina, a self-defense bill’s sponsor argued that “the State should have to prove you did something wrong before they can send you to jail” to await trial in homicide cases. And in Utah, an advocate complained that people should not have to “go through the crucible of a self-defense trial.” Ultimately, after the passage of Florida’s law, more than twenty other states passed some sort of self-defense reform, such as Stand Your Ground, with at least thirteen enacting self-defense immunity.

But the fact that people who lawfully defend themselves are sometimes prosecuted and forced to argue self-defense is unexceptional. It is a truism that self-defense sometimes exculpates—that is precisely why it is an available defense to criminal charges. Singling out self-defense for special treatment as an immunity should have a compelling rationale similar to the ones that justify other prosecutorial immunities. The next Section searches for such a rationale in the legislative debates and commentary.

C.  Searching for a Rationale

A common assertion among advocates for self-defense immunity is that awaiting trial is “not giving the right to self-defense the consideration it deserves.” But why not? After all, awaiting trial is the traditional process and the one afforded other defenses. In his systematic analysis, Paul Robinson identifies dozens of other affirmative defenses that bar conviction. What is the basis for treating self-defense differently than these other defenses? Though legislative debates offer no consistent rationale, four can be teased out: restoring procedural protections for self-defense lost to history, stopping politically motivated prosecutions of self-defenders, vindicating the notion that self-defense is a “natural right,” and reducing defense costs for gun owners. None of these is as strong as the rationale for traditional immunities—an inherent need for pretrial adjudication. Moreover, each is unpersuasive on its own terms.

Some advocates argue that prosecutorial immunity restores self-defense to an exalted place from a bygone era. In Florida, for example, a witness testified that making the prosecutor disprove self-defense before trial “recover[s] a right that we as citizens lost to defend ourselves from criminals.” In Utah, a witness testified that “Utah used to have a robust preliminary hearing procedure” as it relates to self-defense, and that immunity “restores some much-needed balance.”

A related move has been to couple self-defense immunity with Stand Your Ground and then defend both on the basis of Stand Your Ground history. For example, the NRA has said that Stand Your Ground laws, such as Florida’s (which includes an immunity provision), “focus on the narrow issue of whether and to what extent a person who would otherwise have a right to self-defense forfeits that right by not first attempting to flee the confrontation.” With omnibus bills like Florida’s so purportedly reduced, the NRA then asserted that removing the duty to retreat has “a pedigree in American law dating back over 150 years.” Other advocates have similarly ignored everything in recent self-defense legislation other than Stand Your Ground and then defended the entirety on the basis of Stand Your Ground history.

Nostalgia is a staple of gun rights advocacy, so it is unsurprising to see appeals to history when it comes to self-defense immunity. Yet, as shown in Section I.A, there is no basis in Anglo-American legal tradition for immunizing private defensive violence. Treating self-defense as exceptional through immunity is a thoroughly modern innovation.

An alternative rationale is that people exercising lawful self-defense are targeted for “political” prosecutions. Prosecutors have vigorously rejected that narrative, and advocates for immunizing self-defense have failed to offer convincing evidence of political prosecutions, let alone the sort of systemic abuses that would justify a radical change to self-defense law. Advocates for both of the first immunity statutes—in Colorado (1986) and Florida (2005)—could not point to a single example of an improper prosecution. Rather, the chief NRA lobbyist for the Florida law ultimately contended that whether bad prosecutions have been brought is “not relevant.”

In subsequent efforts to immunize self-defense, advocates have invoked the prosecutions of George Zimmerman for the shooting death of Trayvon Martin and Rittenhouse for the Kenosha incident as exemplars of political prosecutions justifying self-defense immunity. Looking to Zimmerman’s prosecution is somewhat ironic given that it took place in Florida after Florida adopted its 2005 immunity provision and Zimmerman opted not to have a pretrial immunity hearing. Furthermore, in both cases the juries reached verdicts only after extensive deliberation. The lead homicide investigator in the Zimmerman case recommended charges but was initially overruled. Many perceived the declination of charges as reflecting racial bias, as Martin was an unarmed Black teenager. A special prosecutor ultimately brought charges and a trial was held. The law considered by Zimmerman’s jury did not include how initial aggressors have a limited right to self-defense, since the judge declined to instruct the jury on the initial aggressor doctrine; perhaps that would have made a difference in the verdict. Others have argued that prosecutors in both cases made strategic errors that may have affected the outcomes. In the Zimmerman trial, half of the jurors reportedly wanted to convict but changed their minds.Deliberations in both cases extended over multiple days before the jurors returned not guilty verdicts.

Of course, in an ideal world, prosecutors would have perfect clarity into guilt and innocence, and prosecutions that result in acquittals after trial would never be brought. That, of course, is not realistic and is the reason why affirmative defenses and trials exist. Moreover, in light of the radical nature of the change wrought by singling out self-defense for immunity, if political prosecutions are the justification, then advocates should put forth more and better examples.

Another rationale that advocates raise is that self-defense is philosophically or morally distinct as a natural or human right. The Republican Party platform refers to the right of self-defense as “God-given.” And the argument that self-defense is a justification and not an excuse is often explained by referencing moral philosophy. But these understandings of self-defense as a natural, divine, or human right have long existed in harmony with adjudication at trial. Blackstone, for example, referred to self-defense as a natural right, but he believed, as described above, that self-defense is squarely a jury question. Saying that self-defense is a natural right does not rationalize treating it as an immunity any more than it rationalizes erasing the common law elements of necessity and proportionality that have long guided self-defense decision-making.

That leaves the fourth explanation, which perhaps arises most often: that gun owners should not have to pay typical criminal defense costs if they have a claim of self-defense. The NRA’s former executive director noted that “the legal fees . . . can easily top $50,000.” A representative of a gun rights advocacy group in Wyoming expressed a similar view: “We don’t want to have a gun owner bankrupted by the criminal process just because he had to use a firearm in self-defense.” And in Utah, an advocate said, “I have people calling me all the time [and saying] I’m afraid it will ruin me if I have to defend myself.” The legislative sponsor of the Utah bill recounted how a person leaving a gun carry class remarked, “I would rather die than financially ruin my family” by using a gun in self-defense.

The cost of criminal defense is a concern for all defendants, not just those asserting that violent conduct was justified as self-defense, and cost typically is not a sufficient rationale for prosecutorial—as opposed to civil—immunity. If self-defense, alone among affirmative criminal law defenses, is to be immunized, it warrants a much stronger rationale than cost saving for gun owners. This is especially true in light of the costs incurred as a result of self-defense immunity that are discussed in the next Part.

III.  THE COSTS OF IMMUNIZING PRIVATE VIOLENCE

The previous Section showed how the usual arguments put forth to support self-defense immunity are thin. It also is important to consider whether immunizing private violence has costs that further undercut exceptional treatment of defensive force. This Part contends that it does: immunizing self-defense can lead to more unlawful violence with less legal oversight; diminish the jury, thereby inviting less accurate and less legitimate outcomes; and introduce inefficiency into the criminal justice process.

A.  More Unlawful Violence (and Increased Impunity)

The message that self-defense immunity sends is troubling: that people can engage in defensive violence that they believe is lawful with less legal oversight. Both logic and data suggest that this message could bring about more assaults and homicides because of the impunity it signals—and in fact provides. Frederick Schauer has observed that “[q]uite often, officials who are immune for one reason or another from formal legal sanctions violate the law with some frequency.” One can expect the same result from self-defense immunity, except for a much larger swath of the population; relatively few people receive official immunity, but everyone is entitled to assert self-defense when defending against criminal charges.

Rafi Reznik has recently argued that the modern understanding of self-defense as a justification, not an excuse, can signal societal acceptance of the alleged offense conduct in a way that promotes more violence; immunity sends an even more powerful signal. As Reznik describes, in the dominant view, a justification indicates that “the wrongfulness of the act is negated.” Excuses, on the other hand, do not negate the wrongfulness of the conduct but “negate the blameworthiness of the actor.” The upshot is that “[j]ustifying self-defense,” as opposed to excusing it, “can . . . amount to an encouragement and it can even amount to an imperative.” Reznik argues that self-defense should be considered an excuse, which it was under English common law. On the ground, however, the trend is going in the opposite direction: jurisdictions are granting immunity to self-defenders, which goes even further down the path toward encouraging the use of violence than considering self-defense a justification.

This trend is especially problematic because people are often wrong about the lawfulness of defensive force. One study found, for example, that a majority of self-reported defensive gun uses are likely illegal. People “view [a] hostile encounter from their own perspective; in any mutual combat both participants may believe that the other side is the aggressor and that they themselves are acting in self-defense.” A particular incident from the summer of the Rittenhouse shooting is exemplary.

Two months before the Rittenhouse shooting, Mark and Patricia McCloskey stood outside their St. Louis, Missouri, mansion as racial justice protesters marched nearby. Both were captured on video screaming angrily and wielding firearms: Mr. McCloskey, an AR-15–style rifle, and Ms. McCloskey, a handgun that she pointed at one protester after another. In Missouri, it is a crime to “exhibit[], in the presence of one or more persons, any weapon readily capable of lethal use in an angry or threatening manner.” A local prosecutor charged the couple with violating that statute. In their defense, the couple asserted that their conduct was justified to protect themselves and their property.

Speaking at the 2020 Republican National Convention (the McCloskeys, like Rittenhouse, became celebrities on the political right for their gun use), Mr. McCloskey, a lawyer, expressed outrage that the prosecutor “actually charged [them] with felonies for daring to protect [their] home.” Then, in a remarkable move, Missouri’s attorney general urged dismissal of the local charges on the basis of the sentiment underlying immunity: “Missourians should not fear exposure to criminal prosecution when they use firearms to defend themselves and their homes from threatening intruders.” In the end, however, the couple effectively conceded that they were not lawfully defending themselves when they pled guilty to the crimes of assault and harassment, thereby waiving any claim for self-defense. In other words, despite their confident assertions that they were legally justified in their actions, they ultimately admitted that they had no legal justification for their conduct.

Unlawful defensive force imposes an especially troubling risk to Black men and women, like many of those marching in front of the McCloskey house, who are mistaken as threats all too frequently. Data has consistently shown that Black people are more likely to be misperceived as a threat than white people. According to L. Song Richardson and Phillip Atiba Goff, this is in part because Black people “serve as our mental prototype (i.e., stereotype) for the violent street criminal.” A prosecution and trial can separate out biased and unreasonable threat perceptions from unbiased and reasonable ones better than any individual can in the moment. And getting it right is important for ensuring a fair and just implementation of criminal law.

Well-intentioned people can have flawed perceptions of lawfulness, but encouraging restraint for defensive violence through the threat of prosecution and punishment is even more important for those who are ill-intentioned. For some people, “genuine and sanction-independent obedience [to the law] is rare.” In that circumstance, “coercion through the threat of sanctions emerges as the principal mechanism for securing the obedience that turns out to be so often necessary.” Immunity lessens the law’s constraining force and risks that someone prone to violence will construe immunity as a license to commit violence.

In this regard, it is notable that a study of the effects of Colorado’s 1986 immunity law found that those invoking immunity “used force (sometimes deadly force) as much out of anger as self-defense.” Moreover, the legal change primarily benefited defendants other than the intended beneficiaries—homeowners confronting stranger intruders. In the years immediately following the enactment, the only “strangers” who intruded into homes and faced defensive force triggering immunity were police officers.

Unfortunately, more recent empirical studies on the impact of changes to self-defense law do not distinguish between the effect of various simultaneous changes, such as Stand Your Ground, presumptions, and immunity. Several such studies have shown that self-defense reforms that include an immunity provision correlate with more violent crime. One study found that in the decade following Florida’s 2005 legislation, “monthly rates of homicide increased by 24.4% and monthly rates of homicide by firearm by 31.6%.” Another found that the law was associated with a 44.6% increase in adolescent firearm homicides. In February 2020, the U.S. Commission on Civil Rights released a report finding no evidence of crime deterrence and an increase in homicide rates in states that adopted such laws. A commissioner recommended rejecting self-defense immunity because it “remove[s] incentives to mitigate or reduce the use of deadly force by protecting the claimant regardless of the collateral consequences.” Yet, as noted, the power of these studies as regards the impact of self-defense immunity is limited and, hopefully, future empirical studies will seek to isolate the effect of self-defense immunity.

A corollary to the signals sent by self-defense immunity is that sometimes immunity can in fact hinder or prevent a conviction of someone who engages in unlawful violence. The analysis of cases soon after Colorado passed its self-defense immunity law in 1986 found that the statute likely led to an acquittal in one case that would otherwise have been a probable conviction, as well as decisions not to prosecute in others. The district attorney for a single county in Kansas has reported “declin[ing] to file charges against thirty-three people based on self-defense immunity,” thirty of which were deemed homicides by the coroner. Three additional cases were charged by the district attorney but dismissed on self-defense immunity grounds by a judge.

Those arguing in support of self-defense immunity do not contest, and implicitly concede, much of this analysis. They acknowledge that the risk of having to defend against a prosecution causes gun owners to hesitate before deploying lethal force, and they seek to reduce such hesitation. However, a cost of immunizing self-defense is to transform the signals sent by conventional self-defense law in a way that likely leads to more unlawful, and at times discriminatory, violence. Furthermore, immunizing self-defense erects an obstacle to achieving a basic goal of the criminal justice system: punishing those who commit crimes of violence.

B.  Fewer Juries in Matters of Community Importance

Another consequence of granting a defendant immunity is to disempower a jury from deciding facts surrounding a properly charged crime. The institution of the jury has long played a central role in self-defense cases. The jury is well-equipped to resolve disputes about the lawfulness of violence. Moreover, and importantly in the context of self-defense, the community’s involvement through the jury legitimates the law and promotes acceptance of outcomes as well as community healing.

Today, the jury is most often discussed solely in the context of defendants’ rights, but the jury’s importance to society is actually far deeper. At the nation’s founding, Anti-Federalists were adamant about protecting the institution of the jury because, even more than protecting the defendant, the jury integrated “the people in the administration of government.” As Laura I. Appleman has put it, “the right of the jury trial” is about “the participation of the citizenry in [the] rule of law.” This feature of the jury—as a key means of community involvement in the law’s implementation—is reflected in the fact that a defendant has no federal constitutional right to waive a jury trial, even if a defendant can demand one. Prosecutors and courts generally can demand jury trials even over the defendant’s objection. Today, as in the past, there is a “strong preference for jury trials on all elements of a criminal case.”

Accuracy is one important interest served by this longstanding commitment to juries, because “[j]uries . . . are considered the best deciders of fact.” This is in no small part because juries “are more representative of their communities than judges . . . . They better represent various races, socio-economic classes, various levels of formal education, differing religions, and a broader spectrum of political engagement than do judges.” This is especially true when the task is assessing “matters reflecting their communities’ values,” like self-defense.

Self-defense is inherently fact-based, calling for answering difficult questions about the reasonableness of a defendant’s perception of—and violent response to—a threat. Evaluating the lawfulness of self-defense calls for an assessment of whether defensive force was reasonably necessary and proportionate to a reasonably perceived threat. Criminal law scholars devise complex classifications in an attempt to capture the permutations of defensive confrontations and how they intersect with the law of self-defense, but it is impossible to resolve self-defense claims through any sort of rote analysis. It is necessary to apply community values and experiences to assess reasonableness, and judges, unlike juries, are often removed from both. Simply because a jury is comprised of a cross-section of the community, the jury will incorporate perspectives and experiences that lead to a fair resolution of disputed facts more so than a single judge who is likely insulated from the circumstances that gave rise to the violence.

Moreover, importantly, community resolution of the difficult factual questions that go into self-defense can legitimate the law and promote acceptance of outcomes. Precisely because “juries have the power to incorporate societal norms and values into their decisions . . . citizens can view these determinations as legitimate and as not influenced by the political leanings of government-employed judges.” That sense of legitimacy, in turn, can help a community accept a case’s outcome and move past the trauma of community violence.

For example, after the killing of Trayvon Martin, the quick decision not to prosecute George Zimmerman led to mass protests across the country. Many thought that the declination of charges suggested that “the criminal justice system was indifferent to Trayvon Martin’s death and was disinclined to try to provide justice.” The fact that Martin was a Black teenager triggered speculation that race was part of the reason for not immediately prosecuting Zimmerman. When a special prosecutor subsequently charged Zimmerman, the move brought great relief. Martin’s mother commented that “[w]e simply wanted arrest, nothing more, nothing less, and we got it.” Although many people who wanted a prosecution may have been disappointed by the jury verdict of not guilty, that the process was followed, and that the decision was rendered by a jury certainly lowered the temperature of the earlier protests.

Conversely, a prosecution’s dismissal because of immunity sends a very different signal to the community. Victims and family members can never know how a jury of their peers would decide on the legality of defensive force. Indeed, a homicide case in Utah elicited the opposite reaction after the defendant was discharged because of self-defense immunity. A family member of the victim of the alleged homicide exclaimed in court: “We all feel the justice system has no doubt failed us.” Another said: “This has forever changed my outlook on the system and the faith that I once had that justice would prevail.” Similarly, in Kansas, after a prosecutor declined to bring homicide charges against juvenile detention officers citing a self-defense immunity law, the victim’s family viewed the decision as “yet another instance of an unarmed Black teenager killed by law enforcement with impunity” and without “even an ounce of accountability.” Likewise, a community partnership expressed “outrage[]” at the declination of charges, viewing it as a “blatant disregard for the life” of the victim.

The denouncements above demonstrate that self-defense immunity can not only prevent a community from healing, but can also undermine the rule of law and faith in the judiciary. In this regard, it is notable that the criticism in such cases is not at the legislature for passing a self-defense immunity bill, or at the governor for signing it, but rather at the “justice system” that “no doubt failed.” Moreover, under the law of self-defense, the harm caused by defensive violence is supposed to “remain[] a legally recognized harm which is to be avoided whenever possible,” and the conduct underlying self-defense is supposed to “remain[] generally condemned and prohibited.” Immunity dilutes the force of such legal values and erodes trust that the judicial system will enforce them.

C.  Inefficient Mini-Trials

One counterargument to concerns about self-defense immunity is that it will only weed out rare, egregious prosecutions. In some places where self-defense immunity is already enacted, the defendant has the burden of proving self-defense at an immunity hearing, or, in the alternative, the prosecutor must only show probable cause that self-defense did not justify the defendant’s violence. In those places, most self-defense cases might still proceed to trial. This, however, raises a question about judicial economy.

To be sure, the likely trajectory for self-defense immunity is for legislators to strengthen it, similar to how Florida recently placed the burden on prosecutors to disprove self-defense by clear and convincing evidence at a pretrial hearing. Since Florida has led the way for NRA-backed initiatives to be subsequently passed elsewhere, it is no surprise that when Utah passed its self-defense immunity law in spring 2021, a legislative sponsor said the law “basically copie[d] and paste[d]” the clear and convincing evidence standard “from Florida[’s] statute.” Furthermore, even in jurisdictions with lesser prosecutorial immunity standards currently, immunity still sends troubling signals that could increase violence.

Setting aside these concerns and focusing narrowly on the argument that immunity will have little impact on prosecutions outside of rare cases, a question arises: Why undertake an expensive immunity hearing that will mirror the eventual trial at all? Two goals of the rules governing criminal procedure are to “secure simplicity of procedure” and “to eliminate unjustifiable expense.” Self-defense immunity runs counter to those goals.

In this regard, it is helpful to contrast self-defense with other pretrial issues discussed above, which generally implicate evidence that is both clear-cut and distinct from proof of guilt or innocence. Whether too much time has passed between criminal conduct and a prosecution so as to violate a statute of limitations, for example, may call only for simple arithmetic unrelated to the alleged offense conduct. The same could be said for speedy trial issues. Determining whether a pending prosecution is substantially the same as an earlier one, thereby violating double jeopardy protections, calls for a comparison of the two prosecutions. And determining whether diplomatic immunity attaches often only requires inquiring into the defendant’s status as a diplomat and whether the sending state has waived the immunity.

Yet proving or disproving whether self-defense exculpates requires consideration of the same witnesses and evidence that will be introduced at trial to prove the charged crime. Indeed, this is implicit in affirmative defenses (like self-defense), which contend that something happening at the time of the alleged offense justified or excused the underlying conduct. Resolving the lawfulness of self-defense ahead of trial would call for delving into the circumstances surrounding the charged offense and receiving testimony from the same witnesses of the alleged crime who will testify at trial. Self-defense immunity hearings, when they do not result in a dismissal, involve “mini-trials of the evidence in advance of the actual trial” that criminal procedure typically seeks to avoid.

To be sure, adding costs and inefficiencies is not always inappropriate. Many scholars agree that grand juries are ineffective at eliminating bad prosecutions and that the plea bargain system that is used to resolve the vast majority of criminal prosecutions creates injustices. Some scholars and advocates have thus suggested reforms that would be costly, like enhancing internal prosecutorial screening or devising something akin to summary judgment for criminal procedure. But self-defense immunity is extrinsic to that broader conversation, which is about how to improve the pretrial process for all issues bearing on guilt and innocence, and for all defendants. Self-defense immunity grants a benefit for one defense championed by powerful lobbyists. That may explain why self-defense immunity is passing in legislatures, but it hardly rationalizes the costs.

CONCLUSION

A central goal of this Article is to show that the exceptionalism reflected in self-defense immunity laws is not rooted in history, tradition, or longstanding priorities of criminal law and procedure. Self-defense has always been an affirmative defense, embedded in a system of defenses and vindicated through the same criminal justice process as other defenses. Those pursuing self-defense immunity have thus far failed to put forward a compelling rationale for a radical departure from legal tradition. Self-defense should remain unexceptional within the system of criminal law defenses to avoid the unwarranted harms that can come from immunizing private violence.

96 S. Cal. L. Rev. 509

Download

* Associate Professor, SMU Dedman School of Law; Fellow, Brennan Center for Justice at N.Y.U. School of Law. Many thanks to Hillel Bavli, Joseph Blocher, Jake Charles, Guillermo Jose Garcia Sanchez, Chris Jenks, Cynthia Lee, Pamela Metzger, Darrell A. H. Miller, Adam Sopko, Jenia Iontcheva Turner, and Cynthia Ward, as well as participants in the U.C. Davis Law Review’s 2021 symposium, for helpful comments and suggestions. Tiereney Bowman, Robert Larkin, Maggie Gianvecchio, Darby O’Grady, Meredith Palmer, and Nick Salinaro provided excellent research assistance.

Suing SPACs

In 2020, the financial world became transfixed by a massive increase in the number of firms going public through special purpose acquisition company (“SPAC”) transactions. A SPAC is a publicly traded company formed solely for the purpose of raising money from investors and choosing a merger partner, thereby bringing the target company public. SPAC shareholders vote on the proposed transaction, but also have the option to redeem their shares for the price paid plus interest prior to the merger. SPACs have always been controversial; they make risky ventures available to unsophisticated investors, may involve acute conflicts of interest, and do not make the rigorous disclosures required in standard IPOs.

Is litigation a solution to these problems? The SPAC boom, as many commentators predicted, precipitated a deluge of lawsuits. Although several studies examine the SPAC transactions themselves, this project is the first comprehensive study of SPAC-related litigation. Using a dataset of all SPAC transactions completed since 2014 and all SPAC-related lawsuits filed since 2017, I assess the prevalence and characteristics of these lawsuits. I find that the probability that a SPAC transaction will generate a lawsuit appears to be unrelated to the returns on the deal, the size of the merger, the industry of the target, and various proxies for SPAC quality. However, I find a negative association between the likelihood of litigation and redemption rate. This is surprising because it means that the SPAC transactions more likely to generate lawsuits are those in which the SPAC shareholders choose to keep, rather than redeem, their shares, presumably signaling greater confidence in the quality of the deal. I argue that many of these lawsuits are opportunistic and may be of questionable quality. I further argue that these lawsuits are an inadequate substitute for the liability that firms face in connection with standard IPOs.

INTRODUCTION

Beginning in 2020, the financial world has been transfixed by a striking rise in the number of firms going public through special purpose acquisition company (“SPAC”) transactions. Although SPACs have been in and out of vogue for several decades, the most recent explosion is unprecedented. In 2020, 53% of all initial public offerings (“IPOs”) were SPACs, and in the first quarter of 2021, that percentage rose to 62%, when SPACs issued more than $30 billion per month in equity.

A SPAC is essentially a publicly traded shell company whose sole purpose is to merge with a private company, thereby bringing it public. The SPAC raises money from investors in an IPO and has a limited period (often about two years) to search for a target. Once the SPAC identifies a merger partner, the target merges into the SPAC in a “de-SPAC” transaction, thus going public while itself avoiding the time-consuming and expensive IPO process. Upon the announcement of a merger, SPAC shareholders are given the opportunity to redeem their shares for the issuing price plus any interest accrued—usually about ten dollars. Low redemption rates are thought to signal confidence in the quality of the merger—shareholders think the transaction is a good one, and thus elect to retain their shares. High redemption rates, by contrast, may signal that shareholders and the market in general are skeptical of the deal.

SPAC transactions have been controversial for most of their existence. They allow young, risky firms to access the public markets—and potentially, unsophisticated investors. De-SPAC transactions are conducted on tight timelines by management who will lose everything if the merger does not close. They may settle for subpar targets or terms, skimp on diligence, or even engage in outright fraud rather than risk losing a deal and returning all the IPO proceeds to the shareholders. Moreover, shareholders that see the transaction through face a risk of significant dilution that they may not understand.

Is litigation a solution to these problems? It would not be crazy to think that surely the heightened risks that investors face in these transactions should be fully disclosed and that SPACs should face dire consequences if these disclosures are incomplete and inaccurate. But by design, the heavy hammer of strict liability under section 11 of the Securities Act and Exchange Act (“section 11”), which penalizes misstatements in standard IPOs, is generally inapplicable to SPACs. This has not deterred plaintiffs, however; top legal blogs and press articles began to bubble with anecdotal reports of a deluge of SPAC-related lawsuits in the spring of 2021. But while numerous commentators have examined the characteristics of the SPACs constituting the boom, so far there has been no comprehensive study of the litigation they have generated. What kinds of claims comprise this flood, and what is their role?

In a first attempt to answer this question, I analyze a sample of 230 SPAC transactions and 150 lawsuits (generated by 62 of those transactions). My sample encompasses the SPAC boom through the second quarter of 2021, which, by all accounts, seems to be the height of the frenzy. I find that the majority of claims arising from de-SPAC transactions are, perhaps unsurprisingly, so-called “merger objection” claims. These are brought either under state law, claiming violation of managers’ fiduciary duties, or under section 14 of the Securities and Exchange Act, alleging fraud in the proxy materials by which management solicited shareholders to vote for the merger. Collectively, these merger objection claims comprise over 60% of the claims in my sample, and most are filed before the merger closes. The second main class of claims comprising the SPAC litigation deluge are broader claims under Securities and Exchange Commission (“SEC”) Rule 10b-5, the all-purpose workhorse of securities litigation, which punishes material misstatements or omissions in connection with the purchase or sale of a security. These are typically brought after the closing of the de-SPAC transaction and comprise roughly 30% of the claims in my sample.

Do these lawsuits target, improve, or deter bad SPACs? Although these lawsuits are in their early days, based on my sample, they appear in fact to target quite good SPACs. In examining these transactions and claims, my main finding is an intuitively odd one: that the likelihood that a de-SPAC transaction will generate a lawsuit—any kind of lawsuit—is negatively related to redemption rate. Transactions with higher redemption rates are commonly thought to be qualitatively worse mergers and might seem likely to generate more lawsuits. My sample, however, shows the reverse: that the probability that a transaction will generate litigation rises as redemption rates decline. This finding is robust to the inclusion of a panoply of controls. An equally puzzling finding is that based on my sample, the probability of being sued appears to be unrelated to deal size, target industry, returns of the merged company, managerial savvy, or various other proxies for deal quality.

What explains this surprising relationship between lawsuits and redemption rates? There are two potential explanations. One cheery possibility is that SPAC-related lawsuits may target fraudulent transactions in which misrepresentation by the management has had the effect of inducing investors to keep, rather than redeem their shares. If this is true, SPAC-related lawsuits may be fulfilling precisely the function for which they are designed: if brought after the merger, they are vindicating the interests of shareholders who did not redeem their shares because they were lied to. And if brought before the merger, these lawsuits may halt bad deals or prompt managers to improve transactions, thus inducing shareholders to keep their shares. If these stories are true, SPAC-related lawsuits are spectacularly successful and should be encouraged.

 But there are also less rosy explanations for the negative relationship between litigation probability and redemption rate. Lower redemption rates—even if they legitimately signal a stronger transaction—could mean larger classes, more purported damages, and higher lawyers’ fees, and may thus be tempting targets for lawsuits after the merger. And in the notoriously dysfunctional ecosystem of merger litigation, challenges to SPACs brought before the deal closes may be filed indiscriminately. Even more damning, they may target fraudulent transactions but nonetheless fail to induce shareholders to redeem. I argue that this is likely because these lawsuits are of low quality, as many merger challenges infamously are. I find that the eventual returns on deals that generated merger objection claims are no better than those that did not, and all of the mergers in my sample closed. There thus appears to be no evidence that these lawsuits are improving de-SPAC transactions or weeding out the bad ones. Roughly half of the merger objection cases in my sample were voluntarily dismissed without litigation within the sample period. The few settlements that are publicly disclosed are for paltry fees. And they are generally filed prolifically and sloppily by a handful of entrepreneurial plaintiffs’ firms composed of fewer than ten lawyers each.

The merger litigation in my sample is the product of a deliberate swap whereby SPACs substitute liability under the merger regime for the heavy cudgel of section 11 liability. In general, we are most concerned about misrepresentations precisely when firms are new to the market and have no history for shareholders to assess; this explains the rigidity of section 11 of the Securities Act, which provides virtually strict liability for misstatements and omissions in a registration statement, which firms file when they conduct an IPO. Because they are shells when they go public, SPACs have nothing to disclose in a registration statement and thus avoid exposure under section 11. To assess the effects of this “SPAC litigation swap,” I compare pre-closing merger objection claims in my SPAC sample to section 11 claims in a sample of IPOs over the same period. I find that in substituting liability under the Securities Act for liability under the merger regime, SPACs have, in terms of sheer volume, fallen from the frying pan into the fire. The number of merger objection claims in 2020 and 2021 is equal to nearly 70% of the number of de-SPAC transactions (as opposed to less than 9% of the IPOs that drew section 11 claims across my sample). But while the merger challenges in my sample have so far disclosed only a few settlements with a mean of roughly $200,000, the mean section 11 settlement in my sample—including cases that were dismissed or are ongoing—is roughly $4.5 million. All of this suggests that merger litigation, especially before the deal closes, may not perform the same policing function as section 11 liability. These cases are pervasive and cheap to settle, meaning that shareholders may not take them seriously as a signal of misconduct, and thus do not redeem their shares. Moreover, the penalties these lawsuits impose are likely too low to induce managers to be truthful.

If SPACs have problems in need of correction, it is unclear whether private litigation is up to the job. Pre-closing merger objection cases, which constitute much of the SPAC litigation deluge, seem to reflect an effort to collect many small fees for small plaintiffs’ lawyers and are unlikely to result in better deals. And although their ultimate impact is unclear, even some lawsuits brought after the merger, largely under Rule 10b-5, may be calculated to procure a high fee for plaintiffs’ counsel, rather than targeting true misconduct. If the SPAC structure constitutes a deliberate bargain by which public investors receive the opportunity to invest in potentially high-risk-high-reward ventures normally unavailable to them in return for less information, perhaps this is no cause for concern. On this view, SPAC litigation may be working as designed; expensive section 11 lawsuits are virtually unknown and are replaced with merger objection claims. Many of these claims, though brought in high volume, are low impact, and the nominal fees to plaintiffs’ counsel are a “tax” that all deal participants are willing to pay. However, if we view SPACs as vehicles for debuting firms to the public market that should disclose information of the same quality to their investors as is required in other IPOs, SPAC-related litigation currently appears unlikely to adequately fulfill this role. While government intervention could be an avenue to remedying any deficiencies in SPACs, reform of shareholder litigation—most urgently, pre-closing merger challenges—is necessary to help police future market innovations.

This Article proceeds as follows. Part I provides some brief background on SPACs and the litigation they have generated. Part II describes my data. Part III discusses descriptive statistics and empirical analysis, and Part IV investigates potential explanations for the odd inverse relationship between litigation probability with redemption rate. Part V compares merger liability in SPACs to section 11 liability in IPOs. Part VI assesses policy implications.

I. BACKGROUND

A special purpose acquisition company, or SPAC, is a public company formed by a sponsor that, following its IPO, has no operations except to search for a non-public firm to merge with and thereby bring public. SPACs originated in the 1980s as “blank check” companies, which often made speculative investments and were considered penny stocks. The SEC adopted tighter controls on blank check companies in 1990 with the Penny Stock Reform Act and Rule 419. While SPACs have evolved in various ways (in some cases, specifically to avoid the strictures of Rule 419), many of their defining characteristics remain the same. In an IPO, SPACs typically sell units priced at $10. Most units consist of a share and a warrant entitling the holder to buy some percentage of a share for $11.50 at a date five years after the completion of the merger. SPACs hold the proceeds from the IPO in an escrow account, in which they accumulate interest. The sponsors of the SPAC have no access to the escrow account. Sponsors usually retain 20% of the shares (called the “sponsor promote”) as compensation, which are contingent on completing a merger.

SPACs cannot form already having identified a merger target, and usually set an 18-to-24-month deadline to find and merge with a target company. Once a target is identified, shareholders of the SPAC are informed and must vote to approve the merger. Independently however, shareholders must also decide whether to keep or redeem their shares; those who redeem are entitled to the price paid for the units, plus a pro rata share of any interest accrued in the trust. Notably, shareholders who choose to redeem their shares can nonetheless both vote for the transaction and keep their warrants (which are unbundled from the shares to trade separately following the IPO). If, however, the SPAC managers fail to identify a merger target in the specified time period, the trust must be liquidated and paid out to shareholders, and sponsors receive nothing (although extensions of time are sometimes negotiated with shareholders). Since it is not certain how many shareholders will choose to redeem their shares prior to the merger, de-SPAC transactions are negotiated to include a minimum amount of cash, and to meet this need, SPAC sponsors frequently arrange for private investment in public equity (“PIPE”) funding.

The popularity of the SPAC skyrocketed in 2020, raising as much cash in that year as in the entire preceding decade, and the first quarter of 2021 alone saw more SPACs created and more money invested than in the entirety of 2020. This meteoric rise has slowed since the second quarter of 2021, although many previously formed SPACs are still shopping for merger partners, and new SPACs are continuing to form. The slowdown is likely the result not only of increased regulatory attention, but of heightened overall scrutiny of SPAC performance.

A. Accounts of the 2020–2021 SPAC Boom

Several expert accounts have painted damning portraits of the SPACs that are the product of the most recent bubble. Using a sample of all forty-seven SPACs that merged between January 2019 and June 2020, Michael Klausner, Michael Ohlrogge, and Emily Ruan found that the structure of SPACs creates substantial costs, misaligned incentives, and on the whole, losses for investors who own shares at the time of SPAC mergers (that is to say, those who do not redeem their shares). These authors evaluated claims about the advantages SPACs offer over traditional IPOs: that they provide a less expensive, faster, and more certain route to the public markets; that they provide a path to being a public company that is often unavailable to firms engaged in complex or uncertain businesses that may be difficult to value; and that they provide more equal access to ordinary investors than alternatives such as private equity. The authors found that SPACs fail on all fronts, largely because of the dilution built into the SPAC structure, and that SPACs tend to drop by at least one third of their value within a year of the merger.

Minmo Gahng, Jay R. Ritter, and Donghang Zhang, in examining SPACs from 2010 to 2020, similarly found that SPAC returns tend to be negative, though cash-weighted returns are less so. They also found that going public via a SPAC transaction is much more expensive for a private firm than undertaking a traditional IPO. However, they found that sponsors take large haircuts and underwriters of SPACs surrender substantial commissions for weak deals, meaning when a SPAC experiences high redemption rates after announcing a merger.

Finally, in a concise overview of the SPAC market as of the summer of 2021, Max H. Bazerman and Paresh Patel also suggested that the SPACs of 2020–2021 are not monolithic in type and have in fact been evolving over the course of the boom. The authors suggested that although not all SPACs are successful, they offer financing opportunities “that compete with later-stage venture capital, private equity, direct listings, and the traditional IPO process [and provide] an infusion of capital to a broader universe of start-ups and other companies, fueling innovation and growth.” They noted that although SPACs had a “questionable start” as blank check companies and have existed as a “niche” “cottage industry” for most of the intervening years, the format has gone mainstream.

B. The SPAC Litigation Deluge

The structure of a SPAC lends itself to various critiques, and thus, legal claims. De-SPAC transactions occur, by definition, under significant time pressure. Moreover, neither the management (via the sponsor promote) nor the underwriter (via commissions) of the SPAC are compensated at all unless a transaction actually occurs. This setup creates incentives for sponsors to close a transaction—perhaps any transaction—rather than liquidate the trust. Additionally, the twenty-four-month time limit means that the SPAC may lack the means or opportunity to conduct sufficient diligence on the target or adequately evaluate the full universe of merger options. Moreover, as the end of the time limit approaches, targets may have increased leverage in negotiations with the SPAC, and though it may serve the sponsors and managers to close the deal before the deadline, shareholders might be better served by extending the deadline or waiting for a more attractive deal. PIPE financing may give rise to additional conflicts of interest by allowing a merger to close at the cost of substantial dilution for the original SPAC shareholders.

Failure to disclose these conflicts could lead to lawsuits under the securities laws. Moreover, SPACs are vulnerable to lawsuits alleging that sponsors breached their fiduciary duties in pursuing the merger, and they may face greater judicial scrutiny of these claims because the SPAC directors may not be disinterested. The de-SPAC transaction may be subject to lawsuits both before and after the merger under Rule 14a-9, which governs the content of proxy disclosures. Broadly, SPACs (and their targets, once the transaction closes) are also public companies and therefore subject to the reporting requirements and fraud prohibitions of the federal securities laws, and thus, false statements and inadequate periodic disclosures may give rise to liability.

Practitioner reports raising alarm about rising SPAC-related litigation began in earnest in spring 2021. This coincided with multiple announcements that the SEC would scrutinize SPACs more closely. Various journalists and industry participants have reported informal tallies, but there has so far been no deep dive into the characteristics and drivers of these lawsuits. In the recent surge of SPAC litigation, which claims predominate, and why? What are the characteristics of de-SPAC transactions that generate lawsuits versus those that do not? Who are the plaintiffs and firms bringing these lawsuits? More importantly, does this surge in litigation reflect SPAC quality, and does it offer a remedy to any of the potential negative outcomes arising from the controversial features of the SPAC format? This Article makes a first attempt at answering these questions.

II. DATA

To assess the recent increase in SPAC litigation, I compose two samples: a sample of lawsuits and a sample of transactions. For my sample of lawsuits, I search Bloomberg Law for dockets in federal district courts, Delaware Chancery Court, and New York Supreme Court containing any references to special purpose acquisition companies. I then manually screen the complaints and dockets of these lawsuits to determine whether the lawsuit arises from a de-SPAC transaction, and if so, what claims are brought. I also gather information on plaintiff law firms and outcomes (if any) from these dockets. My search extends from January 1, 2017, to June 30, 2021.

Next, I compose a sample of all SPAC transactions from January 1, 2014, to July 31, 2021, from SPACInsider and the SEC EDGAR website. I use this range to account for several factors. First, although most claims in my sample are securities claims that have relatively short limitations periods, there are a non-trivial number of contract claims whose statute of limitations is significantly longer. However, my 2017 docket search captured very few lawsuits involving SPACs, and only one involved a transaction that occurred before 2014. I include lawsuits that were filed before June 30, 2021, but many of the cases in my sample are merger objection cases that are filed before the closing of the transaction; accordingly, I extend my transaction sample an additional month. The SPACInsider database furnishes redemption rate, IPO proceeds, implied enterprise value, and warrants issued in the IPO for the majority of these transactions (for those that are omitted, I use SEC EDGAR). I search the websites of the target firm for each transaction to find the target’s age at the time of the de-SPAC. To assess the quality of the sponsors, I code the number of SPACs each management team has participated in at the time of the de-SPAC transaction, which I source from SPACInsider or internet searches. Finally, I take data on returns and dividends for all merged companies from Compustat Daily.

III. DESCRIPTIVE STATISTICS AND EMPIRICAL ANALYSIS

In this Part, I first compare the characteristics of de-SPAC transactions that generated lawsuits to those that did not. I then examine the attributes of the lawsuits in my sample. Finally, I examine the associations between lawsuits and various transaction characteristics.

A. Comparing Transactions that Generated Lawsuits to Those that Did Not

De-SPAC transactions that are subject to at least one lawsuit differ along a few key dimensions from those that are not. These descriptive statistics are reported in Table 1. First, the redemption rates of the transactions that generated lawsuits are significantly lower. The average percentage of shareholders that opted to redeem their shares in de-SPAC transactions that generated lawsuits is roughly 21%; the rate is more than double in de-SPAC transactions that generated no lawsuits at nearly 45%. Perhaps unsurprisingly, the de-SPAC transactions generating lawsuits have significantly higher implied enterprise values: roughly $2.15 billion as opposed to roughly $1.4 billion for de-SPAC transactions that did not draw lawsuits. Transactions generating lawsuits also originated from SPAC IPOs with higher proceeds than those which did not generate lawsuits, with averages of roughly $369 million and $256 million, respectively. Transactions generating lawsuits offer a smaller percentage of a warrant (that can later be used to purchase a share) with each unit sold in the IPO: sued transactions offer an average of 0.44 warrants per unit, as opposed to unsued transactions, which offer an average of nearly 0.55 warrants per unit. Consistent with industry commentary, the age of the target firm is lower for de-SPAC transactions that generate lawsuits with an average of twelve years versus nearly forty-three. However, the median target firm ages are nine and ten years, respectively, and therefore this difference is not significant. Finally, de-SPAC transactions generating lawsuits take about one month longer on average to complete, at 22.4 versus 21.36 months. Although I compare and report the average three-day, seven-day, fourteen-day, thirty-day, and ninety-day post-merger returns for de-SPAC transactions that were sued versus those that were not, there is no statistically significant difference.

Table 1.  Characteristics of Sued and Unsued De-SPAC Transactions

 

Sample of Sued De-SPACs

Sample of Unsued De-SPACs

 
 

Mean

Median

Standard Deviation

Obs.

Mean

Median

Standard Deviation

Obs.

t-stat

Redemption Rate (%)

21.41995

.4683396

31.41641

62

44.83333

50.95

38.0758

168

4.3270

Implied Enterprise Value ($m)

2150.54

1283.5

2739.276

62

1399.343

925

1814.378

165

-2.3949

IPO Proceeds ($m)

369.3452

250

347.367

62

255.9107

229.5

178.7456

168

-3.2349

Warrants

.4442623

.5

.2057031

61

.5477912

.5

.2635458

166

2.7721

Target age at SH vote (years)

12.09677

9

11.54849

62

42.92216

10

218.1146

167

1.1107

Months to complete

22.40323

24

2.58913

62

21.3631

24

3.745537

168

-2.0148

3-day returns

-.0467387

-.0716895

.1475693

28

-.0051028

-.0234741

.1691097

81

1.1585

7-day returns

-.049506

-.1012345

.2070382

53

-.0304111

-.054142

.3861426

142

0.3418

14-day returns

-.0338234

-.0839047

.2580448

54

-.0118041

-.0599034

.653251

141

0.2403

30-day returns

-.0358618

-.1304813

.3650061

29

-.0495576

-.0575954

.2386171

77

-0.2258

90-day returns

-.0484598

-.1514094

.4531122

36

-.0102963

-.0514618

.4307399

84

0.4379

Notes: The sample of sued transactions includes all lawsuits involving a de-SPAC transaction filed between January 1, 2017, and June 30, 2021, located on a Bloomberg Law search of all federal, Delaware Chancery, and New York Supreme Court dockets. The full sample includes all de-SPAC transactions completed between January 1, 2014, and July 31, 2021, sourced from SPACInsider.

