Plea Bargaining’s Uncertainty Problem

Article - Volume 101 - Issue 3


In 2003, prosecutors offered Weldon Angelos a fifteen-year plea deal to resolve drug and gun charges. Angelos turned down the offer, was convicted at trial, and was sentenced to fifty-five years in prison.[1] In 2013, prosecutors offered Joseph Tigano a ten-year deal if he pleaded guilty to growing 1,400 marijuana plants. Tigano refused. After trial, he was sentenced to twenty years.[2] In 2020, Brian Whitton was charged with arson and offered a five-year plea deal.[3] He went to trial and was sentenced to fifteen years.[4]

Modern criminal justice scholarship is increasingly focused on plea bargaining, and with good reason.[5] The Supreme Court tells us that plea bargaining “is the criminal justice system,” accounting for 95% of criminal convictions.[6] Critics assail the “trial penalty”—“the substantial difference between the sentence offered prior to trial versus the sentence a defendant receives after a trial”[7]—as the driving force of this development.[8] The conventional account is that, to avoid draconian post-trial punishments, defendants have little choice but to plead guilty. Those who don’t succumb to the pressure, like Angelos, Tigano, and Whitton, serve as cautionary tales. They receive long sentences primarily, it seems, for taking their cases to trial.

Due to the prevalence of guilty pleas, the trial penalty’s primary impact is not long sentences. Instead, the risk of an extreme post-trial sentence pressures defendants to forego trial and plead guilty. Instances of this more common application of plea bargaining receive less notice but are just as impactful.[9] For example, with three prior “strikes” alleged in an indictment against him, Troy McAlister faced life in prison as he sat in jail awaiting trial on robbery charges. But in 2019, San Francisco voters elected a new District Attorney: Chesa Boudin, a former public defender who ran “on a progressive platform centered on ending mass incarceration.”[10] Two months after Boudin took office, his prosecutors offered McAlister a deal: Plead guilty and be released in a few weeks. McAlister’s attorney opposed the deal, contending that McAlister “would have likely prevailed at trial.”[11] Nevertheless, McAlister pled guilty and was freed the next month.[12] In 2018, the Philadelphia District Attorney’s office, led by progressive icon Larry Krasner, negotiated two plea deals with a similar dynamic: a 13-year deal for a defendant facing a life sentence for a murder committed during a carjacking;[13] and a 3.5-year deal for a defendant facing 40 years in prison for shooting a store owner with an AK-47 during a robbery.[14]

Stories like these make it unsurprising that “there is no part of the criminal process more pilloried than the plea bargain.”[15] Still, as Dan Epps notes, “scholars have struggled to pinpoint why exactly plea bargaining is problematic.”[16] As a result, commentators have, so far, united only behind their distaste for the phenomenon, rather than specific reforms.[17]

The complexity begins with a recognition of the trial penalty’s inevitability. Defendants plead guilty when they perceive a benefit to doing so. And if defendants like Troy McAlister receive a benefit, those who are convicted after trial suffer a trial penalty by comparison. The only way to eliminate this dynamic would be to prohibit guilty pleas. Yet as George Fisher has documented, plea bargaining triumphed in America because it is an inevitable outgrowth of an adversary system characterized by independent prosecutors, crowded dockets, and severe penal laws.[18] Sporadic attempts to abolish plea bargaining have been short-lived, and other countries where plea bargaining is less prevalent are moving towards, not away from, the American model.[19]

Further complicating the picture, the most immediate impact of abolishing plea bargaining could be to increase the severity of a notoriously punitive system.[20] The best evidence of this comes from Alaska, a jurisdiction that forbade plea bargaining in 1975. During the short time the ban was enforced, guilty pleas continued at almost the same rate (without concessions from the prosecution) and “[s]entences became more severe.”[21] As in Alaska, real-world efforts to eliminate plea bargains typically stem from a desire to increase (not decrease) sentencing severity.[22] New York’s infamous Rockefeller Drug Laws paired severe mandatory sentences with strict limits on plea bargains.[23] The 1982 “Victim’s Bill of Rights” enacted by California voters similarly prohibited prosecutors from “[p]lea bargaining in any case in which the indictment or information charges any serious felony.”[24] As in Troy McAlister’s case, these prohibitions are typically ignored.[25] But these futile efforts to block plea deals like the lenient one McAlister received illustrate both the practical difficulty of abolishing the practice and the dangers to criminal defendants of doing so.[26]

Confronted with these complexities, reformers often retreat to generic solutions. The most common is reducing sentence lengths. Prodding judges to impose shorter sentences,[27] legislators to eliminate mandatory minimums,[28] and prosecutors to charge fewer and less serious offenses[29] all have the effect of blunting plea bargaining’s sharp edges.[30] But improving plea bargaining is a coincidental side effect of these laudable reforms. Fewer cases and shorter sentences mitigate every coercive aspect of the criminal justice system. It is the equivalent of shutting off the water to fix a leak. It works, but it masks an inability to diagnose the problem.

This Article attempts to isolate the problem with American plea bargaining. Specifically, it strives to identify something about plea bargaining that is both objectionable and distinct from background problems of American criminal law enforcement. The second part of this formula is critical. Critics often denounce plea bargaining for flaws it does not create. For example, the most common critique, illustrated in the opening paragraph, condemns plea bargaining because X defendant was offered five years, went to trial, and received twenty. But this ignores that the long sentence driving the injustice in these examples is a product of the background law—a stark baseline that could become a more (not less) likely outcome in plea bargaining’s absence.

A similar conflation of problems arises when commentators highlight cases where later-exonerated (innocent) defendants accepted a plea deal to avoid a trial penalty.[31] Again, these critiques apply the broader system’s afflictions to plea bargaining. The pressure created by an attractive plea offer is easily criticized if we posit that the innocent defendant would have been acquitted at trial. But there is no way to know. The problem facing an innocent defendant is a wrongful arrest and prosecution, not a subsequent plea offer. And since we don’t know whether someone who pled guilty would have been convicted at trial—resulting in even harsher punishment—we can’t know if the guilty plea mitigated or aggravated the broader system failure.[32]

Perhaps the most compelling critique is the broadest one: that plea bargaining generates perverse incentives for lawmakers, judges, and prosecutors. And the prevalence of plea bargaining certainly alters the American criminal justice landscape. But the incentives plea bargaining generates point in different directions and overlap with independent incentives for statutory and charging severity. Thus, there is no guarantee that reforming plea bargaining by targeting these incentives would lead to better outcomes or greater lenience.[33]

Perverse incentives, excess charging, sentencing severity, and the prospect of the conviction of the innocent are core problems that resonate throughout the American system.[34] Other problems include overbreadth, inconsistency, corruption, poor defense lawyering, and socioeconomic and racial bias. But these problems would exist without plea bargaining. That’s the analytical difficulty for scholars evaluating the much-maligned phenomenon. Since parties bargain in the system’s shadow, those bargains reflect the system’s flaws. Only by separating the flaws that plea bargaining reflects from those it creates can we isolate the problem with plea bargaining.

The first step is distilling plea bargaining to its essence. A plea offer introduces an artificial decision point for the defendant fighting criminal charges.[35] The difficulty in identifying the problem with plea bargaining is explaining what’s wrong with offering defendants this choice. A rigorous assessment of this question cannot focus with the artificial clarity of hindsight on the subset of cases with the worst outcomes. Analysis must extend throughout the process and include the benefits for defendants of pleading guilty, not just the costs.

A broader lens reveals that, at least in terms of its direct effects, plea offers hurt defendants when defendants choose wrong. Contrary to its name, the trial penalty is not a penalty that arises when a defendant goes to trial. Some defendants faced with an exorbitant gap between a plea offer and a potential post-trial sentence suffer no penalty at all. Most obviously, defendants who are acquitted at trial suffer no trial penalty.[36] Others, like Troy McAlister, turn the broad gap into a substantial “plea discount” by accepting the offered deal. Filling out the universe of possible outcomes reveals that plea bargaining harms rather than helps when defendants choose a more punitive option.

If the sharpest critique of plea bargaining is that defendants sometimes choose “wrong,” it becomes tempting to downplay the problem. As Judge Frank Easterbrook points out, we allow people to make bad bets in other contexts—“climb mountains . . . fail to exercise, eat fatty foods, smoke cigarettes”—even when the stakes are high.[37] Rejecting a beneficial plea deal may be most analogous to declining lifesaving medical intervention.[38] Still, there is a difference between tolerating bad choices and making those choices inevitable. In American plea bargaining, “wrong” choices are not only likely to occur but, in some circumstances, almost guaranteed. The danger is highest when the desirability of a plea offer is difficult to assess ex ante—a common occurrence in light of the dearth of information available to criminal defendants.