            

In Table 2, I report the most common industries represented among the target firms of sued and unsued de-SPAC transactions, based on Fama-French 48 industry classification. The business services industry supplies the target firms for many de-SPAC transactions that are sued and many that are not: 26.67% and 24.85% respectively. While 13.3% of the target firms in sued transactions come from the automotive industry, only 5.45% of the target firms that do not generate lawsuits are in this industry. Similarly, a higher percentage of target firms that draw lawsuits are involved in electrical equipment (8.33%) than those that do not (1.82%). Conversely, transactions involving petroleum and natural gas targets comprise 4.85% of the unsued transactions, but none of the transactions generating lawsuits. While 10.3% of unsued transactions involve pharmaceutical targets, only 3.3% of the sued transactions do. It is interesting to note that across almost all industries reported in this table, redemption rates are higher, sometimes significantly, for the transactions that are not sued than for those that are (the exception is the electrical equipment industry, which furnished just 1.82% of the target firms for unsued de-SPAC transactions).

Table 2.  Industries of Target Firms in Sued and Unsued De-SPAC Transactions Based on Fama-French 48 Classification

 

Sample of Sued De-SPAC Transactions

Sample of Unsued De-SPAC Transactions

Industry

Freq.

%

Mean Redem. Rate

Mean Ent. Value

Mean IPO Proc.

Mean Target Age

Freq.

%

Mean Redem Rate

Mean Ent. Value

Mean IPO Proc.

Mean Target Age

Pharmaceutical Products

2

3.33

7

405

93

3

17

10.30

50

615

124

7

Business Services

16

26.67

36

2571

463

14

41

24.85

43

1832

324

69

Automobiles and Trucks

8

13.33

4

2820

512

7

9

5.45

18

2161

258

29

Retail

4

6.67

16

1213

278

19

7

4.24

71

507

131

15

Petroleum and Natural Gas

0

0

    

8

4.85

32

955

312

7

Entertainment

3

5

52

760

149

4

6

3.64

53

1137

266

15

Electrical Equipment

5

8.33

41

1291

294

26

3

1.82

20

964

185

18.3

Banking

3

5

27

2689

397

9

6

3.64

43

3367

241

15

Other

19

31.67

    

68

41.2

    

Notes: This table includes only industries that comprise roughly 5% or more of the target firms in either sample.

B. Incidence and Attributes of SPAC-Related Lawsuits

In Tables 3, 4 and 5, I report the incidence and characteristics of the lawsuits in my sample. Table 3 reports lawsuits by claim, court, and year. The most prominent observation is the dearth of SPAC-related lawsuits in 2017, 2018 and 2019, the dramatic rise from 2019 to 2020 (from nine lawsuits to forty-six), and the near-doubling of that number in the first half of 2021 alone (to eighty-nine). Lawsuits filed in federal and New York Supreme Courts are largely responsible for this spike (although Delaware has seen a surprising rise—from zero to thirteen lawsuits—in the first half of 2021). The rise in lawsuits generally corresponds with the increase in total SPAC transactions, although the lawsuit curve is shorter and steeper in 2020 and 2021. Notably, the increase in lawsuits corresponds with a sizable decrease in the average redemption rate for all de-SPAC transactions in 2020 and the first half of 2021.

I further tabulate the claims for each lawsuit (I note here that many lawsuits have more than one claim). I report only the most common claims, which are Rule 10b-5 claims, Rule 14a-9 claims, state law claims for breach of fiduciary duty, and contract claims. The most dramatic jump has been in claims brought under state fiduciary law in 2020 and 2021, followed by Rule 10b-5 claims. Rule 14a-9 claims experienced a less pronounced jump, and contract claims have remained low and relatively stable over all the years of the sample. section 11 claims have predictably been virtually nonexistent.

Table 3.  Lawsuits for Sued Transactions by Main Claim Types (January 1, 2017–June 30, 2021)

Year

10b5 Claims

Sec. 11 Claims

14a Claims

State Fid. Duty Claims

Contract Claims

Total Lawsuits

Federal Court

Del. Ch.

Court

NY Sup.

Court

Total de-SPACs

Mean Redem.

Rate of Total de-SPACs (%)

2017

1

0

0

1

2

4

1

1

2

13

47.39

2018

0

0

0

0

1

2

1

1

0

22

54.05

2019

5

2

8

0

0

9

9

0

0

28

64.96

2020

9

0

11

30

8

46

21

0

25

64

37.45

2021 (as of June 30)

38

1

18

42

4

89

56

13

20

85*

22.98

Total

53

3

37

73

15

150

88

15

47

212

 

Notes: Each individual lawsuit may have more than one main claim type. All are tabulated here. Less common claim types, such as unjust enrichment and corporate waste, have been omitted from this table. *Reports transactions closed as of July 31, 2021

       Figure 1 illustrates the timing of these claims. Eighty percent (88) of the claims brought under Rule 14a-9 and state fiduciary duties are brought before the de-SPAC merger, meaning that these claims seek injunctions. The remaining 20% (22) of these claims are brought after the merger for damages. Conversely, 92% (49) of the Rule 10b-5 claims in my sample are brought after the merger.

Figure 1. SPAC Litigation Timeline

I also report in Table 4 the transaction characteristics associated with each type of claim. Transactions associated with contract claims have by far the highest average redemption rate at 35.44%, the lowest average SPAC IPO proceeds at $247.87 million, and take the longest to complete at 23.4 months. Redemption rates for other claim types range from 13.44% to 18.73%, and IPO proceeds from $384.68 million to $412.18 million. Transactions targeted with contract and state fiduciary claims are associated with the highest implied enterprise values, at $2.612 billion and $2.521 billion; Rule 14(a) and Rule 10b-5 claims target transactions with median $2.146 billion and $2.048 billion respectively. Warrants range between 0.4 and 0.49 per unit, and average target age from 9.55 to 13 years.

Table 4.  Transaction Characteristics by Claim Type

 

10b5

14a

State Fiduciary Duty

Contract

 

Mean

Med.

Obs.

Mean

Med.

Obs.

Mean

Med.

Obs.

Mean

Med.

Obs.

Redemption Rate (%)

14.5

.1

20

18.732

.1

25

13.44197

.1

39

35.44367

24.11834

10

Implied Enterprise Value ($m)

2048.775

1046.5

20

2146.336

1267

25

2521.595

1570

39

2612.36

1504

10

IPO Proceeds ($m)

412.18

236.9

20

402.452

280

25

384.6821

280

39

247.87

230

10

Warrants

.4710526

.5

19

.49

.5

25

.3970085

.333

39

.4

.5

10

Target age at SH vote (years)

9.55

7.5

20

10.6

5

25

13

10

39

12.9

9

10

Months to complete

21.75

24

20

22.8

24

25

22.69231

24

39

23.4

24

10

Finally, in Table 5, I tabulate the plaintiff law firms associated with the total number of lawsuits, as well as with each type of claim. I classify plaintiff firms as top plaintiff firms, entrepreneurial SPAC firms, entrepreneurial emerging firms, and top defense firms. Top plaintiff firms are those that appear on the Legal 500 list for securities plaintiff litigation. These are Berman Tobacco, Bernstein Litowitz, Grant & Eisenhofer, Labaton Sucharow, Pomerantz, Quinn Emmanuel, and Robbins Geller. Entrepreneurial SPAC firms are those that brought 5% or more of the lawsuits in my sample. These are Brodsky & Smith, Monteverde & Associates, Moore Kuehn, Pomerantz, Rigrodsky Law, and Robbins Geller. Entrepreneurial emerging firms in my sample are those identified by Klausner and Heglund. The emerging entrepreneurial firms in my sample are Glancy Prongay & Murray, Kahn Swick & Foti, and the Rosen Law Firm. Finally, top defense firms in my sample are King & Spaulding, McDermott Will & Emery, and Williams & Connolly.

       The entrepreneurial SPAC firms lead the pack, with more than twice as many lawsuits as any other category. Notably, the primary main claims driving this result are Rule 14a-9 proxy fraud claims at twenty-one and, even more pronounced, claims under state fiduciary law at fifty-three (hereinafter I refer to these collectively as “merger objection claims”). Rule 10b-5 claims are relatively stable across the three categories of traditional plaintiff firms, with nineteen by top plaintiff firms, sixteen by entrepreneurial SPAC firms, and thirteen by other entrepreneurial emerging firms. The firms traditionally serving defendants, by contrast, are almost exclusively involved in SPAC-related lawsuits by virtue of contract claims, which form a small minority of the main claim types that I tabulate. These claims typically involve disputes among management, or between sponsors/management and PIPE investors, which likely accounts for the involvement of non-plaintiff firms.

Table 5.  Lawsuits by Firm

 

Total Lawsuits

10b5

14a

State Fiduciary Duty

Contract

Top Plaintiff Firms

27

19

4

4

1

Entrepreneurial SPAC Firms

82

23

23

55

0

Entrepreneurial Emerging Firms

14

13

4

3

0

Top Defense Firms

7

0

0

1

5

 C. Empirical Analysis

Table 6 shows the results of OLS regressions in which the dependent variable is a dummy equal to one if the de-SPAC transaction generates at least one lawsuit, and the independent variable is redemption rate. I control throughout for industry using a dummy variable equal to one if the target firm of the de-SPAC transaction is in an industry found by other studies to be particularly vulnerable to securities litigation, including the biotechnology, computer hardware, electronics, retail, or computer software industries. I also control for the log of implied enterprise value of the transaction to assess whether lawsuits are driven by the estimated value of the deal. I control for the target age in years to test anecdotal claims that plaintiff lawyers target de-SPAC transactions involving younger firms. I also control for several proxies for the quality of the de-SPAC merger, including the log of the IPO proceeds of each transaction and the number of warrants per unit issued in the IPO. Finally, I use the number of SPAC transactions the management team had completed prior to the closing of the transaction as a proxy for quality of the management team. All regressions include robust standard errors. To control for the possibility that plaintiffs simply brought more lawsuits at the same time due to some unobserved trend, all regressions include year fixed effects.

The coefficient on redemption rate is negative and statistically significant in all specifications. The coefficient on redemption rate is relatively stable across specifications and is significant at the 5% level in the final specification which includes the most controls (it is significant at the 1% level in several other specifications). The likelihood of a lawsuit rises roughly 0.26% for every 1% decrease in redemption rate. In other words, a de-SPAC transaction with a 25% redemption rate is 13% more likely to generate at least one lawsuit than a transaction in which 75% of the shareholders redeem their shares in advance of the merger.

Table 6.  Probability of a Lawsuit

 

(1)

(2)

(3)

(4)

(5)

(6)

 

Lawsuit Dummy

Lawsuit Dummy

Lawsuit Dummy

Lawsuit Dummy

Lawsuit Dummy

Lawsuit Dummy

Redemption Rate

-0.00309***

-0.00270**

-0.00281**

-0.00277**

-0.00262**

-0.00262**

 

(-4.05)

(-3.18)

(-3.30)

(-3.23)

(-3.01)

(-3.00)

Vulnerable Industry Dummy

0.0153

0.0414

0.0472

0.0472

0.0553

0.0575

 

(0.25)

(0.64)

(0.73)

(0.73)

(0.84)

(0.86)

Log Enterprise Value ($m)

 

0.0431

0.0465

0.0312

0.0259

0.0274

  

(1.45)

(1.56)

(0.75)

(0.61)

(0.63)

Target Age (years)

  

-0.000277**

-0.000257*

-0.000222

-0.000222

   

(-2.68)

(-2.24)

(-1.92)

(-1.91)

Log IPO Proceeds

   

0.0298

0.0313

0.0338

    

(0.51)

(0.51)

(0.55)

Warrants

    

-0.138

-0.142

     

(-1.14)

(-1.17)

Repeat Player

     

-0.00703

      

(-0.29)

Year Fixed Effects

Yes

Yes

Yes

Yes

Yes

Yes

_cons

0.211**

-0.657

-0.714

-0.956

-0.784

-0.850

 

(3.01)

(-1.08)

(-1.17)

(-1.23)

(-0.89)

(-0.94)

r2_a

0.0658

0.0714

0.0758

0.0724

0.0750

0.0710

N

230

227

226

226

223

223

Notes: T statistics in parentheses – * p < 0.05, ** p < 0.01, *** p < 0.001.

I then bifurcate the sample based on claim type. Table 7 shows the results of OLS regressions in which the dependent variable is a dummy equal to one if the de-SPAC transaction generates at least one merger objection claim (either a state fiduciary duty claim or a Rule 14a-9 claim) after dropping transactions that generate lawsuits but no merger objection claims.

Table 8 shows the results of OLS regressions in which the dependent variable is a dummy equal to one if the de-SPAC transaction generated at least one claim under Rule 10b-5 after dropping transactions that generate lawsuits but no Rule 10b-5 claims. I use identical controls as in Table 6, and all regressions include year fixed effects. Although the samples are small, the coefficients on redemption rate are very similar to the one I find in Table 6. A 1% increase in redemption rate corresponds with a 0.25% decrease in the likelihood of a merger objection claim and a 0.22% decrease in the likelihood of a Rule 10b-5 claim.

Table 7.  Probability of a Merger Objection Claim

 

(1)

(2)

(3)

(4)

(5)

(6)

 

Merger Objection Dummy

Merger Objection Dummy

Merger Objection Dummy

Merger Objection Dummy

Merger Objection Dummy

Merger Objection Dummy

Redemption Rate

-0.00294***

-0.00267**

-0.00275***

-0.00266**

-0.00254**

-0.00254**

 

(-4.07)

(-3.31)

(-3.38)

(-3.24)

(-3.08)

(-3.06)

Vulnerable Industry Dummy

-0.0458

-0.0244

-0.0199

-0.0199

-0.0122

-0.0122

 

(-0.80)

(-0.40)

(-0.33)

(-0.33)

(-0.20)

(-0.19)

Log Enterprise Value ($m)

 

0.0312

0.0343

0.00555

0.00668

0.00670

  

(1.09)

(1.19)

(0.14)

(0.16)

(0.16)

Target Age (years)

  

-0.000200*

-0.000159

-0.000144

-0.000144

   

(-2.45)

(-1.69)

(-1.51)

(-1.51)

Log IPO Proceeds

   

0.0581

0.0499

0.0500

    

(1.02)

(0.84)

(0.84)

Warrants

    

-0.103

-0.104

     

(-0.87)

(-0.87)

Repeat Player

     

-0.0000783

      

(-0.00)

Year Fixed Effects

Yes

Yes

Yes

Yes

Yes

Yes

_cons

0.213**

-0.415

-0.468

-0.967

-0.769

-0.769

 

(3.07)

(-0.71)

(-0.79)

(-1.29)

(-0.88)

(-0.87)

r2_a

0.0952

0.0965

0.0992

0.0990

0.101

0.0963

N

215

212

211

211

209

209

Notes: T statistics in parentheses – * p < 0.05, ** p < 0.01, *** p < 0.001.

 

Table 8.  Probability of a Rule 10b-5 Claim

 

(1)

(2)

(3)

(4)

(5)

(6)

 

10b-5 Dummy

10b-5 Dummy

10b-5 Dummy

10b-5 Dummy

10b-5 Dummy

10b-5 Dummy

Redemption Rate

-0.00237***

-0.00223***

-0.00231***

-0.00232***

-0.00218**

-0.00221**

 

(-3.74)

(-3.42)

(-3.48)

(-3.49)

(-3.21)

(-3.22)

Vulnerable Industry Dummy

-0.0277

-0.0134

-0.00745

-0.00756

-0.00170

0.00214

 

(-0.64)

(-0.30)

(-0.16)

(-0.17)

(-0.04)

(0.04)

Log Enterprise Value ($m)

 

0.0166

0.0188

0.0263

0.0252

0.0287

  

(0.77)

(0.86)

(0.97)

(0.91)

(1.03)

Target Age (years)

  

-0.000181*

-0.000190*

-0.000167*

-0.000167*

   

(-2.48)

(-2.41)

(-2.19)

(-2.19)

Log IPO Proceeds

   

-0.0141

-0.00571

-0.00207

    

(-0.27)

(-0.11)

(-0.04)

Warrants

    

-0.0531

-0.0629

     

(-0.52)

(-0.62)

Repeat Player

     

-0.0150

      

(-1.25)

Year Fixed Effects

Yes

Yes

Yes

Yes

Yes

Yes

_cons

0.170**

-0.165

-0.201

-0.0896

-0.190

-0.306

 

(2.88)

(-0.38)

(-0.46)

(-0.13)

(-0.23)

(-0.37)

r2_a

0.0956

0.0952

0.0974

0.0926

0.0867

0.0851

N

188

185

184

184

181

181

Notes: T statistics in parentheses – p < 0.05, ** p < 0.01, *** p < 0.001.

In Table 9, I drop de-SPAC transactions that generated no lawsuits and regress the number of lawsuits per transaction on redemption rate, including the same controls. There is no statistically significant association between the number of lawsuits and redemption rate, although interestingly, the number of lawsuits is negatively related to the vulnerable industry dummy at the 10% significance level in every specification. Less surprisingly, the number of lawsuits per transaction is positively related to the log of the implied enterprise value of the merger at the 10% significance level in the specification including all controls. It seems, accordingly, that while the likelihood of being sued at all is unrelated to deal size, in the subset of transactions that draw at least one lawsuit, large deals draw more litigation than small deals. I note, however, that the number of observations is small, and therefore the predictive power of these regressions may be limited.

Table 9.  Number of Lawsuits in Sued De-SPAC Transactions

 

(1)

(2)

(3)

(4)

(5)

(6)

 

Lawsuits

Lawsuits

Lawsuits

Lawsuits

Lawsuits

Lawsuits

Redemption Rate

-0.00740

-0.00256

-0.00156

-0.00190

-0.00162

-0.00146

 

(-1.18)

(-0.42)

(-0.29)

(-0.35)

(-0.31)

(-0.29)

Vulnerable Industry Dummy

-0.905*

-0.894*

-0.858*

-0.849*

-0.862*

-0.823*

 

(-2.34)

(-2.50)

(-2.40)

(-2.33)

(-2.33)

(-2.32)

Log EnterpriseValue ($m)

 

0.525**

0.578***

0.626*

0.648*

0.675**

  

(3.28)

(3.76)

(2.49)

(2.66)

(2.75)

Target Age (years)

  

-0.0132

-0.0117

-0.0127

-0.0143

   

(-0.64)

(-0.47)

(-0.47)

(-0.51)

Log IPO Proceeds

   

-0.100

-0.210

-0.128

    

(-0.21)

(-0.42)

(-0.25)

Warrants

    

-0.590

-0.710

     

(-0.64)

(-0.72)

Repeat Player

     

-0.109

      

(-0.83)

_cons

2.788***

-8.318*

-9.301**

-8.370

-6.413

-8.418

 

(7.55)

(-2.47)

(-2.96)

(-1.42)

(-1.02)

(-1.22)

r2_a

0.0485

0.123

0.113

0.0982

0.0792

0.0693

N

62

62

62

62

61

61

Notes: T statistics in parentheses – * p < 0.05, ** p < 0.01, *** p < 0.001

       In Table 10, I regress three, seven, fourteen, thirty, and ninety-day returns for the merged firms in my sample on number of lawsuits, controlling for the age of the target firm to assess whether there is a relationship between number of lawsuits and the ultimate success of the merger. There is no statistically significant association in any specification. Finally, to assess whether the likelihood of a 10b-5 lawsuit is related to returns on the merger, in unreported results, I regress a dummy equal to one if the transaction generated a 10b-5 claim on three, seven, fourteen, thirty, and ninety-day returns, controlling for firm age. There is no statistically significant association with returns in any specification.

Table 10.  Returns

 

(1)

(2)

(3)

(4)

(5)

 

Day 3 Returns

Day 7 Returns

Day 14 Returns

Day 30 Returns

Day 90 Returns

Total Lawsuits

-0.0117

0.00131

0.0176

0.0110

-0.0109

 

(-1.18)

(0.10)

(1.12)

(0.47)

(-0.43)

Target Age (years)

0.00000827

0.0000567*

0.0000455*

0.000174

0.0000264

 

(0.69)

(2.21)

(2.28)

(0.21)

(0.45)

_cons

-0.00853

-0.0520**

-0.0628***

-0.0583

-0.0170

 

(-0.50)

(-3.09)

(-3.85)

(-1.71)

(-0.40)

r2_a

-0.00569

-0.00888

0.00496

-0.0155

-0.0154

N

108

194

194

105

120

Notes: T statistics in parentheses – * p < 0.05, ** p < 0.01, *** p < 0.001. Returns are winsorized at the 1% level.

IV. THE ODDNESS OF THE SPAC LITIGATION DELUGE

The punchline result from my sample is an intuitively paradoxical one: that the likelihood of being sued increases for SPACs with lower redemption rates. One might more naturally expect to see lawsuits arising from de-SPAC transactions of lower quality; a high redemption rate would signal that the SPAC investors lack confidence in the proposed merger, and thus that the deal is likely to be worse. Intuitively, it seems that this should generate more lawsuits, not fewer.

There are several possible explanations for this result. First, it is possible that the transactions drawing lawsuits do, in fact, involve fraud. The fraud could induce investors not to redeem, meaning that the low redemption rates reflect the fact that shareholders were lied to, rather than the quality of the deal. This thesis is complicated somewhat by the fact that roughly half of the claims in my sample are merger objection claims that were brought before the transaction closed and shareholders redeemed their shares. In this scenario, it is possible that the lawsuit revealed the fraud and actually induced truthful disclosures that strengthened investors’ confidence in the deal, thereby inducing them not to redeem. Finally, it is possible that that plaintiffs are not targeting SPACs involving fraud, and the negative association between litigation likelihood and redemption rate arises for reasons unrelated to the quality of de-SPAC deals as measured by redemption rate. I assess each of these possibilities below.

A. SPAC Litigation Targets Fraudulent Transactions

One possible explanation for the negative association between the likelihood of a lawsuit and redemption rate is that the transactions that are challenged involve misrepresentations. If the management lies to its shareholders about the quality of its target and diligence, the shareholders may be confident that the deal is a good one and decline to redeem their shares—but this confidence is misplaced. The discovery of such misrepresentations could give rise to lawsuits even where shareholder redemption rates are low.

1. Post-Closing Lawsuits: (Mostly) Rule 10b-5

This seems like a straightforward explanation for SPAC-related lawsuits that are brought after the transaction closes. The majority of such claims in my sample are brought under Rule 10b-5. These claims generally allege a sequence of misconduct that began before the merger and continued through the de-SPAC transaction. Although the recency of the litigation makes it difficult to assess, there are some early signs that these cases may have merit. While half of the merger objection lawsuits in my sample were voluntarily dismissed within the sample period, only a single lawsuit containing a 10b-5 claim was dismissed (out of fifty-three 10b-5 claims). The 10b-5 lawsuits appear to target the de-SPAC transactions with the youngest target companies and the shortest completion timelines, potentially indicating risky business combinations or subpar diligence. Moreover, this is the area in which highly ranked plaintiffs’ law firms have become involved in SPAC-related litigation. The top ranked firms in my sample filed nineteen Rule 10b-5 claims, as compared with four 14(a) claims and four fiduciary duty claims. However, because so few outcomes are available as a result of the longer litigation timeline for these cases, it is difficult to evaluate their quality. The Rule 10b-5 claims in my sample also usually involve defendant SPACs with the highest IPO proceeds (often associated with higher quality deals). The average redemption rate of deals that draw 10b-5 lawsuits is only 14.5%. And finally, I find no statistically significant association between 10b-5 claims and returns on deals in any period. While this finding could be affected by the size of the sample and the availability of information, it may suggest that such lawsuits do not focus on the deals in which investors made out the worst.

There also appear (perhaps unsurprisingly) to be links between Rule 10b-5 claims and public enforcement regarding SPACs. Rule 10b-5 claims began to climb in earnest in the first half of 2021, when the SEC announced that it would be scrutinizing SPAC deals more carefully. This may simply indicate that plaintiffs’ lawyers are jumping on enforcement guidance to make their claims more plausible. However, of the three de-SPAC transactions that have so far generated SEC enforcement actions, two had previously generated multiple Rule 10b-5 lawsuits. The merger between SPAC Diamond Peak Holdings and Lordstown Motors drew SEC and Department of Justice probes in July 2021. The deal, which was completed in October of the previous year, had already generated six Rule 10b-5 lawsuits by the end of the sample period. Similarly, the merger between SPAC VectoIQ and Nikola, which culminated in fraud charges against the combined company’s CEO in July 2021, had drawn three Rule 10b-5 lawsuits during the sample period since its closing in June 2020. Because the SEC is resource constrained, it is commonly thought that the charges it pursues have a high likelihood of merit. That Rule 10b-5 lawsuits arose in significant quantity for these deals before the SEC inquiries suggests that plaintiffs’ lawyers in those specific cases were onto something and that at least in some instances, these lawsuits actually penalize misconduct.

Though they comprise a minority of claims brought after the transaction, I note that there are also some early signs that at least some of the post-closing merger objections in my sample may have teeth. These claims consist of Rule 14a-9 claims alleging proxy fraud and state fiduciary duty claims (often, they are bundled together). Traditionally, post-closing lawsuits for breach of state fiduciary duty are brought by target shareholders alleging that they received an insufficient price for their shares. Analogous claims brought by SPAC shareholders, however, are acquirer shareholder claims, alleging that SPAC directors breached their fiduciary duties by allowing the SPAC to overpay for the target. In the absence of conflicts, such lawsuits are subject to the business judgment rule. This may explain why relatively few of the post-closing fiduciary duty lawsuits in my sample are couched in these terms. Rather, most of the post-closing fiduciary duty claims in my sample allege that the SPAC directors breached their duty of oversight by permitting the fraudulent statements alleged in the proxy.

One notable exception is In re Multiplan Stockholder Litigation in the Delaware Chancery Court. The case arose out of a transaction between Churchill, the SPAC, and Multiplan, a healthcare data analytics provider. The crux of the SPAC shareholders’ allegations was that Multiplan’s largest customer, on which it was dependent, was forming an in-house competitor and planned to shift its business away from Multiplan, and that Churchill’s board deliberately failed to disclose this information in the proxy. The complaint alleged that the SPAC directors, who were all appointed by and had previous ties with the SPAC sponsor, violated their fiduciary duties by “issuing a false and misleading proxy, harming stockholders who could not exercise their redemption rights on an informed basis.” In a widely cited opinion, Vice Chancellor Lori Will denied the directors’ motion to dismiss the claims against the SPAC board and sponsor, finding that the claims were direct, rather than derivative, and therefore plaintiffs need not plead demand futility, and that the board conflicts meant that the transaction would be reviewed under the demanding entire fairness standard. If these determinations hold after trial, they will likely be powerful tools for SPAC investors alleging that they were lied to in connection with a de-SPAC merger. The entire fairness standard only applies where there is a conflict of interest, but SPAC sponsors may often be conflicted, and where sponsors appoint boards with which they have financial ties, as they often do, the threat of entire fairness review is likely to have bite. Moreover, unlike Rule 10b-5 lawsuits, which often take years to resolve, fiduciary duty cases in the Delaware Chancery Court are likely to move relatively quickly, suggesting that these may be the cases most likely to curb SPAC misconduct in the immediate future.

2. Pre-Closing Lawsuits: Merger Objection Claims

Pre-closing merger objection claims are a different beast from most Rule 10b-5 litigation. Much merger objection litigation is disposed of before the closing of the transaction. Thus, the effects of pre-closing merger objection litigation on de-SPAC transactions may, in theory, be quite direct; specific deals may be halted, or management may release additional information pertaining to specific aspects of the deal, giving shareholders greater confidence (or not, as the case may be). Rule 10b-5 litigation, by contrast, is notoriously slow, often taking years to be resolved.

The characteristics of transactions that drew the Rule 14a-9 and state fiduciary duty claims in my sample are reported in Table 4. I note that state fiduciary duty claim defendants boast the lowest mean redemption rate of the claim types in my sample at 13.44% (Rule 14a-9 claims correspond with slightly higher redemption rates, at 18.73%). De-SPAC transactions that are targeted by state fiduciary duty lawsuits have a higher mean enterprise value but lower IPO proceeds than those of Rule 14a-9 claims. They also issue fewer warrants, and the target companies are slightly older. All this indicates that along most metrics (redemption rate, warrants, and target age), lawsuits alleging breaches of state fiduciary duties tend to target de-SPAC transactions that generally appear to be higher quality deals.

If fraud explains the association between rising litigation likelihood and falling redemption rates, the merit of the pre-closing claims in my sample is complicated to assess. The redemption rate in de-SPAC transactions is tabulated very near the closing of the transaction, so redemption rates should, in theory, incorporate any information revealed in a pre-closing lawsuit. Accordingly, these lawsuits could be having one of two effects. First, it is possible that the pre-closing merger objection claims in my sample are halting some bad transactions and lowering redemption rates in deals that close by giving investors greater confidence in the deal. This could be the result of the supplemental disclosures that many defendants issue in response to merger objection litigation. These supplemental disclosures could correct any misrepresentations or omissions and make investors feel good about retaining their shares. In the best case, these disclosures could prompt heightened diligence in management’s pursuit of the deal. The second possibility is less rosy; merger objection claims could target fraudulent transactions, but nonetheless fail to induce investors to redeem, either because investors do not perceive these lawsuits to credibly signal fraud, or because SPAC managers, undeterred by the monetary and reputational consequences of these lawsuits, continue to lie to their investors. While merger objection claims are theoretically a promising means of improving de-SPAC transactions before they close, I argue that they do not appear to be fulfilling this role. If fraud causes investors to sue but prevents shareholders from redeeming, I argue that the more plausible explanation for the negative association between the likelihood of pre-closing merger litigation and redemption rate is that these lawsuits fail to induce investors to redeem their shares, or SPACs to be more truthful in their disclosures.

Merger objection litigation over the last decade has been aptly described as “schizophrenic.” At its peak in 2013, a whopping 96% of announced mergers gave rise to litigation by investors, and 60% of the lawsuits were filed in the Delaware Chancery Court. These lawsuits drew widespread criticism on the ground that they largely lacked merit and that settlements were generally for pre-merger supplemental disclosures that did not benefit shareholders, but did generate lucrative fees for plaintiffs’ lawyers. In January 2016, the Delaware Chancery Court cracked down on these lawsuits in what proved to be a landmark decision, In re Trulia. In that decision, the court declared its refusal to approve settlements in merger cases that do not provide “[m]eaningful [b]enefit” to shareholders.

The literature has demonstrated that the effect of Trulia, in addition to other concurrent cases raising the bar that plaintiffs must meet in merger objection cases, was a plummet in the number of merger objection cases filed in Delaware. To prevent plaintiffs from merely fleeing to other state courts, Delaware’s legislature also allowed for the adoption of forum selection bylaws prohibiting lawsuits in other states. However, these provisions do not extend to Exchange Act claims in federal courts, and accompanying the drop in Delaware cases was a marked increase in merger-related lawsuits filed in the federal courts under Rule 14a-9. By 2018, one account found that only 9% of merger-related lawsuits were brought in Delaware, while 87% were brought in federal court.

In addition to circumventing Delaware’s strictures on merger litigation jurisdictionally, plaintiffs’ lawyers have also found creative ways to avoid the constraints on settlements for window-dressing disclosures and lawyers’ fees. Following Trulia dicta, many defendants have begun to make voluntary supplemental disclosures in response to merger objection lawsuits, thus rendering the claims moot, and paying plaintiffs’ counsel a mootness fee as long as those disclosures were of benefit to shareholders (even if not material to the vote). Various commentators, including courts, have expressed skepticism about this practice, claiming that the supplemental disclosures in contemporary merger litigation confer no real benefit on shareholders. Nonetheless, a recent study has found that at least 63% of merger cases were disposed of with a mootness fee and that such fees appear to have entirely replaced formal settlements in merger objection cases in federal court.

To investigate whether pre-closing merger objection claims improve de-SPAC transactions, I run unreported regressions in which the dependent variables are the three-day, seven-day, fourteen-day, thirty-day, and ninety-day returns on the deal, and the independent variable is a dummy equal to one if the transaction generated at least one pre-closing merger objection claim, controlling for target age. There is no statistically significant association in any specification. Thus, there is no evidence that transactions that draw a pre-closing merger objection claim perform better in the first three months than transactions that draw no lawsuits.

To assess the role of the supplemental disclosures that may be prompted by merger challenges, I use SEC EDGAR and internet searches to see whether defendant SPACs voluntarily released additional disclosures in response to pre-closing lawsuits. In my sample, defendant SPACs issued supplemental disclosures in response to pre-closing merger objection lawsuits 46.15% of the time. The average redemption rate of the de-SPAC transactions that were sued prior to the merger that did not issue supplemental disclosures is 19.11%, while the average redemption rate for those that did issue supplemental disclosures is 6.78%. However, this difference is not statistically significant. The rate of supplemental disclosures in my sample appears to be substantially lower than that which other studies have documented regarding deal litigation in federal courts. Half of the merger objection cases in my sample were voluntarily dismissed by the end of the sample period. Roughly half of those dismissals involved defendant SPACs that appear to have made no supplemental disclosures at all.

I also use SEC EDGAR and internet searches to examine the fees paid by SPACs in connection to these lawsuits. These fees are generally immaterial to the defendant companies that pay them and therefore are often not widely disclosed. They also involve dismissal before certification of a putative class and are therefore not subject to approval by courts (unsurprisingly, none of the merger objections in my sample involve on-the-docket settlements). Very few of the transactions in my sample disclosed mootness fees paid in connection with lawsuits; the average amount was $200,000. This is in line with estimates from other studies, which have found that mootness fees range in amount from $200,000 to $450,000 but have declined in more recent years.

There are two likely overlapping possibilities for why these lawsuits might not induce SPAC shareholders to redeem their shares, even if they target transactions that involve misrepresentations. First, shareholders may not believe that these lawsuits credibly signal fraud. Their ubiquity may mean that investors simply do not take these lawsuits seriously. The disclosures they prompt may not seem to add anything important. Previous literature has concluded that, even when they are made, the supplemental disclosures of yore did not influence shareholder voting outcomes, and the current disclosures required in exchange for mootness fees need not even be material. The amounts that SPAC defendants pay to make these lawsuits go away may not seem to investors like a penalty to remedy serious wrongdoing, but an added tax to make sure the deal closes on time.

There are other reasons that these pre-closing merger objections may not signal to investors that something is truly amiss. More than half (57%) of the merger objection lawsuits brought in New York Supreme Court are against SPACs that have a forum selection clause in the charter or bylaws delegating Delaware as the forum for any fiduciary duty litigation. Of these, 76% were voluntarily dismissed within the sample period without litigation, suggesting two possibilities: (1) the plaintiffs filed first and researched the defendant firms later, discovering the forum selection provisions after the lawsuits were filed and then voluntarily dismissed them; or (2) the plaintiffs privately demanded fees from the defendant SPACs in return for withdrawal that were lower than the cost of litigating the forum selection provision, suggesting that the fees, and thus the settlement value of these cases, were very low. Neither option inspires confidence in the quality of these lawsuits.

Furthermore, the most frequently filing law firms’ litigation patterns in my sample suggest that these cases may not represent credible signals of misconduct. I report the plaintiff firms in my sample in Table 5. The three most frequently-filing plaintiffs’ firms in my sample—Brodsky Smith, Monteverde & Associates, and Rigrodsky Law—account for just over 40% of the lawsuits in my sample. These frequent filers appear to be small plaintiffs’ firms of between five and eight lawyers. Their prolific but largely unlitigated filings have been noted in other studies, leading commentators to remark that “these law firms appear to be more interested in collecting mootness fees than in actively litigating the cases that they file.” Moreover, the complaints filed in the New York Supreme Court, as noted by other commentators, purport to seek an injunction blocking the merger, but do not include the separate motions for a preliminary injunction that would actually be necessary to put off the shareholder vote. This account of entrepreneurial SPAC firms is consistent with many assessments in the literature of plaintiffs’ lawyers’ behavior in the securities and corporate context. One study postulates that the passage of the Private Securities Litigation Reform Act of 1994 (“PSLRA”), and the higher expenses involved in meeting its pleading standards, pushed many small firms away from 10b-5 class actions and toward corporate litigation, such as that challenging mergers. Others note the rise in dismissals (used as a proxy for low-quality cases) concurrent with the rise in lawsuits brought by “emerging firms” in the securities arena generally. Still others have noted that plaintiffs’ firms have proven resilient to attempts to crack down on nuisance litigation and have circumvented these efforts, particularly in the merger litigation context, by filing in federal courts or jurisdictions outside Delaware and by evading requirements for more useful disclosures by voluntarily dismissing cases in exchange for mootness fees.

The second reason that these lawsuits may not induce shareholders to redeem even if they target truly fraudulent transactions is that the fees paid to plaintiffs’ attorneys may be unlikely to deter such fraud. Unscrupulous SPAC managers could pay $200,000, issue supplemental disclosures, and continue to lie about the quality of the target or the diligence they conducted in order to close the transaction and receive their 20% promote. This may contrast with more traditional IPO litigation under section 11, as discussed in Section III.B.

Many of the regulatory and scholarly concerns for the welfare of SPAC investors revolve around issues that might be mitigated with improved disclosure, such as conflicts of interest, speedy deal process, and dilution. Pre-merger deal litigation could theoretically be a mechanism well-suited for providing greater transparency to investors. However, there are reasons to think that even if these cases are targeting fraudulent transactions, they fail to induce SPAC shareholders to redeem. If true, this explanation is a strong condemnation of merger litigation generally, and even more so in the SPAC context: in transactions that are known to be risky to investors, that are by their nature fraught with conflicts of interest, and in which fraud could plausibly be rampant, the type of lawsuit that, in theory, could be of the most preemptive benefit to shareholders is being leveraged by a small number of small firms for small fees at investors’ expense.

B. SPAC-Related Litigation Is Unrelated to Transaction Quality

Of course, it is possible that the SPAC-related lawsuits in my sample do not target fraudulent transactions, and that the lower redemption rates of the transactions that appear to draw lawsuits are not the product of misrepresentations. If this is the case, it appears that plaintiffs may simply have chosen lower-redemption SPACs to sue. The probability of being sued appears to be related neither to the size of the deal, nor the returns, nor the industry, nor the management. Accordingly, in the absence of fraud, we can say little more than that these lawsuits do not appear to be logically related to common metrics for transaction quality, including redemption rate.

There is some reason to think that not all of the low-redemption deals which appear more likely to draw lawsuits involve fraud. My sample suggests that the boom year for redemptions was 2019, with a mean redemption rate of 64.96% across all de-SPAC transactions completed that year. However, only nine SPAC-related lawsuits were filed. The mean redemption rate declined to 37.45% in 2020 as lawsuits climbed to forty-six and even further in the first half of 2021 to 22.98%, as lawsuits exploded at a whopping eighty-nine. Significantly, the first half of 2021 was also the time when the SEC, under the then-new Biden administration, began publicly to take aim at SPACs. The end of March and early April 2021 saw a highly publicized cluster of SEC releases and statements specifically geared at SPACs, addressing, among other issues, key filing issues, accounting considerations, and liability risk under the securities laws. By the end of April, reports were circulating that the SEC was considering new guidance on SPACs. If fraud is the explanation for low redemption rates, then conversely, SPACs reporting high redemption rates are the ones that told the unpromising truth about their prospects and diligence, causing investors to redeem. But according to my sample, these candid, high-redemption SPACs proliferated in earlier years, while SPACs reporting the lowest redemption rates (resulting, presumably, from lying to their shareholders) did their lying concurrently—at least in part—with the announcement that regulatory scrutiny of SPACs was significantly on the rise. It seems counterintuitive that SPACs would be honest when no one was looking and engage in fraud just as the SEC announced that it would be looking very carefully.

To be sure, there is not complete alignment of timing—the SEC’s statements did not begin until April, meaning that SPACs closing earlier in 2021 would have no reason to be more truthful with their investors (other than the ascension of a more pro-regulation administration) than the SPACs before them. However, there may be some explanations for the relationship between redemption rate and lawsuit probability that do not involve fraud.

1. Post-Closing Lawsuits: Standing and Class Size

One potential explanation for the negative relationship between the probability of a lawsuit and redemption rate may be putative class size. In a post-closing lawsuit, fewer redeeming shareholders could mean a larger potential class of investors, which would presumably result in more damages, greater pressure to settle, and a larger potential payout, and therefore a more lucrative lawsuit for plaintiffs’ counsel.

This line of reasoning operates slightly differently for the different types of claims in my sample. Most directly, shareholders have standing in Delaware to challenge a merger under laws of fiduciary duty only if they owned stock at the time of the challenged transaction and throughout the litigation. This means that shareholders who redeem their shares before the merger are clearly out. Similarly, private plaintiffs are presumed to have standing under Rule 14a-9 if they were injured in connection with a proxy solicitation and if “reliance of some shareholders on the statement was likely to affect [how they voted].” Shareholders in SPACs can vote for the transaction and still redeem their shares, but in such a situation are unlikely to have been injured.

Plaintiffs have standing under Rule 10b-5 if they purchased or sold securities in connection with manipulative or deceptive conduct. Such a class could encompass investors who bought the shares of the merged company on the secondary market in reliance on misleading statements issued before or in connection with the merger. Nonetheless, a large class of non-redeeming shareholders who bought their shares based on rosy statements about the deal could also form the basis of a lucrative class for an action under Rule 10b-5.

If class size is the explanation for the negative relationship between litigation probability and redemption rate, then these cases may be opportunistic, rather than targeting the worst transactions. Larger classes likely involve larger damages, create the shadow of greater potential liability, and thus are likely to induce greater pressure to settle and probably a larger payout for plaintiffs’ lawyers. This may also be an explanation for the greater involvement of top tier plaintiffs’ lawyers that does not rely on merit; these firms generally have relationships with institutional investors who are well-positioned to be lead plaintiff in Rule 10b-5 class actions if they want to be and may deliberately target these cases based on potential payoff.

2. Pre-Closing Lawsuits: Litigation as a Substitute or Complement for Redemption

One possible explanation of the negative relationship between litigation and redemption rate in de-SPAC transactions for pre-closing merger challenges that does not involve fraud may be that SPAC shareholders are increasingly using litigation as a substitute or complement for exercising their redemption rights.

On its face, using litigation as a substitute for redemption makes little sense. The SPAC shareholders’ redemption right is essentially a free put option, allowing shareholders to costlessly back out of the transaction. Moreover, many redeeming shareholders do not entirely surrender their claims to the company should the merger prove to be a good one—they may keep their warrants and vote for the transaction, even if they choose to redeem their shares.

To be sure, freeriding on a lawsuit is also a costless option. There appears to be little downside to a nonplaintiff shareholder to watching such a lawsuit unfold. In the context of pre-closing merger objection litigation, any supplemental disclosures may improve deal quality (although, as discussed, this is unlikely), and although the sample is small, my results indicate that share prices of the merged company do not appear to suffer as a result of pre-closing merger challenges. But while the occasional shareholder may actually believe that a merger objection claim will increase the value of their post-merger shares, this appears unlikely based on the supplemental disclosures issued in my sample and given the dysfunctionality of much merger litigation generally. Indeed, it appears that many of these lawsuits are driven by plaintiffs’ lawyers in exchange for nominal fees rather than by SPAC shareholders. Accordingly, it seems generally improbable that the negative relationship between the likelihood of a lawsuit and redemption rate is the result of shareholders treating these options as substitutes.

It might also be possible that sophisticated investors, who may be more likely to redeem, are also more likely to sue as a means of extracting the most possible value from their shares. Although the most direct monetary gains would result if redeeming investors could bring post-closing lawsuits for damages, they generally lack standing to do this. It is also possible that sophisticated investors who redeem might bring pre-closing lawsuits in the hopes of improving the transaction and thus upping the value of their warrants (which they can retain through redemption). I note, however, that the vast majority of merger objection claims in my sample are brought by individual investors, rather than the institutional ones more likely to engage in sophisticated strategy. It is therefore not clear that litigation is being used as a complement to the redemption right intended to make money.

C. Summing Up the SPAC Litigation Deluge

If pre-closing merger challenges produce immaterial disclosures in exchange for nuisance fees, there does not appear to be a plausible interpretation of the negative relationship between lawsuit probability and redemption rate that redeems these lawsuits. If these challenges target high-quality transactions, they are unhelpful. It seems highly unlikely that they are a common or logical substitute or complement for redemption. And if they target fraudulent transactions, they seem to be a failure under the very circumstances where they could, in theory, be most useful. Proponents of merger litigation may argue that these cases provide an opportunity for plaintiffs to force SPACs to improve or abandon their acquisitions and are therefore valuable. This potential indubitably exists and may represent the optimal result for investors, who could redeem their shares, or even, in the best case, benefit from a better deal without the injury of a bad transaction necessitating after-the-fact recompense. But I argue that merger objection lawsuits as they play out on the ground are not performing this role and are therefore undesirable, irrespective of whether they target fraudulent or high-quality deals.

Lawsuits brought after the merger are somewhat more difficult to characterize, in part because the vast majority have not been resolved. If low redemption rates are the product of misrepresentations, these lawsuits are likely meritorious. They also probably are effective deterrents of bad conduct; large class actions are among the more terrifying prospects to corporate management, and perhaps even more so these days for SPACs, which are facing increasingly narrow D&O policies to insure against such lawsuits. Class actions under Rule 10b-5 in particular are likely to be drawn-out affairs, and decreased insurance coverage may mean that the costs of such litigation may be more likely to fall on the defendant firm. But if the low redemption rates are the product of truthful disclosures and signal high-quality deals, the post-closing lawsuits challenging them may be opportunistic. Plaintiffs’ lawyers may be targeting these cases not because there is an indication of misconduct, but because a lower redemption rate means a larger class, more extensive purported damages, greater settlement pressure, and a higher potential payout.

Accordingly, it appears at this early stage that many of the lawsuits comprising the SPAC litigation deluge—the merger challenges brought before the merger—are accomplishing little other than a minimal payout for a few small plaintiffs’ firms. Claims brought after the merger closes may effectively punish fraud. But it is also possible that even some of these lawsuits may be brought for opportunistic, rather than meritorious reasons. And if this is the case, they may be creating a significant cost for the transactions that appear to be of the highest quality. Fully disentangling these competing explanations—that SPAC-related litigation targets fraudulent transactions, or in fact targets quite good transactions—is a complex and perhaps impossible task. But there is at least a credible argument that much of this litigation may be failing to improve SPAC transactions and may be brought for reasons unrelated to SPAC quality.

V. THE BIGGER PICTURE: COMPARING SPAC MERGER LITIGATION TO SECTION 11 LITIGATION

SPACs are engineered specifically to avoid the expensive and time-consuming IPO process. More specifically, they are engineered, in some respects, specifically to avoid liability exposure under the securities laws, which is particularly stringent for IPOs. Since the targets of SPACs go public through a merger instead of an IPO, they are governed by the regime for mergers instead of IPOs. In this section, I explore how, at this early stage, these different liability regimes stack up against each other.

A. Problems Avoided by SPACs: Securities Act Liability and the Forward-Looking Statement Safe Harbor

There are two main reasons commonly cited for the reduced liability exposure faced by SPACs. These are liability under section 11 and liability for forward-looking statements.

The Securities Act of 1933 governs liability for primary offerings, including IPOs. A company issuing securities is strictly liable under section 11 of the Securities Act for material misstatements and omissions in its registration statement, the instrument by which it must register new shares for sale with the SEC. The company’s underwriters, officers, and directors are liable as well; although they may benefit from a “due diligence defense,” in practice, this defense is available only under a very demanding standard. Section 11 liability is virtually nonexistent in connection with SPACs. There are several reasons for this. First, the SPAC IPO occurs when the SPAC is a shell company and therefore has almost nothing in the way of operations or financials to disclose. Second, SPACs could face section 11 liability in connection with the shares that they issue to target shareholders or PIPE investors in connection with the merger. However, once these shares mix in the market with shares issued in the IPO, it is very difficult to “trace” them to the registration statement of a particular offering. Such “tracing” is a requirement for the standing of a purported section 11 plaintiff, meaning that these cases are likely to be rare.

The second reason commonly cited for going public via SPAC rather than via IPO is the safe harbor for forward-looking statements. The PSLRA allows issuers to make projections and forward-looking statements without fear of liability for such statements under the securities laws, as long as the statements are accompanied by meaningful cautionary language. This safe harbor is unavailable in IPOs. The ability to make forward-looking statements may be particularly valuable to pre-revenue SPAC targets, who do not yet have historical earnings with which to entice investors, or to firms in high-tech industries which require extensive start-up capital. However, the demise of this advantage may be imminent; regulators have hinted that the safe harbor may not apply to SPACs after all, and the House of Representatives has released draft legislation explicitly eliminating the safe harbor for SPACs.

B. Merger Liability for IPO Liability: The SPAC Lawsuit Swap

In going public via SPAC, firms essentially swap the IPO liability regime for the merger liability regime. What are the effects of this exchange? To examine how these lawsuits stack up against one another, I compare the merger objection lawsuits and unsued SPACs in my sample to a sample of IPOs from January 1, 2016, to July 31, 2021, matched with a sample of section 11 class actions from January 1, 2017, to June 31, 2021. I gather my sample of section 11 class actions from the Stanford Securities Class Action Clearinghouse and the sample of non-SPAC IPOs from Zephyr.

Descriptive statistics are tabulated in Table 11. My total sample consists of 701 IPOs, 60 of which draw section 11 claims. Although the rate varies year over year (it spikes to over 17% in 2019, and the data for 2021 covers only the first half of the year, coming to nearly 3%), the average percentage of IPOs across the sample that draw a section 11 claim is 8.56%. This is far lower than the percentage of SPACs in my sample that drew merger objection claims during the same period (either under state corporate law or Rule 14a-9), which is 51.89%. Moreover, the average across the sample is understated since there are very few merger objection claims in the first three years of my sample; the average percentage in 2020 and 2021, when these lawsuits began to appear in earnest, is 67.69%.

The consequences of these lawsuits also appear to be dramatically different. The average settlement for section 11 lawsuits in my sample is roughly $4.5 million. In calculating this average, I include cases that were dismissed (where the settlement amount equals zero) and cases that are still ongoing (where the settlement amount is currently zero but may be significant when the case is resolved). This accounts to some degree for low settlement values in the last few years of the sample. But the $4.5 million average, though likely substantially understated, is still much larger than the settlements that have so far been disclosed for merger objections in my sample, which average $200,000. Indeed, the likely reason that so few of such settlements are publicly available is that they are so small as to be immaterial to the issuers and therefore need not be disclosed.

There is also a divergence in the plaintiffs’ firms that bring these cases. Section 11 lawsuits, across my sample, involve a top-tier plaintiffs’ firm 63.3% of the time, and this number is relatively stable year over year (ranging from 60% to 69%). Top-tier plaintiffs’ firms account for only 7.27% of the merger objection claims in my SPAC sample. Finally, the most commonly sued industries in the IPO sample are software and programming (21.67% of the sample) and biotechnology and drugs (11.67%). These are among the industries found in other studies to be particularly vulnerable to securities litigation. By contrast, SPACs that draw lawsuits are generally not in these industries.

Table 11.  Section 11 Lawsuit Descriptive Statistics

Year

Section 11 Lawsuits

Mean Settlement*

Percent Top Plaintiffs’ Firms

Total Non-SPAC IPOs

Percent Non-SPAC IPOs Sued Under Section 11

2017

13

$15,600,000

69.23%

155

8.38%

2018

11

$6,445,455

63.64%

125

8.8%

2019

20

$813,750

60.0%

115

17.39%

2020

11

$0

63.64%

137

8.03%

2021

5 (as of June 31)

$0

60.0%

169 (as of July 31)

2.96%

Total

60

$4,571,841

63.3%

701

8.56%

Notes: *Mean settlement calculations include settlement values for lawsuits that were dismissed and lawsuits that are ongoing.

So, what can we make of all this? In terms of sheer litigation volume, SPACs appear to have fallen from the frying pan into the fire. Though designed in some respects specifically to avoid litigation under section 11, SPACs draw merger objection claims with dramatically greater frequency, and as previously discussed, the bulk of these claims are brought before the merger. However, insofar as information is available, these claims appear to settle for far lower amounts, meaning that for individual SPACs, the increased probability of a lawsuit may be outweighed by the apparently low probability of having to pay an expensive settlement. The majority of section 11 claims are brought by top-tier plaintiffs’ law firms, which may indicate that they are stronger claims than the merger objections, which are brought by a handful of quite small entrepreneurs.

For IPOs, the policy choice has been to wield the heavy cudgel of virtually strict liability against those perceived to have the best knowledge about the newly public firm. This choice was made on the basis that firms new to the market, about which little is known, without any public history and potentially few profits for investors to assess, could be prime candidates for fraud. Investors in such firms have no source of information other than the firm itself, and Congress created the rigid liability provisions of the Securities Act to make that information as accurate as possible. These provisions create the specter of extraordinary costs for those who violate them and are thus commonly regarded as “well suited to deter misreporting.” Under the current legal framework, SPACs do not face this standard. Rule 10b-5 plaintiffs must plead specific allegations of, at the least, recklessness. Rule 14a-9 litigants must prove negligence to prevail, which may be a more demanding task than it appears. Under Delaware law, acquirers’ challenges to mergers under state corporate law are usually reviewed under the deferential business judgment rule. Although there are good arguments that SPAC mergers, in view of potential management conflicts, should be reviewed under the more demanding “entire fairness” standard, any of these standards is harder work for plaintiffs than the virtually strict liability of section 11.

None of these concerns motivating the rigidity of section 11 liability are any less salient in SPACs, and given the conflicts commonly cited with respect to the SPAC form, they are arguably more intense. Yet, the merger litigation that replaces the Securities Act liability in the SPAC context appears to be significantly less costly to issuers—and indeed, that is probably one reason that the SPAC form is popular. It may be that merger litigation simply does not cost enough to induce SPACs to make the disclosures that would be optimal for investors in newly public firms.

But although most SPAC merger litigation may be insufficiently costly compared to Securities Act litigation, it is simultaneously too costly in that it is indiscriminate. Section 11 plaintiffs are constrained by unyielding standing rules and generally have a fleeting one-year limitations period in which to make their claims. These restrictions manifest in the relatively low number of IPOs that draw section 11 lawsuits—in my sample, less than 9%. Although the limitations periods for much merger litigation are not much longer (often two or three years), standing rules are broader. More importantly, the current regime for merger litigation generally allows for quick settlement for small fees without judicial oversight, meaning that no party involved has incentives to put the brakes on these lawsuits; plaintiffs may bring them irrespective of the merit of the case, defendants find it less expensive to pay than to challenge, and courts are not in a position to curtail the fees that are unwarranted or heighten those that are not.

A high percentage of SPACs in the most recent years have drawn merger objection lawsuits. If these SPACs are fraudulent, $200,000 in fees is likely too low to create effective deterrence. If they are not, then many SPACs may be paying a “deal tax” that others have argued, in the standard merger context, is inappropriate.

VI. POLICY IMPLICATIONS

Proposals for improving SPACs abound; academics, regulatory agencies, and even members of Congress are busily gathering information and proposing reforms. Many of these proposals involve heightened disclosures of the various conflicts and relationships inherent in the structure of SPACs. These include, to name a few, sponsor compensation and the incentives sponsors have to close even a subpar deal, the stake that sponsors will have in the merged company, information about the diligence and negotiations that occurred during the merger process, and the probability and extent of dilution for shareholders who continue to hold their shares after the merger.

But perhaps these potential hazards are features, rather than bugs, of the SPAC structure. SPACs, say their proponents, “offer investors and targets a new set of financing opportunities that compete with later-stage venture capital, private equity, direct listings, and the traditional IPO process. They provide an infusion of capital to a broader universe of start-ups and other companies, fueling innovation and growth.” Of course, this breadth comes at a cost. SPACs are common for “firms [that] are speculative, have enormous capital requirements, and can provide only limited assurances on near-term revenue and viability.” Such investments are obviously risky. But one might assert that investors in SPACs understand this, or if they do not, they should. On this view, investors get to participate in these ostensibly high-reward deals in no small part because they lack the protections of a traditional IPO, and this bargain is a reasonable one that should be left intact.

The success of SPAC-related litigation depends on the camp into which one falls. Much of the SPAC litigation deluge is in fact composed of pre-closing merger challenges that probably generate little but low fees for plaintiffs’ counsel. But if SPAC investors accept less information as part of a deliberate bargain for the opportunity to invest in ventures not normally available to them, then perhaps this litigation is functioning more or less as intended, and pre-closing merger challenges represent a minor deal tax that all parties involved are willing to pay.

If, on the other hand, SPACs are not functioning as intended—if the informational goals of SPACs are analogous to those of standard IPOs, and shareholders have not knowingly bargained for them to be otherwise—then it appears that much SPAC-related litigation is not serving this goal. Rather, the merger challenges brought before the lawsuit are not promoting meaningful disclosure, and even a percentage of the smaller proportion of lawsuits brought after the merger may target transactions that can produce the biggest payout for plaintiffs’ counsel, rather than targeting true misconduct. A natural conclusion of this line of thinking is that the system of private enforcement is currently ill-suited to correct the deficiencies of SPACs and that greater public intervention, such as government enforcement actions and more stringent regulation, may be necessary.

But this early sample of SPAC-related lawsuits has broader implications for the utility of shareholder litigation as a policing mechanism. The consolidation of the U.S. public equity markets has been the source of commentary for some time. The traditional IPO process has become the province of “large, late-stage companies.” Concurrently, small- and medium-cap companies—which have “[t]raditionally . . . delivered both higher risk and higher returns”—have become far less accessible to the average investor. The SPAC boom may be in part interpreted as a testament to retail investor appetite for such high-risk, high-return opportunities; retail investors eying the profits of venture capital and private equity moguls when the markets are rallying and money is flowing freely may well think, “Why not me?” Inaccessibility is a two-way street—smaller, growing firms that cannot access the liquidity of public markets could be limited in their opportunities for “economic growth, hiring, and wealth creation.”

While Congress and the SEC have undertaken efforts to make the traditional IPO process more inclusive, the trend toward fewer, larger IPOs has proven persistent. It is thus not surprising—and is perhaps commendable—that private actors have experimented with alternative paths to the public markets. SPACs may not be the optimal result of these experiments. But in the absence of regulation that actually results in IPOs for more emerging, high-growth companies, such experimentation is likely to continue. Strong policing mechanisms for these experiments are necessary. Ex ante regulation, by its nature, is one or several steps behind such experiments, as SPACs aptly demonstrate. Regulation by agency enforcement is controversial, and even if it were not, agencies are resource-constrained and cannot catch everything. Shareholder litigation under broad, existing causes of action, such as Rule 10b-5, proxy fraud, and breach of fiduciary duty, constitutes a critical backup protection for investors in the face of private ordering experiments.

That some experiments, like SPACs, are designed to avoid certain types of liability (such as section 11) makes the efficacy of the remaining causes of action even more important. But if the SPAC-related litigation so far is anything to go by, it is unclear whether shareholder lawsuits are currently up to the job. Based on my sample, a very small subset of these lawsuits—post-closing merger challenges in the Multiplan mold—have resulted in standards that could improve a substantial shortcoming of the SPAC structure. The results of another 30% of these lawsuits—Rule 10b-5 claims—are as yet undetermined, but there are both meritorious and opportunistic explanations for these lawsuits, and the end results may lie somewhere in the middle. By far the largest bloc, comprising roughly half the claims in my sample, consists of pre-closing merger objections that are likely to perform little, if any, effective policing function.

This is most unfortunate, because in theory, these lawsuits could be among the most immediate and effective checks available against predatory structures innovated by private markets. The main advantage to merger objection claims is that they can be brought before a transaction closes, potentially saving shareholders from the consummation of a damaging deal. The supplemental disclosures issued in response to many merger objection claims could remedy prior misstatements or omissions and might force managers to do more comprehensive diligence or find better deals. The key advantage of a merger objection lawsuit is its immediacy; unlike most lawsuits, which can only be brought after the fact, merger challenges can directly affect the challenged transaction.

But based on my early sample of SPAC-related litigation, it is not at all clear that pre-closing merger objections are having this effect. They are so numerous and inexpensive that shareholders could easily be forgiven for not taking them seriously as a signal of real wrongdoing by the management. The disclosures they provide do not appear to induce shareholders to redeem. They do not appear to be related to the returns on the final deal. And the monetary penalties they exact are often too small even to be disclosed and likely would not deter unscrupulous SPAC managers from continuing to lie. All this aligns with the conclusions that have been swirling among scholars and judges in recent years; that “these suits are not being filed with the expectation of obtaining a meaningful recovery for the plaintiff class but rather in order to obtain a quick disclosure and mootness fee.”

This is a problem bigger than SPACs. In general,

the time frames in which an M&A transaction must close usually discourage defendants from attempting to defeat pre-merger litigation on the merits, even when that litigation is abusive. And, even if settlement costs are minimal in comparison to the size of the M&A transaction, transaction costs associated with litigation end up being visited on shareholders, for no or little appreciable benefit.

Correcting the broader environment that has allowed pre-closing merger litigation to become so dysfunctional is beyond the scope of this Article. It is worth observing, however, that even if pre-closing merger objections have different significance in the SPAC context than in standard mergers, and even if SPACs are a phenomenon of the moment, pre-closing merger challenges are worth saving, if not for SPACs, then for future transactions and innovations. The first best solution for SPAC-related litigation, and indeed, for many other transactions, would be to save them.

CONCLUSION

The litigation produced by the recent SPAC boom to date is intuitively odd: the likelihood of being sued is greater for SPACs whose shareholders elect to retain their shares, which suggests that they have greater confidence in the success of the merger. The likelihood that a de-SPAC transaction will generate a lawsuit does not, however, have to be related to the size of the deal, the experience of its managers, or virtually any other proxy for quality. This could mean that these lawsuits are, in fact, targeting fraudulent transactions, and the low redemption rates of these transactions reflect the fact that the shareholders have been lied to. Alternatively, it could mean that these lawsuits target low-redemption (and presumably higher quality) deals for some other reason, which may be opportunistic and unrelated to merit. Though these competing explanations are difficult to disentangle for the post-closing claims in my sample, I argue that the pre-closing merger cases in my sample do not appear to be meritorious, irrespective of whether they target fraudulent or non-fraudulent transactions. They do not appear to be an adequate substitute for section 11 liability, and if SPACs are in need of reform, private litigation may not be the optimal solution.

 

 

96 S. Cal. L. Rev. 553

Download

J.D. Duke Law Lecturing Fellow. For helpful comments, I am grateful to David Berger, John Coates, Patrick Corrigan, Elisabeth de Fontenay, Deborah DeMott, Jessica Erickson, Jill Fisch, Gina-Gail Fletcher, Zohar Goshen, Joe Grundfest, Sharon Hannes, Kobi Kastiel, Michael Klausner, Minor Myers, Michael Ohlrogge, Jim Park, Alex Platt, Adam Pritchard, Barak Richman, Tony Rickey, Amanda Rose, Michael Simkovic, Steven Davidoff Solomon, Andrew Tuch, Andrew Verstein, and participants in the Conference for Empirical Legal Studies, American Law & Economic Association Annual Meeting, Berkeley Law, Accounting & Business Workshop, BYU Deals Conference, and National Conference of Business Law Scholars. Errors are my own.

Co-Creating Equality

When a creative work has many co-creators, not all of whom contributed equally, how should they split ownership? In the absence of a contract, copyright law has long adopted an all-or-nothing answer to this question: if you are deemed to be a “co-author” you get an equal split; otherwise, you get nothing. Because the privileges of co-authorship are so great, courts have erected an onerous barrier to qualifying as a “co-author”: you must have had “control” over the whole collaborative work. This barrier has been criticized both for being arbitrary and for unfairly resulting in lesser contributors going unrecognized and uncompensated. But removing this barrier—in the context of the longstanding rule granting co-authors an automatic equal split—risks unfairly diluting majority contributors. So, in deciding whether to remove the barrier, we have to balance the perceived unfairness of minority contributors going uncompensated against the perceived unfairness of majority contributors being diluted. In this Article I will show that this question of perceived fairness can be answered empirically.

To determine creators’ revealed preferences for how to treat lesser contributing collaborators, I assemble two datasets. Both datasets are in the core copyright domain where the default co-authorship rules are most relevant: co-songwriting. First, I construct a dataset of the songwriting contribution levels, writing credits, and copyright registrations of every band that has a certified Gold Record and writes its own songs. This is over one thousand music groups. Second, by cross-correlating data released by the principal performance rights organizations in response to recent antitrust probes, I estimate royalty splits between the co-writers of over 1.2 million songs. The studies in this Article are the largest and most comprehensive investigation into joint authorship to date that accounts for parties’ contributions.

My primary finding is that the typical behavior of creators is to credit everyone involved in writing as a co-author, even the lesser contributors. A secondary finding is that co-authors who were lesser contributors typically share equally in songwriting royalties. Main contributors thus choose to share much more with lesser contributors than they would be compelled to under current law in the absence of a contract. This revealed preference suggests that we can adopt a more inclusive legal criterion for co-authorship—and in particular remove the arbitrary and exclusionary barrier that to be a co-author one must have control over the entire work—and that we can do so without violating creators’ own sense of fairness. Beyond copyright law and the music industry, these findings have implications for the design of incentives in intellectual property law and creative collaboration more broadly.

INTRODUCTION

Over the last few decades, people have been collaborating to produce creative works more than ever before, from songs to software. But in the absence of an explicit contract, the law of co-authorship now recognizes as “co-authors” only those contributors who have control over the whole collaborative work. This control doctrine has been broadly criticized: the doctrine seems arbitrary; it may leave lesser contributors entirely uncompensated; and it may obscure the identities of the lesser co-creators.

Under current law, the question of who counts as a co-author has a lot riding on it. This is because there is a longstanding rule granting co-authors an automatic equal split of royalties. Unless they have a contract that says otherwise, the default “equal split” rule is that joint authors share license proceeds equally.

Thus under current law, outcomes tend to be binary. Either you have control over the entire work, in which case you count as a co-author and by default get an equal split; or you do not, in which case no matter how much you contributed to the final product you are not a co-author and by default receive nothing.

There are three paths forward. One is to maintain the status quo, leaving the control doctrine in place, at the risk of unfairly leaving minor contributors entirely uncompensated (among other problems). A second choice is to remove the control doctrine for co-authorship, while also removing the (more long-standing) “equal split” rule, and make the royalty share proportional to the level of contribution. A third option is to remove the control doctrine but leave the equal split rule in place, at the risk of unfairly diluting the shares of the main authors.

If the wrong choice is made, we risk disincentivizing creative collaboration, either by giving rise to credit allocations that creators view as unfair and demotivating, or at any rate by imposing the transaction costs of having to contract out of them. The important question is then what is fair as judged by the creators to whom the default rules apply, not as judged by the general public or by legal scholars. Joint authorship default rules are uniquely relevant for co-songwriting; they are preempted by work for hire or other practices in most other contexts. For this reason, in this Article I will attempt to answer the following empirical question: as revealed by their actions, what do the majority of co-songwriters view as a sensible way to grant co-authorship and split songwriting license proceeds? In particular, how do they treat lesser contributing collaborators?

To pursue these questions, I assemble two datasets. First, I construct a dataset of the songwriting contribution levels, writing credits, and copyright registrations of over one thousand music groups—every band with a certified Gold Record that primarily releases its own songs, since certifications began to the present (1959–2021). I explore how co-author-crediting decisions are associated with features of the creative context including members’ levels of songwriting contributions, number of collaborators, geography, genre, and era. Second, I extend beyond music groups and explore royalty splits between co-writers for over 1.2 million songs by leveraging the data sharing undertaken by the American Society of Composers, Authors and Publishers (“ASCAP”) in response to its ongoing antitrust action. I find that lesser contributors are most often credited as co-authors, and as co-authors they most often share equally in songwriting royalties. The results support the position that calls for rethinking the control doctrine while retaining joint authorship’s equal split default rule.

Part I reviews the history of the equal split rule, the phases of the development of the control doctrine in joint authorship law, and the competing proposals for its revision. Part II makes the case that empirical data can guide our way forward. Parts III and IV present the methodologies and results of two studies. Part V discusses the legal implications of the results.

I. CONTEXT

A co-author who licenses a joint work—for example, allowing a song to be used in a television show—must give their co-authors a share of that money. If the co-authors have a contract between them, they receive the agreed amount. Otherwise, each co-author is entitled to an equal share, even if the parties have indisputably made unequal contributions to the joint work.

A. The Equal Split and the “Intent to Merge”

This was not always the case. For the first century after copyright laws appeared in the United States, there was no default rule dictating how to split co-authorship profits, or concerning co-authorship generally. Copyright registration was relatively unimportant for the first century of the United States under the Constitution, for reasons both philosophical and economic. When co-authorship did arise, co-authors settled among themselves whether or not they would share licensing income. The first judicial decision on accounting between co-authors was not until 1874. The court in Carter v. Bailey held that a co-author could license their joint work without their co-authors’ consent and without having to redistribute that licensing income among any co-authors of the work not party to the licensing transaction. The rationale behind this was that the left-out co-authors were equally able to do the same if they put forth the effort, because a license to one person does not use up intangible property. Each co-author owned an undivided interest in the entire work, and so had the full right to license that property so long as they did not interfere with their co-authors’ rights to do the same. This mirrored the situation with patent co-owners.

In the decades that immediately followed Carter, the copyright landscape in the U.S. was transformed by a boom of popular culture and means for its dissemination. By the early twentieth century, entertainment industries were expanding, changing, or being freshly created. Co-authorship grew from the minority case to a common way of producing creative works in important domains. Moreover, dissemination of works was so improved that the circumstances that justified the holding in Carter—that there was no duty to account because one co-author’s licensing of the joint work would not impair the excluded authors’ ability to license it—quite clearly no longer applied. Technologies like radio and film could reach the entire market for a joint work, leaving nothing for excluded authors and thereby effectively destroying the value of the work’s copyright for them.

This rule first shifted in 1915 with Maurel v. Smith. In Maurel, three parties agreed to jointly author an operatic work. The plaintiff wrote the plot (the “scenario”), while the first defendant H. Smith translated the plot into script and claimed to have made substantial modifications. The second defendant added lyrics for musical numbers without reference to the plot. The defendants then registered the copyright for the work without including the plaintiff, blocking her from exploiting her property, as under the 1909 Copyright Act an author’s rights followed from registration. The court held that the license proceeds were held in a constructive trust for the excluded co-author, in effect creating a duty to account, because there had been no other way for the excluded co-author to obtain license revenue. Cases following Maurel, however, interpreted it as establishing a general duty for co-authors to account to each other for profits. Moreover, they were to share equally in those proceeds, at least in the absence of a contract stating otherwise. This position has been treated as settled law since it was established more than one hundred years ago. Courts have not given consideration to uneven shares.

Importantly, Maurel established that contributions to the joint work do not need to be balanced; co-authorship would be found so long as the collaborators share a “common design.” One defendant had argued that the plaintiff should not be considered his co-author because substantial changes had been made to the plot she contributed, in essence asserting that his contributions to the totality were significantly greater than the plaintiff’s. On this point, Judge Learned Hand adopted the view put forth in the English case Levy v. Rutley, that as long as the parties had agreed upon a common scheme for the joint work there “can be no difficulty in saying that they are joint authors of the work, though one may do a larger share of it than the other.” Having established that the plaintiff was indeed a co-author with the defendants, Judge Hand found that the copyright was the resulting res of their three contributions, “and by every equitable rule the defendants hold any legal rights they have upon trust in the same proportion.” In so declaring, Judge Hand applied the common law of tenancy in common to the parties’ relationship, one of equal ownership by default.

But tenants in common can refute their equal undivided shares by showing evidence of unequal financial contributions to the purchase of the common property. In the case of a joint work, the analog would be to adjust the co-authors’ shares to reflect some proportion of their inputs, whether the relative quantity of their creative contribution or value created. Judge Hand did not address this feature of tenancy in common explicitly, but effectively shut out its application in the case when he took up the claims of the second defendant. The second defendant, R. Smith (brother of defendant H. Smith), argued that he did not need to share with the plaintiff profits resulting from a separate publication of the song lyrics he had written for the opera. Judge Hand took the view that the lyricist could not claim the opera played no role in the later sales success of his lyrics. In a consequential declaration, Judge Hand found determining the contribution of the whole to the success of the part in this manner was not possible. For this reason, the lyricist would be required to split any profits from the separate sale of lyrics equally with his co-authors in the whole opera.

Judge Hand (and decisions accepting the logic of Maurel) treated that reasoning as sufficient to implicate the converse scenario as well: that the particular contribution of any co-author to the success of the whole could not be measured, leading to the generalized pronouncement that when “several collaborators knowingly engage in the production of a piece which is to be presented originally as a whole only, they adopt that common design . . . and unless they undertake expressly to apportion their contributions, they must share alike.” The underlying causal premises of Judge Hand’s reasoning is that when all authors’ contributions are necessary for a work’s value, the degree to which each comparatively adds to that value cannot be assessed. Each contribution has its effect at the level of whether or not it is made (that is, categorical) because the contributions are mutually contingent; in this sense all contributions are equally responsible for the work’s total value. This suggests an equal split because it cannot be said that one person’s contribution is “more” of a cause than another’s.

In Maurel, there was disagreement about the parties’ proportional contributions, as the defendants sought to justify excluding the plaintiff from the copyright registration by minimizing her contribution. But the decision suggests that Judge Hand considered the parties’ true contributions to be of comparable magnitude. It may be for this reason that, after rejecting the possibility of determining the responsibility of each party separately for the opera’s success, Judge Hand did not find it helpful to entertain the alternative of dividing royalty rights according to some measure of each author’s direct inputs. But whatever the reason, this position has been enforced even when parties did not dispute that the co-authors’ contributions to the joint work were not equal. In Sweet Music, Inc. v. Melrose Music Corp., the assignee of a co-author requested a three-quarter share in a song’s renewal copyright, on the basis that he had written “half the words and all the music.” In the absence of “evidence indicating that the ownership was intended as other than an undivided one-half interest for each of the co-authors,” the court applied the default rule.

When the Copyright Act of 1976 was passed, it left unchanged court-made law on accounting responsibilities of joint authors to one another. Courts had not yet addressed the question of how comparatively lesser a collaborator’s contribution could be while still being fairly entitled to the equal benefits of authorship. No court had given serious reconsideration to Maurel’s assertion that co-authors are entitled to equal shares of proceeds in the absence of a contract. The Copyright Act codified the criteria for joint authorship in Maurel, requiring that collaborators have an intent to merge their contributions into a unitary whole.

B. Gatekeeping Against Lesser Contributors: From “Intent to Merge” to “Intent to Be Co-Authors”

For the first decade after the Copyright Act of 1976 was passed, most courts followed a literal reading of the statute—and, per legislative history, the common law precedent—to decide joint authorship claims. The longstanding rule was that co-authors share equally in the benefits of co-authorship regardless of their relative contributions. Co-authorship rewards were potentially high, but the intent to merge standard for minting co-authors was low. Lesser contributors, with whom a work’s more significant authors may not have intended to collaborate, or whose contributions were quantitatively small in comparison to their co-author’s, were being granted co-authorship at the district court level, seeding frustration in the Second Circuit.

Beginning in the 1990s, courts heard a series of cases about creative works arising from joint efforts in which the disparities between the collaborators’ contributions were stark. Under the intent to merge statutory standard, they would nevertheless have been equal co-authors. Courts’ gut reaction to these cases was that equal co-authorship would be unfair. Primary creators, it was perceived, would not want to share equal proceeds with collaborators who had made lesser contributions to the work; if forced by the law to do so, they would be disincentivized to collaborate out of fear of sharing authorship.

But equal co-authorship was well entrenched in the law. Modifying collaborators’ co-authorship shares to reflect their relative contributions was not an option under consideration. Instead, courts granted a prerogative to greater contributors to share, or not share, co-authorship with a work’s lesser contributors. This was given effect by adding a mutual intent to be co-authors requirement to the statutory intent to merge and necessary independently copyrightable contributions. These cases proposed that evidence of the parties’ subjective intentions to be co-authors could be inferred from, for instance, how the work was billed or credited.

C. Introduction of the Control Doctrine

While the intent to be co-authors test raised the bar for co-authorship, it did not foreclose it. A new requirement for co-authorship, a need to have control over the whole work, was invented. In the Seventh Circuit, it was framed as additional and necessary evidence of the mutual intent test for joint authorship. In the Ninth Circuit, it was framed as a new test of authorship, without which, as before, there could be no question of joint authorship—regardless of the extent of one’s copyrightable contribution. In Aalmuhammed v. Lee, appellant Jefri Aalmuhammed served (without a contract) as an Islamic consultant on the Warner Brothers film Malcolm X. In addition to these services, he made comparatively minor scriptwriting and directorial contributions that were included in the completed film. These contributions would have been independently copyrightable. All creative contributors intended that their contributions were to be merged into the whole, satisfying the statutory intent test. The panel voiced concern that dominant authors would be deterred from beneficial collaboration if they had to share the benefits of authorship with a co-author whose contributions were substantially less, strongly implying that on policy grounds they sought a construction of authorship that would exclude Aalmuhammed. As evidence against the existence of a mutual intent to be co-authors, the Ninth Circuit adopted the control concept introduced by the Seventh Circuit. Focusing on Spike Lee’s control over including Aalmuhammed’s contributions in the film, control was elevated as the most important factor needed to find there had been an intent to be co-authors.

Since Aalmuhammed, in the absence of a contract, lesser contributors’ joint authorship claims have turned on evidence establishing that they exercised control. As the joint authorship test for the Ninth Circuit, it has been applied in songwriting joint authorship cases, more often than not, although the former was discouraged by the Aalmuhammed court in dicta. In practice, control has meant control over the whole work, though a minority position has found control over “separate and indispensable elements of the completed product” to meet the control requirement. More recently, there are also signs that the Aalmuhammed-like concept of control is resonating with other circuits and trumping creative contribution considerations. But the general consequence of the expanding influence of the control standard is that lesser contributors are blocked from the equal sharing of authorship, and lacking authorship, have no entitlements in the absence of a contract. In the two circuits most consequential for the copyright industries, there is now a trend toward finding works to be single-authored.

The intent to be co-authors test, which replaced the intent to merge test, and the control doctrine, itself designed to further strengthen the intent to be co-authors test, are widely regarded by scholars as suboptimal. Courts aspired to recognize the interests of both lesser and greater contributors, but existing law was understood to force an all-or-nothing choice between the two groups. The consequences of exclusion from co-authorship for lesser contributors include leaving them uncompensated for their work. Also of concern is the disordering effect of these standards on author identification, a central goal of copyright law. Goldstein criticizes the control standard as being “both overinclusive and under-inclusive.” It allows contributors primarily of non-copyrightable expression, such as film producers, to be recognized as authors. At the same time, it complicates the identification of the authors of most other multi-authored works.

D. Scholarly Positions

Scholars generally agree that lesser contributors who make copyrightable contributions should be counted as co-authors. Beyond that, scholars are divided into two camps with respect to how the law ought to treat them vis-à-vis their majority contributing co-authors.

One camp would retain the equal split default while granting equal shares to lesser co-authors. Goldstein would apply the plain language of the statute and allow equal ownership to all contributors of independently copyrightable material if they intended to merge their contributions in a unitary whole. If Warner Brothers does not want to share equal ownership in Malcolm X with Aalmuhammed, it should not fail to negotiate with him for the value of his services—and be more careful in the future. Goldstein levels similar criticisms of the extra-statutory introduction of the intent to be co-authors and control requirements to the Copyright Act’s intent language. While the statutory intent (intent to merge) approach “will sometimes give an economic interest to a contributor . . . who probably did not intend to receive it,” the Copyright Act should not be distorted to protect the economic interests of dominant contributors. A better outcome is for dominant contributors to bear the burden of adjusting shares via contract to avoid an undesired equal split.

The other camp of scholars focuses on courts’ unease with non-equal contributors receiving equal authorship rights as the source of courts’ statute-distorting jurisprudence. This camp would endorse lesser co-authors receiving a lesser split. The shared assumption of this camp—that if lesser contributors reap financial rewards greater than their contributions seem to merit, majority contributors will be disincentivized and creative production will suffer as a result—is rarely questioned. The inefficiency of equal pay is widely accepted. There is an assumption that fairness has an important role to play in economic productivity. But there is more to the psychology of linking compensation to contributions than simply motivating cool-headed, rational workers. The sentiment held by many is that a contribution-based default should be established because fairness requires a correspondence between inputs and outputs.

There are two main approaches, which hold in common that it is possible to make adjustments to existing joint authorship law. The proposed implementation which has attracted the most scholarly support is a rebuttable presumption of equality. This may be because it receives textual support in the legislative history and is drawn from the analogy of joint authorship with real property tenancy in common. In such cases, while undivided equal shares are the default, co-owners may rebut that presumption by showing that unequal contributions had been made to the purchase price. This would allow unequal shares to be awarded if there is evidence the co-authors’ contributions were unequal. Aalmuhammed’s contributions to Malcolm X could be determined by experts to have been responsible for some small fraction of the film’s success, and he could be compensated accordingly. A smaller group of scholars, perhaps relying on the silence of the Copyright Act as to shares in the copyrighted work, argue that proportionality should be the default rule in all joint authorship cases. For these scholars, fairness, as they believe it to be perceived, is a paramount concern. If songwriters prefer to split equally even when a co-author makes a lesser contribution, it would be a challenge to this notion of fairness.

II. WHAT SHOULD BE DONE?

How should we choose whether to reject the intent to be co-authors and control doctrines and instead include lesser contributors as co-authors? And if we do decide to include lesser contributors as co-authors, what would be the most efficient rule for splitting revenue between joint authors: retaining the equal split, or revising it to be contributions-based? With respect to the first issue to be decided, it is possible, if unlikely, that most creators would prefer lesser creators to be excluded from co-authorship and simply paid for their services. Often in collaboration situations where the parties do not have a contract or the contract is silent about co-authorship shares, which is when the default joint authorship rules apply, a main creator will not have funds to pay lesser contributors in advance for services, and the unevenness of contributions may not be clear until the joint work is complete.

With respect to the second question, some would argue that well-established common law default rules are presumptively efficient because they have been accepted by parties across contexts over time. The control doctrine did not arise to thwart the equal split rule for nearly a century. On the other hand, the equal split rule’s detractors regard as self-evident that it is unfair and unpopular. The suboptimality of the equal split default feels like a frictionless assumption. Proportional compensation is the norm in wage labor contexts, which are related but distinct.

The traditional view is that an efficient default reflects the preferences of “most contracting parties—or perhaps most contracting parties in a given industry.” The universe of possible contracting parties in this case is those creative collaborators who would potentially contract with one another over the division of license proceeds. This universe consists principally of
co-songwriters. The core copyright industries include literature, music, theater, film, the media, photography, software, visual arts, and advertising. Collaborative creative production has been on the rise across all of the core copyright industries, but it nevertheless accounts for a very small proportion of output (<0.5%) in most of the visual arts (for example, fine art, sculpture, and photography). The rate is higher, albeit still very low, in literature and theater. In music, however, the co-authorship rate of songs is 37%. Copyright works are much more likely to be produced through collaboration in the film, software, and music industries, and in academia. However, in all but the music industry, the joint authorship default rules are for the most part inapplicable.

There are two reasons for this. First of all, only authors can be joint authors. This means that whenever collaborators’ works are officially “authored” by their employer, the default rules are not relevant. Work for hire arrangements are not generally practiced in the music industry, but they are the norm in film and software. (The stakes for Warner Brothers in Aalmuhammed illustrate why.) Second, in academia, journal articles, which comprise the bulk of academic publishing, as a matter of course reassign royalty streams from creators to publishers. The open access movement opposes assigning copyrights to paywalling publishers, but even under an open access model the publishing royalties would still be inconsequential for authors, albeit for a different reason. These practices—work for hire and publisher assignments—cover two paths to corporate ownership of copyrightable work and most contexts of creation in film, the media, software, and advertising.

A. The Unique Relevance of Co-Songwriting

Songwriting is not included in the categories of creative work covered by work for hire. Typically, musicians who are signed by record labels will be paid an advance against future royalties. However, in those cases, this is not an assignment of ownership of their songwriting copyrights to the labels. So co-songwriters, as a matter of course, will be either contracting out of, or relying on the default rules for, joint authorship to specify the allocation of songwriting royalties.

Songwriting is one of the copyright domains in which the parties are the least likely to be thinking in legalities at the time of creation, or even in terms of industry norms: songwriters have only limited knowledge of other songwriters’ split practices. Unlike scholarship, commercial filmmaking, or software development, songwriting can be (and often is) undertaken by a handful of teenagers in a garage band for whom default rules in the absence of contract are particularly relevant. Moreover, 68% of music groups are primarily composed of friends or family members. Other domains where joint authorship rules tend to apply do not approach the volume of creation—and therefore individual works which are potentially subject to contracting about royalty splits—of co-songwriting. There are over 750,000 people who have co-written a song in the ASCAP repertory alone.

Finally, joint authorship rules are financially consequential for co-songwriters. The economic importance of songwriting royalties is not an argument in favor of focusing on co-songwriters to set the default joint authorship rules. But it is an additional reason why we should be interested in efficiently setting the joint authorship rules more generally. In the history of popular music, the importance of songwriting royalties to musicians has waxed and waned. Songwriting royalties became meaningful for musicians after the 1960s transition to solo artists and bands who wrote their own songs and who, by the 1970s, had the bargaining power to control their own publishing revenues. Beginning in the band era of the 1960s, songwriting royalties for the most successful songs were worth millions. The revenue for artists derived from songwriting has declined again with the advent of streaming. Nevertheless, songwriting royalties are a significant portion of total income for many successful musicians. This became particularly clear during the COVID pandemic which eliminated touring income for over a year for many artists. A survey prior to the COVID pandemic of working musicians found that, in the most recent decade, indie rock bands have earned about 21% of their gross income from songwriting royalties and advances on those royalties. Across other genres this figure was approximately 8% of income, rising to 39% for people who identified as composers, whether performing or not. And songwriting royalties often provide a measure of financial security for musicians. Unlike money from tours, for instance, the checks keep coming in once a musician’s most active career years are over. Songwriters can also receive substantial payouts for licensing their compositions for film, TV, commercials, or video games. Fees for licensing a song for television play can be as low as $1,000 per year to more than $100,000 per year for well-known hits.

If you combine the considerations of (1) collaborations not covered by employment, work for hire, and where there is no standard practice of assignment of copyright, with (2) the volume of collaborations and works produced, there is a strong case to be made that default joint authorship rules are primarily relevant for co-songwriting. It may be that by volume, default joint authorship rules that are efficient for co-songwriting are efficient for most co-authorship cases to which those rules apply in general.

If fairness in labor contexts means applying the proportionality principle, we would expect musicians to find the equal split default rule unfair. But there is no evidence suggesting that songwriters tend to take issue with Judge Hand’s view that they must “share alike.” There has been no movement against it. Proposed amendments to the Copyright Act have never sought to revise the equal split rule, nor have music industry representatives testifying before Congress problematized it. This suggests that the equal split rule has persisted not merely because it is precedent, but because either very little is at stake, or because it leads to what collaborators consider a fair result.

B. Identifying Efficient Joint Authorship Rules Is an Empirical Task

If one agrees with the prevailing view that an efficient default reflects the preferences of most contracting parties, then determining the most appropriate default relies on identifying the preferences of those parties. This can be achieved by looking at what contracting terms those individuals actually agree to, an empirical question. Moreover, in the case of co-songwriting, this is answerable. Unlike many questions governed by default terms in contract law which may have a remote chance of becoming operational, if a song earns any songwriting royalties that the co-songwriters want to collect, they will have to face the question of with whom and how to divide them up.

Co-songwriters create a record of who shares co-authorship of their joint work when they register a song with a performance rights organization (“PRO”). The individuals completing a PRO’s registration form must specify how royalties are to be split between those credited as writers: there is no default split as far as the PROs are concerned. PROs then channel the royalties they collect to a song’s listed co-writers in accordance with that information. The major PRO repertories are public and include nearly the full set of co-authorship crediting decisions made by co-songwriters in the United States.

I am interested in the co-authorship crediting decisions and royalty split choices of collaborations where contributions are uneven. Ideally, I could identify all songs resulting from uneven contributions and see if all contributors, including lesser contributors, are typically credited as co-authors. That information is not attainable. However, in this Article I construct a unique database including songwriting contribution levels for over 1,000 music groups with certified Gold Records—every band with a Gold Record that primarily writes its own songs, from the first certifications to the time of writing (1959–2021) (the “Gold Record bands”). The songwriting process of each band in the Gold Record database is coded for the contribution levels of its band members based on publicly available information.

Because the membership of these most popular bands is well-known, we can determine by looking at their PRO-registered songwriting credits whether lesser contributors are most often counted as co-authors. PRO credits were obtained for bands in the Gold Record database from the song repertories of the American Society of Composers, Authors and Publishers (“ASCAP”) and Broadcast Music, Inc. (“BMI”), which together represent 90% of the U.S. market in public performance rights (Study 1).