Some uncertainty is, of course, unavoidable. But sometimes, there is so much uncertainty in evaluating a plea offer that the defendant’s decision resembles a high-stakes gamble, not an informed choice. Revisiting the medical analogy, it is as if instead of a trusted doctor proposing a treatment option along with accompanying insight into its benefits, you received the proposed treatment from an adversary without sufficient information to determine whether it will make you better or worse. This is the strongest objection to American plea bargaining. In certain scenarios, defendants are forced to stake their lives on propositions no more predictable than the spin of a roulette wheel. Such a dynamic is toxic to any rational system of justice.[39] And defendants’ inability to rationally determine whether to plead guilty or go to trial, and the distortion of outcomes that results, is a problem that plea bargaining creates rather than reflects.

This Article contends that it is the uncertainty that plea bargaining introduces—not the pressure to forego trial,[40] the incentives plea bargaining generates,[41] or the danger to the innocent[42]—that represents the phenomenon’s unique, problematic contribution to American criminal justice. We should, of course, pursue reforms to mitigate unwarranted severity and reduce trial inaccuracy. These are distinct problems with clear, direct solutions.[43] But to the extent we seek to fix plea bargaining, the focus should be on reducing uncertainty.

Centering uncertainty as the distinct problem with American plea bargaining helps to identify constructive reforms. Importantly, the uncertainty that matters is not uncertainty about the post-trial outcome. In a jury system, some uncertainty is unavoidable. The uncertainty that matters is uncertainty about whether the defendant should accept a plea offer. Making that decision as informed as possible will not only eliminate the distasteful spectacle of plea-bargaining roulette but should also lead to more consistent plea deals, reducing unfairness (especially along race and class lines) and unwarranted severity.[44]

This Article diagnoses the problem with plea bargaining in three parts. Part I criticizes the overwhelmingly negative academic consensus about plea bargaining not because it is overly grim, but because it targets the wrong villain—often blaming plea bargaining for background flaws of the broader system. Part II uses this insight to identify weaknesses in the common specific critiques of plea bargaining. Plea bargaining is flawed. But plea bargaining’s appeal is not that it works well. Rather its appeal is that, in many cases, it permits an outcome that is preferable to the non-negotiated alternative. Critiques that highlight flaws in plea bargaining while overlooking the similar (or worse) flaws in that alternative miss this critical point. And they risk generating facially appealing solutions that magnify, rather than mitigate, background injustices. Finally, Part III presents uncertainty as the distinct problem of plea bargaining and proposes ways to improve plea bargaining by minimizing uncertainty. Identifying uncertainty as the unique problem introduced by plea bargaining offers two concrete benefits. First, it provides a clear path to improve the plea-bargaining process—an important goal since that largely unregulated process has become the dominant mechanism for resolving criminal cases. Second, by more accurately pinpointing the locus of injustice in the many circumstances where plea bargaining appears to be, but is not, the villain, the analysis directs reform energy to the places where it can be most effective.

I. Plea Bargaining’s Modest Appeal

Plea bargaining is not only the primary mechanism for resolving criminal cases in the United States. It is also a thorny theoretical puzzle that bedevils both criminal justice scholarship and legal doctrine. Yet neither the existence of plea bargaining nor its ascendance is new. Guilty pleas dominated in this country as early as the nineteenth century.[45] By the 1960s, the trend was old news.[46]

Plea bargaining benefits a host of important constituencies. Legislatures see more of their laws enforced at lower financial cost.[47] Judges alleviate docket pressures and reduce the prospect of reversal on appeal.[48] Police and prosecutors reduce their caseloads, avoid trial losses, and obtain cooperation from defendants.[49] Defendants avoid harsher punishments.[50] Witnesses and victims get a final and uncontested judgment. Citizens are spared the obligation of jury service. That this broad spectrum of benefits spreads over many interested parties explains how plea bargaining became so formidable.[51] In American criminal courts, there is little doubt that “plea bargaining has won.”[52]

The legal academy offers a last bastion of resistance. Perhaps the most forceful critique comes from John Langbein, who famously compared plea bargaining to torture. According to Langbein, like medieval inquisitors, “[w]e coerce the accused . . . to confess his guilt”:

[O]ur means are much politer; we use no rack, no thumbscrew, no Spanish boot to mash his legs. But like the Europeans of distant centuries who did employ those machines, we make it terribly costly for an accused to claim his right to the constitutional safeguard of trial. We threaten him with a materially increased sanction if he avails himself of his right and is thereafter convicted. This sentencing differential is what makes plea bargaining coercive. There is, of course, a difference between having your limbs crushed if you refuse to confess, or suffering some extra years of imprisonment if you refuse to confess, but the difference is of degree, not kind. Plea bargaining, like torture, is coercive.[53]

Langbein’s characterization of plea bargaining as coercion has become the dominant view in the legal academy.[54] But questions remain, as exhibited by an influential exchange in the early 1990s across the pages of the Yale Law Journal. There, Bill Stuntz and Bob Scott noted the “puzzling” contrast that “[m]ost legal scholars oppose plea bargaining” while “most participants in the plea bargaining process, including (perhaps especially) the courts, seem remarkably untroubled by it.”[55] Federal Judge Frank Easterbrook illustrated the point, blandly explaining in his short contribution that “[p]lea bargains are preferable to mandatory litigation . . . because compromise is better than conflict.”[56] Stephen Schulhofer offered the traditional academic critique, arguing that “[p]lea bargaining is a disaster. It can be, and should be, abolished.”[57]

Modern academic commentary on plea bargaining coalesces around Langbein’s theme. Defendants, the modern consensus suggests, have little choice but to plead guilty and accept the severe punishment prosecutors desire.[58] The critique should sound familiar. It resonates with critiques of the criminal justice system more generally: prosecutor power,[59] excess severity,[60] and the danger of conviction of the innocent.[61] Indeed, this overlap hints at the primary weakness of the commentary: plea bargaining’s critics often treat background problems of American criminal law as flaws generated by plea bargaining.

The conflation is clear even in Langbein’s celebrated indictment of plea bargaining as torture. Langbein and modern critics are correct that coercion is a key ingredient of American plea bargaining. Defendants do not accept plea bargains because they want to be punished. Defendants plead guilty because they perceive the alternative (a harsher post-trial punishment) to be worse. But the coercion that powers this dynamic is not generated by plea bargaining. It is the product of baseline American criminal law enforcement. Coercion arises when the police officer arrests and the judge denies pretrial release or commands the defendant to appear in court at a future date. Coercion looms on the horizon in the form of a trial, a judge empowered to impose a lengthy prison term upon conviction, and corrections officers authorized to carry out that sentence.[62] That’s the coercion that pushes defendants to plead guilty. And that coercion exists with or without plea bargaining. Thus, Langbein correctly identifies the forces at play but mistakes their source. Plea bargaining isn’t torture, and plea offers aren’t the modern analogue to the thumbscrew and Spanish boot. Today’s analogue to Langbein’s medieval torture chambers is prisons and jails.

Take, for example, Jeannie Suk’s example of a domestic violence defendant:

As the case continues, the defendant may fear losing his job because of the days he must take off to make repeated court appearances. A plea bargain that immediately ends the case, takes jail off the table, often reduces the charge down to a violation, and leaves no criminal record is similar enough to dismissal that defendants may readily accept.[63]

The plea deal Suk describes does not seem like coercion. It offers a way out. Similar scenarios play out across the severity spectrum. Plea bargains are especially attractive to defendants who are jailed pending trial. In those circumstances, even defendants facing serious charges—like Troy McAllister (discussed in the Introduction)—may be offered a “time-served” deal.[64] If the defendant accepts, he is freed. McAllister was certainly coerced to plead guilty, but background law and procedure, not the time-served plea deal, is the source of that coercion.

Even some of the most nuanced analyses fall into the trap of mistaking background realities of American criminal law for flaws in plea bargaining. For example, Stephanos Bibas famously critiques plea bargaining on the ground that plea bargains do not necessarily track the results of a potential trial—i.e., “the shadow of trial.”[65] Bibas defines “the shadows of trials,” as “the influence exerted by the strength of the evidence and the expected punishment after trial.”[66] Within these parameters, Bibas is clearly correct. He identifies non-trial influences on plea bargains like: “[p]oor lawyering, agency costs, and lawyers’ self-interest,” “bail rules and pretrial detention,” and lumpy sentencing laws.[67] Notice, however, that these non-shadow-of-trial influences are not introduced by plea bargaining. They are part of the baseline realities of American criminal law. Thus, Bibas’s critique requires only that we recognize that plea bargaining is not conducted solely in the shadow of trial.[68] It occurs in the shadow of all aspects of a non-negotiated alternative.