If the answer to whether lesser contributors are counted as co-authors is “no,” then the consequence of the control doctrine that lesser contributors are excluded from co-authorship status will have been shown to align with creator preferences in this case. This would weaken at least one key objection to the control doctrine (see Figure 1).

If the answer is “yes,” for which I make the case, the next question is whether, as co-authors, lesser contributors typically share equally in the benefits of co-authorship. For this Article, I conducted two studies to investigate this question (Study 2a and Study 2b). In Study 2a, I estimated the royalty splits of a third of the uneven Gold Record bands which include all members as co-authors. First, I compiled statements by the bands themselves or those close to them (managers, for example) disclosing the bands’ royalty split practices. The second method used the repertories of ASCAP and BMI to infer the splits of Gold Record bands. In Study 2b, I used this same methodology to infer the splits of 1.2 million co-written songs.

If the results of Study 2 support that “yes,” typically lesser contributing co-authors do receive an equal split of royalties, then the existing equal split rule is presumptively the most efficient default.

If the analysis in Study 2 suggests that “no,” lesser contributing co-authors typically do not receive an equal split of royalties, then the proposals of the second camp of scholars for revisiting the equal split default ought to be debated further.

Figure 1.  What Is Creators’ Preferred Treatment of Lesser Contributors?

C. Relevance of Empirical Data

Can the transaction cost implications of these studies help guide us between an equal split and proportional split rule? Even if the results of Study 2 suggest that we should retain the equal split default, some might question if there is any added value to empirical preference data over and above existing transaction cost arguments in favor of the equal split. Even if preference data were to show that creators prefer a proportional split, the argument might go: an equal split default would still be most efficient because the transaction costs of a proportional split default are too high. I argue that it seems unlikely that the transaction costs of implementing a proportional default split are as high as some contend. The thought is that a contribution-based rule would entail a considerable fact-finding burden and require jurists to assign percentages based on their subjective appraisals of each co-author’s contributions. Unlike in copyright infringement cases, where courts already make such appraisals, in joint authorship disputes the value of a work is not necessarily known. This position rests on evidentiary assumptions that are not compelling. Assigning joint authors uneven shares in a song involves approximate and imprecise appraisals.

Yet those aspects of making proportional attributions would not be unique. Courts are accustomed to carving up responsibility in contexts of nebulous causality to arrive at liability determinations in common law torts. English courts have awarded proportional shares in cases where authorship was found in the absence of contract and contributions were unequal, and ownership is proportional by default in Germany. If there are reasons to think American jurists would face unique obstacles to determining proportional ownership shares, they have yet to be raised. In general, while degrees of interdependence vary across collaborative contexts, almost as a rule people succeed in translating comparative contributions into pay differences within a tolerated margin of error. If most collaborators must contract out of the default regime because an equal split does not match their royalty distribution preferences, then on the whole a proportional split default rule could be more efficient. The pre-existing arguments against the proportional split and in favor of the equal split are weak. These arguments do not fare well if you believe the central preference contention of the advocates of a contributions-based split. The aggregate transaction costs of contracting out of the default rule are the most significant potential drag on efficiency. For this reason, it is important to know both if creators—in this article particularly songwriters—think lesser contributors deserve to be co-authors, and as co-authors, how much they think they deserve.

III.  WHO IS A CO-AUTHOR?

A. Study 1: Are Lesser Contributors Considered Co-authors?

1. Methodology

When songwriters collaborate to write a song, but one of them contributes more than the other, will they all still be credited as co-authors? To answer this question, I focused on a large group of songwriting collaborators: music groups that write their own songs and have one or more Gold Records. As popular music groups, they are often on record about which of their members contribute to writing their songs and how much of a contribution those members make to writing their songs. The 1,003 bands in this study include Gold Record awardees from across the full sixty years (1959–2021) during which Gold Records have been awarded by the Recording Industry Association of America (“RIAA”). The RIAA certifies albums and singles based on sales.

The Gold Record bands have previously discussed their songwriting processes in numerous interviews. Such interviews were the principal basis for identifying bands in which all the members contribute to songwriting but do not contribute evenly (uneven contributions bands). These bands occupied a middle ground between, on the one hand, groups in which all members made more or less even contributions to songwriting (even contributions bands), and, on the other hand, bands in which some members did not contribute to songwriting at all (some members do not contribute bands). The research and coding processes were highly labor intensive, taking several hundred hours over which thousands of sources were screened and compiled.

In addition, unless all of a band’s members made independently copyrightable contributions to songwriting, the band was classified as one in which some members do not contribute. The copyrightability of songwriting contributions was assumed when quotes labeled band members as songwriters or confirmed members’ involvement in songwriting in general terms. When members’ particular contributions were described, case law (interpreting the Copyright Act) was the primary basis for assessing their copyrightability. Band member contributions which included the elements of a musical work—lyrics, melody, harmony, and rhythm—were assumed to be sufficiently original and treated as copyrightable. Most bands were classified as some members do not contribute(555/1,003, or 55%), followed by uneven (258/1,003, or 26%); it was least common for all members to contribute evenly to songwriting (190/1,003, or 19%).

Next, I assessed whether uneven bands credited all members as co-authors of their songs. A song’s writers are listed in several places: liner notes, on PRO registrations, and in United States Copyright Office (“USCO”) registrations. I consulted PRO registrations and validated that they correspond to USCO registrations. A band was classified as one in which all members are co-authors (true/false) if, for a majority of its songs, all the members of the band in the year the song was released were credited as writers. For 38% of the 1,003 bands in the study, it was true that all members are co-authors.

Other factors beyond writing contribution may influence whether or not lesser contributors receive co-authorship credit. For this reason, I collected data on a number of factors. I designated the lowest number of members the band had during its active years as a representative band size and identified each band’s genre of music, region of origin, and decade of formation (see Table 1). I also evaluated whether co-author inclusion changed over time.

Table 1.  Descriptive Statistics for Gold Record Bands 1959–2021

Variable

N = 1003

All members are co-authors

379 (38%)

Songwriting

 

Even

190 (19%)

Uneven

258 (26%)

Some members do not contribute

555 (55%)

Decade

 

1960s and earlier

89 (8.9%)

1970s

129 (13%)

1980s

234 (23%)

1990s

288 (29%)

2000s

177 (18%)

2010s and later

86 (8.6%)

Members

 

2

215 (21%)

3

220 (22%)

4

349 (35%)

5

158 (16%)

6+

61 (6.1%)

Genre

 

Rock

478 (48%)

Hip Hop, R&B, Gospel, Jazz

190 (19%)

Pop

151 (15%)

Metal

65 (6.5%)

Country

44 (4.4%)

Latin

29 (2.9%)

Punk

21 (2.1%)

Electronic

17 (1.7%)

Reggae

8 (0.8%)

 

Region

 

Northeast

170 (17%)

Midwest

84 (8.4%)

West

265 (26%)

South

204 (20%)

Non-USA

280 (28%)

2. Results

From 1959 to 2021, uneven bands typically included all members as co-authors (140/258 or 54%). This tendency strengthened over time, specifically throughout each of the last three decades, even while accounting for genre, region of origin, and representative band size (p < 0.01). Most uneven bands with fewer than five members include all members as co-authors. Among bands with five or more members, a considerable majority (66%) have some members who do not contribute to songwriting, and there were too few even or uneven contribution bands of that size to investigate co-author inclusion rates and whether copyrightable contributions by all members suffice for co-authorship credit or whether some other factor, perhaps control, is required.

To pursue the trend of increasing co-author inclusion over time, I focused on uneven bands formed after 1990 (n=175). A significant majority of these groups credit lesser contributors as co-authors (111/175, or 63%). This preference was less common with larger bands, but otherwise was not associated with other factors.

Table 2.  Logistic Regression for Co-Author Inclusion: UnevenContributions Bands 1959–2021

Variable

log(OR)

95% CI

p-value

Decade (ref. cat.: 1980s)

     

1960s and earlier

-0.30

-1.6, 0.92

0.6

1970s

-0.12

-1.3, 1.0

0.8

1990s

1.2

0.35, 2.1

0.007*

2000s

1.2

0.34, 2.1

0.007*

2010s and later

1.5

0.39, 2.6

0.009*

Members (ref. cat.: 4)

     

2

1.2

0.21, 2.3

0.022*

3

-0.04

-0.81, 0.73

>0.9

5

-0.73

-1.5, 0.05

0.067

6+

-1.4

-3.0, 0.05

0.075

Genre (ref. cat.: Pop)

     

Rock

0.61

-0.25, 1.5

0.2

Hip Hop, R&B, Gospel, Jazz

-0.80

-1.9, 0.27

0.15

Country

-1.7

-3.6, 0.01

0.060

Metal

-0.09

-1.3, 1.1

0.9

Punk

-0.47

-2.7, 1.7

0.7

Electronic

-1.4

-3.4, 0.57

0.2

Reggae

13

-167, NA

>0.9

 

Region (ref. cat.: Northeast)

     

Midwest

-0.51

-1.9, 0.82

0.5

West

-0.07

-0.91, 0.77

0.9

South

0.36

-0.57, 1.3

0.5

Non-USA

0.11

-0.80, 1.0

0.8

Notes: *p < 0.05; **p < 0.005.

Table 3.  Logistic Regression for Co-author Inclusion: UnevenContributions Bands 1990–2021

Variable

log(OR)

95% CI

p-value

Songwriting (ref. cat.: Some members do not contribute)

     

Even

2.9

2.4, 3.5

<0.001**

Uneven

1.8

1.4, 2.1

<0.001**

Decade (ref. cat.: 1980s)

     

1960s and earlier

-0.94

-1.7, -0.20

0.016*

1970s

-0.33

-1.0, 0.27

0.3

1990s

0.50

0.06, 1.0

0.026*

2000s

0.74

0.23, 1.2

0.004**

2010s and later

0.90

0.23, 1.6

0.009*

Members (ref. cat.: 4)

     

2

0.91

0.43, 1.4

<0.001**

3

0.02

-0.42, 0.45

>0.9

5

-0.61

-1.1, -0.11

0.017*

6+

-1.5

-2.6, -0.56

0.003**

Genre (ref. cat.: Pop)

     

Rock

0.37

-0.14, 0.89

0.2

Hip Hop, R&B, Gospel, Jazz

-0.79

-1.4, -0.16

0.014*

Country

-1.0

-2.0, -0.09

0.034*

Metal

0.27

-0.52, 1.0

0.5

Latin

-1.5

-3.0, -0.18

0.032*

Punk

0.87

-0.29, 2.0

0.13

Electronic

-0.50

-1.8, 0.90

0.5

Reggae

1.0

-1.6, 4.1

0.5

Region (ref. cat.: Northeast)

     

Midwest

-0.68

-1.4, 0.04

0.069*

West

0.01

-0.49, 0.51

>0.9

South

0.11

-0.43, 0.65

0.7

Non-USA

0.26

-0.26, 0.78

0.3

Notes: *p < 0.05; **p < 0.005.

             

A significant majority of bands (75%) did not deviate from the practice of crediting or not crediting all members as co-authors which they had established in the year of their first release. Of those 25% of bands that switched credit practices, most of those that changed did so to credit all members as co-authors rather than vice versa.

3. Open Questions

Study 1 investigated co-author inclusion practices, but the data did not contain details on how songwriting royalties are allocated between the band members who are credited as co-authors.

If creators do consider it unfair for lesser contributors to receive equal co-authorship benefits, then we might expect that even though bands’ lesser contributors are most often treated as co-authors, they receive a smaller share of songwriting royalties than their co-authors who made the most significant contributions. This would be the practice predicted by the scholars advocating for a proportional split default. Study 2 was designed to test this hypothesis.

IV. DO LESSER CO-AUTHORS RECEIVE LESS ROYALTIES?

The scholars arguing for a proportional default split would predict that when collaborators’ contributions to a joint work are unequal, they prefer an unequal royalty split. Study 2 tests this hypothesis not only for music groups, but for co-written songs in general. Study 2a investigates the songwriting royalty percentage splits of the uneven contributions bands in the Gold Record database using two different approaches. Study 2b adapts one of these approaches to estimate the proportion of uneven co-authored songs in general, using co-written songs in the ASCAP repertory.

A. Study 2a: Gold Record Bands Royalty Splits

Study 2a compiles uneven Gold Record bands’ and their managers’ disclosures about their royalty splits, along with statements about their splits from other reliable sources. If the proportional preference hypothesis is correct, the results should indicate that when co-authors’ contributions are uneven, they prefer to split songwriting royalties unevenly as well.

1. Methodology

First, a coder searched for information on the songwriting royalty split practices of every uneven band in the Gold Record database that includes all members as co-authors (n=140). Split information was found for 21% of these bands that met a standard of “reasonably certain” source reliability (29/140); on this basis their splits were classified as equal or unequal.

Second, to gain further insight into how uneven bands split songwriting royalties, I consulted the ASCAP repertory. In 2015, ASCAP began to disclose the percentages of song royalties under the control of its repertory, likely in preparation for an antitrust review by the U.S. Department of Justice. ASCAP is the only PRO that currently discloses the writers’ share of royalties it controls. I used ASCAP’s disclosures on the proportion of royalties under its control to code each song by the uneven bands as either split equally, or not. A coder searched the ASCAP repertory for songs by the uneven bands that include all members as co-authors (n=140) and with at least one member on whose behalf ASCAP does not collect royalties. Twenty-two bands (22/140, or 16%) were identified for which split data could be inferred.

The basis for inferring how a song’s royalties are split involved dividing the total writer’s share by the number of credited writers for the song, then evaluating whether the percentage due to each writer under an equal split arrangement was a factor of the percent controlled by ASCAP. If so, the song was classified as split equally; otherwise, the song was classified as not split equally. For example, for a song with two writers, if ASCAP controls 50% of the writer’s share of royalties, it follows that the non-ASCAP writer is receiving the remaining 50%. On the other hand, if ASCAP controls more or less than 50%, the writers are splitting unequally.

2. Results

The two approaches for identifying royalty splits used in Study 2a—compiling the splits of bands that have publicly disclosed their arrangements and inferring splits using the ASCAP repertory—converge in suggesting that approximately 80% split equally. Both methods suggest a majority of uneven bands nevertheless split royalties equally: the results of the online searches found that 83% (24/29) of disclosed splits were equal among band members; the PRO data suggests that 77% (17/22) of the bands split equally.

Using these two strategies, Study 2a was able to account for 33% of the uneven Gold Record bands that include all members as co-authors (46/140). The two methodologies produced results for different bands, with the exception of six bands for which results were found both ways. For these six bands, the different methodologies led to the same assigned split. These results indicate that most lesser contributors receive an equal split in the context of music groups. Study 2b investigates whether this is also true of co-written songs more generally.

B. Study 2b: All Co-Authored Songs

To go beyond how bands split royalties and to infer whether lesser contributors for all co-authored songs tend to be rewarded equally, I applied the strategy of inferring splits based on the proportion of royalties controlled by ASCAP to all eligible co-authored songs in the ASCAP repertory.

1. Methodology

The ASCAP repertory contains nearly 7 million songs. More than 2.5 million of these songs are credited to two or more authors. However, royalty splits can only be inferred in cases where a song’s royalties are controlled by both ASCAP and another PRO. There were a few steps involved in identifying those songs, and ultimately royalty split data was obtained for 1,237,764 works; this covers 92% of the songs ASCAP only partially controls, amounting to 48% of all co-authored songs in the ASCAP repertory.

Each song’s royalties were classified as split equally or split unequally, by the same methodology adopted in Study 2a. To account for potential misclassifications of unequally split songs as equally split songs, I estimated that a reasonable lower bound for the total proportion of equal split songs should include an 8% downward adjustment on the result of this classification method. This gave an estimate of the overall proportion of equally split songs in the all co-authored songs ASCAP data.

The next step was to estimate the proportion of those equally split songs that were produced through uneven contributions by collaborators. As a guide, I referenced the proportions of bands including all members as co-authors for Gold Record bands whose songs were composed by members’ contributions which were even, uneven, or without contributions by all members. These two datasets are similar at a high level based on the available data.

2. Results

A significant majority (63%) of the co-authored songs split royalties equally. To answer the research question, I needed to identify the subset of the ASCAP songs that were written with uneven contributions; however, the full dataset also includes songs written with even contributions, gift credits, and sampling. As a guide for inferring the size of the subset of interest, I refer to Study 1 for the relative shares of the three categories of songwriting contributions of the Gold Record bands including all members as co-authors: even (30%), uneven (44%), and some members do not contribute (26%). Relying on these proportions, I first assume that of the 63% of equal split ASCAP songs, 30% were written with even contributions. The remaining 33% equal split ASCAP songs I assume to have been written with uneven contributions. I assume that 26% of the unequal ASCAP songs credit as writers individuals who did not make a copyrightable contribution (based on the proportion of some members do not contribute bands which did the same). This leaves 11% of the unequal split ASCAP songs assumed to be uneven contributions songs (37% unequal minus 26%), and suggests that the proportion of unevencontributions songs that is equally split is 75%. A further adjustment should be made to account for the possibility of “hidden inequality” described above, which adjusts the estimated proportion downward by 8% to 67%.

This number is lower than the approximately 80% in Study 2a. One reason for this is that I have not taken into account sampling. Sampling inflates the number of total co-written songs, and specifically the number of unequal songs.

There is no difference in equal splitting between name combinations that only appear once (66%) versus at least ten times (65%). Number of collaborations for a set of writers was negatively associated with equal splitting, but nevertheless majority preference for equal splitting remained significant even with collaborative groupings with over one hundred credited works together (57%). A higher number of credited writers was negatively associated with equal splitting. The majority preference for equal splitting flips with five or more writers (27%).

V. GENERAL DISCUSSION

In this final Part, I review some limitations of the studies and how the findings answer the research questions, along with caveats on the support for the equal split rule. I discuss the legal implications and contributions of this research and preview other work focusing on the mechanisms that potentially drive these co-author inclusion and equal split practices.

A. Limitations

There are several limitations to the data and results of these studies; here I consider some of the most noteworthy. First, whereas a comprehensive database could be constructed to investigate the question of co-author inclusion (Study 1), this was not possible for exploring how royalties are split (Study 2). Instead, there are the convergent lines of Studies 2a and 2b. Each study involved making assumptions during data collection or analysis, which were necessary on account of data or resource limitations. To the extent these assumptions are not supported, the studies’ results and implications may be affected.

The results of the Gold Record database analysis may reflect survivor bias. Only a small fraction of all music groups earns Gold Records; it may be that less successful music groups are less likely to include minor contributors as co-authors. Although it cannot be ruled out, this seems unlikely because bands’ co-author inclusion choices are early business decisions in the groups’ lives, made before the arrival of great success, and are most often stable.

Songs are short by convention, which precluded investigating preferences when disparities in contributions are of the magnitude present in Aalmuhammed. Still, as discussed, the control doctrine has been applied more often than not regardless of the magnitude of the parties’ contributions, and, as argued, co-songwriting is the most important domain where the default rules are likely to apply in practice.

Figure 2.  Studies 1 & 2: Research Questions and Results

B. Summary of Findings

The control doctrine has resulted in the exclusion of lesser contributors from joint authorship when there was no contract between the collaborators. This, along with other problematic effects, has caused some scholars to propose returning to the pre-control legal regime and retaining the equal split rule. Others propose putting in place a proportional split default rule to better address the concerns with the control doctrine, and implicitly, align with creators’ assumed allocation preferences.

To identify the best choices, Studies 1 and 2 sought to uncover the actual preferences of co-songwriters. I have argued that the best option for co-songwriters is presumptively the best rule on the whole for those collaborating to produce joint works.

1. Are Lesser Contributors Counted as Co-Authors?

First, I explored whether or not lesser contributors in co-songwriting are typically rewarded with co-authorship (Study 1). To observe the leap from contributor to co-author, I constructed the Gold Record database of 1,003 bands—every band that both has a Gold Record and writes its own songs. I found that the inclusion of lesser contributors as co-authors has steadily increased for the past six decades, becoming the typical practice (63%) beginning in the 1990s. I also found that 75% of bands never changed their initial practice of including or not including all members as co-authors. This is consistent with a stable level of songwriting contributions. When bands did shift from their initial co-author crediting choices, it tended to be in the direction of more, rather than less, co-author inclusion. This could be explained by an increase in members’ songwriting contributions, but it also raises questions for future investigation about a possible ratcheting effect, whereby it is hard to renegotiate a generous initial co-author inclusion practice.

The strength of the co-author inclusion result is notable given the heterogeneity of the data, which includes bands from subgenres as diverse as Norteño, Metalcore, Eurodance, Contemporary Christian, Children’s Music, and Punk Rock. The music created by these groups appeals to a wide spectrum of people’s tastes.

This practice runs counter to the de facto exclusion of lesser contributors under the control doctrine. Additionally, to the extent that control is incompatible with recognizing more than a few authors because it is understood as a right to override others’ creative choices, the results are not consistent with co-songwriters sharing that view: songs created through the uneven contributions of four collaborators typically credit all involved.

Although this preference data strengthens the case against the control doctrine, I find support for limits on grants of co-authorship even for contributors who have made copyrightable contributions. When group size exceeds four contributors, having made a copyrightable contribution—even an equal one—is no longer treated as sufficient for co-authorship in most cases. Thus, considerations beyond contributions do play a role in deciding who is granted co-authorship credit for a joint work, at least in collaborations at scale. This is consistent with the view in Aalmuhammed that “as the number of contributors grows and the work itself becomes less the product of one or two individuals, . . . the word [‘author’] is harder to apply.” Future research might investigate what these other considerations may be. Nevertheless, in the case of co-songwriting, the findings suggest large collaborations are uncommon. On the whole, the concepts and consequences of the control test are at odds with the revealed preferences of this significant creator population. Because it cuts against these revealed preferences, the transaction costs of the current legal regime are presumptively high to the extent that parties seek to contract out (though the data does not report contracting rates).

Future research may explore what accounts for this inclusivity trend. It may be due in part to an increase in the kinds of bands that tend to include all contributors as co-authors, such as smaller bands. There have also been changes in music industry standards and practices, which may have played a role. First, mechanical songwriting royalties doubled—and continued to increase over time—after the Copyright Act of 1976. As songwriting royalties increased in value, more band members may have sought a piece of the pie. Including all members as co-authors may have facilitated agreements to feature the strongest songs on albums without generating animosity. Second, there were changes in what contributions were viewed as “songwriting.” In early Rock, the melody writer was considered the writer of the song. Early Rock was rooted in the Blues, and the rhythm portion of the song was often an uncopyrightable standard. Third, musicians have arguably become savvier as Rock culture has developed. In early Rock, the only legally sophisticated party was often the producer, studio owner, label owner, or bandleader. It was in that person’s best interest to limit the number of writers on the song and, if possible, add themselves as writers.

2. Do Lesser Contributing Co-Authors Receive Less Royalties?

The converging lines of evidence from Studies 2a and 2b suggest that royalties are typically split equally, even with lesser contributing co-authors (over 70% of the time). In Study 2a, the songwriting royalty splits of a third of the uneven contributions Gold Record bands were inferred based on information disclosed by the bands’ inner circles or ASCAP; in Study 2b, splits were inferred for co-songwriting more broadly, based on a sample of more than a quarter of co-authored songs registered with U.S. PROs. Perhaps surprisingly, the underlying preference assumptions of scholars who have argued the law should be revised to favor contribution-based splits were not supported. Instead, the results throw support behind the proponents of the existing equal split rule.

As with initial co-author inclusion choices, there was stability in initial split choices. Co-authors who wrote together only once were as likely to split royalties equally as co-authors who collaborated one hundred times. 

C. Legal Implications

Study 1 affirms the conventional wisdom, endangered by the control doctrine, that contributions are the most important consideration in granting co-authorship. Songwriting contributions overshadowed differences of genre, region, band size, and era. The copyrightability considerations standards have been on somewhat shaky ground as control has expanded to trump other factors in co-authorships tests. These results provide additional support for pushing back against that trend and reasserting a contributions-centered way of thinking about who is a co-author.

At the same time, the results imply that once co-authors are recognized, their comparative contributions are often disregarded when they allocate proceeds among themselves. Co-authorship practices appear as a step-function, rather than a matter of degree. Rather than provide a mandate for establishing a contributions-based default split rule, the findings suggest that the equal split rule is consistent with preferences and does not deter grants of co-authorship to lesser contributors. Taken together, Studies 1 and 2 suggest that equal co-authorship for lesser contributors is the practice of a plurality of the involved co-songwriters. On this basis, the studies lend support to the statutory intent to merge copyrightable contributions standards, or, at least, support the suggestion by Professor Mary LaFrance, and others, that a rebuttable presumption of intent is appropriate.

A return to the intent to merge co-authorship standard would predictably reach undesirable results in certain industries, most notably the film industry and editorial relationships. This section discusses two possible responses.

First, any default rule will not satisfy all situations. Parties can contract out of the default rule, and it is reasonable to expect that sophisticated parties will do so. Inevitably some may fail to put in place a contract through pure oversight. But it is inefficient to change the rule when such oversights are exceptional. In other contexts, when parties failed to contract around default rules, there have been costly settlements that have no doubt subsequently increased attention to such matters down the road.

Second, future research might explore a parameter-based approach to co-authorship determinations. For example, the number of collaborations was associated with co-author inclusion and royalty split. There has been considerable theoretical debate over the extent to which defaults should be “tailored” (that is, modified) such that they lead to different results for different parties based on some relevant characteristic. This has been recognized to potentially lead to more efficient outcomes by better reflecting decisions that parties would actually make in a costless environment. Default tailoring comes with costs: the cost to lawmakers of identifying and implementing the most efficient tailoring, and the cost to parties of interacting with a default more specific to their circumstances than a one-size-fits-all rule. Thus, the most efficiently tailored rule is likely one that reflects a discrete set of common practices, thereby reducing information costs. This may be why previous proposals to tailor the equal split default have focused on industry-based tailoring as a commonsense default unit. This Article has suggested that the number of collaborators is one potential parameter; if the size of collaborative groups predicts split preference across genres and regions, it is worth investigating whether it is a useful predictor across industries. Pursuing this and other leads may bring about gains in efficient tailoring over and above what is possible with industry-based rules.

D. Why Equality: Mechanisms

The data presented suggest that when it comes to relative rewards, co-songwriters, on balance, prefer equal over contributions-based allocations—even in unequal collaborations. What can account for this surprising preference for equality? A systematic investigation of the preference mechanisms at work is out of the scope of this Article. I conclude here by priming intuitions connected with the theoretical framework I use elsewhere for that purpose.

Context often determines which of a few interpersonal allocation rules we find to be most psychologically rewarding. There are many contexts in which we find equality—where everyone is rewarded in the same quantities—to be the most psychologically rewarding state of affairs. Under other circumstances, we may prefer contribution-based remuneration. A context can be defined, for example, by who is involved or what goals are involved. There are patterns to when we are likely to have these preferences, which have been studied under various research traditions in distributive justice. Earlier, I noted that the members of a large number of bands are friends or family members. Theories that can provide a way of thinking about the influence of parties’ close relationships on the allocations they prefer when distributing a resource—such as royalties—among themselves, are most conducive for understanding this equality preference.

CONCLUSION

The trends in co-songwriters’ preferences revealed by this research offer the first comprehensive look into decisions made every year, hundreds of thousands of times, offering a point of reference for creators. In songwriting, co-authorship matters. Not only does it confer status; it also confers money. For some musicians, co-authorship royalties are their retirement plan. As the lead singer in an all-female Punk Rock band explained songwriting credits, “we’re straight up talking business here.”

And yet, in this Article I have looked at this “business” decision of how to share co-authorship and found that short-term financial self-interest is not the end of the story: main songwriters share songwriting credit, and songwriting royalties, much more generously than the standard by which lesser contributors would be rewarded under the control test. This revealed preference suggests that we can adopt a more inclusive legal criterion for co-authorship while retaining the equal split default and that we can do so without violating creators’ own sense of fairness.

APPENDIX

Table A1.  Logistic Regression for Gold Record Bands 1959–2021: Are All Members Included as Co-authors?

Variable

log(OR)

95% CI

p-value

Songwriting (ref. cat.: Some members do not contribute)

     

Even

2.9

2.4, 3.5

<0.001**

Uneven

1.8

1.4, 2.1

<0.001**

Decade (ref. cat.: 1980s)

     

1960s and earlier

-0.94

-1.7, -0.20

0.016*

1970s

-0.33

-1.0, 0.27

0.3

1990s

0.50

0.06, 1.0

0.026*

2000s

0.74

0.23, 1.2

0.004**

2010s and later

0.90

0.23, 1.2

0.004**

Members (ref. cat.: 4)

     

2

0.91

0.43, 1.4

<0.001**

3

0.02

-0.42, 0.45

>0.9

5

-0.61

-1.1, -0.11

0.017*

6+

-1.5

-2.6, -0.56

0.003**

Genre (ref. cat.: Pop)

     

Rock

0.37

-0.14, 0.89

0.2

Hip Hop, R&B, Gospel, Jazz

-0.79

-1.4, -0.16

0.014*

Country

-1.0

-2.0, -0.09

0.034*

Metal

0.27

-0.52, 1.0

0.5

Latin

-1.5

-3.0, -0.18

0.032*

Punk

0.87

-0.29, 2.0

0.13

Electronic

-0.50

-1.8, 0.90

0.5

Reggae

1.0

-1.6, 4.1

0.5

 

Region (ref. cat.: Northeast)

     

Midwest

-0.68

-1.4, 0.04

0.069

West

0.01

-0.49, 0.51

>0.9

South

0.11

-0.43, 0.65

0.7

Non-USA

0.26

-0.26, 0.78

0.3

Notes: *p < 0.05; **p < 0.005.

     

96 S. Cal. L. Rev. 607

Download

Fellow, Stanford Law School. For comments on versions of this Article, I thank Lisa Ouellette, Paul Goldstein, Robert MacCoun, Samantha Zyontz, Joel Shor, Anna Lewis, and Adam Brown. For helpful feedback, I also thank the participants of the 2022 Copyright Scholarship Roundtable at Columbia Law School, the Stanford Legal Research in Progress Workshop 2020, the Intellectual Property Scholars Conference 2020 at Stanford Law School, and the Works-in-Progress Intellectual Property Colloquium 2020 at Santa Clara University School of Law. Thank you to Matthew Kelleher and William Button for outstanding research support.

The Limitations of Applying the Stored Communications Act to Social Media

The advent of social media has increasingly affected how people live and communicate. Millions of Americans use social media every day, and the numbers continue to grow. The motivation to post on social media is multifactorial and includes a desire to stay connected, find others with shared interests, change opinions, and encourage action, but posting also serves to boost one’s self-esteem and self-worth. However, posting on social media creates a serious risk of self-disclosure, with people revealing more intimate details online than they would in more traditional settings without really appreciating the privacy issues and potential negative consequences related to such disclosures.

As social media use continues to grow, its use as a tool in police investigations has also increased. Both the content and metadata associated with social media posts now routinely aid law enforcement authorities in finding patterns and, importantly, in establishing timelines in criminal investigations. Thus, there is an urgent need to revise the existing laws governing stored communications—to better adapt them to these new, evolving technologies and improve the legal framework governing online privacy rights. This Note argues that various aspects of the Stored Communications Act (“SCA”) are outdated and that thirty-six years after it was enacted, it is time for an update that reflects the changing landscape of evolving technological advances.

The Note explores how the internet and social media use have evolved over the years and explains why the SCA no longer sufficiently protects consumers from government acquisition of their information. Particular emphasis is placed on the novelty of social media “Stories,” a technology unlike any that Congress could have imagined when it enacted the SCA in 1986. The Note examines the history of the SCA—with a focus on the Fourth Amendment, the Electronic Communications Privacy Act, and Supreme Court cases addressing the applicability of the Fourth Amendment to various forms of communication technology—before analyzing the SCA in detail, and looks at how law enforcement agencies can obtain these communications for use in criminal investigations. The Note concludes by arguing that the SCA needs to be revised to more adequately apply to today’s social media technologies since their content, and non-content, does not easily fit into the currently delineated categories. Revising the SCA would afford greater protection to consumer communication rights: not only would the SCA better apply to modern technology, but it would also be more readily applicable to future emerging media technologies.

INTRODUCTION

The rise of social media has significantly impacted the way people live and communicate, and the trend toward extensive social media use will likely only continue to grow. According to a Pew Research Center study, seven in ten Americans use social media. On average, people spend an estimated two and a half hours on social media platforms over the course of their day, and “[a] majority of Facebook, Snapchat and Instagram users say they visit these platforms on a daily basis.” More specifically, 69% of Americans use Facebook, 40% of Americans use Instagram, and 25% of Americans use Snapchat. These percentages represent a significant number of people—approximately 230 million, 133 million, and 83 million, respectively. Further, social media users make extensive use of the “Stories” feature, with one billion Facebook Stories being posted daily and five hundred million daily active users of Instagram Stories worldwide. The motivation to post on social media is multifactorial and includes a desire to stay connected, find others with shared interests, change opinions, and encourage action, but posting also serves to boost one’s self-esteem and self-worth. These desires create a serious risk of self-disclosure on social media, with people revealing more intimate details online than they would in more traditional settings without really appreciating the privacy issues and potential negative consequences related to such disclosures.

Just as social media has become popular with the American public, it is also becoming increasingly utilized as a tool in police investigations. A 2012 survey showed that four out of five law enforcement agents used social media to gather intelligence during investigations. Not only do authorities look online for public information, but they also request access to private data directly from social media providers—which can help them build their criminal cases. For example, after finding photos and comments “glamorizing alcohol abuse” on a woman’s MySpace page, prosecutors were able to use them as evidence and advocate for a longer sentence for her vehicular manslaughter conviction. Since people are less inhibited when it comes to social media disclosures, they often share details of their lives and more controversial opinions than they may in other forums. After these once private thoughts are stored electronically, they become more easily accessible to investigators. Not only can the content of social media posts aid criminal investigations, but the related metadata alone “can help law enforcement authorities to find patterns, establish timelines and point to gaps in the data.” Therefore, social media metadata can be just as easily used to gather information on a suspect as the actual content of a post. Because the trend toward extensive social media use will likely endure, there is an urgent need to revise the laws governing stored communications—to better adapt them to these evolving technologies and improve the legal framework governing online privacy rights.

This Note argues that various aspects of the Stored Communications Act (“SCA”) are outdated and that thirty-six years after it was enacted, it is time for an update that reflects the changing landscape of evolving technological advances. Part I of this Note explores how the internet and social media have evolved throughout the years and explains why the SCA no longer affords sufficient protections against government acquisition of consumer information. It discusses the evolution and expansion of social media platforms. Particular emphasis is placed on the novelty of social media Stories, which are unlike any technology that Congress could have imagined when they enacted the SCA in 1986.

Next, Part II examines the history behind the SCA to explain why the law was initially passed by Congress, with a focus on the Fourth Amendment, the Electronic Communications Privacy Act (“ECPA”), and Supreme Court cases addressing the applicability of the Fourth Amendment to various forms of technology. Part III analyzes the SCA in detail, focusing on the distinctions made between the different types of internet service providers (“ISPs”) and the different aspects of communications (content versus non-content data). It looks at how the content and non-content information—for example, metadata including a user’s identity, location, and other data not part of the main substance of the communication—can be obtained by law enforcement in the course of a criminal investigation.

Part IV argues that the SCA cannot be easily applied to social media today because it does not fit within the categories delineated in the SCA. Most importantly, it highlights how (1) social media content does not easily fit into either of the SCA’s currently defined categories because Congress could not have anticipated the advances in the technologies that exist today; and (2) “non-content” is not fully defined in the statute, and therefore lends itself to being more easily obtained in some situations as opposed to others. Finally, Part V suggests ways in which the SCA can be revised to more adequately apply to social media today and ultimately protect the right to privacy guaranteed by the U.S. Constitution.

I. INTERNET PRIVACY AND EVOLVING TECHNOLOGY

Americans are entitled to their right to privacy, which on third-party ISPs such as Facebook and MySpace is protected by the SCA. One problem with the SCA, however, is that it is dated. Although the internet was invented in the 1960s, it was not widely used until 1983, when computers on different networks were finally able to easily communicate with one another. When the SCA was enacted in 1986—just three years later—Congress had only a limited experience with internet use and the potential privacy problems it could create, and had certainly not envisioned the extensive modern use of social media. This partially accounts for some of the weaknesses in this legislation and why the SCA is often difficult to apply to social media today.

A. Evolution of Social Media Platforms

Social media is defined as “forms of electronic communication . . . through which users create online communities to share information, ideas, personal messages, and other content.” This definition implies that social media could not exist without the internet, and that it depends on user-generated content. While it can be said that social media began in 1971, when the first email was sent, for many people social media really began in the late 1990s or early 2000s—years after the SCA was enacted—with the advent of messaging services such as AOL and MSN Messenger. MySpace, arguably the “most popular and influential” of the early social media platforms, was later launched in August 2003, and it allowed individuals to interact by commenting on each other’s profiles and sending private messages. It was the largest social media platform until Facebook, created in 2004, overtook it in 2008. Facebook has now grown to be the largest social media platform in the world with almost three billion monthly active users.

The number and types of social media platforms have grown extensively. Today, other prominent social media platforms include Instagram and Snapchat. Instagram was launched in 2010 and is a platform focused on sharing photos and videos. Snapchat was created in 2011 and gained its popularity from users’ ability to send each other pictures or videos (“Snaps”) that disappear shortly after being opened. These platforms allow users to share content with their friends, some of which they believe to be “private,” visible only to those friends they allow to see it. However, the widespread use of these platforms has created new issues with how the government can legally access and use these communications.

B. Emergence of Stories on Social Media Platforms

The continued evolution and development of new information sharing functions on social media platforms have created multiple issues concerning user privacy rights. For example, in 2013, Snapchat began to allow people to share “Stories” that are displayed for twenty-four hours before becoming inaccessible. Stories are a collection of individual Snaps that are played in the order in which they were created and allow users to share their entire day in a narrative manner. Today, Stories are also available on a variety of other social media platforms, including Facebook and Instagram. Part of the reason why Stories are so successful is because they are only available temporarily, so people can post small daily updates or silly images that they only want visible for a short period of time. Therefore, users reasonably believe that their content will remain private and then disappear, becoming permanently inaccessible. Another reason for the success of Stories is that “social media [S]tories tend to be more spontaneous” than an individual’s carefully curated feed, making it feel more “casual.” As a result, these Stories can be extremely useful to law enforcement, as they can provide a less filtered view of an individual’s daily life and a timeline for the posted events. Thus, the challenge becomes balancing users’ right to privacy with the government’s need for access to information in order to investigate criminal offenses.

As it exists now, the SCA does not provide an adequate statutory framework for protecting communications on the various aforementioned social media platforms and, importantly, does not specifically address new advances in technology such as transient Snapchat and Instagram Stories. Since the SCA does not adequately protect individuals from unlawful searches of their private social media data, there is a need for Congress to reform the statute to accommodate evolving technology.

II. HISTORY OF THE STORED COMMUNICATIONS ACT

A. The Fourth Amendment

The Fourth Amendment to the Constitution protects “[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” While the meaning of “search” is not immediately defined by the Amendment, the Supreme Court has held that “[a] ‘search’ occurs when an expectation of privacy that society is prepared to consider reasonable is infringed” and that “[i]f the inspection by police does not intrude upon a legitimate expectation of privacy, there is no ‘search.’ ” Thus, when it comes to physical searches, the meaning of the Fourth Amendment is well understood, whereas what constitutes a search in the digital context is more uncertain.

In Olmstead v. United States, the Supreme Court held that wiretapping did not violate the Fourth Amendment because the lack of physical trespass and seizure of anything tangible meant there was no search or seizure. Because the Court refused to expand the Fourth Amendment to protect telephone communications, the government could legally intercept citizens’ communications as long as they did not physically enter their homes. Olmstead was later overruled by Katz v. United States, indicating a change in ideology that afforded citizens protection of their privacy even without a physical search. Because Katz held that a physical intrusion was not necessary to invoke the Fourth Amendment, online searches—which lack physical intrusions—can still violate the Fourth Amendment.

B. The Electronic Communications Privacy Act

In light of these changing viewpoints on the applicability of Fourth Amendment protections, Congress enacted the ECPA in 1986 in an effort to adapt the doctrines of the Fourth Amendment to the various emerging technologies. The SCA, which provides privacy protections to stored electronic and wire communications, is one part of the ECPA. The ECPA was created with the purpose of protecting American citizens from “the unauthorized interception of electronic communications.” Congress recognized a need to “update and clarify Federal privacy protections and standards in light of dramatic changes in new computer and telecommunications technologies.” Rightly, Congress worried that due to these advances, personal communications could be intercepted by individuals who had no right to obtain them, and thus felt it was important to enact the ECPA. However, the scope of the ECPA did not fully anticipate the impact of the growth and extent of social media.

C. Supreme Court Cases Addressing the Fourth Amendment and Technology

More recently, the Supreme Court heard a series of cases that addressed the applicability of the Fourth Amendment to newer technologies. In each of these cases, the Supreme Court Justices grappled with applying the existing legal framework, indicating that it is time for a change. In Justice Sotomayor’s concurring opinion in United States v. Jones, she emphasized that in the absence of a physical trespass, a Fourth Amendment search occurs “when the government violates a subjective expectation of privacy that society recognizes as reasonable.” She also argued that “it may be necessary to reconsider the premise that an individual has no reasonable expectation of privacy in information voluntarily disclosed to third parties” because “[t]his approach is ill suited to the digital age.” Justice Sotomayor’s statements highlight the need to reevaluate the applicability of the current legal framework to new technologies.

Two years later, in Riley v. California, Justice Roberts acknowledged that because technology enables modern cell phones to contain and potentially reveal a wealth of private information, cell phones require greater privacy protections than would be necessary for a traditional search. Four years after Riley, the Court once again addressed warrantless searches in Carpenter v. United States, this time through the collection of cell phone records from a third party. Again, Justice Roberts recognized the need for stronger privacy protections, stating that “a warrant is required in the rare case where the suspect has a legitimate privacy interest in records held by a third party,” such as the cell site records indicating the defendant’s location and movements. The government had acquired this information pursuant to a court order issued under the SCA, which was obtained based on evidence that the information might be relevant to the ongoing investigation. Finding this burden of proof—requiring only that the information might be relevant, which is lower than the probable cause required to obtain a warrant—to be unacceptable, the Court held that to access these cell site records, a warrant was required. The differing standards of proof required to obtain warrants and court orders to access records from these new technologies illustrate that sometimes the SCA troublingly affords lesser protections to individuals’ private information.

III.  THE STORED COMMUNICATIONS ACT

The SCA was enacted to regulate electronic and wire communications that are stored on third-party servers and therefore governs the interaction between government investigators and administrators of third-party service providers. It was meant to expand the privacy protections afforded by the Fourth Amendment to digital content, clarifying its applicability. However, the SCA regulates retrospective communications, meaning it only applies when the government seeks to obtain information already in a provider’s possession. Additionally, the SCA only applies to two types of ISPs: providers of electronic communication service (“ECS”) and providers of remote computing service (“RCS”). An ECS is defined as “any service which provides . . . the ability to send or receive wire or electronic communications;” email and cell phone service providers would therefore be examples of ECS providers. An RCS, on the other hand, is defined as any service that provides to the public “computer storage or processing services by means of an electronic communications system.” Thus, once an email has been received but not deleted or a voicemail has been left in storage for later review, email and cell phone services are treated as RCS providers. Because ECS and RCS providers are afforded different levels of protection, it is important to be able to appropriately categorize modern ISPs to determine how much protection users’ communications will be given.

While transmitting communications and storing communications are different functions, this distinction matters less today, as many modern ISPs provide both services. In 1986, however, Congress was concerned about businesses such as hospitals and banks using remote computing services to store records and process data. Thus, they felt the need to create the RCS category to address this concern. Generally, the SCA prohibits disclosure of both content and non-content data of customer communications, but the SCA provides exceptions to this rule. These exceptions, which are discussed below, are divided between § 2702, which regulates voluntary disclosure, and § 2703, which regulates required disclosure.

A. Disclosure of the Contents of Social Media Posts

1. Voluntary Disclosure of Customer Communications

Section 2702(b) details the nine circumstances in which a provider may voluntarily disclose the contents of a customer’s communications. These exceptions include allowing the contents to be disclosed “to an addressee or intended recipient of such communication” and “with the lawful consent of the originator or an addressee or intended recipient of such communication.” For the most part, the communications can be disclosed only with the permission of the sender or intended recipient, which protects the user, or without their permission in the case of an emergency, such as a missing child. Therefore, while individuals are generally protected against voluntary disclosures of their private information by ISPs, it does not mean that the government is unable to obtain this information; it can be compelled through required disclosure under § 2703.

2. Required Disclosure of Customer Communications

Should the government decide that obtaining an individual’s communications is essential for building a criminal case against them, the disclosure of those communications is governed by § 2703. This is where the largest privacy threat to social media users lies, as ISPs are then legally required to turn over the contents of customer communications to law enforcement. How the government goes about getting this information under § 2703, however, depends on a variety of factors, beginning with whether the ISP is categorized as an ECS or an RCS.

If the government requires information from an RCS, there are three ways for it to compel disclosure. First, the government can compel disclosure without notifying the customer if “the governmental entity obtains a warrant issued using the procedures described in the Federal Rules of Criminal Procedure (or, in the case of a State court, issued using State warrant procedures . . . ) by a court of competent jurisdiction.” Alternatively, if the government provides notice to the customer, it can compel disclosure by using either (1) “an administrative subpoena authorized by a Federal or State statute or a Federal or State grand jury or trial subpoena;” or (2) “a court order . . . [obtained] under subsection [2703](d).” Warrants place a higher burden on the government in order to obtain the requested information, while subpoenas and court orders are more easily obtainable. Thus, allowing the government to choose the second or third method to avoid having to obtain a warrant shifts the burden to the individual, who then must object to the subpoena or court order to protect their private information.

Required disclosure from an ECS, on the other hand, is even more complicated because it also considers information about the age of the communication. If the communication is 180 days old or less, the government may only compel disclosure “pursuant to a warrant issued using the procedures described in the Federal Rules of Criminal Procedure (or, in the case of a State court, issued using State warrant procedures . . . ) by a court of competent jurisdiction.” If the communication is more than 180 days old, however, the government can compel disclosure with either a warrant or, if prior notice is provided, a subpoena or court order. In effect, this makes it easier for investigators to obtain older communications, with no explanation as to why the 180-day mark is significant; thus, in this situation, users are arbitrarily afforded less protections.

B. Disclosure of the Non-Content Data of Social Media Posts

1. Voluntary Disclosure of Customer Records

Section 2702(a)(3) prohibits ECS and RCS providers from “divulg[ing] a record or other information pertaining to a subscriber to or customer of such service . . . to any governmental entity.” However, § 2702(c) provides an exception to this rule: “A provider . . . may divulge a record or other information pertaining to a subscriber to or customer of such service . . . as otherwise authorized in section 2703.” Therefore, while the SCA prevents ECS and RCS providers from voluntarily disclosing non-content information to governmental entities, as with content, the government can still obtain the information by utilizing § 2703’s required disclosure provision.

2. Required Disclosure of Customer Records

Section 2703(c)(1) states that a governmental entity can require an ECS or RCS provider to disclose a record or other information when the governmental entity “obtains a warrant issued using the procedures described in the Federal Rules of Criminal Procedure (or, in the case of a State court, issued using State warrant procedures . . . ) by a court of competent jurisdiction”; “obtains a court order”; “has the consent of the subscriber or customer”; “submits a formal written request relevant to a law enforcement investigation concerning telemarketing fraud”; or “seeks information” under § 2703(c)(2). Section 2703(c)(2) allows ECS and RCS providers to disclose the name; address; telephone connection records (or records of session times and durations); length of service and types of service utilized; subscriber number; and “means and source of payment” when the governmental entity “uses an administrative subpoena authorized by a Federal or State statute or a Federal or State grand jury or trial subpoena or any means available under [§ 2703(c)](1)].” Again, governmental entities are able to obtain varying amounts of private information about customers from ECS and RCS providers with either a warrant or a court order, sometimes even with only a subpoena. Even more troubling, § 2703(c) does not require the government entity receiving the records or information to provide notice to the customer. Thus, subscribers’ privacy may be being infringed without their knowledge, providing them with fewer opportunities to protect themselves.

IV. SOCIAL MEDIA AND THE STORED COMMUNICATIONS ACT

Prior to 2010, no court had specifically addressed whether social media platforms were within the jurisdiction of the SCA. In order for the SCA to apply to social media platforms, these ISPs must be considered either ECS or RCS providers. The District Court for the Central District of California was the first to examine whether social media platforms were ECS or RCS providers in Crispin v. Christian Audigier, Inc. The district court held that because the three social media platforms in question provided either private messaging or email services, they qualified as ECS providers. This demonstrated that the SCA could be applied to social media platforms and can, therefore, be used to control the release of social media communications. While Crispin made it clear that Facebook, Instagram, and Snapchat would be governed by the SCA, it remains unclear whether these platforms qualify as an ECS, an RCS, or both, in the context of specific functions. As a result, which regulations should be applied when the government seeks to obtain users’ content (or non-content) from social media platforms during a criminal investigation remains uncertain.

A. Obtaining Contents of Social Media Posts

1. Obtaining Contents from Private Social Media Accounts

The SCA only applies to communications that are not “readily accessible to the general public.” Thus, it is important to understand how a user’s varying privacy settings on social media platforms can affect the applicability of the SCA. Facebook, Instagram, and Snapchat each have varying features that provide users with controls to limit who can see the content they have posted on their individual accounts, in some instances allowing the users to limit who can view individual posts as well, and the ability to block other users from viewing their content. Accordingly, should a user want their social media content to be private, they have the ability to set those limits using the social media platform settings.

In Crispin, the court held that “[u]nquestionably, the case law . . . require[s] that [user content] be restricted in some fashion . . . [to] merit protection under the SCA.” Therefore, if a user sets their content visibility to anything other than public, it qualifies as private. This was confirmed in Ehling v. Monmouth-Ocean Hospital Service Corp., in which the District Court of New Jersey found that “when users ma[d]e their Facebook wall posts inaccessible to the general public, the wall posts [we]re ‘configured to be private’ for the purposes of the SCA.” Similarly, in Facebook v. Superior Court (Hunter), the Supreme Court of California held that social media posts that were configured to be public fell within § 2702(b)(3)’s lawful consent exception, which allows ISPs to disclose a user’s content with the user’s consent. By this logic, if a user’s content is visible to the public, they are consenting to the RCS provider’s disclosure of their content. The SCA, therefore, does not protect social media content that is posted publicly because consent is an exception to the prohibition of voluntary disclosure under § 2702. The Hunter court also held that “restricted communications sent to numerous recipients cannot be deemed to be public—and do not fall within the lawful consent exception.” In other words, even if social media communications are limited to a large group of people, that does not mean these posts are considered public. According to the Ehling court, “the critical inquiry is whether Facebook users took steps to limit access to the information . . . . Privacy protection provided by the SCA does not depend on the number of Facebook friends that a user has.” By restricting one’s content with privacy settings, a social media user can therefore take advantage of the SCA’s privacy protections and make it more difficult for the government to obtain their content—by requiring them to get a warrant, for example—for use in a criminal case, but not all users are that savvy or careful.

Based on this jurisprudence, it should not matter how broad the user’s privacy settings are—as long as the individual specifically took steps to limit who can view their content, it becomes protected from voluntary disclosure. This is not foolproof, however, because, as discussed earlier, disclosure may still be permitted if authorized by § 2703. This remains problematic because, as Justice Sotomayor stated in Jones, a Fourth Amendment search online occurs when the government violates a “subjective expectation of privacy[,]” and one could argue that when an individual invokes privacy settings, they reasonably expect that their content will be kept private. If obtaining individuals’ social media data constitutes a search, then under Justice Roberts’s logic in Carpenter, a warrant should be required because social media content can contain lots of information about a person’s day, including their location and movements, like the cell site records in Carpenter. Therefore, it stands to reason that all searches of private social media content should require a warrant, which is not currently the case under the SCA.

2. Social Media: Does Disclosure of Its Content Follow ECS or RCS Regulations?

As previously discussed, the SCA has different standards for an ECS than for an RCS—the government can more easily obtain communications from an RCS, whereas obtaining communications from an ECS depends on how long ago the communications were created, thus emphasizing the importance of properly categorizing each social media platform. In Crispin, the court found that social media platforms can be characterized differently depending on the state of the messages: before the messages have been opened, ISPs operate as ECS providers, but once the messages have been opened and retained, the ISPs operate as RCS providers. This creates significant complexity and results in variability in how the SCA is applied to each social media platform, given the different standards between RCS and ECS providers and the difficulty in determining which standard will apply.

The Crispin court acknowledged that Facebook wall posts and MySpace comments “present a distinct and more difficult question” as to whether the social media platforms are acting as ECS or RCS providers. On one hand, the court stated that Facebook and MySpace were ECS providers with respect to wall posts and comments because they were being held for “backup purposes once read.” Here, the court relied on Snow v. DIRECTV, Inc., in which a district court found that because electronic bulletin board services (“BBS”) did not have temporary, intermediate storage, they were actually storing the information for backup purposes and thus were an ECS. The court analogized Facebook and MySpace wall posts and comments to BBS, concluding that these posts and comments were also being stored for backup purposes since they were not deleted after being read, and thus the social media platforms should be considered ECS providers.

On the other hand, the court also said that Facebook and MySpace could be considered RCS providers with respect to wall posts and comments because they maintained these communications not only for storage, but also for display purposes, as users wanted their friends to be able to see the communications. The court relied on Viacom International Inc. v. YouTube Inc. in this instance, analogizing Facebook wall posts and MySpace comments to private YouTube videos. In Viacom, the court found that YouTube was an RCS provider because it stored videos on behalf of its subscribers. Thus, the Crispin court concluded that Facebook wall posts and MySpace comments, like YouTube videos, can be stored for the purpose of allowing other users to view the content, thus making Facebook and MySpace RCS providers, like YouTube. Ultimately, the court did not rule whether Facebook and MySpace were ECS or RCS providers with respect to wall posts and comments, remanding the case for further development. This complexity demonstrates how ill-suited the SCA currently is to protect individuals’ privacy on social media platforms, as there is no clear and consistent way to apply it. Further, the arguments made in Crispin emphasize just how arbitrary the distinction between an RCS and ECS provider can be when it comes to social media platforms. Because social media platforms do not fit neatly into either category, courts can come to different conclusions as to how these ISPs should be regulated, thus leading to uncertainty regarding the protection of privacy rights of social media users. This arbitrariness can be explained by the fact that the SCA was written in 1986, as articulated in Konop v. Hawaiian Airlines, Inc.:

[T]he ECPA was written prior to the advent of the Internet and the World Wide Web. As a result, the existing statutory framework is ill-suited to address modern forms of communication like [social media platforms]. Courts have struggled to analyze problems involving modern technology within the confines of this statutory framework, often with unsatisfying results.

The Konop court’s words make clear that the SCA has become outdated because Congress was unable to foresee the problems that would arise for privacy protections resulting from not yet existing communication technologies. This is further supported by the fact that the Crispin court was unable to make a decision regarding the status of Facebook and MySpace with respect to wall posts and comments, given the limitations in clearly and consistently applying the SCA to communications on the various social media platforms. Courts’ inability to readily place certain features of social media platforms into existing categories highlights the inadequacy of the SCA in affording privacy rights to users of the prevalent modern technologies and supports that now is the time to change the SCA to clarify its applicability and afford stronger protections for various types of social media communications by creating more appropriate categories that these ISPs can be classified into.

3. Challenges in Applying SCA Content Disclosure to Stories

Stories are a relatively new feature of social media platforms, having only been in existence since 2013. Like with the aforementioned difficulty in generally applying the SCA to social media platforms and user content, Stories, which disappear within twenty-four hours, provide another example that highlights the limited applicability of the current statutory framework under the SCA to modern communication technologies. From a privacy perspective, the good news is that most of these posts are removed from ISPs’ servers as soon as the twenty-four hour period is up. Since the content is no longer on the social media platform’s server, it is not possible for ISPs to disclose this content—even pursuant to a court order, subpoena, or warrant—because the content would no longer be in storage. However, concerns remain for any content that remains saved on the server, which might still be obtainable for criminal investigations under the current SCA.

In addition, both Facebook and Instagram Stories can be saved in Story Archives, and Snapchat Stories can be saved in Memories. This content, therefore, could feasibly be disclosed to the government under the SCA if the proper exceptions and procedures were met. Because part of the appeal of Stories is that posts are only available for twenty-four hours, users likely do not think about how long their content is maintained in storage. Rather, many incorrectly assume that the content has been permanently deleted when the twenty-four hours expire. The problem here is that if Stories are governed by current ECS rules, once Stories are more than 180 days old, they can be obtained with notice and a subpoena or court order. This goes against the intent underlying Justice Robert’s opinion in Carpenter because one could similarly argue that individuals who post Stories believe they have a reasonable expectation of privacy in these Stories that are now only available for their own view, yet they can, in fact, still be obtained with lesser protections than a warrant. Therefore, even though the SCA was intended to extend the protections of the Fourth Amendment to online communications, currently it does so unsuccessfully, particularly in the case of Stories.

Because Stories are so new, there have not been many cases addressing how the SCA applies to them. In Facebook, Inc. v. Pepe, the District of Columbia Court of Appeals considered an allegedly sent “disappearing Instagram ‘Story’ ” for the first time. The court found that the Instagram Story was content under the SCA, and that because James Pepe was an “addressee or intended recipient” under § 2702(b), Facebook was permitted to disclose any Instagram Stories that were responsive to the subpoena. However, this addressee or intended recipient exception would not apply if the government were seeking disclosure in a criminal case, as the individual who posted the Story would likely not have invited a government official to view their private Facebook, Instagram, or Snapchat Story. Thus, the inquiry then shifts to consider whether social media platforms are acting as RCS or ECS providers when it comes to Stories.

One could analogize Stories to Facebook wall posts and MySpace comments when applying the SCA to social media Stories. Following the Crispin court, this would mean that ISPs offering Stories could be considered either RSC or ECS providers. The first argument is that Facebook, Instagram, and Snapchat act as ECS providers when individuals post Stories because the individual is “sending” the electronic communication to the people who they have allowed to view it. This would follow from analogizing Stories to wall posts or comments that are in “backup” storage. As per Crispin, if the messages are being stored on the servers solely because they were not deleted, then they are in backup storage and, thus, should be governed by ECS rules. Unfortunately, users do not usually think about deleting this type of content because they know that once it disappears, no one else can see it. However, what they often fail to realize is that these communications are then considered to be in backup storage, meaning they can still be disclosed to the government under the SCA.

Alternatively, Facebook, Instagram, and Snapchat could be considered RCS providers because they are simply storing the Stories on the server for others to view. In Crispin, wall posts were compared to YouTube videos that were stored for the purpose of allowing other users to view the content. Arguably, Stories are also stored for the purpose of allowing others to view them, not simply because they have not been deleted. Therefore, even though a Story disappears after twenty-four hours, the user can reshare the content from their Archive, similar to changing a YouTube video’s settings to modify who can view it at any point in time.

On the other hand, Stories could also be analogized to private messages, which further complicates the analysis of SCA protections, particularly when considering the reasoning in Crispin, which stated that when a message is unread, the ISP acts as an ECS, but once the message has been read, the ISP then acts as an RCS. Stories can be viewed by whomever the user allows, depending on their privacy settings, meaning that at any given point in time, the Story might have been viewed by a portion, but not all, of the potential audience. Thus, is the Story considered “unread” until all possible viewers have seen it, or does it switch to being “read” once at least one individual has viewed it? Alternatively, a Story could be “sent” while it is available for viewing by others but then switched to “read” once the twenty-four hours are up.

Whether or not a Story is considered to be an ECS or an RCS function directly impacts how law enforcement agencies can obtain its contents since the content of a Story would only be protected with a warrant if it were governed by ECS rules and 180 days old or less. Otherwise, Stories could be obtained with either a subpoena or a court order, making them easier to acquire for criminal investigations. These types of questions have not yet been adequately addressed by courts, and because Stories have qualities of both RCS and ECS communications, it is not possible to consistently predict whether RCS or ECS rules should govern in individual cases. The difficulty in determining how to appropriately apply the SCA to Stories supports the need for the proposed changes to the SCA.

B. Obtaining Non-Content Data From Social Media Posts

1. Applying SCA Non-Content Disclosure to Social Media Platforms

Disclosure of non-content data stored by social media platforms is different from disclosure of content in that non-content disclosure does not depend on whether the provider is an ECS or an RCS. While content is defined as including “any information concerning the substance, purport, or meaning of that communication,” non-content is not well-defined. The SCA does, however, define some non-content data that can be obtained with only a subpoena, including the user’s name, address, and telephone number. This stems from the third-party doctrine, which states “the
Fourth Amendment does not prohibit the [government from] obtaining . . . information revealed to a third party.” This creates an exception to the reasonable expectation of privacy that is protected by the Fourth Amendment: once an individual voluntarily shares information with a third party, they lose any reasonable expectation of privacy in that information. It can be assumed, however, that non-content data is any information that is not the main substance of the communication, including the metadata incorporated in the communication, for example, the user’s identity, location, payment information, and telephone number. This is problematic because under § 2703(c), non-content data can sometimes be easily obtained by the government with a court order. Because the SCA does not explicitly state which types of non-content data can be obtained with a court order and which require a warrant, a lot of discretion is left to police officers and the courts.

“Some non-content information, particularly associational information and location information, is inherently expressive, capable of directly exposing intimate details of an individual’s life.” In the age of social media, people are constantly posting images and videos online; when people take photos, for example, the image files contain metadata that includes the time and date when the image was taken, along with the exact location where the photograph was taken. Facebook, Instagram, and Snapchat collect a lot of information about an individual’s daily life, including sensitive location information. Like wireless providers, Facebook, Instagram, and Snapchat are all able to collect individuals’ locations from Bluetooth signals, wireless networks, and cell towers. Additionally, these platforms also store information such as the location, date, and time at which the photograph or file was created. This information could be used in a criminal investigation to pinpoint the time and place where a crime occurred or where a suspect was located at a particular time, making it highly valuable for the government when charging someone with a crime. Thus, it is important to afford this information the highest level of protection.

Because social media is a newer phenomenon, most courts have yet to address the issue of obtaining non-content data, which can include time and location information from a social media platform. In In re Application of the United States of America for an Order Pursuant to 18 U.S.C. § 2703(d), a magistrate judge ordered Twitter to turn over information
pertaining to multiple subscribers; this information included “records
of user activity . . . including the date [and] time” as well as
“non-content information associated with the contents of any communication . . . [including] IP addresses.” The Virginia district court held that because § 2703(d) requires the government to show only “reasonable grounds” that the records sought are relevant and material to an ongoing criminal investigation, and because the third-party doctrine applies to IP address information, the court order was valid. The court differentiated IP addresses from beeper monitoring because IP addresses are shared with all internet routers when a user accesses Twitter, while tracking a beeper allowed the government to monitor inside a private residence, which was not otherwise open for visual surveillance. While this case clarified what one district court believed the SCA means for IP addresses, it does not help to clarify how the SCA applies to exact location information such as the metadata embedded in Facebook, Instagram, and Snapchat posts.

However, courts have addressed the issue of whether obtaining location information from a wireless carrier constitutes a search under the Fourth Amendment. In Carpenter, the Court held that a court order obtained under § 2703(d) was not a permissible means of acquiring a defendant’s historical cell-site location information (“CSLI”) from a wireless carrier. The Court found that individuals have a reasonable expectation of privacy in their physical location, and when the government accessed CSLI from the wireless carriers, it violated the defendant’s reasonable expectation of privacy. As a result, the Court held that the government “must generally obtain a warrant supported by probable cause” before acquiring records containing location information.

Because the SCA was intended to extend Fourth Amendment rights to online communications, it might be acceptable to infer that obtaining location information from social media platforms would also require obtaining a warrant supported by probable cause. However, the Carpenter Court articulated that its decision was “narrow” and that it does not “address other business records that might incidentally reveal location information,” which means that the metadata contained in the photos and videos posted on social media may not require the government to obtain a warrant, which could compromise people’s privacy rights. As Justice Sotomayor pointed out in her concurrence in Jones, “it may be necessary to reconsider the premise that an individual has no reasonable expectation of privacy” in the information they disclose online. “This approach is ill suited to the digital age, in which people reveal a great deal of information about themselves to third parties in the course of carrying out mundane tasks.” Justice Sotomayor is right: in the digital age, individuals post a wealth of information online that they expect—as a result of their privacy settings—to be visible only to those they choose. Thus, it is time to reconsider the notion that revealing this information to third-party social media platforms means that the government should be able to easily obtain their locational information because there is no “reasonable expectation of privacy.”

2. Challenges in Applying SCA Non-Content Data Disclosure to Stories

Stories provide users with the unique opportunity to create information that can qualify as both content and non-content data at the same time. When an individual posts their Story online, they are able to add “stickers,” which can indicate to those viewing the Story the exact location of the individual and the date and time the Story was posted, among other things. Thus, when a user posts a location in their social media Story, it actually appears as part of a graphic. In this sense, it would appear to be content because it is part of the image. On the other hand, since it is a location, Instagram will likely also collect that information separately from the content. It would then appear that, in this situation, the location information would be both content and non-content data at the same time; how then should a court determine whether a subpoena, court order, or warrant is required to compel the information from Instagram? Unfortunately, this is unclear under the current statutory framework of the SCA.

Former CIA agent Michael Morell admits that “[t]here’s a lot of content in metadata” and that “[t]here’s not a sharp difference between metadata and content . . . It’s more of a continuum.” If even the government accepts that it is difficult to distinguish between content and non-content data, then the SCA should not be differentiating between the two and allowing weaker protections for non-content data when, in fact, it may reveal information just as sensitive as content. Because the SCA was created prior to the creation of social media, it does not account for the overlap in the types of information that can be obtained from non-content and content data. This is another reason why the SCA needs to be rewritten: to clarify and remove the ambiguity of how sensitive non-content information can be disclosed.

V. REVISING THE STORED COMMUNICATIONS ACT

A. Requiring Warrants for All Compelled Content Disclosures

While the SCA provides some protections for private communications on ISPs, the statute needs to be updated and better tailored so that it is applicable to all the various nuances of modern technologies. Currently, the strongest protections are afforded to unretrieved emails and other temporarily stored files that are 180 days old or less. All other communications can be more easily obtained with a subpoena combined with prior notice. Under the Federal Rules of Criminal Procedure, a subpoena “may order the witness to produce any books, papers, documents, data, or other objects the subpoena designates.” This is even less protective of an individual’s right to privacy than having to obtain a court order, which requires that the “governmental entity offers specific and articulable facts showing that there are reasonable grounds to believe that the contents of a[n] . . . electronic communication . . . are relevant and material to an ongoing criminal investigation.” To obtain a warrant, on the other hand, there must be “probable cause to search for and seize a person or property.” This places a heavier burden on the government and thus ensures that social media users are not losing their right to privacy without stringent protections, which should be the goal of any such legislation.

Because the line between defining a social media platform as either an ECS provider or an RCS provider is so unclear, applying existing laws can lead to variable results that negatively impact users’ privacy rights. As previously discussed, under the SCA, the same ISP can be treated as an ECS for some functions, but an RCS for others; this leaves users with inconsistencies in the treatment of their personal communications, which can infringe on their privacy. Importantly, whether a social media platform is characterized as an ECS or an RCS has a direct impact on the stringency of the procedures that law enforcement must follow to obtain the content. Further, although the SCA does not specifically differentiate between public and private social media accounts, because the SCA was only intended to cover private communications, it inadvertently creates counterintuitive privacy protections. For example, in Crispin, the court held that opened private messages on Facebook and MySpace were covered by RCS rules, while ECS rules covered restricted wall posts and comments. Effectively, this meant that wall posts and comments, which can arguably be seen by all of an individual user’s friends, were afforded greater protections than private messages, which are typically only seen by the sender and the intended recipient. This is counterintuitive because it means that less private communications receive greater protection than more private communications.

Consequently, there is a clear need for Congress to reform the SCA now, and as a first step, require warrants for all communications, regardless of whether an ISP is characterized as an RCS or ECS. Warrants provide the strongest protection for social media users, and when it comes to individual liberties, the government has an obligation to preserve these liberties with the broadest legal protections possible. This is especially important considering the case law, which argues that individuals have a right to be protected under the SCA if they took steps to protect their content. By requiring warrants for the disclosure of all social media communications, the SCA would be able to provide the strongest statutory framework to protect users’ privacy and prevent the unjust use of their social media content against them in criminal court.

B. Removing the Differentiation Between RCS and ECS

The previously highlighted variability and liability in characterizing social media platforms as RCS providers in some instances and ECS providers in others has become even more problematic with the recent emergence of social media Stories. If Stories are analogized to emails or private messages—because the user posts the Story with the intention that others will see it and it will be gone shortly after the message is read—they would be governed by ECS rules, similar to the private messages in Crispin. Alternatively, Stories considered analogous to YouTube videos—because they are stored for only a limited number of people to view—would be governed by RCS rules. The courts have yet to address whether Stories should be governed by ECS or RCS rules, but there are arguments for both sides because Stories do not fit neatly into either category.

Because the SCA was not created to accommodate these newer technologies, it would be more effective to revise the SCA categories rather than attempting to fit new technologies into the existing categories. Because social media platforms offer various functions that involve both message transmissions and electronic storage, the language of the SCA needs to be amended to eliminate the distinction between RCS and ECS altogether. Orin Kerr suggested doing this by identifying that the SCA applies only to “network service providers,” which would encapsulate the current definitions of ECS and RCS and then apply the SCA rules to different types of files held by the network service providers. This would alleviate the difficulty of determining which rules apply to social media providers in different situations and would further clarify privacy rights for users by establishing when and how their content is protected. Importantly, this would also provide consistency and give users a better understanding of their rights online, which may, in turn, influence what information they choose to post on social media—especially if they know it could later be used against them in a criminal case. Without this clarity, social media users do not know whether their content is protected and what steps they need to take to protect their private communications, which may, consequently, have a “chilling effect” on their conduct.

C. Requiring Warrants for All Compelled Non-Content Data Disclosures

As technology has grown and evolved, the distinction between content and non-content data has continued to blur. This is particularly true when individuals include the date, time, and location of their posts in the actual post or Story. When Facebook, Instagram, and Snapchat collect that information, it becomes non-content data, some of which can be disclosed pursuant to only a subpoena, and some of which requires either a court order or a warrant. One way to address this issue would be to require warrants for all compelled disclosures of non-content data. This is in line with the suggestion to require warrants for all compelled disclosures of content.

By requiring warrants for compelled disclosures of non-content data, criminal investigators would then have to show probable cause before obtaining the information, which is the highest standard available. In Carpenter, the Court acknowledged that individuals have a reasonable expectation of privacy regarding their physical location. Unlike cell-site records, social media platforms do not collect information on users every time their phone pings a cell tower. Instead, locations are collected when individuals post to social media. Therefore, it is currently unclear whether location information would always be protected by a warrant under the SCA.

While it is true that some non-content data records reveal more than others, advances in metadata analysis have shown that assembling disparate pieces of metadata can lead to larger discoveries. Thus, although one might argue that it would be better to specify which types of records require a subpoena, which require a court order, and which require a warrant, this practice would be difficult to consistently implement. Rewriting the SCA to guarantee that such non-content metadata is protected by the highest protection affordable would ensure that social media users are provided their First Amendment rights.

D. Removing the Distinction Between Content and Non-Content Data

Perhaps a simpler solution to this problem of differentiating between content and non-content data would be to eliminate the distinction altogether. The distinction comes from Ex parte Jackson, in which the Court held that “a distinction is to be made between different kinds of mail matter,—between what is intended to be kept free from inspection, such as letters . . . and what is open to inspection, such as . . . printed matter, purposely left in a condition to be examined.” The Court held that mail can only be opened and examined under a warrant because otherwise it would constitute an illegal search. Thus, content is what is “intended to be kept free from inspection,” as it is sealed away, and non-content data is what is left in the open.

When the Court first created this distinction in Ex parte Jackson, it made sense to differentiate between the information on the outside of an envelope, which could be openly seen by others, and the content that was stored within an envelope. However, trying to apply that logic to social media now no longer makes sense because the distinction between content and non-content data has become so blurred. For example, when a user posts a picture of their dog on their Instagram profile, they can include a geotagged location to where the photograph was taken. Is the location still non-content data because it is not the “substance” of the post, or is the location content because the user is using it to indicate where the picture was taken and, therefore, it is part of the description? If the latter were true, it would then arguably be content.

If the same information can be considered both content and non-content, it does not make sense to allow law enforcement to obtain the same information with lesser protections solely because they can argue that it is non-content data. Eliminating the distinction between non-content and content data would remove the uncertainty and enable social media users to be confident that all aspects of their posts would be protected.

CONCLUSION

The Ninth Circuit had it right when it said, “until Congress brings the laws in line with modern technology, protection of the Internet and websites such as [social media platforms] will remain a confusing and uncertain area of the law.” Social media platforms, as a whole, do not fit nicely into the existing ECS and RCS categories that Congress created when drafting the SCA in 1986. Some functions of social media platforms lead to the platform being treated as an ECS, while other functions lead to the platform being treated as an RCS. In other instances, it is difficult to determine whether a specific function indicates that the social media platform is acting as an ECS or an RCS. As a result, the SCA can be inconsistently applied to disclosures of social media content. Most importantly, certain functions on social media are arbitrarily afforded stricter protections than others, solely because of how they are inconsistently categorized under the current SCA. The rationale for affording communications greater protections when they are classified as an ECS that is 180 days old or less versus the fewer protections afforded to an ECS that is more than 180 days old or as an RCS is unclear. As a result of these arbitrary distinctions, law enforcement has an easier time searching an individual’s private social media, which may only require a subpoena or court order, than it would going through someone’s diary, which requires a warrant.

Further complicating the application of the SCA to social media today is the fact that in the age of social media, it is becoming more difficult to distinguish content from non-content data. When Congress drafted the SCA, it attempted to apply the Fourth Amendment to online communications and therefore made a distinction between content and non-content data; however, the difference between what constitutes content—analogous to what is contained inside an envelope—and non-content—analogous to what is on the outside of an envelope—in the digital context has become difficult to discern. Courts have also considered the third-party doctrine when determining what information could be obtained with a subpoena, reasoning that because the information had been disclosed to a third party, the user had no reasonable expectation of privacy. However, social media users disclose a variety of personal information when signing up for an account, often including, at a minimum, their name, birthdate, and email address, and their posts include lots of additional metadata. The privacy of these data is critical to define because they can be used by law enforcement to piece together where an individual was at the time they posted to social media or where an individual was when the content they posted was retrieved. Whether this very sensitive information should require a warrant or a lesser means to be retrieved by law enforcement is not currently clearly defined in the SCA.

The ECPA—which includes the SCA—was enacted to protect citizens from having their electronic communications intercepted without the proper authorization, but these protections need to change in response to evolving communication technologies. This legislation was intended to extend Fourth Amendment protections to new technologies, but because social media technologies have evolved so rapidly since 1986, the SCA no longer truly affords the intended protections. For citizens to be protected against unreasonable searches of their digital media, Congress needs to restructure the existing legislation to properly address how communication technologies have evolved over the past thirty-six years. Not only can one social media platform function as both an ECS and an RCS provider under the current SCA definitions, but it is now also difficult to determine whether a specific social media function, such as Stories, which has properties of both, should be governed by ECS or RCS rules. Further, there is now duplication of content and non-content data, making it difficult to clearly differentiate them and ensure that all of this personal information is being adequately protected under the SCA.

To ensure the protection of constitutional privacy rights and prevent private social media communications from being unfairly used against their creators in court, Congress should require that all compelled disclosures be governed by the same rules as the Fourth Amendment; that is, it should require that there be a warrant and “probable cause.” If all compelled disclosures were to require a warrant, then equal protections would be applied in all situations, as the standard would be consistent across physical and digital searches; this would help ensure that defendants’ due process rights were not violated. Further, because the distinctions between an ECS and RCS, as well as content and non-content data, are no longer appropriate, it would be advantageous for Congress to revise the SCA to better align with modern technologies by drawing the necessary delineations based on the functions being used, not on the specific type of provider. This way, the SCA would not only better apply to modern technology, but it would hopefully also better apply to future emerging technologies.

 

96 S. Cal. L. Rev. 707

Download

Executive Senior Editor, Southern California Law Review, Volume 96; J.D. Candidate 2023, University of Southern California Gould School of Law; M.S. Clinical Research Methods 2020, Fordham University; B.A. Psychology 2015, New York University. My thanks to my parents, Marlene and Lee Allen, and Jennifer Guillen for their input and support throughout the note-writing process. I would also like to thank my Note advisor, Professor Eileen Decker, for her guidance, and the members of the Southern California Law Review for their hard work and thoughtful suggestions.

Battle of the Opinions: Conflicting Interpretations of False Opinions and the Falsity Standard Under the False Claims Act

Congress has let loose a posse of ad hoc deputies to uncover and prosecute frauds against the government . . . . [Bad actors] may prefer the dignity of being chased only by the regular troops; if so, they must seek relief from Congress.

INTRODUCTION

What most people probably do not realize is that approximately ten percent of all government spending is lost to fraud, which amounts to hundreds of billions of dollars annually. It should be of no surprise then that public attitudes toward government spending are mixed. With the recent COVID-19 pandemic, government spending and the number of fraudulent schemes have both reached unprecedented levels. This alone is quite alarming from a policy perspective. Furthermore, in combatting this widespread fraud, the government has had to consider an important legal issue, which also happens to be a philosophical concern that permeates life and introduces uncertainty into the legal system.

The distinction between fact and opinion seems quite obvious, but there is more to this dichotomy than meets the eye. Most individuals intuitively understand that facts have an objective basis in reality whereas opinions are merely one’s own subjective interpretation of some matter. It follows that facts can be proven or disproven using an objective metric and that facts can reinforce or contradict any given claim. But what about opinions? Can they be “true” or “false” in the same sense? Can the substance of their truth be invalidated by other opinions? Do opinions gain an elevated legal status if they inevitably result in life-or-death consequences for another individual?

The circuit courts have recently grappled with these difficult questions in the context of Medicare-related claims under the False Claims Act (“FCA”), a civil anti-fraud statute. To prevail on an FCA claim, plaintiffs must prove, inter alia, falsity; that is, the defendant made a false claim for government payment. The FCA, in its current iteration, does not provide guidance on the standard for proving falsity. Normally, this would not present an issue because “absent other indication, ‘Congress intends to incorporate the well-settled meaning of the common-law terms it uses.’ ” However, claims for government payment or reimbursement are sometimes based only on a subject matter expert’s evaluation. This is particularly true in the medical field, where doctors are required to treat patients using their clinical judgments. Thus, proving falsity in these cases necessarily entails disproving expert opinion. Given the subjective nature of opinions, common-law developments have not been uniform, and circuit courts have entrenched themselves on different sides of the aisle.

On one side are circuit courts that believe that the FCA requires proof of an “objective falsehood.” This seems to be the traditional interpretation, with many courts at the district and appellate levels dismissing plaintiffs’ claims when they failed to establish that a defendant’s representation was objectively false. Most recently, the Eleventh Circuit, in United States v. AseraCare, Inc., considered when the hospice provider certifications regarding a patient’s “terminally ill” status can be considered false under the FCA. In its holding, the court determined that claims cannot be false based on “a reasonable disagreement between medical experts.”

Approximately six months after the Eleventh Circuit’s ruling, the Third Circuit, in United States v. Care Alternatives, explicitly rejected the objective falsity standard in favor of a subjective falsity standard, whereby expert testimony challenging a physician’s judgment can be adequate evidence of falsity. The Ninth Circuit seemingly followed suit in Winter ex rel. United States v. Gardens Regional Hospital & Medical Center, Inc. when it proclaimed that a party stating an FCA claim does not need to plead an objective falsehood. The defendants in both cases petitioned the Supreme Court for writs of certiorari; unfortunately, on February 22, 2021, the Court rejected the petitions without comment, leaving the question unaddressed and prolonging the circuit split.

This Note explores the aforementioned circuit split and scrutinizes the decisions under various frameworks given the statutory gap regarding falsity under the FCA. In doing so, it will consider relevant common law guidance and regulations and focus on the courts’ adherence to precedent and principles. Few doctrinal analyses on the falsity element of the FCA have been conducted, and to my knowledge, this is the one of the first to propose that (1) the recent disagreement over objective falsity is a nontraditional three-way circuit split, and (2) the falsity standard needs to be flexible to accommodate various controlling regulations and statutes. This Note then argues that the Ninth Circuit has correctly elucidated the issue: courts should not focus on the objective or subjective falsehood standard but rather on the context and circumstances of each case.

Part I of this Note provides a foundational understanding of the FCA, the healthcare industry, and falsity in common law contexts. This includes the FCA’s legislative history, qui tam claims, statistics regarding recovery, medical decision-making, Medicare hospice benefit (“MHB”), and history of objective falsity cases. Part II discusses prior Supreme Court and appellate decisions that provide a useful framework to analyze the circuit split. Part III analyzes the three central cases that have contributed to the recent circuit split: United States v. AseraCare, Inc., United States v. Care Alternatives, and Winter ex rel. United States v. Gardens Regional Hospital and Medical Center, Inc. Part IV recommends that courts analyze falsity under the Tenth Circuit and Supreme Court’s common law test defined in United States ex rel. Polukoff v. St. Mark’s Hospital and Omnicare, Inc. v. Laborers District Council Construction Industry Pension Fund. Part IV also argues that FCA-intersecting statutes and regulations have impliedly allowed for both objective and subjective falsity standards to exist. Furthermore, Part IV suggests that the issue may be more efficiently addressed by the legislature than the courts and contextualizes the problem within the broader whistleblower policy debate.

I. BACKGROUND

A. The False Claims Act and Its Legislative History

Originally enacted in 1863 at the request of President Abraham Lincoln, the FCA is America’s first whistleblower law and currently one of the strongest whistleblower laws in the United States. The FCA allowed the federal government to combat widespread fraud committed by defense contractors against the Union Army during the American Civil War. In a congressional session statement, Senator Jacob Howard noted that “shells for the use of the Army . . . have been filled not with the proper explosive materials . . . but with saw dust” and that “[a]rms have been supplied which, on examination and use, have turned out to be useless and valueless.” The original Act contained criminal and civil penalties for wrongdoers. But the crucial feature of the Act that allows for its effective implementation is the qui tam provision, which enables private citizens to bring suits on behalf of the government; this essentially “empower[s] . . . ordinary citizens to act as private attorneys general.” Claimants in these qui tam actions, known as the “relators,” are incentivized by the fact that they receive a portion of the recovered damages. Relator is the term found in the FCA statute because the term whistleblower was not in use at the time of statutory enactment. Although the two terms are synonymous, courts and parties often prefer to use the term relator. Congress believed that it was necessary to “set[] a rogue to catch a rogue” due to the resource constraints that the government would have faced if it investigated and inquired into every business dealing involving its contractors. Senator Howard declared that this provision was “the safest and most expeditious way I have ever discovered of bringing rouges to justice.” Those convicted under the original version of the statute were liable for double the government’s damages in addition to a $2,000 penalty for each false claim. Relators would have received fifty percent of the total damages.

Nonetheless, since its inception, the FCA has been amended by Congress several times. Given that the Act was made for the purposes of deterring fraudulent profiteers of war while rewarding those who were upstanding, it was only fitting that the statute would be abused and tested during a subsequent major conflict, World War II. Then Attorney General Francis Biddle pursued criminal action against a host of defense contractors using the criminal provision of the FCA. Concurrently, groups of petitioners filed civil complaints against the same contractors and undoubtedly attempted to piggyback off the government’s work in the hopes of gaining a piece of the settlement. This parasitic exploitation of the Act did not go unnoticed, and Congress amended the FCA in 1943. The amendment reduced the relator’s guarantee of fifty percent of recovered damages to a maximum of ten percent. The recovery limit for relators was also capped at twenty-five percent in cases in which the United States did not join. Most importantly, Congress removed relators’ ability to file suits if “the United States, or any agency, officer or employee thereof” possessed evidence or information of the fraud. This alteration single-handedly eliminated the majority of qui tam FCA cases.

Approximately forty years later, Congress caught wind of reports of rampant fraud committed by federal contractors. In 1986, the FCA experienced almost a complete reversal of the strict prohibitions which chilled qui tam cases. The “any prior government knowledge” proscription was replaced with the substantially less restrictive “public disclosure of allegations or transactions” qualification. In addition, recovery for successful relators increased marginally, and liability for perpetrators of fraud increased from double damages to treble damages.

The most recent iteration of the FCA occurred in 2009, when Congress made a somewhat subtle amendment to the statute which limited the scope of claims encompassed by the FCA. A “material to a false or fraudulent claim” element was added. In essence, the wording of the prior FCA iteration allowed one of the critical elements to be met if the government simply paid or approved a fraudulent claim. The new requirement, however, adds a materiality aspect; that is, the government’s decision to pay or approve a claim must have been predicated on a falsity.

This current version of the FCA specifically penalizes, among other offenses, (1) knowingly presenting, or causing to be presented, a false or fraudulent claim for payment, and (2) knowingly making, using, or causing to be made or used, a false record or statement material to a false or fraudulent claim. FCA claims are broken down into the following requirements: falsity, causation, knowledge, and materiality. The statute provides functional definitions for knowledge but offers no guidance on the definitions of falsity.

B. The False Claims Act in the Twenty-First Century

The importance of the FCA in combatting fraud in the twenty-first century should not be underestimated. Approximately ten percent of all government spending is lost to fraud. During fiscal year 2020, the government spent over $6 trillion dollars. Accordingly, the government stands to lose hundreds of millions of dollars each year. Like cases in other areas of law, the majority of FCA cases settle or are dismissed before trial. Nonetheless, the number of FCA cases and associated monetary payments have substantially amplified in recent years. More than 4,000 new cases have opened since 2015. In 2020 alone, qui tam relators and the government filed 922 new FCA suits and subsequently obtained more than $2 billion dollars in recovery and settlements. With the onset of the COVID-19 pandemic, the Department of Justice has already begun investigating and prosecuting the spike in COVID-19 recovery-related programs. Fraud cases are more prevalent now than ever, and the FCA creates a necessary foundation with which to combat these issues.

Although historically used to uncover and deter military-based fraud against the federal government, the FCA in the current era has undergone a drastic shift, not based on the substance of law but rather due to policy shifts in healthcare law. The rapid expansion of the healthcare sector and burgeoning government programs are likely responsible for this shift. Over eighty percent of fraud cases against the government are now related to healthcare. Furthermore, healthcare-related FCA cases account for more recovery than FCA recovery from all other sectors combined.

C. Medical Decision-Making and Medicare Hospice Benefits

Given that all three cases contributing to the circuit split concern Medicare-related fraud, a general discussion of fraud within the medical practice area is warranted. Fraud in the medical industry is not novel. In particular, the Federal Bureau of Investigation has noted that health care fraud causes several billions of dollars in losses each year. Although there are a variety of factors that contribute to the prevalence of health care fraud, the subjectivity inherent in medical decision-making is a prominent one. There will almost always be another medical professional who does not agree with the course of action taken. Moreover, the medical industry is unique in that the medical opinions of physicians sometimes lack the objective proof to reinforce their actions and regulations often give deference to medical judgments.

The MHB presents a fitting example of a controlling statutory restriction that specifically grants physicians this deference. Due to the growing number of aging individuals enrolled in Medicare, Congress passed the MHB in 1983. The MHB allows Medicare beneficiaries to forego traditional curative care in favor of electing interdisciplinary palliative treatment. However, eligibility is based on a written confirmation of a “terminally ill” prognosis by a physician or medical director. Terminally ill is defined as “a life expectancy of 6 months or less if the terminal illness runs its normal course.” This certification must include a written narrative explaining the clinical findings and be accompanied by clinical information and other documentation. Once these conditions are met, Medicare and Medicaid programs will provide payment to hospice providers for costs incurred under the Social Security Act. The regulations have directly acknowledged the “uniqueness of every Medicare beneficiary” and that predicting someone’s end-of-life is not an “exact science.” Accordingly, certifications may be renewed by the physician for additional sixty- or ninety-day periods.

Following the MHB’s establishment, FCA cases alleging hospice fraud have increased dramatically. This includes two of the three circuit split cases. Predictably, most cases are initiated by whistleblowers in qui tam suits, as foreseen by the legislature. In 2016, the MHB provided hospice care to more than one million individuals, and Medicare reimbursed over $16 billion for hospice care. Moreover, unlike FCA litigation in other areas of law, FCA litigation in connection with the MHB demonstrates a unique scenario that has perplexed the courts: stratification of the FCA by a purposefully deferential statute.