Plea bargaining introduces an alternative to the non-negotiated baseline. Thus, Bibas’s central claim that “plea bargaining adds another layer of distortions that warp the fair allocation of punishment” misses an important nuance.[69] Plea bargaining is a reaction to the flaws Bibas identifies and, consequently, is just as likely to mitigate as to aggravate those distortions. Take Bibas’s examples: Poor lawyering? Even an incompetent attorney can communicate a standard plea deal to the client.[70] And the deal becomes all the more valuable as the attorney’s skills decrease. Pretrial detention? Often the only way for defendants to cut short a period of pretrial detention is a plea bargain that ends delay, allowing a defendant to exit dangerous jail conditions and return to work and family obligations.[71] Lumpy sentences? Defendants are more likely to be able to smooth out the lumps that matter—like mandatory sentences or immigration consequences—through plea bargaining as opposed to trial.[72]

Thus, Bibas is right that plea bargains do not occur solely in the shadow of trial. Plea bargains are negotiated in the shadow of reality. That means the full force of the criminal law’s many failings: poor lawyering, congested courtrooms, harsh collateral consequences, pretrial detention, and the practical toll of an ongoing case on witnesses, victims, and defendants. And above all, it means the severity of the non-negotiated alternative. The shadow is long and oppressive. But it makes little sense to pretend that when these flaws influence the parties’ negotiations, plea bargaining is to blame.

None of this is meant to suggest that plea bargaining works well. It is the ultimate contrast gainer. The only thing plea bargaining has going for it is all the flaws in the alternative. Plea bargaining is like the picnicker in the woods who, seeing an angry bear approaching from a distance, puts on running shoes and begins stretching. The picnicker’s companion (trials) says, “are you crazy, you can’t outrun a bear.” The picnicker responds, “I don’t need to outrun the bear, I just need to outrun you.”

Assuming that we are stuck with plea bargaining, it is important to recognize the practice for what it is. Plea bargaining endures precisely because there are so many flaws in the alternative (delayed resolutions via formal, often overly punitive trials). Thus, pointing out the flaws that emerge during plea bargaining is neither a compelling critique nor a direct pathway to reform. If plea bargaining is not the source of these problems, it is rarely the right place to fix them. Most problems in plea-bargained cases stem from background factors: draconian laws, poor lawyering, congested courtrooms, incessant delays, and pretrial detention. Plea offers don’t create these problems. And without a plea deal, defendants have no choice but to suffer their full force.[73] A plea bargain presents an option. To the extent this alternative improves the defendant’s circumstances or leaves them unchanged, influential critiques like those made by Langbein, Schulhofer, and Bibas fall flat. Plea bargaining isn’t perfect. It doesn’t have to be. It just has to outrun the competition.

II. What’s (Not) Wrong with Plea Bargaining

Part I makes the modest case for plea bargains. Plea bargaining is a mechanism for negotiating alternatives to a deeply flawed baseline. This Part applies that insight to the most common arguments against plea bargaining, each illustrated through a quotation leading off the pertinent section. Critics assail plea bargaining because it creates trial penalties, often with stark gaps between plea offers and potential post-trial sentences, incentivizes severe laws and overcharging, and jeopardizes the innocent. The discussion below identifies the weaknesses in each of these criticisms. These are all real problems. But to a large extent, critics see the reflection of the broader system’s flaws in plea bargaining and, like Langbein, mistake their source.

A. The (Inevitable) Trial Penalty

“Every day, in virtually every criminal court throughout the nation, people plead guilty solely as a consequence of a prosecutor’s threat that they will receive an exponentially greater post-trial sentence compared to the pre-trial offer.”[74]

The most common critique of plea bargaining focuses on the trial penalty. Critics define the trial penalty as “the substantial difference between the sentence offered prior to trial versus the sentence a defendant receives after a trial.”[75] This difference, critics contend, coerces defendants to accept punishment on the prosecutors’ terms—making the trial penalty “one of the most lethal tools in the prosecutor’s kit.”[76]

A simple example illustrates the trial penalty dynamic. Imagine the following law:

Anyone who drives more than five miles per hour over the speed limit shall be sentenced to five days in jail. Those who show remorse by pleading guilty shall be sentenced instead to a $5 fine.

The likely impacts of the law would be that (1) far fewer people would contest speeding tickets, and (2) a tiny percentage of speeders would go to trial, lose, and be sentenced to five-day jail terms.

The speeding hypothetical illustrates that the trial penalty primarily affects those who do not go to trial. In those cases, the penalty operates as a “threat[]”[77] made here in the statute itself, but more typically by either the prosecutor[78] or a judge,[79] that pressures defendants to plead guilty.[80]

The most intriguing insight about the trial penalty is that it seems inevitable in any system that allows defendants to avoid trials.[81] Defendants plead guilty when they perceive doing so to be preferable to the alternative. But if a defendant benefits from pleading guilty, a similarly situated defendant who goes to trial (and loses) suffers a penalty. Empirical analysis reflects this reality: “Nationwide data evaluating sentences for the same offense type show that guilty plea sentences are the least punitive, with jury trial sentences the most punitive, and bench trial sentences in between.”[82]

We would see trial penalties even if there were no plea bargaining and even without prosecutors. Defendants who come before the court on the charges for which they were arrested by police (or complaints of private individuals) could still expect to receive lower sentences by pleading guilty. Judges do not need prosecutors (or statutes) to threaten trial penalties. Judges can signal the prospect of leniency explicitly or even implicitly by establishing a practice of post-guilty-plea lenience.[83] Thus, it is little surprise that a trial penalty remained even when plea bargains were banned in El Paso, Texas. Analysis found a continuing “sentence differential between defendants who pleaded guilty and those who were convicted at trial” as “guilty plea defendants received shorter prison terms than jury trial defendants.”[84] Seeking mercy from the court is a time-honored tactic.[85] Evidence of the phenomenon can be found in American case reports as far back as 1814, when a militiaman threw “himself on the mercy of the court” in an unsuccessful effort to avoid punishment for failing to heed a call to fight the British in the War of 1812.[86] The only way to eliminate these kinds of trial penalties would be to require all defendants to go to trial—a solution that seems both unlikely and potentially harmful to many defendants.

The inevitability of a punishment differential between defendants who confess guilt, cooperate, and seek mercy and those who are convicted after trial suggests that the mere fact of a trial penalty is not the problem with plea bargaining.[87] But this insight raises a doctrinal paradox. The inevitability of a trial penalty coexists awkwardly with the defendant’s constitutional right to go to trial. As the Supreme Court has stated, “To punish a person because he has done what the law plainly allows him to do is a due process violation ‘of the most basic sort.’”[88] Thus, the Constitution would seem to prohibit trial penalties.

Courts and policymakers resolve this tension by allowing defendants to receive benefits for pleading guilty (plea discounts) but prohibiting them from being penalized for going to trial (trial penalties).[89] The strongest argument to support this odd state of affairs highlights various mitigating factors that arguably correspond with pleading guilty. For example, at sentencings after a guilty plea, defense attorneys leverage the notion that a person who accepts responsibility should be punished less severely than an equivalent offender who does not.[90] Again, this is not a modern innovation. In 1907, the Supreme Court of Oregon extolled guilty pleas that allowed the public to benefit from “any reduction of sentence . . . that may have been possible by the defendant showing a desire to be reformed, and throwing himself on the mercy of the court . . . .”[91] A modern exemplar of this practice can be found in the Federal Sentencing Guidelines, which reward “acceptance of responsibility”—typically operationalized through a guilty plea—with a substantial sentencing concession.[92] American courts reject legal challenges to provisions like this on the ground that they “clarif[y] a tradition of leniency extended to defendants who express genuine remorse and accept responsibility for their wrongs.”[93] A similar articulation can be found in an Advisory Committee Note to the Federal Rules of Criminal Procedure, which states: “Where the defendant has acknowledged his guilt and shown a willingness to assume responsibility for his conduct, it has been thought proper to recognize this in sentencing.”[94]

The courts’ acceptance of plea discounts and (purported) rejection of trial penalties introduces a theoretical dilemma. How can we distinguish a plea discount from a trial penalty? As a matter of logic, if defendants who plead guilty receive a benefit for accepting responsibility, those who insist on a trial receive an equivalent penalty upon conviction.

Scholars try to distinguish trial penalties from plea discounts in two ways. The first is through the motivations of the government officials involved.[95] But this only shifts the question from one location to another. Just as it is difficult to distinguish a trial penalty from a plea discount, it is hard to determine whether the prosecutor or judge is trying to reward defendants who plead guilty or punish those who go to trial. The speeding statute set out at the beginning of this subpart illustrates the point. Does the statute create a trial penalty, a plea discount, neither, or both?