D. An Undisputed Era of Objective Falsity

Objective falsity was widely considered to be the standard before the new Third and Ninth Circuit holdings challenged the status quo; a considerable number of courts, including the Third Circuit itself, have previously recognized this standard. This ostensibly established standard derived from a mix of healthcare- and non-healthcare-related FCA claims, which likely solidified its acceptance and promulgated its spread across jurisdictions. Some of these cases were decided as early as 2005 and are briefly explained below to illustrate the formerly unified landscape which has been shattered by the circuit split.

In United States v. Prabhu, the District of Nevada held:

To establish falsity under the FCA, it is not sufficient to demonstrate that the person’s practices could have or should have been better. Instead, plaintiff must demonstrate that an objective gap exists between what the defendant represented and what the Defendant would have stated had the Defendant told the truth.

The government alleged that the physician’s claims for pulmonary rehabilitation and simple pulmonary stress tests were false due to insufficient documentation. The government interpreted the American Medical Association’s guidance publication to require specific measurements and a written report for a simple stress test. However, the record indicated that Medicare failed to issue specific guidance regarding the precise type of documentation needed to provide care and that there was no physician writing documentation requirement. In light of these facts, the parties’ contentions, and the “general confusion” among the government and its own experts, the court believed that “reasonable persons can disagree regarding the billing requirement[]” and the physician’s documentation practices fell within “the range of reasonable medical and scientific judgment.” Furthermore, the government did not establish a concrete violation of a “controlling rule, regulation, or standard” when the physician provided pulmonary rehabilitation services. As a matter of law, the government failed to establish falsity, and the court granted the motion for summary judgment.

In United States ex rel. Wilson v. Kellogg Brown & Root, Inc., the Fourth Circuit determined that “[an] FCA relator cannot base a fraud claim on nothing more than his own interpretation of an imprecise contractual provision.” The relators claimed that the defendant contractor, their former employer, falsely certified that it would uphold its contractual duties by maintaining military vehicles in “good appearance” when “it would not, and later did not, abide by those terms.” The court outright rejected this assertion because “[i]t is well-established that the FCA requires proof of an objective falsehood.” The court also found no evidence of this claim, as the United States government—the actual party to the contract—never expressed dissatisfaction with the contractor’s performance. Relying solely on their interpretation of imprecise maintenance provisions, the relators failed to state a valid falsity claim under the FCA.

In United States ex rel. Yannacopoulos v. General Dynamics, the Seventh Circuit decided that “[a] statement may be deemed ‘false’ for purposes of the False Claims Act only if the statement represents ‘an objective falsehood.’ ” The relator contended that amendments to a contract between a company and Greece were “reverse false claims,” false statements used to conceal, avoid, or decrease an obligation to pay or transmit money or property to the government. However, the relator simply relied on his interpretation of the terms of agreement without proof of any evidence. As a result, the court affirmed the district court’s motion for summary judgment.

In United States ex rel. Wall v. Vista Hospice Care, Inc., the Northern District of Texas ruled that “[a] testifying physician’s disagreement with a certifying physician’s prediction of life expectancy is not enough to show falsity.” The relator asserted, inter alia, that defendant hospice service providers improperly enrolled and sought reimbursement from Medicare and Medicaid for patients who were not eligible for hospice care. Although the relator presented a medical expert’s testimony that ninety percent of the records were ineligible for certification, it was not sufficiently linked to the corporate scheme to falsify records and thus did not create a triable “fact issue as to falsity.”

II. BUILDING AN ANALYTICAL FRAMEWORK

Important cases have discussed how opinions relate to the FCA, when opinions may be considered false in the context of medical necessity, and the two theories of falsity. The totality of these cases provides an analytical framework with which to analyze the circuit split and are discussed below:

 A. When Opinions Can Be False

As a prelude to the circuit split, the Supreme Court in Omnicare, Inc. v. Laborers District Council Construction Industry Pension Fund addressed the issue of when opinions can be false. The case involved Omnicare, the largest pharmacy services provider for nursing home residents in the United States, and its filed registration statement with the Securities and Exchange Commission (“SEC”). The filing included two statements of opinion. First, the company believed that its “contract[ual] arrangements with other healthcare providers, . . . pharmaceutical suppliers and . . . pharmacy practices [were] in compliance with applicable federal and state laws.” Second, the company believed that its “contracts with pharmaceutical manufacturers [were] legally and economically valid arrangements that [brought] value to the healthcare system and the patients.” The plaintiffs, pension funds that purchased Omnicare stock, alleged that the company’s statements were materially false based on later lawsuit filings from the government stating that the company received payments from drug manufacturers in violation of anti-kickback laws. In addition to claims of materially false representations regarding legal compliance, the complaint maintained that none of the company’s officers and directors possessed reasonable ground to believe that the opinions offered were truthful and complete. In support of this, plaintiffs pointed to the fact that one of Omnicare’s attorneys previously warned of a contract that carried a heightened risk of liability under anti-kickback laws.

The district court granted Omnicare’s motion to dismiss on the grounds that the statements about a company’s belief regarding its legal compliance are only actionable if those who made the statements knew they were untrue at the time. The court thus concluded that the plaintiffs’ complaint failed to meet the standard because there were no allegations stating that Omnicare’s officers knew they were violating the law.

On appeal, the Sixth Circuit reversed the district court’s holding. The court acknowledged that the opinions related to legal compliance, rather than “hard facts.” Nonetheless, the court proceeded to explain that the plaintiffs simply had to allege that the opinion was objectively false and were not required to contend that an Omnicare employee “disbelieved [the opinion] at the time it was expressed.”

After granting certiorari, the Supreme Court addressed the following two issues: (1) when an opinion may constitute a factual misstatement; and (2) when an opinion may be considered misleading by the omission of discrete factual representations. On the first issue, the Court held that sincere statements of pure opinion are not “ ‘untrue statement[s] of material fact,’ regardless [of] whether an investor can ultimately prove the belief [was] wrong.” Relying on common law principles, the Court illustrated two examples that provided exceptions to when statements of pure opinion can be false. These exceptions include when (1) the speaker does not actually hold the opinion, or (2) the opinion contains a false, embedded fact. On the second issue, the Court ruled that opinions may be misleading when a registration statement omits material facts about the issuer’s inquiry into or knowledge concerning a statement of opinion and if those facts conflict with what a reasonable investor would take from the statement itself. The Court does note, however, that an opinion is not necessarily misleading if it omits facts that “cut[] the other way” and analyses into this inquiry should always consider context.

Thus, the Supreme Court effectively recognized that individuals make false opinions when (1) they do not actually hold the opinion; (2) the opinion contains a false, embedded fact; (3) they are aware of facts that would preclude such an opinion; or (4) they are not aware of any facts that would justify the opinion.

B. Opinions Analysis in the Context of Medical Necessity

In United States v. Paulus, the Sixth Circuit conducted an Omnicare-based analysis in the context of a medical case without explicitly referencing the case. In Paulus, a cardiologist was criminally prosecuted for health care fraud and false statements. Specifically, the cardiologist exaggerated the extent of arterial blockages in his patients in order to perform and bill for medically unnecessary coronary stenting procedures. The crux of this case depended on the interpretation of angiograms, with the plaintiff using the testimony of nine doctors to testify that the level of blockage differed from what the defendant had reported. The defendant responded by pointing out the subjectivity of angiogram interpretation, including data from multiple studies.

During trial at the district court level, the jury convicted the cardiologist of healthcare fraud and making false statements. However, the court directed a judgment of acquittal and subsequently ordered a new trial. The court reasoned that the degree of arterial blockage was a matter of “subjective medical opinion,” and thus the cardiologist’s angiogram interpretations “could be neither false nor fraudulent.”

On appeal, the Sixth Circuit reversed because it believed that clinical judgments can trigger FCA liability when an individual (1) asserts an opinion they do not truly believe, or (2) has knowledge of facts that contradict their opinion. The court reasoned that “[t]he degree of stenosis is a fact capable of proof or disproof.” The court then likened the deliberate inflation of blockages on an angiogram to the telling of a lie, which infers the commission of a fraud when paired with the billing of a more expensive procedure. In its analysis, the court essentially utilized the first two false opinion definitions described in Omnicare: (1) not honestly holding an opinion, and (2) an opinion containing a false, embedded fact. The court thought it to be clear that angiograms are facts and implied that angiogram interpretations are obviously not facts “capable of confirmation or contradiction.” Accordingly, the court pivoted to the idea that the cardiologist did not give an opinion but instead misrepresented facts by lying about the results. Thus, the court reversed the trial court’s judgment and reinstated the jury’s verdict.

C. Factual Versus Legal Falsity

In United States ex rel. Polukoff v. St. Mark’s Hospital, the Tenth Circuit identified and distinguished between two types of falsities, factual and legal falsity, prior to conducting a falsity analysis under the FCA. In this case, a relator, the former co-worker of the defendant, sued the defendant-cardiologist as well as two hospitals under the FCA. The complaint alleged that the cardiologist performed thousands of medically unnecessary cardiac surgical procedures and fraudulently certified otherwise to receive reimbursement under the Medicare Act. Central to this claim was the Centers for Medicare and Medicaid Services’ (“CMS”) “reasonable and necessary” requirement for surgeries. Industry guidelines indicated when performing surgeries would be appropriate for specific types of patients, which the cardiologist allegedly ignored. Instead, he misrepresented on the certifications that he had performed them in accordance with the guidelines. Thus, this representation was false under the FCA.

The district court granted the defendants’ motion to dismiss. The court reasoned that “Medicare does not require compliance with an industry standard as a prerequisite to payment. Thus, requesting payment . . . does not amount to a ‘fraudulent scheme.’ ” Moreover, “because [o]pinions, medical judgments, and conclusions about which reasonable minds may differ cannot be false for the purposes of an FCA claim,” the relator failed to state a claim under the FCA.

On appeal, the Tenth Circuit reversed and remanded because it fundamentally disagreed with the district court’s narrow interpretation of the FCA’s reach. The court read the FCA broadly so as to encompass “claims for medically unnecessary treatment.” Another reason the court presented was “that an allegedly false statement constitut[ing] the speaker’s opinion does not disqualify it from forming the basis of FCA liability.” To support this reasoning, the court looked to its bifurcated understanding of falsity in a previously decided case. The court held that “false” may indicate factually false or legally false. Factually false claims are express claims that are not based in fact (for example, seeking payment for services that were never provided or submitting incorrect information), whereas legally false claims cover instances where an individual certifies compliance with applicable legal requirements when, in fact, the individual knew there was no compliance. Since the relator’s complaint alleged non-compliance with Medicare regulations, the court’s straightforward, logical analysis of legal falsity was as follows: (1) “[a] Medicare claim is false if it is not reimbursable;” (2) “a Medicare claim is not reimbursable if the services provided were not medically necessary;” and (3) in order for a claim to be medically necessary, “it must meet the government’s definition of ‘reasonable and necessary,’ as found in the Medicare Program Integrity Manual.” The procedures, certified by the cardiologist, did not comport with the government’s definition of the phrase, and thus the certifications were false under the FCA.

III.  ANALYSIS

While most articles have divided the circuit split issue between objective and subjective falsity, further inspection demonstrates that the circuit split is not binary. All three cases in the circuit split look to the statutory language of the FCA. The Eleventh and Third Circuit interpretations directly conflict, as they arrived at an objective and subjective falsity standard, respectively, after contemplating the same regulations surrounding the MHB. The Ninth Circuit case did not involve the MHB but instead considered the statutory language of Medicare programs and the CMS’s definition of “reasonable and necessary.” Although the Ninth Circuit fundamentally employed the same analysis as the Eleventh Circuit, it explicitly rejected the Eleventh Circuit’s objective falsity standard and implicitly adopted the subjective falsity standard. Thus, three distinct standards have emerged from the case law.

First, this Section will discuss the Eleventh Circuit’s analysis and decision in United States v. AseraCare, Inc., which establishes a higher burden of proof at the summary judgment stage for relators and the government. Second, this Section will examine the Third Circuit’s holding in United States v. Care Alternatives and why it chose to critique and explicitly depart from the Eleventh Circuit’s adoption of the objective falsity standard. Third, this Section will consider the Ninth Circuit’s more even-handed analysis in Winter ex rel. United States v. Gardens Regional Hospital & Medical Center, Inc. and why it refused to adopt a rigid falsity standard.

A. United States v. AseraCare, Inc.

In AseraCare, the government intervened in a qui tam suit filed by three former AseraCare employees against AseraCare, claiming that the hospice provider had a practice of knowingly submitting unsubstantiated Medicare claims in violation of the FCA. These reckless business practices allegedly enabled the provider “to admit, and receive reimbursement for, patients who were not eligible for [MHB],” resulting in the “misspending” of millions of Medicare dollars. The court noted this case as falling under the “false certification” theory of FCA liability (in other words, when there is a false implication of having complied with a legal requirement).

To establish its case, the government first identified over 2,000 hospice patients for whom AseraCare had billed Medicare. The government then narrowed this population to a subset of 223 patients and retained a physician to directly review these patients’ medical records and clinical histories. Acting as the government’s primary expert witness, the physician, relying on his own clinical judgment, opined that 123 out of 223 patients were ineligible for hospice benefits at the time AseraCare received reimbursements from Medicare. Critically, the government’s case was substantially weakened when its expert witness conceded that he was unable to affirmatively say whether AseraCare’s medical expert, or any other physician, was wrong about the accuracy of the prognoses at issue. Furthermore, the expert witness (1) never testified that no reasonable doctor could have concluded that the patients were terminally ill at the time of certification, and (2) changed his opinion concerning the eligibility of certain patients over the course of the proceeding.

A brief recitation of the procedural posture and history is warranted so as to provide context for the appellate court’s analysis. Following discovery and analysis of relevant patient records, AseraCare moved for summary judgment on the grounds that the government failed to adduce evidence of falsity under the FCA. In its motion, AseraCare specifically asked the district court to apply the “reasonable doctor” standard; that is, “the government must show that a reasonable physician applying his or her clinical judgment could not have held the opinion that the patient at issue was terminally ill at the time of certification.” Even though the district court found this standard convincing, it declined to apply it and denied the motion. The district court then bifurcated the trial into two phases, one on the falsity element and the second on the remaining FCA elements. This limited the government’s ability to rebut AseraCare’s expert testimony during the first phase. Nonetheless, the dueling expert testimony was a critical component of trial. The government’s expert and AseraCare’s expert diverged in how they approached analysis of patient life expectancy. The government’s expert used a “checkbox approach” to assess terminal illness by comparing patient records to medical guidelines. By contrast, AseraCare’s expert did not formulaically apply guidance and used a more “holistic” approach. At the trial’s conclusion, the district court provided the following jury instruction: “A claim is ‘false’ if it is an assertion that is untrue when made or used. Claims to Medicare may be false if the provider seeks payment, or reimbursement, for health care that is not reimbursable.” Thus, the jury had to decide which expert was more persuasive, with the less persuasive opinion being deemed a false opinion. In its answers to special interrogatories, the jury found that AseraCare had submitted false claims for 104 of the 123 patients at issue.

Following this partial verdict, AseraCare moved for judgment as a matter of law, contending that the district court articulated an incorrect legal standard in its instruction. The court agreed that it had committed reversible error in its instruction and ordered a new trial. The court believed it should have advised the jury of two “key points of law,” which were not previously acknowledged: (1) “the FCA’s falsity element requires proof of an objective falsehood”; and (2) “a mere difference of opinion between physicians, without more, is not enough to show falsity.” The court then considered summary judgment sua sponte and concluded that the government could not prove the falsity element as a matter of law because the government “presented no evidence of an objective falsehood for any of the patients at issue.” Summary judgment was granted in AseraCare’s favor, and the government appealed.

On appeal, the government’s core argument was that competing expert testimony regarding patients’ medical records supporting a terminal illness prognoses was enough to raise a factual question for the jury. In contrast, AseraCare contended that the determinative inquiry was whether the certifying physician exercised genuine clinical judgment. If so, the accuracy of such judgment cannot be false as a factual matter. The Eleventh Circuit immediately recognized that “the standard for falsity [was] in the context of the Medicare hospice benefit, where the controlling condition of reimbursement is a matter of clinical judgment.” Accordingly, the Eleventh Circuit was tasked with considering how the FCA intersects the scope of hospice eligibility requirements.

The Eleventh Circuit initially evaluated whether the falsity claim was a legal or factual falsity. The court concluded that the case concerned a legal falsity claim because “[t]here is no allegation that the hospice services AseraCare provided were not rendered as claimed.” Then, the court identified the following two “representations,” which may form the legal basis for an FCA claim: (1) the “representation by a physician to AseraCare that the patient is terminally ill in the physician’s clinical judgment”; and (2) the “representation by AseraCare to Medicare that such clinical judgment has been obtained and that the patient is therefore eligible.” The court found that the government’s allegations only referred to the first representation. The first representation, however, made it such that the government’s FCA case rested entirely on the question of when a “physician’s clinical judgment regarding a patient’s prognosis [can] be deemed ‘false.’ ”

To answer this question, the court heavily relied on applicable regulations and the text of the MHB statute due to the “dearth of controlling case law.” The court looked to the plain meaning of the entire statue and regulations instead of focusing on specific words. The general requirements were that (1) hospice providers must submit a certification claim for patients, (2) the certification must be in writing, (3) the certification must be based on clinical judgment, (4) clinical information and other documentation supporting the prognoses must accompany the certification, and (5) the reimbursement must be for “reasonable and necessary” payments for managing terminally ill patients. The court subsequently pointed out that several requirements allow for a certain degree of subjectivity. For example, submission of claims must be individually tailored to each patient’s clinical circumstances. Check boxes and standard language used for all patients are prohibited. Furthermore, the subjective and objective medical findings of each patient should be considered. The court believed that this built-in flexibility was fully intended by Congress and that Congress would have used different language if it wanted a more rigid and objective standard. Thus, the court’s role was not to establish a more objective standard against the implied language of the statute and regulations.

Although the court emphasized that the regulations intended for MHB eligibility were to simply be predicated on the procurement of a physician’s clinical judgment, the government sought to elevate the standard such that the underlying information must support, “as a factual matter,” the certification. The court disagreed with this framing of the eligibility requirements, stating that it is not consistent with the text or design of the law. The relevant regulations merely require that clinical information and other documentation supporting the medical prognosis accompany the certification and be filed in the medical record. The court therefore determined that supporting documentation does not have to, standing alone, prove the validity of a physician’s initial clinical judgment. As long as the physician’s interpretation is reasonable, certification requirements are met.

The Eleventh Circuit ultimately concurred with the district court’s holding that a mere difference of medical opinion alone is insufficient to establish falsity under the FCA; however, it also ruled that the district court had gone too far in sua sponte granting summary judgment. The court recognized that reasonable doctors may disagree on a patient’s condition and that neither one could be wrong. As a result, “[a] properly formed and sincerely held clinical judgment is not untrue even if a different physician later contends that the judgment is wrong.” To reach this conclusion, the court relied on and cited to the Supreme Court’s decision in Omnicare. Adhering to Omnicare’s general principles, the court acknowledged that opinions regarding terminal illness can be deemed objectively false in various circumstances. The court finally deferred to the legislature or CMS after the government expressed concerns that an objective falsity standard “will likely prove more challenging for an FCA plaintiff.”

B. United States v. Care Alternatives

Like the AseraCare case, Care Alternatives involved qui tam relators who were former employees of a hospice provider, Care Alternatives. The relators alleged that Care Alternatives admitted ineligible MHB patients and directed its employees to alter the patients’ certifications to reflect eligibility. During discovery, both sides produced extensive evidence, which included dueling expert opinions. The relators’ expert examined nearly fifty patient records and opined that thirty-five percent of patients’ records did not support a certification of need for hospice care. The expert went even further and testified that “any reasonable physician would have reached the conclusion he reached.” Care Alternatives’ expert disagreed and believed that a reasonable physician would have found all of the patients to be hospice-eligible.

At the district court level, Care Alternatives moved for summary judgment based on the finding that the relators could not satisfy the four elements of the FCA claim. In particular, Care Alternatives claimed that relators had not produced sufficient evidence of falsity. The court granted Care Alternatives’ motion “based solely on failure to show falsity.” To reach its conclusion, the court looked to the holding in AseraCare, finding that a “mere difference of opinion between physicians, without more, is not enough to show falsity.” The relators appealed. Thus, the question before the appellate court was whether a reimbursement claim may be considered false under the FCA simply on the basis of conflicting medical expert testimony.

In reviewing the appeal, the Third Circuit began its analysis by discussing the MHB. For the most part, the court agreed with the Eleventh Circuit’s interpretation in AseraCare of the certification requirements for Medicare reimbursement of terminally ill patients. Similar to the AseraCare court, the Third Circuit even noted that “making a prognosis is not an exact science.” However, departing from the Eleventh Circuit’s reading, the court emphasized that this “inexactitude does not negate the fact that there must be a clinical basis for a certification.”

Where the Third Circuit truly departed from the Eleventh Circuit was in its common law analysis of the terms “false” or “fraudulent” under the FCA. Due to the lack of statutory guidance on the meaning of falsity, the court identified, from its prior cases and the Tenth Circuit’s rationale in Polukoff, the following two ways in which a claim may be false: (1) “factually, when the facts contained within the claim are untrue”; and (2) “legally, when the claimant . . . falsely certifies that it has complied with a statute or regulation the compliance with which is a condition for government payment.” As applied to the case before the court, Care Alternatives allegedly made incorrect certifications, which qualified the claim under the legal falsity theory. The court reasoned that the objective falsity standard is at odds with the concept of legal falsity, which is the appropriate standard, and by adopting the prior standard, the district court limited its analysis to factual falsity. The court further held that the district court’s objective falsity standard conflated the knowledge and falsity elements of an FCA claim. Thus, by rejecting the objective falsity standard, the court sought to separate the knowledge and falsity analyses to comply with the text of the statute. Under a legal falsity standard, disagreement between experts as to a physician’s certification may be evidence of falsity under the FCA.

The Third Circuit also considered and rejected the district court’s bright-line rule that a doctor’s clinical judgment cannot be “false.” In doing so, the court acutely relied on the Paulus opinion. Underlying the district court’s decision was the premise that medical opinions are subjective and cannot be false. The Third Circuit sided with the Sixth Circuit’s emphasis on the fact that medical “opinions are not, and have never been, completely insulated from scrutiny.” The Paulus holding suggested that good faith medical opinions are not punishable but dishonest medical opinions may trigger liability for fraud. Consequently, in line with its legal falsity analysis, the court believed that whether an individual acted in good faith or misrepresented a fact, thereby committing fraud, was “exclusively” a question for the jury.

The Third Circuit then went on to explain why it chose to depart from the Eleventh Circuit’s standard. The first issue that the court highlighted was how the Eleventh Circuit framed the falsity question. The court interpreted its sibling court as having construed the clinical information and documentation requirement of the MHB in an overly narrow fashion when it concluded that the supporting documentation requirement is only designed to address the mandate that there be a medical basis for certification instead of considering “whether the clinical information and other documentation accompanying a certification of terminal illness support[s] . . . the physician’s certification.” Therefore, this limited the inquiry to whether there was sufficient evidence of “the accuracy of the physician’s clinical judgment regarding terminality,” which the court understood to exclude legal falsity and only include factual falsity. The court posited that under the legal falsity theory, conflicting medical opinion is relevant evidence of the clinical information and documentation requirements. Furthermore, the court characterized the AseraCare court as coming to the conclusion that clinical judgments cannot be untrue, which it fundamentally disagreed with based on its interpretation of common-law definitions.

Ultimately, the Third Circuit had a drastically different breakdown of the falsity issue as compared to the Eleventh Circuit because it based its entire analysis upon the distinction between what it understood to be factual and legal falsity.

C. Winter ex rel. United States v. Gardens Regional Hospital and Medical Center, Inc.

In Winter, the relator, a registered nurse and former director at Gardens Regional Hospital (“Gardens”), filed a qui tam FCA suit against her former employer. The procedural history of this case is relatively simple compared to those of the aforementioned cases. The relator alleged in a complaint that Gardens submitted Medicare claims falsely certifying that patients’ hospitalizations were medically necessary. In support of this claim, the relator pointed to her own after-the-fact review of admission records. Gardens moved to dismiss the complaint for failure to state a claim, which was subsequently granted by the district court. The district court asserted that to prevail on an FCA claim, plaintiffs must show that a defendant knowingly made an objectively false representation. Thus, a statement that implicates a doctor’s clinical judgment can never state an FCA claim because subjective medical opinions cannot be proven to be objectively false. The relator appealed.

The Ninth Circuit started its analysis by reviewing the medical necessity requirement and the FCA. Medicare reimburses providers for inpatient hospitalization only if the expenses incurred are “reasonable and necessary.” CMS administers the Medicare program and has defined a reasonable and necessary service as one that “meets, but does not exceed, the patient’s medical need, and is furnished in accordance with accepted standards of medical practice for the diagnosis or treatment of the patient’s condition.” Similar to the MHB, the Medicare program allows doctors to form their own clinical judgment based on complex medical factors. However, the language specifically provides that factors must be documented in the medical record and the regulations consider medical necessity a question of fact. Thus, a physician’s certification has no presumptive weight in determining medical necessity and must be evaluated in the context of medical evidence. The court subsequently reasoned that the relator’s allegations fall under the “false certification” theory of FCA liability. Since medical necessity is a condition of payment, every Medicare claim includes an express or implied certification of necessary treatment. Accordingly, claims for unnecessary treatment are false claims. The court stated that many other circuits, including the Tenth in Polukoff and Third in Care Alternatives, reached the same conclusion regarding the scope of FCA claims.

The Ninth Circuit then proceeded to analyze the application of opinions to the FCA by interpreting the language of the statute. The court interpreted the FCA broadly, citing congressional intent and the Supreme Court’s refusal to “accept a rigid, restrictive reading” of the FCA. Due to the lack of statutory guidance on what constitutes a false or fraudulent claim, the court looked to common-law definitions. In doing so, the court referred to treatises and a number of cases, including Paulus and Omnicare, that a subjective opinion may be fraudulent if (1) it is “not honestly held,” (2) it implies the existence of nonexistent facts, (3) the speaker knows facts that would preclude such an opinion, and (4) the speaker does not know facts that justify it. The court additionally explained that the “knowing presentation of what is known to be false” does not mean “scientifically untrue.” Although a scientifically untrue statement is false, it may not be actionable if it was not made with the requisite intent. Likewise, an opinion with no basis in fact can be fraudulent if expressed with knowledge.

The court considered and outright rejected the request from Gardens and amici curiae for the court to hold that the FCA requires plaintiffs to plead an objective falsehood. The court stated that the plain language of the FCA “does not distinguish between ‘objective’ and ‘subjective’ falsity or carve out an exception for clinical judgments and opinions.” The court further noted that policy arguments cannot supersede the “clear” statutory text and it could not engraft that requirement onto the statute. The court therefore held that the FCA does not require plaintiffs to plead an objective falsehood.

Interestingly, the court claimed that the Eleventh Circuit’s decision in AseraCare was not “directly to the contrary.” First, the court noted that the Eleventh Circuit, notwithstanding the language about objective falsehoods, did not consider all subjective statements to be incapable of falsity. Second, the court believed that the Eleventh Circuit narrowly confined the objective falsity standard to the MHB, which granted deference to physician judgment. In the court’s view, its sister circuit did not necessarily apply the standard to a physician’s certification of medical necessity by (1) explicitly distinguishing Polukoff, and (2) explaining that the less-deferential medical necessity requirement remained an important safeguard to its reading of the MHB eligibility framework.

Given that litigation was at the motion to dismiss stage, the court ruled that the relator’s complaint plausibly alleged false certifications of medical necessity. The relator (1) showed correlations between the spike in admissions and timing of the scheme; (2) presented both irregular admission trends and admission statistics; (3) alleged a specific number of false claims, each in great detail; and (4) set forth anecdotal evidence which supported both an inference of knowledge and falsity. The court also plainly dismissed Gardens’ argument and the district court’s characterization of the relator’s allegation as simply being her own competing opinion. First, according to the court, opinions can establish falsity. Second, the court believed that even if the relator’s own evaluations of the medical record were discounted, there were enough facts alleged to suffice the plausibility of fraud.

In sum, while the Ninth Circuit disagreed with the Eleventh Circuit about the objective falsehood standard, it applied the same common law rule regarding when an opinion can be false for the purposes of an FCA claim.

IV. DISCUSSION

AseraCare, Care Alternatives, and Winter highlight a growing tension between the different approaches and standards within the falsity element of the FCA. The hospice context has been the battleground between the Third and Eleventh Circuits, which have attempted to solve the issue of whether dueling expert testimonies, without more, create a triable issue of fact for the jury. Nonetheless, it is quite evident that the imposition of a rigid falsity standard lends itself to application in FCA claims which have no basis in hospice care certifications, as seen in Winter. Furthermore, how courts analyze false opinions according to laws and regulations as well as the intent behind them is of great importance because it forms the conceptual foundation for constructing a proper framework and reaching the most legally sound conclusion. The following questions naturally follow: How should courts analyze false opinions and the falsity standard? And is the objective or subjective falsity standard the more appropriate reading of the FCA statute?

Section IV.A argues that Polukoff and Omnicare provide a comprehensive framework for the courts to categorize types of FCA claims and, if the alleged conduct includes opinions, whether the opinion is false. Section IV.B argues that, given the reach of the FCA, objective and subjective falsity standards are appropriate depending on the applicable regulations. Section IV.C suggests that, as a practical matter based on policy concerns, Congress amend the FCA to create special definitions and provisions for professional medical judgment. Finally, Section IV.D addresses the competing policy trade-off of over-incentivization to file false claims and contextualizes the arguments made in this Note to the broader whistleblower policy debate.

A. The PolukoffOmnicare Common Law Test

Unlike the Care Alternatives court’s factual and legal falsity breakdown, the AseraCare and Winter courts utilized the PolukoffOmnicare common law framework to reach their conclusions; this is the proper way to analyze the falsity element of the FCA. First, the PolukoffOmnicare framework fully encompasses all types of FCA claims. The Polukoff court divides FCA claims into factual and legal claims. The Omnicare decision sets out the four different ways in which an opinion may be false: (1) the actor does not actually hold the opinion; (2) the opinion contains a false, embedded fact; (3) the actor is aware of facts that would preclude such an opinion; or (4) the actor is not aware of any facts that would justify the opinion. The first step in any FCA claim determination should be the Polukoffanalysis. Courts can properly distinguish the entire universe of FCA claims into two categories and decide where the claim before them fits. Moreover, if legal claims are not implicated, the standard automatically defaults to an objective falsehood standard. Courts should subsequently consider whether a legal, FCA claim fits into one of the four Omnicare false opinion types. Regardless of whether courts adopt an objective or subjective falsehood standard, the Omnicareframework remains pertinent because it defines the totality of false opinions. Adhering to this analytical procedure will not only ensure that common law precedent has been properly followed but also unofficially standardize the framework across circuits. As discussed above, the Eleventh and Ninth Circuits identically and correctly applied this framework.

Second, while the Third Circuit correctly relied on Paulus to identify that opinions can be false, it fully ignored when opinions can be false according to the common law; it would not have made this fatal error if it used the Omnicare framework. The Paulus court specifically stated that “opinions may trigger liability for fraud when they are not honestly held by their maker,or when the speaker knows of facts that are fundamentally incompatible with his opinion.” The “when” conjunctions in the statement are critical to understanding the common law reasoning behind false opinions. However, the Third Circuit seemingly disregarded the dependent clauses, so that it could adduce some misconstrued holding from another circuit to support its conclusion regarding subjective falsity. In a similarly reductive fashion, the Third Circuit mischaracterized the Eleventh Circuit’s holding in AseraCare to state that clinical judgments are never false. The primary reason the government was unable to successfully make its case in AseraCare was that, in lieu of available evidence, it solely used an expert witness who was unable to claim that no reasonable physician could have reached the contested conclusions. The Third Circuit, however, conflated the lack of evidence with the Eleventh Circuit’s framing of the issue. In reality, the Eleventh Circuit noted that opinions can be false, as it directly followed and cited to the Omnicare decision. The Third Circuit invoked the common law but never identified any evidence to suggest a false opinion under the Omnicare categories. The Third Circuit completely discounted the Supreme Court’s principle that, as a general rule, sincere statements of pure opinion are not “untrue statement[s] of material fact” even if the speakers are ultimately wrong. Thus, if the certifying physicians in Care Alternatives truly believed that their patients were terminally ill, the Third Circuit, without conducting a proper Omnicare-based analysis, would have controverted existing Supreme Court precedent.

Third, the Third Circuit’s entire analysis is based on its understanding of factual and legal falsity, but the court fundamentally misconstrued the relationship between objective falsity and the factual/legal falsity distinction in the Polukoff holding. The Polukoff decision indicated that factual falsity refers to express claims which are entirely based on fact, whereas legal falsity refers to any claim where legal requirements were not met. Accordingly, a subset of legal falsity claims includes claims where the legal requirement was not met due to negligent, reckless, or deceitful conduct, which implicates some extent of knowledge or lack thereof (in other words, implied claims). The Polukoff court simply demarcated the types of FCA claims which could reasonably be brought by plaintiffs. The Third Circuit, however, proclaimed that objective falsity is incompatible with legal falsity. The underlying assumption with this assertion is that objective facts may only be employed to challenge facts and not opinions. On a theoretical level, this line of logic is problematic because facts are objectively more concrete than opinions. Accordingly, as a matter of law, facts take precedence over opinions in the hierarchy of proof. Although “pure” opinions cannot be rebutted with facts, not all opinions are “pure,” especially those of medical professionals. Generally, professional opinions have some foundation in fact, which essentially places them on a spectrum between fact and opinion. This hybridization makes professional opinions susceptible to dispute by both facts and opinions. Thus, relegating the objective falsity standard to factual claims of falsity severely misses the extent of the standard’s applicability. On a practical level, the Supreme Court in Omnicare codified these observations into common law. For example, an opinion which contains a false, embedded fact is considered a false opinion. False opinions naturally fall under the Third Circuit’s legal falsity umbrella. The Supreme Court stated that if the embedded fact is proven to be false, the opinion is also false. This type of false opinion clearly allows for rebuttal with a contradictory factual finding, so an objective falsity standard is not necessarily improper when applied to legal falsity claims.

Finally, the Third Circuit improperly accused the Eleventh Circuit of wrongfully conflating the FCA’s knowledge and falsity elements; in fact, the Third Circuit was the court that conflated these elements. In AseraCare and Winter, the Eleventh and Ninth Circuits acknowledged that some evidence applies to proving both knowledge and falsity. This necessarily makes it difficult to analyze these elements separately. Even the district court in Care Alternativesrealized this when it rejected the relators’ claim because they failed to establish evidence of the physicians’ underlying knowledge (such that any physician lied or actually believed that the certified patients were not hospice-eligible). The Third Circuit made clear that “in our Court, findings of falsity and scienter must be independent from one another for purposes of FCA liability.” This is a misunderstanding of what common law principles apply. The fundamental distinction between an honest opinion and an opinion made in bad faith is the prerequisite knowledge used in forming the opinion. Therefore, when determining whether an opinion is false, the common law specifically looks to the speaker’s intent, which necessarily implies a knowledge requirement. The Third Circuit clearly subverted Omnicare by ruling that an after-the-fact reasonable disagreement between physicians can show falsity. Furthermore, as a practical concern, the Third Circuit’s falsity and knowledge separation could substantially increase the risk that the juror’s perception becomes tainted. The district court in AseraCare bifurcated its trial because it feared that evidence related to the knowledge element, particularly AseraCare’s flawed admissions policies and certification procedures to determine if a patient was terminally ill, would be inferred by the jury to satisfy the falsity element. This did, in fact, confuse the jury’s analysis of the threshold falsity question. Conceptually, general corporate practices have no bearing on whether a particular hospice claim is false if the medical evidence points to the fact that the patient was terminally ill. Accordingly, the Third Circuit’s interpretation of the knowledge and falsity elements potentially writes the falsity element out of the FCA statute by allowing evidence of knowledge to cloud a jury’s perception of falsity.

B. Courts May Reasonably Reach Different Falsity Standards

The objective and subjective falsehood standards are not necessarily diametrically opposed in the broad legal sense. In remaining true to Congress’s intent, courts have used the FCA “to reach all types of fraud, without qualification, that might result in financial loss to the Government.” In doing so, the FCA has been interpreted alongside other applicable laws and regulations since its inception. Prior to the recent medical FCA cases, this was not an issue because fraud was never predicated solely on subjective professional opinions without a tangible associated fact. In Polukoff terms, the entire realm of FCA claims were factual claims and non-opinion legal claims. Due to the factual basis for these types of claims, courts had to adopt an objective falsehood standard, which slowly resulted in uniformity among jurisdictions. However, as noted by the Winter court, the Supreme Court “ ‘has consistently refused to accept a rigid, restrictive reading’ of the FCA.” The same reasoning can be extrapolated to the falsity standard to the extent that it is an element of an FCA claim. Interactions between the FCA and applicable laws and regulations thus do not inherently allow for a universal falsity standard but rather the possibility of different falsity standards to be adopted in specific circumstances. This flexibility in the legal interpretation of falsity is also the better policy approach that allows for a more robust legal system.

The objective falsehood standard is an appropriate legal interpretation based on CMS’s guidelines and the MHB’s purposeful deference to physician judgment. The MHB and CMS’s guidelines for hospice eligibility repeatedly reference the subjectivity involved in determining terminal illness. First, as a general matter, Congress has not amended the hospice eligibility criteria. Second, the MHB specifically prohibits the use of check boxes and requires a narrative explanation of the diagnosis. Third, the MHB allows for unlimited recertifications. Fourth, the MHB requires physicians to consider subjective and objective medical findings. Finally, the MHB explicitly declared that predicting life expectancy is not an exact science. Taken in totality, these factors show the imprecise nature and complexity of hospice certifications. Thus, based on the lack of any statistical or medical measurement for longevity, medical professionals have been afforded the utmost deference by Congress. If courts were to adopt a subjective falsity standard with regards to hospice care, FCA trials would devolve into a meaningless battle of expert opinions, neither of which may be false. This is exactly what happened at the district level in AseraCare. The objective falsity standard is more sensible and provides a safeguard to trivial FCA claims based on falsity. Congress’s intention was not to allow “rogues” to take a doctor to court based on their certification that a patient had six more months to live simply because they found another doctor who believed the same patient had a remaining life expectancy of six and a half months. By adopting an objective standard, juries are not forced to become a “third doctor” who simply evaluates purely medical judgments.

At the same time, the subjective falsehood standard is an appropriate legal interpretation based on Medicare regulations and CMS’s definition of “reasonable and necessary.” Medicare reimbursements for inpatient hospitalizations are contingent on the provided services being reasonable and necessary, as defined by the CMS. Similar to the MHB, Medicare regulations demand that doctors evaluate complex medical factors to form their clinical judgment. In contrast to the MHB, however, Medicare regulations do not give physicians “unfettered discretion.” The regulations explicitly defer to the accepted standards of medical practice but provide no presumptive weight to a physician’s certification. Expert opinions must be analyzed in the context of medical evidence. An objective falsehood standard is not necessarily incompatible with Medicare hospitalization claims but would be redundant. Since the Medicare program already requires medical evidence for initial certifications, facts are presumably available in every case. The focal point of these cases surrounds the interpretation of these facts. Therefore, medical expert testimony offers more value than just competing medical theory. Testimony effectively provides valuable, logical medical inferences and contextualizes the interpretation of medical data, which is substantially less subjective than end-of-life determinations. Due to the fact that judgments are less rooted in medical theory and more rooted in medical practice, juries are able to make more substantiated findings, as they did so in Paulus and Polukoff. Accordingly, the subjective falsity standard is the more appropriate standard under these circumstances.

C. Legislative Action for the FCA

Given that rejections of the objective falsity standard have occurred exclusively in medical-related FCA claims, the unique challenges associated with the medical realm may be more efficiently handled through legislation. Judicial interpretations of falsity have been effective in filling the statutory gap in the FCA until the current circuit split, where false opinions in the medical context have divided the common law landscape. The crux of the issue is that the medical sector is quite anomalous when compared to other areas of practice, but courts cannot simply apply a medical-specific standard. As a general matter, courts do not derogate legal standards based on the field of application (for instance, the financial sector does not receive a different legal standard from the technology sector simply because the fields are different). Common law seeks to prescribe a set of legal rules and principles that can be consistently applied. Absent some countervailing statute or regulation, the common law is standardized across all fields. The countervailing statute in AseraCare and Care Alternatives was the MHB. The countervailing regulation in Winter was Medicare. However, the plain language of the MHB statute and Medicare regulations grants different degrees of deference to doctors. As a result, it is unclear how courts can adopt one standard without subverting congressional deference to doctors. If courts adopt a bright-line objective falsity standard, they will comply with the MHB and protect medical professionals from frivolous FCA suits but impose a higher standard of proof for plaintiffs, which will prevent valid suits involving false Medicare certifications from getting through trial. On the other hand, if courts adopt a subjective falsity standard, false Medicare certification claims could be handled appropriately through FCA litigation while frivolous claims will be brought against physicians who genuinely certify hospice care, which neither the FCA nor the MHB protects against. As a result, while the Supreme Court can attempt to resolve the circuit split in the future, the resolution may not be desirable as the adopted standard may lack the nuance needed to accommodate both medical laws and regulations.

From a policy perspective, it is also in Congress’s interest to clarify how and when physicians should be held accountable for their clinical opinions. The legislative branch, beholden to the people, creates laws while the judiciary promotes fairness and justice through the interpretation of such laws. Congress originally wrote the MHB and authorized Medicare programs after appreciating the host of factors that go into complex medical decision-making. The objective was to strike a balance between accountability and scrutiny within different medical settings. It is simply not the courts’ job to engage in judicial policymaking that overrides congressional intent. Courts cannot require stricter or looser scrutiny of physician judgments as they see fit and would effectively be doing so by adopting a bright-line standard. Moreover, the majority of FCA claims for the past several years have been from the medical field. Based on this consistent trend and America’s aging population, the composition of FCA claims for the foreseeable future will remain dominated by healthcare-related claims. Thus, it is imperative that Congress provide explicit, meaningful guidance on this issue.

D. Policy Considerations for the FCA

Although this Note primarily focuses on the fraud deterrence aspect of the FCA, a comprehensive discussion would not be complete without addressing the competing policy tradeoff—over-incentivization of whistleblowers to file false or frivolous FCA claims. In 2020, the relator share awards totaled over $300 million with only 672 qui tam claims filed. Although the potential for monetary gain differs for each case, it can be extrapolated from this data that whistleblowers can win hundreds of thousands, if not millions, of dollars if they prevail on a claim. This is a powerful incentive for unscrupulous individuals hoping to profit from this well-intentioned statute. A legal falsity standard certainly eases their ability to do so. However, the hurdles that would have to be overcome by such individuals virtually eliminate the risk of undeserving payouts. First, after the filing of a qui tam complaint, the government is required to investigate the allegations and can move to dismiss if the findings show that the relator has no grounds for the complaint. Even if the government does not move to dismiss, approximately 52% of FCA cases are resolved at this stage by agreed dismissal or settlement. Second, at the motion-to-dismiss stage, the complaint may fail as a matter of law due to lack of specificity. Third, at the summary judgment stage, the relator must show a genuine dispute of material fact. In the event that a legal falsity standard is applied, the relator may not have any issues convincing the court. Nonetheless, nearly 80% of FCA cases were resolved before the summary judgment stage. Thus, it is statistically unlikely for a relator, much less a dishonest one, to even reach this stage. Finally, the relator must prevail at trial. But more than 99% of FCA cases settle or are dismissed before reaching trial due to the high stakes nature of FCA litigation. Consequently, the systemic barriers and costly litigation process should sufficiently dissuade fraudulent rogues and assuage any concerns regarding overburdening of the judicial system.