The second technique for distinguishing impermissible trial penalties from permissible plea discounts is to propose a normative baseline: a “proper” penalty for a given crime. Under this approach, trial penalties arise whenever the post-trial sentence will be higher than the proper penalty. For example, Ben Grunwald explains that “a plea offer represents a trial penalty if the government threatens to pursue charges at trial above the normative baseline, which is defined as the punishment the defendant should receive without incorporating his decision to exercise or waive the right to trial.”[96]

This would be a viable approach to identify and (potentially) regulate trial penalties if there was some way to agree on the baseline. Unfortunately, the normative baseline—the “punishment the defendant should receive”—is contested.[97] Rarely will the prosecution, defense, judge, and outside observers agree on the punishment the defendant should receive (if any) for an alleged crime. Each participant may, in fact, embrace a range of punishments, all of which they consider fair—perhaps dependent on variables like cooperation and remorse. This hopelessly complicates efforts to identify trial penalties by comparing plea offers to a “proper” or “regular” punishment.[98] And when we cannot agree on a subjective baseline, the temptation is to adopt an objective one: the punishment authorized ex ante by the legislature and imposed by a neutral judge based on the facts proven to a jury at trial. That is the approach of the modern courts and, by equating the “proper” penalty with the penalty imposed after trial, it extinguishes the possibility of finding any trial penalties at all.[99]

In sum, so long as we permit defendants to exercise agency in deciding whether and how to contest charges against them, a trial penalty seems inevitable. And trying to distinguish the good from the bad trial penalties is, so far, too slippery a task to yield fruit. This suggests that the trial penalty itself is neither the problem with plea bargaining nor a productive target for plea-bargaining reform. (Subpart II(D) addresses the related but distinct question of whether a trial penalty, defensible in the abstract, becomes indefensible if it is too large.)

B. A Complex Stew of Incentives

“Legislatures have increased the punishment associated with conviction at trial so that prosecutors have leverage to induce guilty pleas.”[100]

“In most cases, prosecutors overcharge not because they seek to impose unduly harsh sentences on defendants, but simply because of the bargaining leverage it provides.”[101]

Another common critique of plea bargaining highlights the perverse incentives it creates for legislators who draft criminal laws and the police and prosecutors who enforce them. To the extent these incentives are the problem with plea bargaining, we might try to reform plea bargaining to minimize those incentives. Of course, the only way to eliminate these incentives is to abolish plea bargaining—a seemingly unrealistic reform.[102] But as discussed below, even that might not make much difference. The incentives created by plea bargaining overlap to a large degree with those that would exist in its absence. That suggests that attempting to eliminate the incentives generated by a plea-bargaining regime would not only be practically impossible but less impactful than critics assume.

The incentives-based critique of plea bargaining takes two forms. First, prominent scholars argue that legislatures enact harsher penalties than they believe are warranted “so that prosecutors . . . could use those penalties as leverage to obtain guilty pleas”[103] and to “make it easier for [prosecutors] to obtain convictions through plea bargaining.”[104] Thus, plea bargaining indirectly generates excess punishment by artificially inflating the severity of criminal law.

The weakness of this argument is that, even if we grant that legislatures approach criminal laws with this level of sophistication, the instrumental goals they seek to achieve are not simply guilty pleas. The most commonly expressed instrumental purpose for criminal laws is to deter crimes—a cause theoretically (although not practically) furthered by extreme punishments that serve as an example to others.[105] And, in fact, there is evidence that legislators view plea bargaining with disfavor in this context. For example, California’s Three Strikes law—the paradigmatic example of a severe criminal law—seeks to discourage (not encourage) plea bargains, stating: “The prosecution shall plead and prove all known prior felony serious or violent convictions and shall not enter into any agreement to strike or seek the dismissal of any prior serious and/or violent felony conviction allegation . . . .”[106]

But beyond that, strategic legislators seek to assist law enforcement by incentivizing cooperation, not guilty pleas. The evidence offered by the scholars quoted above supports this point. Scholars who claim that legislators are trying to induce guilty pleas quote legislative comments that champion a related but distinct goal: “to put pressure on defendants to cooperate in exchange for a lower sentence so evidence against more responsible criminals can be attained.”[107] This cooperation can take a variety of forms, including aiding in halting ongoing criminal activity, undoing the harm of past crimes (e.g., recovering stolen items), identifying accomplices, or even just promising to discontinue the alleged criminal conduct. While a plea deal and cooperation often arise together, they are distinct. Defendants who plead guilty waive their right to a trial but retain the option of declining to cooperate with authorities. And defendants who cooperate do not always plead guilty. Some avoid charges altogether.[108]

A desire to incentivize cooperation creates the same type of incentives as plea bargaining. And these incentives, like those that characterize plea bargains, must be overseen by some official (judge or police officer or prosecutor) with the authority to reward cooperative defendants and punish uncooperative ones. That suggests that even if formal plea bargains were forbidden, a legislature trying to incentivize cooperation would still enact severe laws and intend that they apply in some cases—specifically, to defendants who do not cooperate. Thus, it does seem likely that legislatures attach severe punishments to offenses not just because they believe those punishments are warranted but also for instrumental reasons. But those instrumental reasons go beyond a desire for guilty pleas.

The incentive-based critique of plea bargaining takes a second form, extending to police, judges, and prosecutors. In fact, Grunwald’s argument about “proper” trial penalties,[109] referenced in the prior subpart, can be reframed as a way to prevent prosecutors from overcharging to obtain plea-bargaining leverage. But again, prosecutors and police seek cooperation, not just guilty pleas.[110] It is difficult to conceive how any system could eliminate these incentives. Even if prosecutors were somehow barred from plea bargaining, police officers could present similar options to suspects—“tell us who supplied your drugs/where the body is/who helped you, or you are going to jail/prison/the electric chair.” As long as defendants can choose to cooperate (or not), and police, prosecutors, and judges can choose the harshest options (or not), incentives to enact severe laws and invoke them will persist, even without plea bargaining.

In addition to the desire to encourage cooperation, prosecutors have still other incentives to charge aggressively. In a world without plea bargains, prosecutors would face greater uncertainty in the form of more jury verdicts and, perhaps, more assertive judges.[111] Plea bargains are not the only setting in which criminal justice actors compromise. Not knowing which charges will survive judge and jury screens, aggressive prosecutors could charge broadly to ensure some convictions.[112] As Allison Larsen has written, even juries engage in compromises during their deliberations, and those compromises can look just like the bargains that take place before trial.[113] Finally, given the constitutional requirement of notice in criminal procedure, courts are more likely to permit prosecutors to dismiss charges as a case progresses than to add them close to trial.[114] This creates an additional incentive to err on the side of charging more rather than less.

None of this is to say that plea bargaining does not add incentives to the mix. Instead, the argument here is that plea bargaining is just one of many influences that can distort the charging calculus. And plea-bargaining incentives do not only point toward severity. True, plea bargaining creates incentives to charge high to generate leverage. But, as other scholars have noted, it also creates incentives for prosecutors to undercharge to encourage prompt, low-effort resolutions.[115] As Josh Bowers notes, a high conviction rate, “the most visible rubric of quality job performance,” is “achieved most readily by lenient, quick bargains.”[116]

Given this mix of incentives, the overriding question becomes the degree to which some incentives dominate the others. The answer, of course, will vary by jurisdiction and prosecutor,[117] but some intriguing empirical evidence suggests that plea bargaining does not substantially distort charging practices. Statistical analysis in El Paso, Texas, found little evidence of changes in charging after that jurisdiction prohibited plea bargains.[118] In Alaska, while prosecutors declined more cases after the plea-bargaining ban, the rate of convictions per police referral barely budged.[119] This riddle appears to be explained by the fact that the stricter screening was offset by a reduction in post-charge dismissals.[120] And since sentence severity actually increased, that suggests that the ban changed only the timing, not the substance, of Alaska prosecutors’ charging decisions.[121]

In sum, the stew of incentives that exists without plea bargaining looks a lot like the stew that exists with plea bargaining. This means that even if we could conceive of some new mechanism to do so, it is unlikely that we could manipulate plea-bargaining incentives in the desired direction. Perhaps the most likely outcome is that the system would readjust back to the ex ante equilibrium. That’s what happened in Alaska, particularly with respect to the cases that mattered most: A post-plea-bargaining-ban study there found that “[t]he conviction and sentencing of persons charged with serious crimes of violence such as murder, rape, robbery, and felonious assault appeared completely unaffected by the change in policy [abolishing plea bargaining].”[122]

Further evidence of the practical difficulty of changing incentives through plea-bargaining reform comes from the collective “meh” that met seemingly dramatic statutory restrictions on plea bargaining in New York and California.[123] There is no evidence that these bans altered charging severity in those states, probably because plea-bargaining restrictions are so easily avoided.[124] After all, when the defendant and prosecution reach an agreement, there is typically no one with the necessary incentives and resources to challenge the resulting deal.

While commentators commonly critique plea bargaining as saturated by improper incentives, the identified incentives overlap with those that exist in its absence. And the incentives that plea bargaining creates point in different directions. This suggests that incentives for severe laws, overcharging, and under-prosecution are neither a unique contribution of plea bargaining nor a promising target for plea-bargaining reform.

C. Too Broad Spreads

“As numerous commentators have observed, ‘the gap between post-trial and post-plea sentences’ is often ‘so wide’ that ‘it becomes an overwhelming influence’ in the defendant’s decision to plead guilty.”[125]

Many commentators situate the problem with plea bargaining not in the mere existence of the trial penalty but in overly broad spreads between a plea offer and the post-trial sentence. The argument is that at some point, these spreads create too much of a trial penalty, generating unconscionable pressure on defendants to relinquish their constitutional rights and plead guilty.

One problem with focusing on broad spreads is the difficulty of fashioning a remedy. Spreads can be extended in two ways. First, prosecutors can charge aggressively, increasing the potential post-trial penalty. Existing checks on this kind of overcharging include grand-jury review, judicial dismissal of charges lacking evidentiary support, a guarantee of a jury trial, and judicial sentencing.[126] Scholars typically scoff at these checks.[127] But it is difficult to conceive of further checks on prosecutorial charging that would prohibit prosecutors from charging offenses that plausibly fit the facts alleged. Second, spreads can be extended in the other direction through increasingly lenient plea agreements. Again, absent substituting another decision maker for the prosecutor—an action already allowed through the requisite judicial approval of plea agreements[128]—there is no viable way to forbid such prosecutorial lenience.

The practicality problems do not end there. Shrinking the gap between potential post-trial sentences and plea offers requires control of two variables. But there is no guarantee that either variable can be pinned down, a necessary prerequisite for concrete limits. Parties can agree to plea deals that do not formally dictate a precise sentence—already a common occurrence—leaving informal expectations to fill in unspoken gaps.[129] And plea deals aren’t the only variable that can be hard to nail down. Post-trial sentencing laws usually give judges discretion, and the considerations that inform that discretion can change with developments that arise during trial.[130] That means post-trial sentences will often be unpredictable as well. In this context, trying to restrict plea offers by reference to potential post-trial sentences will be like trying to knit a scarf out of spaghetti.

In addition, reforms that decrease spreads between plea offers and potential post-trial sentences could backfire. In many cases, like those set out in the Introduction of this Article, large spreads appear to foster injustice. But large spreads are like large bets. We can’t evaluate their wisdom by looking only at the ones that went wrong. It is true that when defendants are convicted after trial, the broad spread appears to hurt the defendant. But in other scenarios, such as where defendants accept a plea deal, large spreads can mean increased lenience. And if the defendant is acquitted or sentenced less severely after trial, the broad spread evaporates.

For example, Emily Bazelon chronicles a lenient plea offer from New York City prosecutors in her book Charged.[131] There, “Kevin” considers his options as he faces trial for carrying an unlicensed, loaded firearm, an offense that carries a mandatory 3.5-year prison term. Bazelon praises the prosecutor’s office for offering Kevin a deal where if he pleads guilty, completes counseling, and maintains a period of good behavior, the charges will be dismissed. The spread between the offer and the potential post-trial sentence is vast. But in this case, that was a sign of lenience.[132]

The same dynamic plays out with more serious offenses. For example, in 2018, Jacob Anderson, a Baylor University fraternity president, was charged with rape. The case drew notoriety when Anderson accepted a plea deal to a lesser charge resulting in no prison time, a $400 fine, and three years’ probation.[133] If Anderson had gone to trial and been convicted, he could have served up to twenty years in prison. Anderson faced and avoided a substantial trial penalty, and again, the plea bargain was an instrument of lenience.

If the defendants referenced above had gone to trial and lost, they would have suffered large trial penalties. But the lenient plea offers, and resulting broad spreads, are not the cause of this severity. The source of the severity is the background laws governing the charges brought by a prosecutor and the subsequent judge and jury decisions about those charges. Coercive pressure increases with pretrial detention, and as it becomes more likely the jury will convict and the judge will impose a harsh sentence. But these factors exist even without a plea offer. And against this backdrop, the larger the spread, the more favorable the defendant’s situation becomes. That suggests that broad gaps are not the problem with plea bargaining—at least not in the way suggested by academic critics of penal severity.[134] And, as already noted,[135] even if they were, spreads between plea offers and potential post-trial sentences are an elusive target for reform efforts.

D. Innocent Defendants

“[C]onvicting the innocent is unequivocally easier in a world that permits plea bargaining.”[136]

Many critics of plea bargains focus on the plight of the innocent defendant.[137] For example, in an influential essay titled Why Innocent People Plead Guilty, Judge Jed Rakoff explained that the “gravest objection of all” to American plea bargaining is that “the prosecutor-dictated plea bargain system, by creating such inordinate pressures to enter into plea bargains, appears to have led a significant number of defendants to plead guilty to crimes they never actually committed.”[138] Rakoff is correct that laws that authorize or even mandate severe punishments generate strong pressure on defendants to plead guilty. This is increasingly true the broader the gap between the post-trial sentence and plea offer. And at some point, pairing a relatively lenient plea offer with a severe post-trial sentence creates so much pressure that it can become rational for even the innocent to plead guilty.

And while Rakoff and most commentators focus their critique on defendants accused of serious crimes, the danger that an innocent person would plead guilty arises throughout the severity spectrum. That’s because for minor offenses, plea offers can be exceedingly lenient. If the defendant pleads guilty, there may be no further formal punishment. If the defendant is jailed pending trial, the guilty plea may effectively mean less punishment than a post-trial acquittal.

Critiques like Rakoff’s are helpful in revealing the coercive pressures of American criminal law. But when directed at plea bargaining, they commit the now-familiar misstep of ignoring the alternative. Rakoff points out that studies of the wrongly convicted reveal that “about 10 percent[] pleaded guilty to those crimes.”[139] That’s not good. But in citing this statistic, Rakoff implicitly acknowledges that ninety percent of the group he spotlights were wrongly convicted at trial. While Rakoff’s argument supports a conclusion that guilty pleas present a danger to the innocent, it simultaneously supports a conclusion that trials present an even graver danger. Thus, Rakoff’s attack on pleas is incomplete. He shows that plea bargaining is not perfect, but he has not established that it causes (or aggravates) the problem he identifies: punishment of the innocent.

When an innocent defendant pleads guilty to avoid a looming trial penalty, the outcome is a heartbreaking injustice. But plea bargaining contributes to this injustice only if the defendant would have been acquitted at trial (or if the case would have been dismissed). If the defendant would have been convicted—a real possibility as illustrated by Rakoff’s own statistic—the plea deal mitigates the injustice.[140] Thus, Rakoff has not identified the “gravest objection” to plea bargaining. He highlights an important but distinct problem: the failure of the overall system to reliably separate the innocent from the guilty.

E. Bad Motives

“[Prosecutors should not] ‘pil[e] on’ charges in order to unduly leverage an accused to forgo his or her right to trial.”[141]

Even if it is impossible to regulate the incentives generated by plea bargaining as a general matter, it might be possible to regulate motives at the level of the individual actor. Thus, perhaps motives are the real problem with plea bargaining, and the phenomenon could be improved by checking prosecutors and judges who exhibit the wrong ones. As hinted in the American Bar Association’s hedged guidance quoted above (“unduly”), however, it is difficult to identify the motives that should be eliminated and even trickier to propose how to do so.

The most famous example of improper prosecutorial plea-bargaining motives comes from the Supreme Court case Bordenkircher v. Hayes.[142] There, Paul Hayes was charged with passing a forged $88 check. The prosecutor offered a five-year plea deal but warned that if Hayes turned down the offer, the prosecutor would seek an indictment under the State’s Habitual Criminal Act—a statute that mandated a life prison term for certain repeat offenders. The prosecutor explained during the proceedings that he wanted Hayes to plead guilty to “save the court the inconvenience and necessity of a trial and taking up this time.”[143] Hayes turned down the offer, went to trial, and was sentenced to life.[144]

While standard accounts portray Glen Bagby, the Kentucky prosecutor in Bordenkircher, as the prototypical plea-bargaining villain,[145] it is not clear that the prosecutor’s motive is the factor that makes the case problematic. While crudely expressed, Bagby’s motive for pressuring Hayes to plead guilty echoes widely accepted justifications for plea bargaining: preserving resources, decreasing the burden on witnesses, and so on. The primary mechanism for achieving this goal is rewarding defendants who plead guilty. The unavoidable flip side of rewarding defendants who plead guilty is not rewarding (i.e., penalizing) those who go to trial.

The harsh Kentucky Habitual Criminal Act distorts every aspect of the Bordenkircher case. To isolate the role that a prosecutor’s motives can or should play, it helps to recreate Bordenkircher with similar motives but without that statute. Imagine Hayes is arrested in 2022 in Philadelphia for passing a forged $611 check ($88 indexed for inflation).[146] Let’s assume the crime is caught on camera and so (as turned out to be the case in Bordenkircher) there is little prospect of an acquittal. The prosecutor proposes that Hayes enter a diversion program where he must admit guilt and reimburse the store for the loss. Hayes declines. The prosecutor then files a theft charge. Hayes is convicted and the judge sentences him to five days in jail. Again, the Philadelphia prosecutor proceeded more harshly against Hayes because he refused to admit guilt on the prosecutor’s terms. But now that motive does not seem as problematic. If that is right, the problem with the Bordenkircher prosecutor’s behavior may not be the prosecutor’s motives but the tools he used to act upon those motives.