In addition, while the extracted case law from the described cases does not provide an exactly useful model for FCA litigation in non-medical practice areas, the proposed non-duality falsity standard concept can be utilized to address the broader whistleblower policy debate. Whistleblower laws typically attempt to strike a balance between protecting the rights of whistleblowers and respecting an employer’s rights to remove personnel. Lawmakers must therefore decide who whistleblowers can report information to while still receiving sufficient protections from employer retaliation. This has been the subject of scholarly debate and criticism for years, which still rages on today. In January 2021, the Anti-Money Laundering Act was enacted, which expanded the recipient list for employees of financial services institutions. However, the Act still imposes a rigid report recipient requirement. Similar to how a rigid falsity standard fails to account for the plethora of intersecting laws and regulations, an unduly restrictive recipient list likely cannot match the diversity of situations that whistleblowers find themselves in. It is evident that the issue of rigid standards permeates the whistleblower legal arena. Moving forward, open-ended or flexible standards may provide the nuance necessary to usher in a new era of comprehensive whistleblower reforms.

CONCLUSION

AseraCare, Care Alternatives, and Winter are the first cases to adopt the objective or subjective falsehood standard for FCA claims in the context of medical certifications based on “false opinions.” The specific question at issue is whether dueling expert opinion, without more, creates a triable issue of fact for the jury. The objective falsity standard posits that conflicting opinions are not enough whereas the subjective falsity standard believes contradictory judgments are sufficient. The practical result of courts adopting the subjective standard is that relators and the government are more likely to survive the pleading and summary judgment stage by simply providing dueling expert opinion. An objective falsity standard makes it more difficult for plaintiffs to prevail on an FCA claim. The Eleventh Circuit in AseraCare adopted the objective standard after analyzing the plain language of the FCA and MHB. The Third Circuit in Care Alternatives arrived at the subjective standard after analyzing the same statutes. The Ninth Circuit in Winter implicitly agreed with the subjective standard after it considered the FCA and Medicare regulations. Superficially, there is a clear circuit split over the falsity standard. The two standards are at odds, but each one is applicable in different medical settings based on Congress’s intent and the plain meaning of the governing statutes and regulations. More importantly, however, is not what standard each circuit adopted but how the courts arrived at their conclusions.

While there are countless examples of FCA certifications requiring a medical opinion or exercise of discretion, the above trifecta of cases perfectly contrasts how courts should and should not invoke common law. The Eleventh and Ninth Circuits, while reaching different conclusions, employed the same common law framework and principles in their analyses. They primarily relied on (1) the Tenth Circuit’s Polukoff holding to distinguish factual and legal falsity, and (2) the Supreme Court’s Omnicare decision discussing when opinions may be deemed false. Conversely, the Third Circuit stated that it looked to common law for guidance while (1) misconstruing case law, (2) ignoring common law precedent, and (3) failing to apply common law in its lackluster analysis. Unlike the Third Circuit, courts should utilize the PolukoffOmnicare framework because it categorically constricts the universe of FCA claims into a logical, comprehensive framework with which to analyze false opinions.

The Supreme Court missed an opportunity to at least resolve the analytical differences between the circuits when it denied certiorari for Care Alternatives and Winter. As a matter of policy, the decision to adopt or reject a rigid falsity standard will have wide-ranging consequences, and it should be up to the legislature to insulate or scrutinize physicians for their certifications. Aside from adjudging these exercises of discretion as true or false, the Supreme Court has the responsibility of correcting circuits when they falter in their representation of common law principles. Omnicare is arguably the most relevant common law precedent in terms of providing an analytical framework for determining false opinions. Thus, the Third Circuit’s disregard of Omnicare sets an extremely disruptive example for other courts. Moving forward, the Supreme Court should announce that courts must abide by Omnicare when engaging in an FCA analysis involving opinions.

 

APPENDIX:  COURTS ADOPTING OR REJECTING OBJECTIVE FALSITY

Courta

Case Name

Year

Type of Legal Claim

Adopting Courts

Tenth Circuit

United States ex rel. Morton v. A Plus Benefits, Inc.b

2005

Medicaid

District of Nevada

United States v. Prabhuc

2006

Medicare

Fourth Circuit

United States ex rel. Wilson v. Kellogg Brown & Root, Inc.d

2008

Contract

Seventh Circuit

United States ex rel.Yannacopoulos v. General Dynamicse

2011

Contract

Third Circuit

United States ex rel. Hill v. University of Medicine & Dentistry of New Jerseyf

2011

Research Grant

Third Circuit

United States ex rel. Thomas v. Siemens AGg

2014

Contract

Northern District of Texas

United States ex rel. Wall v. Vista Hospice Care, Inc.h

2016

Medicare/Medicaid

Eleventh Circuit

United States v. AseraCare, Inc.i

2019

Medicare

Rejecting Courts

Third Circuit

United States v. Care Alternativesj

2020

Medicare/Medicaid

Ninth Circuit

Winter ex rel. United States v. Gardens Regional Hospital & Medical Center, Inc.k

2020

Medicare

Notes:  This list merely demonstrates the uniformity of the legal landscape prior to the Third and Ninth Circuit decisions in 2020 and is not intended to show every single jurisdiction that has ruled on the issue. aOf note, the Third Circuit originally adopted the objective falsity standard in 2011, reaffirmed the standard in 2014, and rejected the standard in 2020. This is quite peculiar, as it is the only Circuit that has switched its opinion on the issue. The Tenth Circuit adopted the objective falsity standard in 2005 but moved away from that decision in United States ex rel. Polukoff v. St. Mark’s Hospital. However, the Tenth Circuit has not explicitly embraced a subjective falsity standard. Sources:  bUnited States ex rel. Morton v. A Plus Benefits, Inc., 139 F. App’x. 980 (10th Cir. 2005). cUnited States v. Prabhu, 442 F. Supp. 2d 1008 (D. Nev. 2006). dUnited States ex rel. Wilson v. Kellogg Brown & Root, Inc., 525 F.3d 370 (4th Cir. 2008). eUnited States ex rel. Yannacopoulos v. Gen. Dynamics, 652 F.3d 818 (7th Cir. 2011). fUnited States ex rel. Hill v. Univ. of Med. & Dentistry of New Jersey, 448 F. App’x 314 (3d Cir. 2011). gUnited States ex rel. Thomas v. Siemens AG, 593 F. App’x 139 (3d Cir. 2014). hUnited States ex rel. Wall v. Vista Hospice Care, Inc., No. 3:07-cv-00604-M, 2016 U.S. Dist. LEXIS 80160 (N.D. Tex. June 20, 2016). iUnited States v. AseraCare, Inc., 938 F.3d 1278 (11th Cir. 2019). jUnited States v. Care Alts., 952 F.3d 89 (3d Cir. 2020), cert. denied, 141 S. Ct. 1371 (2021). kWinter ex rel. United States v. Gardens Reg’l Hosp. & Med. Ctr., Inc., 953 F.3d 1108 (9th Cir. 2020), cert. denied sub nom. RollinsNelson LTC Corp. v. United States ex rel. Winters, 141 S. Ct. 1380 (2021).

 

96 S. Cal. L. Rev. 665

Download

* Executive Postscript Editor, Southern California Law Review, Volume 96; J.D. Candidate, 2023 University of Southern California, Gould School of Law; M.P.H. Global Epidemiology 2018, Emory University Rollins School of Public Health; B.S. Integrative Biology 2016, University of Illinois at Urbana-Champaign. I would like to thank Professor Jonathan Barnett for guidance, Professor Eileen Decker for serving as my advisor, and the Southern California Law Review for excellent editorial assistance.

Familial Searches, the Fourth Amendment, and Genomic Control

In recent years, police have increasingly made use of consumer genomic databases to solve a variety of crimes, from long-cold serial killings to assaults. They do so frequently without judicial oversight per the Fourth Amendment’s warrant requirement by using consumer genomic platforms, which store hundreds of thousands or millions of user genomic profiles and enable law enforcement to infer the identity of distant genomic relatives who may be criminal suspects. This Essay puts this practice into context given recent legal and technological developments. As for the law, the Supreme Court in United States v. Carpenter has suggested that technologically driven and expansive datasets may be entitled to the full suite of Fourth Amendment protections. As for technology, we describe here the development of a novel technology that allows users to engage in genomic analysis in a secured environment without making such information available to a third party. Taken together, we present a possible technological solution to ensuring Fourth Amendment protections for direct-to-consumer genomic data.

Download

Lenity and the Meaning of Statutes

Ordinary canons of statutory interpretation try to encode linguistic rules into jurisprudence. Their purpose is to figure out the meaning of a text, and their outcome is to determine the meaning of the text. Both the purpose and the outcome are linguistic.

The rule of lenity is not an ordinary canon of statutory interpretation. The rule of lenity’s outcome is to determine the meaning of a text, giving ambiguous criminal statutes a narrow interpretation, but its purpose is public policy, protecting defendants when ambiguous statutes failed to give fair notice that their actions would be punished. Unlike the ordinary canons of statutory interpretation, lenity encodes into jurisprudence not a linguistic rule, but a policy rule. Thus, a discrepancy arises: lenity’s outcome is linguistic, but its purpose is non-linguistic.

This Article makes the following three contributions. First, it analyzes the nature of the discrepancy between lenity’s purpose and outcome. Second, it demonstrates that this discrepancy leads to doctrinal issues in how the rule of lenity is applied. Sometimes the rule of lenity is over-inclusive: it is applied even when there is no violation of fair notice. Sometimes the rule of lenity is under-inclusive: the rule of lenity fails to protect certain defendants that were misled by ambiguous criminal statutes. Third, this Article argues that we can align lenity’s purpose and outcome by reforming lenity into an excuse in criminal law, and this theoretical reformation will resolve the aforementioned doctrinal issues.

INTRODUCTION

Short-barreled rifles are often used for criminal purposes because their shorter length allows them to be more easily concealed. For that reason, § 5821 of the Internal Revenue Code levies an excise tax on the manufacture of short-barreled rifles, while no such tax is levied on the manufacture of long-barreled rifles.

Thompson/Center Arms, a firearms manufacturer, packaged as one unit the following separate parts that were to be put together by the customer: a shoulder stock, a pistol, and a barrel extension. For convenience, I will call this unit of three parts the “Thompson/Center kit.” Putting the three pieces together—attaching the shoulder stock to the handle of the pistol and the extension to the barrel of the pistol—the customer would end up with a long-barreled rifle. If the customer only attached the shoulder stock to the pistol handle without using the barrel extension, then they would end up with a short-barreled rifle.

Thus, the following legal issue arose in United States v. Thompson/Center Arms Co. Is Thompson/Center Arms liable for the § 5821 excise tax? Does the manufacture of the Thompson/Center kit count as an instance of manufacturing a short-barreled rifle?

The Supreme Court stated that § 5821 is ambiguous about what counts as the manufacture of a short-barreled rifle and that the Thompson/Center kits sat squarely in the penumbra. On one hand, Thompson/Center Arms intended for the kits to be put together into a long-barreled rifle, but on the other hand, the kit made it tremendously easy for consumers to put together a short-barreled rifle regardless of Thompson/Center Arms’s intention. Were the Court to construe § 5821’s language broadly, Thompson/Center Arms would be liable for the excise tax on short-barreled rifles, but were the Court to construe the statute’s language narrowly, Thompson/Center Arms would not be liable.

To resolve whether § 5821 should be given a broad or narrow reading, the Court applied the rule of lenity, which gives all ambiguous criminal statutes a narrow meaning, thus absolving Thompson/Center Arms of liability on the excise tax. This is a surprising application of the rule. The rule of lenity is a rule of statutory interpretation meant to apply only to criminal statutes to protect criminal defendants, yet it was applied in Thompson/Center to determine the meaning of a civil tax statute in favor of a civil plaintiff. Because the company had already paid the tax and was suing for a refund, no criminal penalties were at stake for Thompson/Center Arms.

Thompson/Center’s holding presents a major problem for the administration of tax law. The standard rule in civil law grants deference to an administrative agency’s interpretation of the relevant laws. The rule of lenity runs in the opposite direction, interpreting statutes in favor of the taxpayer over the agency, the Internal Revenue Service (“IRS”). This poses a special danger to the IRS’s enforcement efforts against abusive tax shelters that prey on indeterminacies in the tax law.

Despite this problem, the Court’s hands were bound by a technicality. According to the rule of lenity, criminal statutes should be interpreted narrowly such that uncertainty about the meaning of the statute is resolved in a way lenient to the defendant. Section 5821 is, like tax law generally, a civil statute, but it is also a criminal statute because its meaning has implications for criminal liability. Under § 5871, criminal penalties would be imposed for non-compliance with § 5821. Section 5821 plays a dual role, determining how much tax one is required to pay and, thereby, defining the actus reus for criminal liability. Thus, although Thompson/Center Arms was litigating the civil matter of how much tax it owed, because the outcome of this case might have criminal implications down the road, § 5821 would need to be read narrowly, following the rule of lenity.

Thompson/Center thus establishes that the rule of lenity applies to statutes that serve both a criminal and civil purpose, even if the issue at bar is a purely civil one, because the interpretation of dual-purpose statutes in the civil context necessarily carries over to define criminal liability. Tax laws generally play this dual role since they determine civil tax liability, and criminal penalties are imposed for non-compliance with tax law. Using lenity to narrowly interpret the meaning of a tax statute will both limit the reach of criminal sanctions for tax evasion and also limit the assignment of civil tax liability.

The purpose of the rule of lenity, however, is to protect fair notice for criminal defendants. When statutes are ambiguous, citizens can be misled into thinking that their actions were permitted rather than prohibited. The law fails to communicate the expected standard of behavior. Given the severity of criminal punishment and the moral condemnation that attaches, we ought to be especially concerned about criminal defendants who did not receive fair notice of the law. Thus, when a defendant’s act is a borderline case of an ambiguous criminal statute, the law absolves them of criminal liability as a recognition of its own failure to provide fair notice that such an act would be punished.

Since the rule of lenity was supposed to provide fair notice in punishment, its application to civil tax law, where no punishment is at stake, grossly oversteps its purpose. Even if a taxpayer loses a case determining their civil tax liability, so long as they continue to pay said tax liabilities, they would avoid criminal penalties. I call this overstep of lenity’s purpose the “too much lenity” problem.

On my analysis, the central theoretical issue with the rule of lenity is the discrepancy between the rule’s purpose and outcome. The rule of lenity’s purpose is to ensure fair notice about which actions are punished under the law. The rule’s outcome, as a canon of statutory interpretation, is to determine the meaning of a statute. The rule of lenity has a linguistic outcome, but a non-linguistic purpose. Thus, lenity’s purpose and outcome are not consistent with one another.

This application of lenity as a canon of statutory interpretation, which I call the “semantic rule of lenity,” is incongruous with its normative purpose of fair notice in criminal law, resulting in its encroachment into civil matters where no punishment is at stake. Unlike other canons of statutory interpretation, which aim to figure out the meaning of a statute, substantive canons, like the rule of lenity, aim to implement normative principles, like fair notice. Therein lies the disconnect. The purpose of the rule of lenity does not have anything to do with the ascertainment of meaning, yet the rule ends up determining the meaning of the statute. The resulting problem of too much lenity demonstrates that this disconnect leads to real consequences.

But notice that this is a contingent feature of the rule of lenity. Lenity need not be applied as a canon of statutory interpretation. Its purpose merely requires us to let go those criminal defendants who never received fair notice of punishment. Other legal doctrines that require us to absolve certain defendants of guilt—for example, excuses such as insanity or duress—do not involve determining the meanings of statutes. So why should the rule of lenity perform this odd, dangerous, vestigial function of determining the meaning of statutes? If the proximate aim is to absolve defendants of liability when their actions were not unambiguously criminalized by Congress, we can and ought to do so without invoking the semantics of statutes.

Challenging the standard semantic application of lenity, I will instead argue for the unorthodox position that lenity should be reworked from a canon of statutory interpretation to an excusing condition specific to criminal law. In that way, lenity would be applied in the same manner as the doctrines of duress or insanity, as an affirmative defense to prosecution rather than a canon of statutory interpretation. Without any of the semantic baggage that currently burdens the rule of lenity, excuses can apply in criminal law without extending into civil law and thus avoid the too much lenity problem. For instance, when a taxpayer is just litigating the issue of how much taxes they will have to pay for such-and-such economic transaction because they disagree with the IRS about the meaning of a statute, the courts should use ordinary interpretative principles that would best allow the tax law to serve its function of justly and efficiently collecting revenue. But if that same taxpayer was being tried for tax evasion because the statute at issue was ambiguous—as § 5821 was with regard to Thompson/Center kits—then lenity should be applied as an excuse, an affirmative defense, in order to protect fair notice of punishment.

Viewed top-down, this Article can be understood to present the following argument for my conclusion that lenity should be applied as an excusing condition in criminal law rather than as a canon of statutory interpretation: First, I demonstrate that lenity’s purpose of fair notice of punishment does not match its outcome of determining the meaning of statutes. Second, I analyze three distinct doctrinal problems that stem from this mismatch between purpose and outcome. Third, I solve these problems by showing how the legal system can unite lenity’s purpose and outcome by instituting lenity as an excuse rather than a rule of statutory interpretation. Because of this conceptual harmony, the three aforementioned problems are solved if we apply lenity as an excusing condition in criminal law. Each step presents novel contributions to the literature.

Part I explicates the rule of lenity and justifies the doctrine as upholding the structural rule of law value of fair notice. Fair notice is best understood as a structural consideration about the legal system. The laws must be structured so as to provide a path safe from punishment along which ordinary citizens can walk. In our society, this path is marked by published statutes delineating which acts are permissible and which are impermissible. Fair notice is thus essential to providing a genuine choice to avoid punishment.

Part II shows that lenity’s purpose and outcome are at odds with one another. Whereas ordinary rules of statutory interpretation have the purpose of trying to figure out the meaning of a statute and the outcome of determining the meaning of a statute, the rule of lenity has the purpose of protecting criminal defendants and the outcome of determining the meaning of a statute. Thus, while ordinary rules of statutory interpretation have a semantic purpose and semantic outcome, the rule of lenity has a semantic outcome and a non-semantic purpose.

Part III demonstrates three doctrinal problems that arise from the mismatch between the rule of lenity’s purpose and outcome.

Section III.A presents the linguistic ambiguity problem. To use a stylized example, suppose a statute ambiguously imposes criminal penalties for starting a fire next to a “bank.” Defendant A started a fire next to a financial bank. Defendant B started a fire next to a river bank. Because of the ambiguity, neither Defendant A nor B had fair notice that their actions were prohibited. However, because neither interpretation of the word “bank” lets both defendants go free, the rule of lenity cannot resolve the fair notice problem here. This is the problem of linguistic ambiguity.

Section III.B presents the too much lenity problem, introduced above. In this Section, I consider the impact of a lenity-driven tax regime both in terms of the areas of tax law where lenity is most likely to be applied and its contrast to the deference regime it replaces.

Section III.C demonstrates the problem of higher-order vagueness. Applying the semantic rule of lenity to a vague statute that prohibits a certain category of actions changes the meaning of the statute to prohibit only clear cases of that category of actions. For instance, a statute may say “do not drive dangerously,” but after the court applies the rule of lenity, the statute means “do not drive clearly dangerously.” The problem is that the new meaning that the rule of lenity has assigned will itself be vague. Just as vague predicates have borderline cases of which items qualify as members of the category, there is also vagueness one level up about which items qualify as borderline cases. If the vagueness of “do not drive dangerously” violates fair notice, then construing the statute to mean “do not drive clearly dangerously” will not satisfy fair notice because what counts as “clearly dangerous” is itself a vague matter as some driving is clearly clearly dangerous and some driving only borderline clearly dangerous. The semantic rule of lenity is thereby under-inclusive, creating vagueness at a higher-order but failing to take that second-order vagueness into account for purposes of fair notice.

Part IV connects the legal theory set out in Parts I and II with the doctrinal analyses of Part III to support my ultimate proposal that lenity be provided solely as an excuse in criminal law instead of its current application as a canon of statutory interpretation. In criminal law theory, excuses are most often understood in comparison to justifications, another category of affirmative defense. Whereas justifications typically serve to make an act permissible—for instance, killing another is not morally wrong if done in self-defense—excuses absolve an actor of criminal liability for their wrongful conduct when the actor lacked a genuine choice to follow the law. For instance, a browbeater may have threatened to bust the defendant’s kneecaps unless the defendant commits a criminal act for the browbeater’s benefit. In such a situation, because the browbeater’s coercive threat left the defendant no choice in the matter, the law affords the defendant an excuse of duress.

The semantic rule of lenity functions more closely to justification; by assigning a narrow meaning to a statute, it shrinks what counts as impermissible. The semantic rule of lenity, when it applies, concludes that the defendant’s actions were not prohibited by law. However, I argue that the purpose of lenity instead aligns most closely with that of an excuse. Though lenity may seem an unlikely bedfellow to doctrines such as duress or insanity, I demonstrate that all of these doctrines aim to protect citizens who lacked a genuine choice to follow the law. In cases such as duress, one lacks the choice because of some coercive threat. In cases of lenity, one lacks the choice because one was not given fair notice about which acts would be punished. Although, in contrast to justification, the defendants may have done some prohibited act in these cases, punishing them would nevertheless go against the rule of law principle of preserving a path safe from punishment.

By shedding lenity of its semantic cloak, jurisprudence can avoid the three aforementioned doctrinal problems. Providing lenity as an excuse rather than fixing the meaning of a statute would allow the law to absolve both Defendant A and Defendant B (from the “bank” example above) of criminal liability since both defendants lacked fair notice that their actions would be punished. By restricting lenity to criminal law, taking the form of an excuse stops lenity from creeping into civil law, thereby solving the too much lenity problem. Because the excuse would not determine the meaning of the statute, issues of higher-order vagueness do not require additional iterations of lenity, thereby solving the higher-order vagueness problem.

Part V considers two counterarguments to the excuse of lenity. The first counterargument states that the excuse is unnecessary because strict construction of the civil tax code is good jurisprudence. In response, I analyze the ways that the teleology of tax law is distinct from the teleology of criminal law. Tax law helps citizens figure out how much to contribute to the public fisc as a matter of distributive justice. Unlike criminal law, tax law is not meant to sanction prohibited behaviors—a tax on income, for instance, is not meant to morally condemn those who earn income. Since tax law is not meant to serve as a system of incentives, ex-ante notice is far less important. Furthermore, choosing strict construction over the best interpretation will undo the effort to justly allocate social burdens, to the detriment of the very people who relied on the tax law to serve this function. The second counterargument against the excuse of lenity states that a legislature could satisfy the requirement of fair notice by letting citizens know by statute that the rule of lenity will not be applied to the criminal code. I argue that such a move is tantamount to notifying the public that there will be no fair notice given.

I. THE RULE OF LENITY

At its core, the rule of lenity is a rule of statutory construction that resolves any “uncertainty concerning the ambit of criminal statutes” in favor of the defendant. Often, such uncertainty can arise due to linguistic indeterminacy, the most common type of which is vagueness. In these cases, the meaning of the vague statute is narrowly interpreted to include only clear, prototypical cases of the criminal statute. Such narrow construction is justified by the rule of law value of fair notice. Fair notice allows citizens who wish to avoid punishment to seek safety in reading the statute and choosing to avoid those actions that carry criminal penalties.

A. Lenity’s Outcome: Statutory Construction

Indeterminacy of meaning (linguistic indeterminacy) is a universal feature across natural languages. Statutes, since they are written in natural language, sometimes have indeterminate meanings. Such instances can give rise to what we may call hard cases or legally ambiguous cases, where the statute gives no direction one way or another to those cases that straddle the indeterminacy.

Of course, even when the statute gives no direction, a case at bar cannot go unresolved. One way to resolve a case in which there is no resolution provided by the statute itself is to have what legal theorists call a “closure rule.” A closure rule simply determines which way a judge should rule when the law is unclear one way or another, acting as a tie breaker of sorts. The rule of lenity is often held by jurisprudents to be a paradigm closure rule. When a criminal statute fails to resolve a case because its meaning is indeterminate with respect to the facts at bar, then the judge must assign the statute a narrow meaning that favors the criminal defendant.

In law, by far the most common sort of linguistic indeterminacy arises from vagueness. Consider, for instance, the well-trodden “no vehicles in the park” statute:

NO VEHICLES IN THE PARK ACT: Any person who brings or drives a vehicle into a federal park shall be guilty of a misdemeanor, which may be punished by a fine.

From the language of the statute, no one can deny that driving an automobile into a federal park is prohibited. In contrast, we may be quite uncertain about whether someone who pushed a wheelbarrow into a federal park is criminally liable under the statute since it is uncertain whether a wheelbarrow is or is not a vehicle in this context. The term “vehicle” is vague since it admits of borderline cases where the application of the term “vehicle” is indeterminate.

An instance of a vague predicate is prototypical or core if and only if it is clearly a member of the predicate’s category. A borderline or penumbral case of a vague predicate is an object that is neither clearly a member nor clearly not a member. Thus, a sedan is a prototypical vehicle while wheelbarrows are borderline cases that do not clearly fall into nor outside of the vehicle category.

When it comes to vague statutes, the rule of lenity is best cashed out using this distinction between clear and borderline cases. Under the rule, a vague criminal statute will be assigned a narrow meaning that includes only the clear cases of the vague categories. For instance, the narrow interpretation of “vehicle” includes automobiles but not wheelbarrows. The rule of lenity, like statutory interpretation more generally, is semantic in that it operates to determine the meaning of the statute.

Courts are supposed to employ the rule of lenity in the realm of criminal law. Under such a rule, the destruction of a fish was not found to be a violation of a statute prohibiting the destruction of “tangible objects” in a federal investigation, and transporting a stolen airplane did not count as transporting a stolen “vehicle.” These were, in the eyes of the court, not prototypical cases of the statutes’ language.

B. Lenity’s Purpose: Fair Notice

Courts have typically appealed to fair notice, sometimes referred to as “due-process notice,” as the principal justification for the rule of lenity. Punishment of criminal activity is serious both in the severity of its costs on the punished and in the moral condemnation that attaches to it. For the exercise of the sword of government in doling out punishments to individuals, rule of law is of principal order. A central criterion of rule of law is that those who are subject to the threat of such force be given fair warning that they are under such threat. In our society, such notice is primarily given by the publication of criminal statutes. But publication is only the first step. Statutory notice is fair only when the content of the prohibitions can be readily ascertained from the published statute. Thus, when Emperor Caligula posted new statutes high on the top of Roman columns to prevent the citizenry from reading them, he failed to give fair notice to his citizens.

Posting laws where no one can read them is not the only way to violate fair notice. Notice can be unfair due to a statute’s linguistic indeterminacy. Similar to how linguistic indeterminacy can fail to give guidance to judges on how to rule on hard cases, linguistic indeterminacy fails to provide guidance to citizens on what sort of behavior is prohibited by law. Consider the following illustration of this aspect of fair notice employing hypothetical expectations about a vague statute.

When one reads a vague criminal statute, so one version of the fair notice story goes, one thinks not of the borderline cases, but instead the prototypical cases. For instance, it is most likely that bringing a wheelbarrow into the park never crosses an individual’s mind as they read the words, “Any person who brings or drives a vehicle into a federal park shall be guilty of a misdemeanor.” The mental representation of the concepts conveyed by a statute typically does not include borderline cases. As a result, punishing someone for a borderline violation of a criminal statute would go against the natural reading of the statute. They would not have been given fair notice that their conduct would be subject to punishment but rather misled into thinking that they were following the law by the vagueness of the statute. In order to preserve the important rule of law value of fair notice, the rule of lenity requires a narrow construction of such statutes.

Leading cases on the rule of lenity often explicitly endorse a similar story regarding the expectations that readers of a vague statute are likely to have. For instance, in McBoyle v. United States, the Supreme Court applied the rule of lenity to rule that airplanes were outside the scope of the phrase “motor vehicle” as it was used in a federal criminal statute. The opinion justifies excluding airplanes from the motor vehicle category by stating that the “motor vehicle” phrase “evoke[s] in the common mind only the picture of the vehicles moving on land.” Courts are worried about the lay citizen reading a statute and naturally having only the prototypical instances come to mind.

Importantly, the analysis just described is meant to be focused on the statute itself rather than the defendant. That is, for any given statute, the test is not to see if the defendant in the instant case actually read the statute. Many, perhaps most, defendants have not. Instead, the analysis looks at the statute itself and how the text comes across to the ordinary reader. If the indeterminacy of a statute risks misleading readers, the rule of lenity attempts to limit punishment in such instances by requiring a narrow construction of the statute. The rule of lenity aims to correct a deficiency in the law itself.

H.L.A. Hart’s rule of law account of excusing conditions to criminal liability can provide additional theoretical grounding to the concept and value of fair notice. On Hart’s account, people should be able to avoid law’s sanctions if they so choose. There is an important security provided by knowing that we will be safe from punishment so long as we choose to follow the laws set out for us. However, if we read a statute and naturally think only of the prototypical cases, then we will think that we are following the law when we commit borderline violations of that statute. Punishments for non-prototypical violations of a criminal law statute subvert the safety of choice to follow the law. The park-goer does not think that they violate the “no vehicles in the park” statute when pushing a wheelbarrow across the park gates. If it were not for the rule of lenity, their having read the law and intention to follow it would provide no assurance that they are safe from punishment; the court could arrive at an interpretation that they had never expected by considering a wheelbarrow a vehicle.

Hart’s position here can be understood as a safe path argument. It is a minimal requirement of a legal system that it provide at least one path safe from punishment along which ordinary citizens can walk. The clearest violation of a safe path is the criminalization of both an action and its absence. For instance, suppose that criminal law both required citizens to wear a face mask and forbid citizens from wearing a mask. It may even be the case that both laws, understood separately, are reasonable—perhaps the legislature passed the first law to minimize transmissions of an infectious disease and the legislature passed the second law because the purchase of face masks by laypeople caused a shortage for healthcare workers. However, having both laws at once clearly violates the safe path requirement. There would be no way for a citizen to avoid punishment in a system that punishes both an action and its absence. In this situation, we would say that there is no safe path at all.

The absence of fair notice likewise violates the safe path principle. This is because the presumed safe path for ordinary citizens is the option to read the law and avoid the prohibited acts. When fair notice is violated, for instance by the punishment of non-prototypical violations of law, this safe path is upturned. These citizens’ reading of a vague statute would mislead them into thinking that they are outside the reach of punishment only to have the rug pulled out from under their feet. The state cannot be said to have provided its citizens a genuine choice to avoid punishment because the citizens were misled about which actions would lead to punishment. Vague statutes thus compromise the availability of a path safe from punishment along which ordinary citizens can walk. Lenity aims to protect for citizens a genuine choice to avoid punishment.

In sum, it is a rule of law principle that the government may not punish an individual without having first given fair notice that such actions would be punished. Such a principle protects the ability of citizens to find out which acts are punished and avoid committing such acts. The statutory notice that government gives to citizens is fair only insofar as citizens can naturally discern which of their acts are prohibited from reading the statute. Without lenity, citizens can be misled by a vague statute into thinking that they are safe from punishment. The significance of this rule of law value has been thought by some to endow on the rule of lenity a “quasi-constitutional status” due to its role in protecting fair notice.

II. A MISMATCH BETWEEN PURPOSE AND OUTCOME

Part I, in the process of explicating the rule of lenity, has presented two propositions that deserve further consideration:(1) the rule of lenity determines the meaning of criminal statutes, and (2) the rule of lenity is best justified by the normative principle of fair notice. The second proposition has to do with the rule of lenity’s purpose. The first proposition has to do with lenity’s mechanism; in order to carry out its purpose of fair notice, the rule of lenity stipulates a narrow meaning to a linguistically indeterminate statute. Though each proposition is well-accepted—one would not have trouble finding Supreme Court opinions or law school casebooks that repeat these truths—it nevertheless seems to me that the two propositions are at odds with one another. Lenity’s purpose is normative, but its outcome is semantic.

By definition, to interpret a text is to ascertain its meaning. The rule of lenity is not an attempt to ascertain the meaning of a statute; it instead stipulates a narrow meaning to a statute in order to protect fair notice. Thus, it is odd that the rule of lenity, which does not even purport to ascertain the meaning of a statute, is nevertheless a canon of statutory “interpretation.” If the purpose of the rule of lenity is something other than figuring out the meaning of a statute, then why does it end up determining the meaning of the statute?

This oddity of the rule of lenity may be best understood in contrast to more ordinary canons of statutory interpretation. For example, many canons rely on “maxims of word meaning” or rules of grammar to help piece together the meaning of a text. For these canons, their purpose and outcome are aligned. These rules rely on linguistic premises to ascertain the meaning of a text, so it makes sense that the outcome of applying these rules is to determine the meaning of statutes.

Canons of statutory interpretation can be analytically divided into three categories. While textual canons “[find] meaning from the words of the statute” and reference canons determine “what other materials might be consulted to figure out what the statute means,” substantive canons like the rule of lenity instead implement normative principles external to the task of interpretation like fair notice. Substantive canons, in contrast to the other two types of canons, are not concerned with “finding” or “figuring out” what the statute means. Substantive canons are grounded in normative policy principles rather than interpretative principles.

Putting these distinctions to work, one can only conclude that the rule of lenity is a canon of statutory “interpretation” in name only. The purpose of the rule of lenity is to protect criminal defendants who failed to receive fair notice that their conduct would be punished. Rather than interpreting a text, the rule assigns the words of a statute narrow meaning in order to implement normative principles concerning rule of law values. The canon is not a rule of interpretation properbecause it never seeks to interpret, that is, ascertain the meaning of, a statute. The rule of lenity has a semantic outcome—determining the meaning of a statute—which is flatly inconsistent with its non-semantic purpose.

Notice also that the problem I have outlined here does not depend on any particular theory of statutory interpretation. The discrepancy between the rule of lenity’s purpose and outcome relies only on the distinction between figuring out a meaning and stipulating a meaning. The rule of lenity stipulates the meaning of a statute instead of trying to figure out what the statute means. On no theory of statutory interpretation is providing fair notice for criminal defendants a way of figuring out the meaning of a statute. Providing fair notice is, on its face, neither a way of getting at the plain or ordinary meaning of a text nor uncovering the purpose of a statute, so it cannot be understood as either a textualist or purposive doctrine of interpretation. The discrepancy between the rule of lenity’s purpose and outcome should worry legal scholars of all stripes.

III.  THREE DOCTRINAL PROBLEMS

The theoretical disconnect between lenity’s purpose and outcome just outlined in Part II entails thorny doctrinal consequences. This Part explores three such doctrinal consequences: the rule of lenity cannot handle linguistic ambiguity, the rule oversteps its boundaries and enters civil law, and the rule fails to resolve issues of fair notice that result from higher-order vagueness.

A. Linguistic Ambiguity

Consider the following hypothetical. The word “bank” may refer to either the financial institution (“financial bank”) or the land next to a river (“river bank”). Suppose the Bank Safety Act criminalizes starting a fire within one hundred feet of a bank, and it is indeterminate which of the two meanings should be applied to the term “bank.” As argued in Section I.B, such indeterminacy of meaning violates the principle of fair notice. Suppose further that two defendants are on trial, Defendant A for having set fire next to a financial bank, and Defendant B for having set fire next to a river bank.

The rule of lenity states that an indeterminate text must be interpreted in favor of the defendant. But which one? Giving the statute either meaning will absolve one of the defendants but still condemn the other. The rule of lenity is like the Buridan’s ass unable to choose between two identical stacks of hay. If the court rules that “bank” refers to financial banks, then Defendant A will be held criminally liable, and if the court rules that “bank” refers to river banks, then Defendant B will be held criminally liable. This is because the statute’s indeterminacy arises from linguistic ambiguity rather than vagueness.

Linguistic ambiguity should be understood as distinct from another kind of ambiguity discussed earlier, what one might call legal ambiguity. What judges and practicing lawyers most often mean when they use the term ambiguity is a general kind of uncertainty about the application of a statute. Legal ambiguity can arise for a variety reasons. One such reason for legal ambiguity, discussed in the previous part of this Article, is the vagueness of language. Another reason for legal ambiguity is the kind of linguistic indeterminacy we saw with the two meanings of bank, what I refer to here as linguistic ambiguity.

Vagueness in language concerns how far out to draw the boundaries of certain terms, for instance how broadly we draw the category of manufacturing a short-barreled firearm. When linguists use the term ambiguity, they are instead referring to terms that can have disparate meanings altogether, such as the two possible meanings of the term bank. Put succinctly, vagueness concerns interpretations that differ in degree while ambiguity concerns interpretations that differ in kind. Whereas in cases of vagueness, the court can choose between broad and narrow readings because the narrow reading is a proper subset of the broad reading, in cases of ambiguity, there is no narrow interpretation because neither the river bank meaning nor the financial bank meaning is a proper subset of the other. Either reading of bank holds one defendant culpable while letting the other go free.

Plainly, this result of the rule of lenity is inconsistent with the demands of the rule’s fair notice purpose. Neither Defendant A nor Defendant B had fair notice that their action was punishable because the statute was ambiguous between their two readings. One could read the Bank Safety Act and come away thinking that it permits starting fires next to financial banks or come away thinking that it permits starting fires next to river banks. Given the indeterminacy of meaning, both are natural readings of the statute. The law does not clearly mark the path safe from punishment. Since neither defendant received fair notice, it would be unfair to punish either defendant.

Thus, the rule of lenity’s outcome is under-inclusive with respect to its purpose. Though the rule of lenity’s purpose of protecting fair notice would dictate absolving both defendants of criminal liability, its semantic outcome is unable to provide such a result.

At this point, the astute reader might raise the following objection. Thus far, by focusing on the fact that bank has just two possible meanings, I have obscured a third option that would work best. When it comes to linguistically ambiguous statutes, the objection states, the rule of lenity should say that the statute has no meaning at all. That is, the Bank Safety Act should be construed not to criminalize any behavior because its use of the term bank has no meaning. Following this rule, both Defendants A and B would go free, and the result would thus comport with the demands of fair notice.

In response to this objection, suppose that there is a third defendant, Defendant C. Defendant C started a fire next to a financial bank that happened to be located on a river bank. On either meaning of bank, Defendant C is guilty and, thus, had fair notice their actions were prohibited by law. Defendant C cannot possibly claim that the ambiguity in the statute would mislead someone into thinking that their actions were permissible. If the court construes the Bank Safety Act to have no meaning at all, then it would let Defendant C go, despite the fact that they had fair notice of punishment. The rule would still fail to serve its purpose.

The hypothetical Bank Safety Act demonstrates one way the disconnect between the rule of lenity’s purpose and outcome could lead to its being under- or over-inclusive, but a critic may nevertheless contend that such ambiguities appear rarely in the actual law. When will a reader actually be faced with the term “bank” in a statute and be unable to figure out whether it refers to river banks or financial institutions? Usually, the context and purpose of a statute will make one meaning the clearly right interpretation for a linguistic ambiguity, thereby eliminating any indeterminacy.

In part, I agree with the critic and, in part, I disagree. I concede I have no quantitative measurement of how often courts are faced with linguistic ambiguity, so these cases may indeed be rare. Scholars have noted real examples where the courts have had to interpret linguistically ambiguous statutes, but it is not obvious how often such ambiguities appear. Where I disagree with the critic is that I fail to see how this is a criticism. There is, at minimum, a conceptual problem at issue—the rules and principles of our legal system fail to conceptually form a coherent whole. The hypothetical example I used here lays bare a real incoherence in our legal system. Uncovering this previously unnoticed incoherence deepens our understanding of the rule of lenity. Moreover, even if linguistically ambiguous statutes are rare, the incoherence of the rule of lenity will still have other critical real-world consequences as the next Section of this Article will show.

B. Tax Law’s Rule of Lenity

Incongruous with its purpose to provide fair notice of punishment, the rule of lenity leads to narrow constructions of texts even outside of the criminal context. For instance, in United States v. Thompson/Center Arms Co., the Supreme Court, relying on the rule of lenity, assigned a narrow meaning to the phrase “making of a firearm” with regard to an excise tax levied on the manufacture of firearms under I.R.C. § 5821. The statute’s definition for “firearm” included short-barreled rifles, but excluded pistols and long-barreled rifles. The taxpayer packaged as one unit three parts that could be connected together: a shoulder stock, a pistol, and a barrel extension. As before, let us call this unit of three parts a “Thompson/Center kit.” Putting the three parts together would create a long-barreled rifle, on which no excise tax is laid. Putting just the shoulder stock and pistol together would create a short-barreled rifle on which excise tax is laid.

The Court stated that the manufacture of a Thompson/Center kit was not clearly an instance of making a firearm, but was also not clearly not an instance of making a firearm. Its next move, surprisingly, was to apply the rule of lenity. The Court assigned a narrow meaning such that only clearly making a firearm would count under § 5821. Since making the Thompson/Center kit is not clearly an instance of making a firearm, § 5821 does not here apply. Therefore, under the rule of lenity, the taxpayer need not pay any excise tax on the manufacture of a Thompson/Center kit.

The imposition of tax is a civil matter, not a criminal one. Thompson/Center Arms had paid the excise tax and was merely bringing suit to get a refund of those payments. Recall that the rule of lenity was justified under the context of punishment and the special kind of notice that the harshness of punishment demands. It seems no more appropriate to apply the rule of lenity in a civil matter than it would to apply a beyond a reasonable doubt standard of evidence to civil trials. So why was the rule of lenity being used within the context of tax law? The Court’s winding reasoning proceeds as follows. To begin, § 5871 imposes criminal penalties for nonpayment of the § 5821 excise tax on firearms. I.R.C. § 5821 is, by that fact, both a criminal statute and a tax statute (or “dual-purpose statute”). Therefore, the rule of lenity should apply to § 5821 within the context of criminal law to assign a narrow meaning to the phrase “making of a firearm.” The meaning of a single statute cannot fluctuate depending on what the statute is being used for. This principle of consistency in statutory interpretation is well grounded. (Using technical language of utterances, types, and tokens, it is easier to state this proposition more precisely: though a single utterance type may have multiplicity of meaning depending on context, a single utterance token cannot.) Therefore, if the rule of lenity requires assigning a narrow meaning to “making of a firearm” in the criminal law context, the narrow meaning assigned to “making of a firearm” applies to cases of civil tax law as well.

Thompson/Center stands for the principle that the rule of lenity properly applies to dual-purpose statutes. This abstract principle has left unresolved the concrete questions of exactly how lenity will change the interpretation and administration of civil law. Does lenity apply to all tax statutes? Where lenity does apply, what is its effect, counterfactually speaking? Though there is a lot of uncertainty in this area of jurisprudence, the following Sections analyze these two questions in order.

It should be noted that there are also dual-purpose statutes outside of the tax realm in areas ranging from securities law to environmental law, where violations of civil law can carry criminal penalties. Thus, I intend for my analysis of lenity in tax law to be valuable in itself as well as serving as an illuminating case study for the problem more generally across the variety of dual-purpose statutes in the law.

1The “Willfulness” Requirement

The dual-purpose nature of tax law will serve as the starting point of the inquiry. I.R.C. § 5821 is not the only statute that carries criminal penalties for non-compliance. I.R.C. §§ 7201 and 7203 assign criminal penalties to nonpayment and evasion of any tax imposed under the tax code, Title 26. Similar provisions assign criminal penalties for various procedural violations. So, one might reasonably conclude that the rule of lenity ought to apply to the interpretation of tax laws generally.

However, the Court in Thompson/Center implies that the rule of lenity need not be applied to all tax laws because § 7201 and related statutes can only be violated if the taxpayer acts “willfully.” The willfulness requirement of § 7201 already builds in notice as a pre-condition of punishment since the willfulness requirement is a requirement that the taxpayer know of and understand the law that they are breaking. Though the opinion is not explicit about either the rule or the underlying principle, the Court appears to be taking the position that the willfulness requirement satisfies the requirement of fair notice, so no application of the rule of lenity to the general tax law is required.

The Thompson/Center opinion’s use of the willfulness requirement as a dam against applying the rule of lenity is colorable, but not without cracks. The first crack in the dam is that not all tax statutes require willfulness. Thompson/Center presented just such a case, as § 5871 had no willfulness language. In those instances, it is clear that the rule of lenity should apply. The second crack in the dam is that it is not some necessity of tax law that its violations be punished only if such violations are willful. The willfulness requirement of tax law, as this Section argues, is contestable and contingent.

Generally, ignorance of the law is not an excuse. One is not released from criminal liability for not having known about the existence of a law criminalizing that particular conduct. Justice Oliver Wendell Holmes gave an oft-cited defense of the doctrine, “to admit the excuse at all would be to encourage ignorance where the law-maker has determined to make men know and obey.” Such a deterrence rationale satisfies the utilitarians. Retributivists, in contrast, have tended to consider the mistake of law doctrine a thornier problem. In particular, retributivists have found the lack of an excuse for mistake of law unfair when applied to mala prohibitaoffenses and in cases where the defendant was not culpable for his ignorance of the law.