In fact, there are a variety of scenarios where pressuring a defendant to plead guilty may be acceptable. Imagine a small town where two serial killers have struck independently. The local prosecutor is certain of each defendant’s guilt but only has the resources to prosecute one case. The prosecutor might offer a lenient plea deal, say five years in prison to the first defendant who pleads guilty, leaving the defendants with a difficult decision. As in Bordenkircher, the spread between the post-conviction (potential life) sentence and the plea offer is enormous, and the prosecutor’s motive is to preserve court resources. But now it is not so easy to explain why that motive is inappropriate. Other analogous scenarios might involve efforts to push defendants whose guilt is clear to plead guilty to avoid trauma for vulnerable victims or to cooperate in locating victims’ remains.[147] And once we concede these examples, proponents of plea bargaining can alter the variables to cover virtually any scenario. The two serial killers in a small town become fifty homicides and a hundred robberies facing an overwhelmed district attorney’s office. Or a thousand drug cases. The policy questions become more fraught, but the fight has shifted based on the appropriate punishment (if any) for various crimes and alleged offenders—not the motivations for resolving cases without trial.

Putting aside the alarming substantive severity in Bordenkircher, the other unusual aspect of the case was not the prosecutor’s motives but his timing. The prosecutor sought the additional charge after plea negotiations. This led the Sixth Circuit in the case to reverse, explaining: “Although a prosecutor may in the course of plea negotiations offer a defendant concessions relating to prosecution under an existing indictment, he may not threaten a defendant with the consequence that more severe charges may be brought if he insists on going to trial.”[148] The Sixth Circuit’s rule (rejected by the Supreme Court)[149] would permit the same plea-bargaining dynamic in future cases as long as the prosecutor first sought the “extra” charges and then offered to drop them as part of a plea bargain.

The short-lived Sixth Circuit rule illustrates the problem with attacking background criminal justice problems (like overcharging or severity) through restrictions on plea bargaining. Barring the prosecutor’s tactic in Bordenkircher incentivizes prosecutors to charge high at the outset so that subsequent plea offers become a reward (a plea discount), not a punishment (a trial penalty). This is especially problematic in the large, urban offices that generate most criminal cases.[150] Large offices generally have separate sections of the office responsible for indicting cases.[151] In those offices, the rule adds little protection because busy trial prosecutors are likely reluctant to refer already-indicted cases back to the grand jury section. The Sixth Circuit’s rule turns this unobtrusive practical disincentive into a concrete legal prohibition, increasing the pressure on the initial charging attorney to include all viable charges at the indictment stage—potentially leading to more overcharging, not less.

In sum, it is not clear that the problem in Bordenkircher and similar cases is the prosecutor’s motive. The timing of the offer and the harsh recidivism law invoked make the prosecutor’s motive in the case seem unusually nefarious. But when that motive is isolated, it appears little different from the motives that support the practice of plea bargaining generally. As the Supreme Court has recognized, there is little room, legally speaking, to permit plea bargaining without also tolerating prosecutors and judges whose motives include a desire that defendants plead guilty.[152]

F. The Shadow of Plea Bargaining

“By lowering the price of imposing criminal punishment, plea bargaining gave America more of it.”[153]

Another common critique of plea bargaining is that it increases the overall efficiency of the State’s ability to impose criminal punishment. And that means we end up with too much. A related critique is that plea bargains short circuit the venerable jury-led factfinding process, and so even when defendants benefit from plea bargains, society suffers.

It is generally accepted that plea bargaining enables a larger system of criminal law enforcement than would otherwise be possible and, of course, fewer trials.[154] Interestingly, this common-sense observation depends on a few important assumptions: (1) legislatures would be unwilling to devote sufficient resources to resolve all the cases that would have to be tried without plea bargains; (2) courts would not become more tolerant of procedural shortcuts (or long delays) if every case went to trial; and (3) plea bargaining would not be largely replaced by unconditional guilty pleas and requests for mercy. If the assumptions are met, abolishing plea bargaining would shrink the criminal law’s bloated footprint—and that would clearly be a good thing.[155]

The basic response to these important critiques has been noted already: abolishing plea bargaining seems unlikely and will be harmful to at least a subset of criminal defendants. But even more importantly, this attack on plea bargaining once again represents an indirect effort to address core problems of American criminal law enforcement: aggressive policing, expansive charging, and overly broad and punitive laws. Regulating or abolishing plea bargaining is a tool that could be wielded to try to reduce the scope and severity of criminal law, but it is a blunt tool: the gardening equivalent of removing weeds with a baseball bat.

While all might agree that there are too many criminal cases, consensus on which cases to eliminate may not be as easy to achieve.[156] Restricting plea bargains would likely reduce the number of cases, but doing so could get rid of the “wrong” cases if police and prosecutors gravitate toward easier-to-prove crimes against easier targets rather than trickier, but more impactful, offenses committed by the most sophisticated defendants.[157] Direct mechanisms avoid this problem. A legislature that sought fewer drug cases, for example, can legalize drugs or make violations civil, not criminal, infractions. A district attorney’s office seeking fewer misdemeanor prosecutions can decline such cases. Police wanting to focus on white-collar crimes instead of retail theft can shift investigative resources accordingly. Trying to achieve similar goals through restrictions on plea bargains makes the least sense. One might counter that direct reductions in criminal caseloads are politically difficult to achieve. But even if that is true, there is little reason to expect the same political actors who are reluctant to reduce case volume directly to do so haphazardly by restricting plea bargaining. (A collective refusal to engage in plea bargaining by defendants, rather than the government, presents a different political calculus but seems equally unlikely.)[158]

With respect to the displacement of trials, it is beyond dispute that a plea deal is not as noble a form of dispute resolution as a jury trial: it is “more like a used car sale than the glamorous trials in television shows.”[159] And for those of a historical bent, this is surely not the system the Founders intended.[160] Still, eighteenth-century observers would also be shocked by the sheer breadth of the modern criminal-law-enforcement apparatus. And even if the Founders would have expected more trials in this new reality of pervasive criminal laws, it is hard to argue that criminal defendants must pay the price for this nostalgia.

In sum, there is growing consensus on the need to dramatically ratchet down the scope and severity of criminal law enforcement. But that goal can be more sensibly achieved and rationally tailored directly by decreasing the number and breadth of statutory crimes and reducing the frequency and severity of their enforcement. Eliminating plea bargains and thereby generating even greater dysfunction in the crowded criminal courts presents an alternative tactic to try to reduce the system’s swollen footprint, but the path is uncertain and would almost certainly harm many defendants along the way.

III. Fighting Uncertainty

Having failed to uncover the problem with plea bargaining in the expected places, this Part situates the problem with plea bargaining somewhere new: in uncertainty. It begins by explaining the problem of uncertainty. Then, it proposes two mechanisms for reducing uncertainty and transforming the plea-bargaining process. The goal is to move American plea bargaining closer to a hypothetical world where, whenever defendants must choose between a guilty plea and a trial, the right choice is unmistakable.

A. The Problem of Uncertainty

It is easy to overlook the problem of uncertainty because most commentary views the plea-bargaining dynamic in hindsight, evaluating cases after they have been resolved. For defendants deciding whether to agree to a plea offer, however, uncertainty is key.

Uncertainty’s importance becomes clear when we contemplate scenarios without it. Imagine a defendant knows: (1) that conviction is certain at trial and the sentence that will follow and (2) the sentence that will result from a guilty plea. Presented with two known outcomes, the defendant can select the least punitive. Offering the defendant this choice makes the defendant no worse off and potentially substantially better off. There might still be questions about fairness and equity, but these questions would target the substance (an accepted plea offer or post-trial sentence), not the bargaining process.

In the real world, the defendant can’t be sure what will happen at trial. Thus, the primary contributor of uncertainty is doubt about whether there will be a conviction.[161] Some cases are dismissed prior to trial and others end in acquittal. Even when defendants are convicted, the conviction may be on fewer counts than charged or on lesser offenses. In addition, many, if not most, sentencing laws leave judges with discretion.[162] This means that even if a defendant is convicted at trial, the ultimate sentence may be harsher, comparable, or less severe than the sentence contemplated in a plea deal. Defendants can get a trial discount, not a trial penalty, when they (correctly) decline unfavorable plea offers.

Uncertainty also arises in evaluating the plea offer itself. While some plea offers include a specific sentence that the judge must impose, many do not. Thus, the defendant may not know what the outcome will be after a guilty plea.[163] The universe of potential outcomes can be represented by four possible scenarios that face a charged defendant with the following unknown variables: a post-trial sentence (TS) and a post-plea sentence (PS) of a defendant who either pleads guilty (PG) or goes to trial (T):

This graphic reveals that if the defendant pleads guilty, the resulting sentence may be (1) lower than the trial sentence (plea discount), or (2) higher than the trial sentence (plea penalty). If the defendant goes to trial, the trial sentence may be (3) higher than the plea sentence (trial penalty), or (4) lower than the plea sentence (trial discount). In scenarios one and four, the defendant chooses correctly, selecting less punishment. In scenarios two and three, however, the defendant inadvertently selects more punishment.

The uncertainty seems daunting. But the chart reveals that to avoid the wrong choice, the defendant needs to know only one thing: whether the post-plea sentence (PS) is lower than the post-trial punishment (TS). The defendant does not need to pin down either variable. As long as the defendant can confidently forecast which option will result in less punishment, a plea offer introduces no new harm.