Tax law presents an exception to the rule that mistake of law does not excuse. I.R.C. § 7201 provides that “[a]ny person who willfully attempts in any manner to evade or defeat any tax imposed by this title or the payment thereof shall, in addition to other penalties provided by law, be guilty of a felony.” The “willfulness” requirement of § 7201, as interpreted in Cheek v. United States, is a legislative exception to the rule that ignorance of law does not excuse.

As the court stated in Cheek, the central idea behind this exception is that “[t]he proliferation of statutes and regulations has sometimes made it difficult for the average citizen to know and comprehend the extent of the duties and obligations imposed by the tax laws.” Knowledge of a body of law as complicated as tax law requires either ability and effort devoted to understanding the requirements of tax law or the resources to hire an able person who has devoted time to studying the tax law. When the barrier to knowledge of tax law is so high, it would be unfair to punish individuals who have violated the tax law due to ignorantia legis. The Court’s reasoning thus echoes the aforementioned retributivists’ fairness concerns regarding cases in which defendants are not culpable for their ignorance of the law. For this reason, knowledge of law has been understood to be required for criminal liability across cases interpreting several criminal tax statutes.

Whether ignorance of tax law should be an excuse is a matter of balancing costs and benefits of such a rule. As the law becomes more complex, the unfairness of punishing a mistake of law increases. However, exempting mistake of law cases adds costs to the litigation process and lowers deterrence effects when people who know the law can credibly claim in court that they did not. Whereas legislatures have typically found that the balance tips against allowing ignorance of law as an excuse to criminal liability generally, Congress has found the balance tips in favor of allowing ignorance as an excuse when it comes to issues of tax law.

Reasonable minds, of course, can disagree with Congress about the outcome of the cost-benefit analysis. The cost-benefit balancing is contingent on not only the complexity of law and our valuation of the competing normative principles, but also the positive facts. For instance, suppose that the Treasury Department could provide a pre-populated tax return for low- and middle-income individuals. State-level implementation in California has been successful in providing pre-populated returns for those with simple tax situations, and pre-populated tax returns could plausibly be implemented at the federal level as well, the proposal even having been a part of then-Senator Barack Obama’s presidential campaign platform. Furthermore, most developed nations have return-free filing for low- and middle-income taxpayers, and such a system is not outside the realm of possibility in the United States. If we were to resolve the compliance difficulties currently in our system for low- and middle-income taxpayers, then the case for removing willfulness becomes much stronger, perhaps overwhelming the reasons for keeping the willfulness requirement. If the Treasury were to do all of the legwork for the taxpayer, then complying with the tax law would require no greater intellectual sophistication than following criminal law generally.

Regardless of how one would, from one’s preferred moral and political valuations, balance the costs and benefits, I take it that we all agree that if the balance of reasons weighed against the willfulness requirement, Congress should be able to revise the language of I.R.C. § 7201 (and corresponding criminal tax statutes) to delete the word “willfully.” An amendment by the legislature that ignorance of the law does not absolve one of criminal liability in tax law, which is a matter of retributive justice, should not have enormous implications for the distribution of tax liabilities, a matter of distributive justice. Yet this is precisely the consequence of the too much lenity problem.

2. From Deference to Strict Construction

Following the question of to which statutes the rule of lenity will apply, the second question is what effect such an application will have when the rule does apply to a dual-purpose statute. Recall that the rule of lenity requires finding in favor of the defendant when the law is unclear. This interpretative stance is striking as an approach to tax law. Indeterminacy is a persistent problem for statutes, and the tax code is no exception. Moving from an approach of uncovering the best interpretation of tax statutes to a taxpayer-wins approach in hard cases is a harsh blow to the tax law’s aims. An application of the rule of lenity is particularly harmful to the IRS’s enforcement efforts, as it is in the gaps of legal ambiguity where tax shelters thrive.

This issue becomes the clearest when comparing the rule of lenity to the general doctrines granting deference to the Treasury, and by extension the IRS, in interpreting the tax law. This comparison serves to analyze the counterfactual, the interpretative approach that would govern were lenity not to apply. An examination of the counterfactual brings to light just how starkly lenity contrasts in terms of both its purpose and effect.

Generally, when statutes are ambiguous, administrative agencies are granted deference (often called “Chevron deference”) by the courts in the agencies’ interpretation of the statutes they administer. The deference given to the IRS helps it to fill in the gaps of statutes in a way that comports with the aims of the tax code, collecting revenue in a just and efficient manner.

In contrast, the Department of Justice, which prosecutes federal criminal offenses, receives no such deference in its interpretation of criminal statutes. Instead, it is well established that to afford it deference would be to run completely opposite the rule of lenity. Whereas the rule of lenity is a pro-defendant approach to interpretation, affording deference to the Justice Department would be pro-prosecution. As Justice Scalia has put it, to afford deference to the Justice Department would “turn the normal construction of criminal statutes upside-down” into “a doctrine of severity.”

The doctrines of deference and lenity clearly juxtapose two distinct considerations about the right approach to take with regard to gaps in the law. On the lenity account, legal indeterminacy represents a rule of law failure and, in order to protect a path safe from punishment, citizens who fall under that penumbra cannot be punished. This account makes sense given the role of criminal law in carrying out retributive justice aims of punishment and moral condemnation. On the deference account, the gaps in the law ought to be filled by the expert, policy-driven approach of administrative agencies. This account makes sense given the role of tax law in coordinating distributive justice and revenue-raising functions. By cabining lenity to criminal law and deference to civil law, these opposing doctrines would have been kept aligned to their respective purposes, but under the Thompson/Center holding, lenity would apply to dual-purpose statutes that are being interpreted in the civil context. Even in cases that solely determine civil tax liability, instead of the interpretive regime that would best carry out the purposes of the tax law, the courts must employ a rule built to protect criminal defendants. The rule of lenity is incongruous with its purpose.

Without the kind of policy-driven approach permitted by Chevron, it is hard to imagine that there can be effective policing of tax shelters. In order to distinguish between abusive tax shelters and permissible tax planning, the agencies must look to the general purpose of the tax laws. This is because tax shelters follow the letter of the tax law while going against the fundamental spirit of the tax code. Whereas deference allows the IRS to interpret statutes in line with the spirit of the law, lenity swings much closer to the textualist “letter of the law” interpretation. Foreign jurisdictions applying ordinary meaning textualist approaches to interpretation have struggled to strike down tax shelters, and Thompson/Center threatens the same for the US system.

As with the willfulness dam limiting the statutes to which lenity applies, the Court has partly walled off the deference due to some agency interpretations from Thompson/Center’s assault. Though the tension is not yet fully resolved, the Supreme Court has laid out a middle way between the two competing doctrines for some dual-purpose statutes in Babbitt v. Sweet Home Chapter of Communities for a Greater Oregon, such that not all administrative interpretations will be stripped of deference. According to the middle way, agency interpretations of dual-purpose statutes will still be granted deference if they satisfy fair notice principles. In Sweet Home, the Court noted that the agency interpretations satisfied fair notice because they came in the form of regulations that had been published for twenty years.

But the Sweet Home middle way is limited. Not all administrative interpretations come by longstanding published regulations. Thompson/Center, for instance, presented a case in which no regulations were present. In the spaces where the IRS has not passed longstanding, formal, law-like regulations or has passed regulations with language itself subject to competing interpretations, it appears that fair notice will not have been provided. These gaps are significant. Practitioners (or indeed anyone familiar with the tax system) would vouch for the importance of informal, nonbinding IRS guidance on tax matters. Abusive transactions exploiting legal ambiguities in the tax code are often noticed by the IRS only after a taxpayer has engaged in such transactions. For these cases, Thompson/Center would severely hinder the Service’s efforts in effectuating the purpose of the tax laws by shifting from a deference regime to lenity.

Furthermore, the Sweet Home approach to deference has also drawn academic criticism for failing to coincide with the non-delegation principle, which would confine the morally laden task of drafting criminal law statutes to elected officials in the legislature. Chevron is essentially a delegation doctrine, recognizing the delegation of interstitial lawmaking authority from the legislature to the administrative agencies. Since dual-purpose statutes serve criminal functions, allowing agency interpretations deference essentially puts the agencies in the role of filling in the criminal law and thereby violates the non-delegation principle. Agency deference ought to be limited to civil law just as the rule of lenity ought to be limited to the criminal law.

3. Legislative Solutions to the Too Much Lenity Problem

As I hope to demonstrate in this Article, I think that there are solutions to the too much lenity problem. Before getting to my preferred solution in Part IV, I discuss in this Section a possible legislative response and the difficulty it faces.

One possible response to the problem of too much lenity is for Congress to draft a separate criminal tax code and civil tax code. The problem of too much lenity arises when a criminal tax law refers to the language of a civil tax law. For instance, § 5871 states, “Any person who violates or fails to comply with any provision of this chapter shall, upon conviction, be fined not more than $10,000, or be imprisoned not more than ten years, or both.” The phrase “this chapter” refers to chapter 53 of the Internal Revenue Code, which governs the taxation of machine guns, destructive devices, and certain other firearms. It thereby requires substantive tax laws within chapter 53 to now perform double duty, assigning civil tax liability and serving as part of the criminal actus reus for § 5871.

Separating the two contexts through drafting may seem a reasonable solution at first, but thinking through how such a solution could be carried out leads to a primary difficulty. How could the legislature be able to draft language regarding the violation of tax law without referring to such laws? The content of the crime set out in § 5871 is that someone violated the tax law. And if this violation of the tax law is what we hold to be criminal, then it is hard to see how the criminal statute could be drafted without reference to the civil tax law.

The act/omission distinction partly explains the issue at hand. The distinction is ordinary and, so, should be familiar to most. To water a plant involves carrying out some willed bodily movement, an action. Omissions can best be understood negatively as the absence of a certain act. If you have agreed to water your friend’s plants while they are on vacation, then your failing to do so is an omission—an absence of the act of watering. The law typically criminalizes acts; a major exception is in tax law, where omissions are criminalized.

Consider the language of 18 U.S.C. § 1584, which punishes “[holding another person] to involuntary servitude.” Holding another person to involuntary servitude is an act. The statute reflects the prohibition against involuntary servitude laid out in the Thirteenth Amendment but, importantly, does not directly reference the Thirteenth Amendment. Since § 1584 assigns punishment to an act, it need not refer to any other provision. It can merely replicate the language of the Thirteenth Amendment and punish holding others to “involuntary servitude.” And although this is an instance of replication between the Constitution and a statute, it is not hard to see how the same could be accomplished with replication between criminal law and civil law. The civil code can set out civil penalties for the conduct of such-and-such act and the criminal code can set out criminal penalties for the conduct of such-and-such act without either needing to directly reference the other.

In contrast, I.R.C. § 5871, and tax crimes more generally, punish non-compliance with respect to some legally required conduct, an omission. Since the omission is defined by the required conduct that one is omitting to do, one cannot spell out the omission without reference to the law that sets out the required conduct in the first place; insofar as that required conduct is a matter of civil tax law, that means that the criminal tax law must refer to the civil tax law. I.R.C. § 5871 must refer to § 5821 since § 5821 sets out the required conduct, the omission of which is punishable.

C. Higher-Order Vagueness

To make it easier to talk about the rule of lenity, let us stipulate another law and some facts about language. Suppose that there is a law prohibiting driving dangerously. The safe driving statute reads:

Whoever operates a motor vehicle or motorcycle on the public roads or highways at a dangerous speed, having regard for width, traffic, use, and the general and usual rules of such road or highway shall be fined not more than twenty-five dollars.

The half-fictive statute is based on former Oregon General Code Section 12603, which was upheld as a valid statute in State v. Schaeffer.

Table 1.  Table of Stipulations (Stated Again Infra)

Statute’s Meaning

Analysis of Statute’s Meaning (Includes Borderline Instances)

Citizen’s Mental Representation of Statute’s Meaning (Only Prototypical Instances)

Dangerous Driving

60 mph or faster

70 mph or faster

Clearly Dangerous Driving

70 mph or faster

80 mph or faster

Clearly Clearly Dangerous Driving

80 mph or faster

90 mph or faster

Note: I encourage the reader to refer to this table while working through the following paragraphs. In order to state the problem, some unusual and technical locution must be used, so the graphical component of this table will aid in comprehension.

All reasonable people will admit that what counts as dangerous driving admits of borderline cases and is, thus, a vague predicate. Suppose by stipulation that 60 miles per hour (“mph”) is the cutoff for driving dangerously on Birch Avenue at 10 a.m. on Wednesday—one is dangerous if and only if one is driving at 60 mph or faster. Of course, driving at 60 mph is not prototypically dangerous, it is instead a borderline case. In fact, it is the border! Let us then stipulate that driving on Birch Avenue is clearly dangerous if and only if the car is going 70 mph or faster.

When a person reads the safe driving statute, their mental representation includes only these prototypical, clear instances of dangerous driving, or so the story of fair notice goes. Driving at 60 mph, borderline dangerous driving, never crosses the mind of Average Joe as dangerous as he drives down Birch Avenue at 60 mph. Thus, when Joe goes on trial, the judges apply a rule of lenity. They construe the statute to mean that Average Joe can only be found guilty for dangerous driving if he has driven clearly dangerously, not just borderline dangerously. To do otherwise would be unfair to his natural reading of the statute and violate fair notice as a rule of law value. So a rule of lenity, which caters to expectations, now requires judges to only find a defendant guilty of dangerous driving if the car was moving at 70 mph or faster, for it is these speeds that are clearly dangerous. Joe has not violated the safe driving statute, the court rules.

From here, the story unravels. The key observation is that someone who knows about the rule of lenity will now actually have a narrower realm of expectation. Recall that the rule of lenity, as a canon of statutory interpretation, assigns meaning to the statute. After Joe’s trial, the meaning of the statute changed from prohibiting dangerous driving to prohibiting clearly dangerous driving. So suppose Steve knows that courts have applied the rule of lenity with respect to the safe driving statute because he read the opinion from Joe’s verdict. Whereas Joe read the statute to mean that “dangerous” driving is prohibited, Steve rightly reads the statute to mean “clearly dangerous” driving is prohibited. The ultimate authorities on the meaning of statutes are the courts, and the courts have stated that the safe driving statute means do not drive clearly dangerously. Steve knows from reading the opinion from Joe’s case that if he drives dangerously but only barely so such that he is still a borderline rather than prototypical case of dangerous driving, he will then be outside the ambit of the statute. The rule Joe follows is do not drive dangerously. The rule Steve follows is do not drive clearly dangerously. Since Joe and Steve have different propositional contents for the rules that they are following, they will also have different mental representations. If Steve expects that he will only be in violation of the statute for clearly dangerous driving, he will conjure the mental image of a prototypical clearly dangerous speed, not a borderline clearly dangerous speed. In other words, if mental representations of concepts are just those of prototypical instances, as discussed in Section I.B., then the mental representation that Joe has is of clearly dangerous driving while the mental representation that Steve has is of clearly clearly dangerous driving. The crux of the issue is that “clearly dangerous” is itself a vague predicate—what counts as clearly dangerous driving admits of both clear and borderline cases. This is the recursive phenomenon of higher-order vagueness, vagueness about the borderline cases.

Driving at 70 mph is a borderline case of clearly dangerous driving. Driving at 70 mph, however, is not clearly clearly dangerous driving. It is merely clearly dangerous. The mental representation of dangerous driving that Steve has upon reading the statute with the rule of lenity in mind—the propositional content of which is do not drive clearly dangerously—is driving at 80 mph or greater. Thus, Steve does not expect to be found guilty of dangerous driving when he drives at 70 mph. Applying exactly the same sort of reasoning that justified having the rule of lenity in the earlier case, a court system ought now to adopt a double rule of lenity to deal with the issues caused by second-order vagueness; otherwise, they will violate Steve’s expectations and the rule of law value of fair notice. Steve can be found guilty of dangerous driving only if he drove clearly clearly dangerously—at 80 mph or greater.

Such reasoning can continue ad-infinitum, adding the clearly adverb with each iteration of higher-order vagueness. In order to protect fair notice, there must be the triple rule of lenity, the quadruple rule, the quintuple . . . . But surely this is absurd. Since we plainly ought not adopt an infinite rule of lenity—lest we let many dangerous drivers go free—and fair notice does seem to be an important rule of law value in criminal law, something has gone quite wrong. Citizens who read a statute after the rule of lenity has been applied are failing to receive fair notice of punishment. Call this the “higher-order vagueness problem.”

Many readers, when presented with my argument above, have responded that the court ought to draw clear boundaries in order to avoid the higher-order vagueness problem. On their account, instead of changing the meaning from dangerous to clearly dangerous, the court should instead state something akin to “we hereby stipulate that any speeds at 70 mph or greater will count as dangerous driving for the purpose of the safe driving statute.” Whereas “clearly dangerous” is vague, “70 mph or greater” is a bright line rule. No problem of higher-order vagueness is presented for “70 mph or greater.” Steve, when reading this opinion, should have a clear mental representation that 70 mph driving is prohibited by law.

The problem with such a response is that it fails to notice that this discussion has thus far been using elliptical construction to hide the context dependence of the statute. The safe driving statute states that the notion of dangerous speed must be understood in the context of “width, traffic, use, and the general and usual rules of such road or highway.” Even if the court draws clear boundaries in one context, it leaves the other contexts open. 70 mph is a clearly dangerous speed for driving on Birch Avenue at 10 a.m. on Wednesday. But what counts as a dangerous speed on Grove Street at 8 p.m. on Saturday or MLK Boulevard at 4 p.m. on Tuesday? Surely, the court cannot delineate what counts as dangerous for every width, traffic, use, and the general and usual rules of every road and highway. And what of vague predicates that reject quantification altogether, such as the No Vehicles in the Park statute? How would a court draw up a bright line rule for the meaning of “vehicle”? The courts are severely limited in their ability to draw bright line rules. In most cases, they must simply apply the rule of lenity to restrict the meaning of a vague statute to only its prototypical instances, thus leading to the higher-order vagueness problem.

1. Technical Bookkeeping

For most legal scholars, the above Section should be convincing on its own. For these scholars, I recommend skipping this addendum on the more technical workings of the intuitive story set out above. Those more inclined to debate the theoretical foundations of law may disagree with how I have presented the issues above. Here, I respond to such disagreements.

In the above example of Steve and Joe, some theoretical premises were implicit in how I laid out the example. Premise one, legal realism is false. Premise two, judges assign meaning when applying the rule of lenity. Premise three, there is a fact of the matter about the borders of vague predicates, but such facts are unknowable (in other words, epistemicism). The higher-order vagueness problem is not dependent on these premises. Even if all three premises were false, I would need to revise only the manner in which the problem is laid out, not the substance.

The first two premises get to at what point Steve can rightly have the expectation that the law only punishes clearly dangerous driving. For instance, suppose the first premise is false and legal realism is true. According to legal realism (or, more precisely, legal realism as characterized by H.L.A. Hart), the law is whatever a judge will say it is. If that is the case, then Steve need not wait for the court to actually apply any rule of lenity for he knows they will. Legal realism states that the fact the court will apply the rule of lenity makes it currently the case that the statute has a narrow meaning. And if the future fact that judges will apply the rule of lenity is current law, then Steve should think, even before Joe’s case is heard, that the law prohibits clearly dangerous driving. The only difference here is a matter of timing. Was the meaning of the statute made narrow by the rule of lenity or was it always narrow since the rule of lenity will be applied when the meaning of the statute is litigated? Either way, the problem of higher-order vagueness stands.

Regarding the second premise, recall the earlier argument in Part II that the rule of lenity stipulates rather than figures out what the statute means. Though canons of statutory interpretation typically seek to figure out the existing meaning of a statute, substantive canons like the rule of lenity instead assign meaning to a statute based on normative considerations. The rule of lenity is not a rule of interpretation in substance since it is not concerned with figuring out what the words mean. When the courts are applying the rule of lenity, it is often within the space of indeterminacy, where meaning has run out. That courts change, rather than interpret, the meaning of a statute when they apply the rule of lenity (premise two above) was a key part of how I originally framed the higher-order vagueness problem.

Suppose, arguendo, the second premise is false and that the rule of lenity is a way of uncovering the existing meaning of the statute. That is, the safe driving statute already has a narrow meaning before it is ever litigated, and in litigation, judges are merely uncovering the existing meaning rather than changing the meaning to implement normative principles. This would make the rule of lenity a rule of statutory interpretation in substance. Even so, the higher-order vagueness problem remains. Again, the only thing that changes is that Steve, if he understands the already existing meaning of the statute, should think that only clearly dangerous driving is prohibited without needing to know about Joe’s case. As with the legal realism premise, the only change here is a matter of timing.

Finally, I have been speaking as if there is a definite fact of the matter about the category membership of borderline instances of a predicate and that we do not know such facts. I find the supposition of epistemicism an easy way to talk about vagueness, but its falsity does not solve the higher-order vagueness problem. The higher-order vagueness problem arises from the general features of vagueness that all theories of vagueness must accommodate:(1) vague predicates have borderline cases that cannot be clearly categorized either as or as not members of such predicates; (2) when reading a vague statute, the reader’s mind tends to conjure up only the clear cases and not the borderline cases; and (3) the question of which items are clear or borderline cases of vague predicates is itself infected with vagueness, thus necessitating distinctions between, for example, clearly clearly dangerous driving and borderline clearly dangerous driving. Features (1) and (2) necessitate a rule of lenity to provide fair notice, and feature (3) kicks the problem one level up each time that the rule of lenity is applied such that features (1) and (2) now apply to the higher level. All three features are theory-independent phenomena.

IV. LENITY AS EXCUSE: REVISING THE DOCTRINE

I have thus far noted the discrepancy between the rule of lenity’s purpose and outcome as well as three doctrinal problems that arise from the discrepancy. The rule of lenity cannot resolve cases of linguistic ambiguity. The rule of lenity extends into civil law. The rule creates higher-order issues of fair notice. Further, I have argued that such problems are foundational to the rule of lenity as it is currently applied. If my arguments are sound, then we must revise the jurisprudential approach to indeterminate criminal law at the foundation. But what should such revisions look like? This Part examines the fundamental nature of the courts’ current lenity jurisprudence and how it ought to be rectified in a way that maintains rule of law values.

On my diagnosis, the issue is that the courts have understood the rule of lenity to be a canon of statutory construction. As a canon of statutory construction, it determines the meaning of the statute to which it applies. Call such a doctrine the semantic rule of lenity. The meaning of statutes is not the right instrument by which to implement the demands of notice in punishment. As I have thus far argued in this Article, statutory interpretation is too blunt a tool for the fine purpose of protecting fair notice.

In some sense, it should not be surprising that the semantic rule of lenity runs into technical problems. The originators of the rule of lenity likely did not foresee the three doctrinal problems I have listed here. The rule of lenity, which traces back to sixteenth century England, predates both the advent of the Internal Revenue Code and contemporary linguistics. Ideally, we should like to reconceptualize the rule of lenity such that we avoid the three doctrinal problems while maintaining its function carrying out rule of law values.

The semantic rule of lenity should be replaced by what I will call the lenity excuse. There ought to be an affirmative defense available to defendants in those instances in which the defendants’ actions were within the penumbra of an indeterminate criminal statute without changing the meaning of that statute. Because the new rule would operate as an excusing condition, the mere fact that the law did not unambiguously criminalize a defendant’s conduct would be sufficient to negate any liability for criminal defendants in the same way that duress or insanity would negate liability. In this way, lenity would function like other excuses (such as duress or insanity) that absolve defendants of criminal liability when it would be unfair to punish them.

In the remainder of this Part, I will argue that my proposed revision to lenity would not only be pragmatic, solving the three doctrinal problems that plagued the semantic rule of lenity, but also conceptually fruitful, helpfully tying together the purpose of lenity with that of other excuses.

A. The Categorical Unity of Lenity and Excuse

In order to understand the categorical unity between lenity and excuse, one must first understand two foundational concepts and their relation to one another: affirmative defense and excusing condition. In criminal law, the establishment of an affirmative defense will absolve the defendant of criminal liability even if the prosecution has established case that all elements of the offense are present. A paradigmatic example is the excuse of duress. Suppose, for instance, that a defendant has stolen cash from his friend’s wallet because a thug made a credible threat to kill the defendant unless the defendant stole from his friend and gave it to the thug. Even if the prosecution can establish that all elements of the larceny offense are present, the defendant may appeal to the defense of duress, which absolves a defendant of criminal liability if the defendant was threatened with “unlawful force . . . , which a person of reasonable firmness . . . would have been unable to resist.”

For the second concept, that of excusing conditions, this Article will follow the analysis by H.L.A. Hart. Rather than defining the term, Hart provides a non-exhaustive list of its members: “Mistake, Accident, Provocation, Duress, and Insanity.” Unlike other analyses of excuses, which tend to center their focus around the defendant’s moral responsibility, Hart’s analysis of excusing conditions focuses on their role in protecting liberty. Hart finds excuses to be valuable because they provide for citizens the valuable ability to predict in what instances one will be punished and to avoid such instances through one’s own will.

Return to the duress example above. If the defendant stole from the defendant’s friend because a gunman threatened to kill the defendant otherwise, the duress excuse would absolve the defendant of criminal liability. If there was no such excuse available, then it would be very difficult for a citizen to ensure they avoid punishment. In such a system, the citizen cannot guarantee that they will avoid punishment as a result of two factors working in conjunction. Firstly, to the extent one has no control over what a violent gunman will do, one cannot guarantee that one will not be threatened by a gunman. Second, to the extent that it is near impossible to resist the orders of a gunman, one cannot guarantee that one will not commit the crime the gunman demands. Thus, without a duress excuse, whether or not one will go to jail would depend on the unpredictable whims of a gunman. In such a case, it cannot be said that the individual had a genuine choice to avoid the law’s criminal sanctions. The duress excuse eliminates this worry by ensuring that, in this unpredictable circumstance, one will be saved from punishment.

As argued in Section I.B, fair notice of the criminal laws is also an essential part of citizens’ having a genuine choice to avoid punishment. Fair notice is essential because it gives citizens an opportunity to figure out which actions are subject to punishment under the law. For instance, if the government chose not to publish the criminal laws but instead keep them private, an ordinary person would not have the ability to figure out which actions will be met with punishment. Although Hart himself did not explicitly consider the question of whether or not fair notice doctrines should be understood as excuses, his theory and its implications are clear. In order to protect the choice to avoid punishment, Hart plainly states that citizens must be given the ability to “find out, in general terms at least, the costs they have to pay if they act in certain ways.”

Thus, lenity’s purpose of protecting the choice to avoid punishment aligns the doctrine more closely with the domain of excuse than the domain of statutory interpretation. Given the theoretical unity between lenity and excuse, it may be instructive to look to how other excuses are applied in the criminal law and consider whether lenity should be given the same treatment.

Many excusing conditions, such as duress and insanity, are employed as affirmative defenses under the law. In these instances, we take it as obvious that if what we want to do is free the defendant, then we should do that directly by permitting a defense, rather than indirectly through changing what a statute means. I propose here that the same treatment be given to lenity, allowing for defendants to simply avoid punishment in instances where the statute was indeterminate with respect to the defendant’s behavior without constraining the meaning of that term as the current semantic rule of lenity does.

Excuses are often understood in contrast to another category of affirmative defense: justifications. In the legal context, both serve as affirmative defenses requiring acquittal even where the prosecution has established the case that all elements of the offense are present. However, as a moral matter getting at the theoretical grounding of the doctrines, the two categories of defense diverge on the question of why acquittal is required. Justifications defeat what would otherwise be a prohibition against acting in a particular way. It turns what would ordinarily be prohibited into a permissible act. A standard example is self-defense. Killing another is ordinarily impermissible, but not so if done in self-defense. Someone who kills another in self-defense has done nothing wrong. This contrasts sharply with the nature of excuse, which presupposes the wrongfulness of the act done. To use the duress example, we would say that the defendant did something wrong by stealing, but only did so because the coercive threat left them no choice otherwise. Both excuse and justification absolve the actor of criminal liability, but justifications do so by negating the impermissibility of the actions whereas excuses affirm that the actions were impermissible but absolve the actor for a reason standing outside the wrongfulness of the act itself.

By leveraging the distinction between a justification and an excuse, one can see why the lenity excuse better accords with our intuitions regarding notice as a rule of law value as opposed to the semantic rule of lenity. Like justification, the semantic rule of lenity redraws the lines of what is permissible and what is impermissible behavior. By giving the statute a narrow construction, the semantic rule shrinks what counts as impermissible behavior. But in many cases in which the rule of lenity is applied, we see that the defendants really did do some harmful act, such as disposing of an object that could serve as evidence and transporting a stolen airplane across state lines. Therefore, the semantic rule of lenity fails to accord with the reason why we acquit defendants who conduct non-prototypical criminal activities. The reason we acquit them is that it would be unfair, from a rule of law perspective, to punish an act without clear notification that such an act would be punished, not that the defendants have clean hands.

Under the lenity excuse, a criminal defendant would be absolved of liability for their actions if their actions were not clearly within the meaning of the criminal statute without needing to change the meaning of said statute. The underlying conduct rule, for example, the rule prohibiting dangerous driving, does not change. The law can stand both for the proposition that a citizen acted wrongfully and the proposition that punishing the individual, despite their wrongful acts, would violate our rule of law principles. Refashioning lenity from a canon of statutory interpretation to an excusing condition essentially allows us to have our cake and eat it too. The lenity excuse protects the safe path principle while still maintaining the best interpretation of the statute.

B. Solving the Three Doctrinal Problems

Recall the earlier discussion of the hypothetical Bank Safety Act, which criminalizes starting a fire within one hundred feet of a bank. We stipulated there that “bank” was ambiguous between a river bank and a financial bank and that Defendant A had set fire next to a financial bank and Defendant B had set fire next to a river bank. The semantic rule of lenity does not resolve such a situation since either reading of “bank” would absolve only one of the two defendants of guilt. However, since neither defendant received fair notice that their act would be criminalized due to the linguistic ambiguity of the Bank Safety Act, the fair notice purpose would require absolving both of guilt.

Under a lenity excuse regime, the situation is neatly resolved. The court, for all defendants, need only ask the question whether either one of their actions were unambiguously criminalized by the law. Ex hypothesi, due to the linguistic ambiguity inherent in the statute, neither defendants’ actions were unambiguously criminalized, so both defendants would be absolved of culpability. The same would go for any case of linguistic ambiguity in criminal statutes, mutatis mutandis. Thus, the outcome of the lenity excuse is consistent with what fair notice demands.

The semantic rule of lenity problematically extended past its purpose of fair notice in punishment by applying in purely civil contexts. With the lenity excuse, a statute used in both criminal and tax law contexts can be given the best interpretation rather than the narrowest construction, so tax law’s civil purposes are protected. Nevertheless, the lenity excuse can still apply in criminal contexts, which is the context in which fair notice is required due to the special status of punishment. Although the meaning of an inscription must, for linguistic reasons, stay constant across contexts, no such principle of consistency applies to affirmative defenses at law. The law can, and does, permit defenses in criminal law that are not available in civil law. Insofar as we have a principled reason, namely the special status of punishment, a company like Thompson/Center Arms Co. could leverage the excuse of lenity to avoid punishment for the manufacture of Thompson/Center kits, but it should not have such an excuse when courts are determining its civil tax liability, such as whether the firearms excise tax applies to the production of Thompson/Center kits. This solves the too much lenity problem.

The semantic rule of lenity, by assigning meaning to a statute, led to fair notice problems caused by higher-order vagueness. With the excuse of lenity, there is no need to assign any particular narrow construction to the statute itself. Without any new assignment of meaning, studious potential criminals have no reason to have different expectations of what violates the criminal law after reading an opinion employing the lenity excuse. The court instead affirms, for instance, that borderline dangerous driving of 60 mph is still dangerous, though it absolves the defendant who drove at 60 mph of criminal liability since that would not have been the defendant’s expectation from reading the statute. The sovereign command remains don’t drive dangerously rather than changing to do not drive clearly dangerously, and no double rule of lenity is required. Joe, who drove at 60 mph will be able to benefit from the lenity excuse because his actions were not clearly prohibited by the vague safe driving statute; Steve, who drove at 70 mph will not be able to benefit from the excuse. Steve’s driving was clearly dangerous and, since the conduct rule remains do not drive dangerously as opposed to do not drive clearly dangerously, Steve has received fair notice that his actions would be punished. The excuse of lenity thereby solves the higher-order vagueness problem.

The reader may here object that a problem parallel to the higher-order vagueness problem nevertheless remains. Can Steve not say that he thought himself to be following the law given that there is an excuse of lenity that absolves borderline dangerous driving of culpability? To put the force of the counterargument another way, what is the difference between saying, as the semantic rule of lenity does, that clearly dangerous driving is prohibited and saying, as the excuse of lenity does, that dangerous driving is prohibited but borderline dangerous driving is excused?

The key distinction is that although the semantic rule of lenity is directly construing the statutory conduct rule, an excuse is not meant to guide conduct. It would be quite odd to think that the presence of excuses in the criminal law is tantamount to the law’s saying, “If you are planning to commit homicide, please make sure you are insane or under duress.” Excuses are instead best understood as addressing the government on how to adjudicate questions of criminal culpability. The criminal law permits excuses sotto voce.

This sotto voce feature of excuses may be analogized to the same in statutes of limitations. A five-year statute of limitations on assault, for example, is not to let citizens know that they are permitted to assault others so long as they can lay low for the next five years. If the government were to amend the statute of limitations to seven years, an assaulter cannot complain of unfairness. It would be on its face ridiculous for the criminal to complain, “I assaulted someone yesterday thinking I would only have to hide for five years, not seven. You are treating me unfairly!” Likewise, because excuses are not conduct rules, the rule of law principle that citizens be given fair notice of which conduct is prohibited does not apply to excuses. Therefore, the excuse of lenity does not require a “second-order” excuse of lenity, thereby resolving the higher-order vagueness problem.

C. Other Justifications for the Semantic Rule of Lenity

I have argued in the previous Section that an excuse of lenity best aligns the doctrine with its rule of law purpose of fair notice while the semantic rule of lenity does not. Though the value of fair notice is the most often cited justification for the rule of lenity, it is certainly not the only justification for such a long-standing and august doctrine of criminal law. Ideally, the excuse of lenity would be consistent with these other justifications as well—it would be a shame to throw out any babies with the bathwater. In this Section, I consider the other justifications for the rule of lenity and demonstrate how the excuse of lenity is consistent with such aims.

The first set of reasons significantly different from fair notice for having the rule of lenity includes those that are still closely connected to restricting the scope of criminalization. The rule of lenity “constrains the discretion of law enforcement officials”; it is a speed bump against over-criminalization in the United States; and it protects the (relatively) politically powerless citizens who would have a hard time organizing to change the criminal code. Since the excuse of lenity likewise works to restrict the scope of criminalization by offering a functional near-equivalent of the rule of lenity in the criminal context, all of these purposes are also carried out by the excuse of lenity.

Another category of reasons in favor of the rule of lenity involves the notion that criminal law is solely the province of the legislature, the non-delegation principle. Again, the functional near-equivalence between the semantic rule of lenity and the excuse of lenity within the criminal context will explain why the excuse of lenity can do much of the work that the semantic rule of lenity currently does. Whenever the excuse of lenity applies, because the excuse will be dispositive of the case, the courts need not resolve the indeterminacy of the criminal statute at hand. Courts will need to resolve penumbral issues in dual-purpose statutes, but this will not make a difference for criminal liability since the excuse of lenity will be available when defendants fall into the penumbra. Since defendants can leverage the excuse against vague statutes in court, it will, like the semantic rule of lenity, put the impetus on Congress to draft clearer statutes.

There is one reason in favor of the semantic rule of lenity that does not apply to the excuse of lenity: the semantic rule of lenity has a long history. The rule of lenity originated in sixteenth-century England and has survived in application to the present day. This is indeed a value lost if we were to do away with the semantic rule of lenity, but its importance ought to be put in proper perspective. Though the semantic rule of lenity’s history is long, canons of statutory interpretation are not law and do not have precedential effect.

V. COUNTERARGUMENTS AND RESPONSES

In this Part, I consider some arguments specifically against the existence of the too much lenity problem, the higher-order vagueness problem, and the excuse of lenity. The central counterarguments are (1) the tax law is best construed narrowly in civil contexts, so the “too much lenity problem” is actually a feature, not a bug, of the semantic rule of lenity, and (2) if the legislature were to announce that there is no rule of lenity, then the higher-order vagueness problem dissipates due to the fact that individuals are now on notice that statutes will be construed according to the intent of the legislature. Both arguments are important because they go to the theoretical foundations of this Article.

A. Tax Law Would Be Better Off If the Rule of Lenity Applied

The first counterargument puts forth that the problem of too much lenity is no problem at all since the tax code ought to be subject to strict construction, resolving any indeterminacy in favor of the taxpayer. On this view, it is unfair to tax a citizen without clear say-so by statute. The application of the rule of lenity to the tax law is to be celebrated, not decried. Some European nations, for instance, have strict-construction tax systems favoring taxpayers. Such a response, I contend, fails to comport with the differential attitude citizens should have with regard to the administration of distributive justice and retributive justice.

First, tax law is the government’s most important lever in carrying out principles of distributive justice. Distributive justice concerns how institutions should be designed to fairly distribute the benefits and burdens of societal cooperation. Our progressive income tax system carries out a democratically determined vision of distributive justice under which tax obligations directly correspond to one’s income earned in the marketplace. The statutes provide the skeletal structure for this vision, the corpus of which is fleshed out by the judicial and administrative authority. To undo the interstitial authority is to partly undo the very aims of the tax code.

On this picture of tax justice, to deviate from the best interpretation of a statute in favor of a narrow interpretation of a statute not only undoes what distributive justice would require, but thereby also partly undoes the provision of a valuable moral service by the government. The tax system provides valuable coordination between citizens to hire an expert to tally up what justice requires of them and hold each other to that tally. Since taxpayers have moral reason to pay what justice requires, following the best interpretation of the tax law helps their aims rather than impeding them.

Second, though related to the first point, the purpose of tax law is distinct from the purpose of criminal law in that the imposition of tax does not typically aim to serve a deterrence function. After all, a tax on income is not meant to discourage the earning of income. There is no implicit public moral rebuke attached to civil tax liability as there is for criminal liability. Although there are exceptions, the principal purpose of the tax law is to collect (and sometimes distribute) revenue in a just and efficient manner. Notice is most critical when the statutes are intended to guide citizens’ behavior since ambiguously drafted statutes cannot properly serve this guiding function. After all, a citizen cannot use a statute to guide their behavior if they cannot figure out what the statute means.

To bring out this point, we can think of the perspective of a hypothetical idealized taxpayer with regard to the tax law. The taxpayer understands that they have a moral obligation to contribute a certain amount to the common pool of resources by which we fund the various functions of government. However, it is quite unlikely that the taxpayer could even estimate how much they should contribute if they were to reason purely from philosophical first principles or that they would know much about the content of such first principles. Even if the taxpayer resolves the coarse-grained question regarding their obligation to pay taxes, it is unlikely that they will even be able to approximate an answer to the fine-grained question of how much taxes they are morally obligated to contribute as a matter of justice.

One way for the taxpayer to resolve the fine-grained question is to defer the calculations to appointed experts in the legislature and the Treasury. On this account, the taxpayer can carry out their ordinary business without worrying about what constitutes their fair share contribution and, at the end of the year, rely on the tax law and the aid of administrative officials to figure out what that fair share is given the activities they engaged in and their results. Instead of having to think through how the tenets of John Locke and Jean-Jacques Rousseau apply to him as a citizen, the taxpayer can just fill out an IRS Form 1040-EZ. This sort of division of labor is critical due to the difficulty of answering the fine-grained questions of political morality and the limited resources that citizens have to put towards such inquiry.

For these taxpayers, notice is only relevant insofar as they need the information to pay what they owe. Since the taxpayer is not treating tax liability as a cost or benefit of such-and-such action, clarity in laws is actually more important for the administration of such laws rather than being governed by them. This is also why the tax law can bear such enormous complexity. Whereas conduct rules primarily meant to dictate citizens’ behavior must be drafted simply so that citizens can understand the conduct rules, tax laws can be drafted with greater complexity because they are primarily addressed to administrators and judges who have expertise in tax law. Though idealized for purposes of exposition, this sort of narrative is consistent with both the theoretical work in political philosophy and the empirical research that often, though notably not always, finds that tax rates have no effects or very small effects on taxpayer behavior.

None of this is to deny the proposition that there would be something good provided by having a tax law system where taxpayers can more easily figure out their tax liability. For instance, if a taxpayer does not know how much taxes they will have to pay at the end of the year because the tax laws are too vague, it may lead to the taxpayer over- or under-saving for the forthcoming tax liability. Instead, my argument is merely that the civil tax law lacks many of the features that make fair notice far more important in criminal law. Given these differences—for civil tax law, there is no moral condemnation, no punishment, and no intended deterrence effect—we have good reason to think that tax law ought not follow the stringent fair notice requirements of criminal law.

B. Statutory Notice That Fair Notice Laws Have Been Repealed

The second counterargument contends that if courts were to get rid of the rule of lenity altogether in conjunction with notice of such at the legislative level, then the problem of higher-order vagueness would not arise. Here, the central idea is that the legislation stating that the rule of lenity does not apply to the criminal code would itself stop citizens from forming any expectations about the rule of lenity.

Such an approach has a fundamental problem. A notice that there would be no lenity provided would amount to notice that there is no fair notice. This becomes plain if we recall the fictive story underlying the fair notice doctrine. The reason a citizen needs fair notice is that they may think they are following the law when they are not. To put the point another way, citizens would simply be on notice that they cannot find comfort in their natural understanding of a criminal statute. That fair notice has been abrogated still stands.

The critic might then respond that the legislature ought to impose a statutory single rule of lenity. Given my argument for the higher-order vagueness problem, having just one rule of lenity would be arbitrary—what reason do we have to stop at one rather than two?—but such violations are forgivable. The law is in the business of line drawing and, since the hair-width difference between what is inside the line and outside the line can hardly be a difference-maker, line drawing is often an arbitrary matter.

The bigger issue is that having one rule of lenity does not resolve the fair notice problem so long as the rule of lenity remains semantic in nature. The same conclusion about having no rule of lenity applies to the single rule of lenity. So long as individuals understand that the meaning of a statute has changed from an application of the rule of lenity, then to announce by statute that there will be no more higher-order rules of lenity will only violate individuals’ expectations that their actions are within the bounds of legally permissible behavior. The selection of any n-tuple rule of lenity cutoff is arbitrary and will disrupt fair notice for the n+1th order reader of the statute.

CONCLUSION

This Article has argued that the semantic nature of the rule of lenity leads to three problems in which the rule breaks away from its purpose of providing fair notice in criminal law. The rule of lenity cannot deal with linguistic ambiguity. Some criminal statutes also play civil functions, thereby transferring the strict construction of the rule of lenity from criminal contexts to the civil context. Once the courts construe the meaning of a statute to include just the clear cases, it then creates a fair notice burden regarding the question of what counts as the clear cases, which is itself a vague matter.

To resolve these issues, we ought to replace the semantic rule of lenity with an excuse of lenity. Excuses and fair notice share the common denominator of providing ordinary citizens the safety of choosing to avoid punishment, so having lenity provided as an excuse would more closely align the rule with its purpose. An excuse of lenity would provide the same benefits as the semantic rule of lenity, restricting the scope of criminalization and maintaining criminal law within the province of the legislature, without the drawbacks of having a semantic rule.

 

96 S. Cal. L. Rev. 397

Download

Assistant Professor of Law and Philosophy, University of Southern California. Thank you to Scott Altman, Jody David Armour, Jordan Barry, Thomas Bennett, Jonathan Choi, Robin Craig, Noël Cunningham, William Eskridge, Felipe Jiménez, Mitchell Kane, Gregory Keating, Adam Kern, Daniel Klerman, Yao Lin, Erin Miller, Michael Moore, Clare Pastore, Marcela Prieto, Robert Rasmussen, Emily Ryo, Daniel Sokol, Kevin Tobia, Gideon Yaffe, Yuan Yuan, Jack Whiteley, participants of the New York University School of Law–Lawyering Scholarship Colloquium, participants of the University of Pittsburgh–Law and Language Group, and participants of the University of Southern California Gould School of Law–Faculty Workshop for their invaluable help. Any errors are mine and mine alone.