B. Reducing Uncertainty Through Information

The uncertainty surrounding whether to accept a plea offer can be attacked on multiple levels. The most obvious layer of uncertainty is the potential for an acquittal. Here, discovery and defense lawyering are critical. From the government side, the prosecutor should present all the known information about the case to the defense.[164] While open-file discovery is becoming more common,[165] some resistance to pre-plea discovery stems from a sense that defendants do not need to be told by the government whether they are guilty—the facial determinant of the propriety of a guilty plea.[166] It is true that in most scenarios, defendants know if they committed a charged crime. The themes explored in this Article, however, explain why that argument misses the mark. Defendants do not need pre-plea discovery to decide whether they are guilty. They need pre-plea discovery to decide whether they should exercise their constitutional right to go to trial.

Defense attorneys play an important role in reducing uncertainty through uncovering evidence and advising defendants. While discovery helps the defendant assess the prosecution evidence, the defense attorney can offer guidance on the defense case. This means seeking out information that the government may not possess—including video footage, phone records, physical evidence, and uncovering and interviewing witnesses and the defendant.

Government disclosure and defense investigation decrease uncertainty about the likely outcome at trial. But the defendant does not just need to know about the evidence that will be presented at trial. The defendant also needs to know how the factfinder will react to that evidence. Again, the defense attorney’s guidance will be critical. Yet more is needed. The defense should be able to review statistical data about trials in the jurisdiction. Some jurisdictions will have this information readily available and may already publish it. Others can develop the necessary data-collection procedures. These procedures will be valuable for many purposes and are already underway in some jurisdictions.[167]

For the historical outcome data to be helpful to defendants, it must distinguish post-guilty-plea sentences from post-trial sentences and be broken down by offense and judge. Defense attorneys can scour this data for “comparables,” like real-estate agents setting a price for a home. This information will decrease the uncertainty facing defendants considering plea offers and should, as a beneficial side effect, reduce the prevalence of outlier plea deals.[168]

C. Reducing Uncertainty Through Plea Offers

Providing information about the government’s evidence and the outcomes in analogous cases can take the casino out of plea bargaining. But even with all this information, uncertainty remains. As already noted, however, it is not necessary to resolve all the uncertainty. The goal is to make the right answer to one question—plead guilty or go to trial—unmistakable. This will often depend on the plea offer itself.

Traditionally, commentators urged prosecutors to ensure that plea offers are substantively reasonable and treat similar cases similarly.[169] The analysis presented here suggests an additional consideration: certainty. The prosecutor need not make any plea offer at all[170]—and defendants can always plead guilty without any concessions. But when a prosecutor proposes a plea deal, the prosecutor should only present options that the prosecutor believes unequivocally make the defendant better off. Even if the criminal justice system has unavoidable elements of chance, the prosecutor should seek to minimize, not magnify, those elements. Offering the defendant an opaque plea deal is the equivalent of asking the defendant to pull the arm of a slot machine to determine the sentence. Offering an unfavorable plea deal (i.e., a plea deal that a defendant would be unwise to accept) is even worse—the equivalent of rigging the casino game. Neither practice has any place in a rational system of justice.

The earlier-referenced case of the teenager (Kevin) caught with a loaded gun in New York City illustrates the ability of the prosecutor’s offer to solve the uncertainty problem.[171] There was little doubt about Kevin’s factual guilt. But Brooklyn jurors had been known to acquit in similar cases.[172] Thus, Kevin faced great uncertainty about the post-trial outcome. More information about the outcomes in typical gun-possession cases would give Kevin a sense of the likelihood of success at trial. But that number—say, a 50% trial conviction rate—plus the looming prospect of a mandatory minimum sentence would still have left him unsure. The prosecutor’s plea offer is the critical final ingredient. The prosecutor in Kevin’s case made the plea deal so attractive that despite the uncertainty about what would happen at trial, there was little uncertainty about what mattered most—whether Kevin should plead guilty. Thus, plea bargaining improved Kevin’s circumstances, even though he was unable to realistically forecast the outcome at trial and faced a large trial penalty.

As Kevin’s case illustrates, the plea offer itself can be a critical tool for ratcheting down uncertainty. The exact approach will depend on context. If conviction and a lengthy sentence after trial are likely, most plea offers will satisfy the requisite certainty requirements. Any concession from the prosecution is an advantage over the baseline. At the other extreme, if the evidence is weak, the charges should be dropped, eliminating many high-uncertainty scenarios. The challenging questions arise in the middle.

In many cases, the defendant, advised by counsel, will have a relatively high degree of certainty about the likely trial outcome. In one recent study, researchers presented case vignettes to judges, prosecutors, and defense attorneys and found remarkable similarity in those actors’ estimates of the likelihood of conviction and the plea deal that should result.[173] One defense attorney in another study explains, “most cases” settle based on the chances at trial and since those chances were “usually pretty slight,” a plea deal with a “reasonable” sentence recommendation serves as a way to “buy[] some insurance that the judge would not have had a bad breakfast or decide that your guy was suddenly the worst type of offender” and sentence too harshly.[174] The relatively low percentage of acquittals in most jurisdictions suggests that this may be a widespread perception.[175] If these kinds of plea-bargaining dynamics are the norm (and not the exception),[176] that would explain how judges and lawyers have grown comfortable with the institution of plea bargaining even as scholars increasingly view it as anathema.[177]

To explore the prototypical scenario where a defendant foresees a likely conviction at trial, we can revisit the speeding statute hypothetical. It seems safe to assume that most people who get speeding tickets are speeding. And there is usually little doubt that the government will be able to prove the offense. Still, there may be objections to imposing punishment. Maybe the speed limit is too low. And a five-day sentence seems too harsh. But these are problems with the severity of the substantive law, not the adjudication process. In this example, plea bargaining is blameless. Even with a substantial trial penalty, it is better to have a plea option. Why? The absence of uncertainty. Conviction is near certain. The sentence is mandatory. The defendant has all the information needed to make precisely the correct choice between the options presented. The option to plead guilty in exchange for concessions only helps.

This logic carries through to serious cases where the evidence of guilt is strong. For example, imagine the defendant is caught at the scene of a crime after committing a murder and casually confesses in a recorded interview to police. Assume also that the penalty for murder is a mandatory twenty years to life. Now imagine two possible scenarios:

Scenario One: The prosecutor offers a plea deal of ten years.

Scenario Two: The prosecutor does not offer any plea deal.

The defendant goes to trial, is convicted, and receives the statutory penalty. In Scenario One, the defendant was offered a plea deal and suffered a trial penalty. But, if anything, that defendant was treated better than the defendant who received no plea offer in Scenario Two.

Uncertainty changes the dynamic. Imagine the defendant was not caught at the scene and has a viable defense. Now, the same ten-year plea offer is problematic. How can the defendant decide whether to take the deal? The plea sentence is substantially less than the post-trial sentence, but the defendant might be acquitted at trial.

At first glance, this seems to bring the uncertainty problem into the domain of the familiar concern about large trial penalties. But interestingly, the relationship between the potential trial penalty (the spread between post-trial sentence and plea offer) and uncertainty is not linear. Instead, it unfolds in an inverted v-shape. This can be illustrated with a few more data points for the hypothetical murder prosecution with a mandatory twenty-year sentence set out above. Imagine five possible plea offers:

As the trial penalty[178] increases, uncertainty goes up and then down. The defendant’s choice is easiest in the first and fifth scenarios; difficult in the second and fourth scenarios; and almost impossible in the third scenario. The defendants’ uncertainty about the correct choice, as the trial penalty increases, can be depicted as follows:

The graph illustrates that problematic plea offers arise towards the middle of the inverted v-shape, where it is hardest to determine whether to take the plea deal. On the left side, the plea offer is close to the expected trial sentence (thus the trial penalty is low). Given the real prospect of acquittal, the defendant should reject the offer and go to trial. On the right side, the plea offer is so much better than the expected post-trial sentence that the defendant would be wise to accept it. The middle of the chart is the bazaar. In that broad space where uncertainty is highest, the defendant is presented with choices that resemble those of a Las Vegas casino.

If a defendant is likely to be convicted but has some defense[179]—the typical scenario in cases that go to trial—an uncertainty lens suggests that ideal plea offers would fall on the left or right side of the chart. The offer should involve only a modest discount to encourage a trial resolution, or be so generous as to make the choice to plead guilty obvious. A plea offer in the middle of the graph, by contrast, introduces tremendous uncertainty. The defendant is more likely to choose incorrectly and consequently to be punished for that choice rather than the underlying crime.

D. Applying Uncertainty Analysis: Specific Cases

Having presented a new uncertainty lens for evaluating plea bargaining, this subpart applies that lens to specific scenarios: the plea deals rejected by Weldon Angelos, Paul Hayes, Joseph Tigano, the plea deal accepted by “Kevin,”[180] and the plea deals typically offered to defendants charged with selling drugs in the District of Columbia. The analysis reveals that uncertainty, not potential innocence, severity, or broad spreads, best identifies problematic plea bargains.

The case of Weldon Angelos is often cited to illustrate the evils of modern plea bargaining described in Part II.[181] But it also illustrates the problem of uncertainty. In 2002, police arrested Angelos after he conducted three separate marijuana sales to a police informant. The government offered Angelos the following deal: plead guilty to one count of drug distribution and an accompanying § 924(c) count (a mandatory sentence enhancement for carrying a gun “in furtherance” of drug trafficking).[182] All other charges would be dropped, and the government would recommend a fifteen-year sentence.[183] Angelos declined the offer, went to trial, and was convicted of the drug offense plus three gun enhancements, resulting in a fifty-five-year sentence.[184]

The first step in evaluating the plea-bargaining dynamic in the Angelos case is to assess the likely trial outcome ex ante. For this inquiry, it was critical that Angelos receive full disclosure about the government’s evidence at the time of the plea offer. While there appeared to be little question that the government would be able to convict Angelos of selling marijuana to an undercover informant,[185] the evidence supporting the gun enhancements (an informant’s uncorroborated claim) was thin. Police observed the drug sales but had not seen any guns. The informant’s written post-buy reports did not mention any weapons.[186] Angelos needed to know these facts, along with information about the informant’s credibility (or lack thereof), to assess the government’s case against him.

The next step is the plea offer. The plea offer was harsh, reflecting the prosecution’s apparent belief that Angelos was a “known gang member” and “mid-to-high [level] drug dealer”[187] and his refusal to cooperate.[188] Angelos had no prior record, but the plea offer contemplated fifteen years in prison and included a mandatory gun enhancement. The potential trial outcome for Angelos was even more severe: fifty-five years in prison. Considering all the factors, the plea offer Angelos received falls on the middle-right of the uncertainty-versus-trial-penalty graph. There was high uncertainty about the correct choice and a high trial penalty.

The Angelos case parallels Bordenkircher v. Hayes with respect to uncertainty. Although conventional narratives overlook this aspect of the case, Paul Hayes’s choice was not as clear as is generally reported (five years versus life). Hayes’s brief in the Supreme Court noted that judges in Kentucky do not have to follow bargained-for sentence recommendations.[189] Even if Hayes had accepted the five-year plea deal offered by the prosecutor, the judge could have rejected the prosecutor’s recommendation and sentenced Hayes higher—up to ten years in prison.[190] And the life sentence Hayes actually received after trial left him eligible for early parole release—which Hayes, in fact, received after nine years.[191] These nuances (judicial sentencing discretion and parole release) made Hayes’s ex ante choice much more complicated than the conventional framing. In addition, check forgery can be tricky to prove.[192] Given any realistic chance of acquittal, the plea offer in Bordenkircher becomes strikingly stingy,[193] which helps explain why Hayes turned it down. This suggest that, as in the Angelos case and contrary to conventional wisdom, the problem with the plea bargaining in Bordenkircher was uncertainty. Even in the face of draconian background law, the prosecution’s offer did not clearly make Hayes better off. It presented a spin of the roulette wheel. Like Angelos, Hayes spun the wheel and lost. But it is important to distinguish the severity of the outcomes from the problem of uncertainty.[194] The distinction becomes clear if we contrast these cases with the Tigano[195] case.

Joseph Tigano similarly faced a high trial penalty and an unfavorable plea offer. But there was little uncertainty. Having confessed to multiple federal agents that he was responsible for a large marijuana crop growing in his home, Tigano was almost certain to lose at trial, at which point he would receive a mandatory twenty-year prison sentence.[196] Thus, unlike the choice presented to Hayes and Angelos, Tigano’s dilemma had a clear answer. Kevin’s case offers a similar data point.[197] Kevin faced a high trial penalty but little uncertainty about the wisdom of taking the generous plea deal. In both Kevin and Tigano’s cases, the absence of uncertainty meant that the plea offers unequivocally improved the defendants’ circumstances.

We can fill out the picture with one last type of plea: a common plea deal in the Superior Court in Washington, D.C. Defendants caught selling drugs in D.C. are typically charged with two offenses: distribution of the drugs that they sold, and possession with intent to distribute any unsold drugs found in their possession.[198] Each charge is a felony punishable by up to thirty years in prison.[199] There is usually little question about the government’s evidence. Most cases arise when undercover police officers approach a person they suspect of selling drugs and purchase drugs with prerecorded money. A team of officers then arrests the suspect and seizes any remaining drugs and (usually) the prerecorded money.[200]

A standard plea offer in these “buy-bust” cases is to drop the possession-with-intent charge in exchange for a guilty plea to the distribution charge.[201] Most notably, there is little difference between sentencings after guilty pleas and sentencings after trial. Judges understandably sentence defendants concurrently on the two offenses.[202] The primary difference is that a defendant who pleads guilty can seek leniency for having accepted responsibility. These aren’t the kind of plea deals that appear in media accounts or law-review articles. Yet plea deals with minor concessions offered to defendants expected to lose at trial may be the most common type.[203]

The array of scenarios sketched out above allows for a useful comparison of various plea deals. The following chart shows each of the plea deals discussed above on the uncertainty-versus-trial-penalty matrix.[204] A key for the cases is as follows:

D = D.C. Drug Case K = Kevin T = Tigano

H = Hayes A = Angelos

While the trial penalties and the severity of the plea deals vary throughout, the problematic plea deals are at the top of the chart where—at the time of the plea offer—uncertainty is high. For both Angelos (A) and Hayes (H), it is the high uncertainty that creates a situation where plea bargaining worsens the background injustice of draconian sentencing laws. There is simply no way to know, ex ante, whether these defendants should take the deals offered. The prosecution in those cases presented Angelos and Hayes with the equivalent of a game of chance to determine their fate. In the other three examples, two of which also involve severe mandatory sentences, the casino element is absent. Plea bargaining either unequivocally improved the defendant’s situation or had little impact. For Kevin (K) and Tigano (T), the plea deals are clearly better than the alternative: the “right” choice is obvious, and the defendants’ situations improve through plea bargaining. In the typical D.C. drug case, the plea deal only slightly changes the calculus. A defendant with a plausible defense (for example, a defendant who does not have the marked money on his person at the time of arrest) will go to trial. A defendant with little chance for acquittal will take the modest plea deal. Again, plea bargaining either mildly improves the defendant’s circumstances or leaves them unchanged. Overall, the chart illustrates that uncertainty, not the trial penalty, severity, or other factors, distinguishes scenarios where plea bargaining is likely to improve defendants’ predicaments (D, K, T) from those where it worsens them (A, H).


The American criminal justice system suffers from a cascade of problems. There is a problem of severity, where judges impose, or statutes dictate, sentences that are disproportionate to the underlying offense. There is a problem of overcharging, where prosecutors charge too many and too severe offenses. There is a problem of inconsistency, where severity is not evenly applied due to distortions, including those introduced by race and class. And there is the problem of innocence, where defendants are convicted of crimes they did not commit. These problems are like diseases infecting the criminal justice system. This Article situates plea bargaining as a symptom of these diseases, not their cause. In fact, in many applications, plea bargaining functions to mitigate the system’s many flaws. But plea bargaining sometimes makes a unique contribution to the dysfunction: increased uncertainty.

Identifying plea bargaining’s distinct contribution to American dysfunction produces two important insights. First, it provides a blueprint for improving the largely unregulated plea-bargaining process—the dominant mechanism for resolving criminal cases. This is important not just because of the potential benefits of these reforms but also to displace other common plea-bargaining reform proposals that could aggravate rather than mitigate injustice. Second, and perhaps more important, by revealing that plea bargaining is rarely the primary source of the problems plaguing the criminal justice system, the analysis helps to funnel limited reform energy to the places where it can be most effective.

  1. . United States v. Angelos, 345 F. Supp. 2d 1227, 1230–32 (D. Utah 2004).
  2. . United States v. Tigano, 880 F.3d 602, 605 (2d Cir. 2018); Brief for Defendant-Appellant at 38, United States v. Tigano, 880 F.3d 602 (2d Cir. 2018) (No. 15-3073), 2017 WL 880549, at *38 (recounting plea offer); Brief for Appellee at 2, United States v. Tigano, 880 F.3d 602 (2d Cir. 2018) (No. 15-3073), 2017 WL 2418187, at *2 (describing sentence).
  3. . Defendant’s Motion for Judgment of Acquittal at 7, United States v. Whitton, No. 2:19-cr-00224, 2020 WL 8458727 (E.D. Wis. Mar. 19, 2020).
  4. . United States v. Whitton, No. 2:19-cr-00224, 2020 WL 8408949, at *1 (E.D. Wis. July 17, 2020) (announcing sentence); United States v. Whitton, No. 2:19-cr-00224, 2020 WL 1991585, at *1 (E.D. Wis. Apr. 27, 2020) (describing facts); Press Release, U.S. Dep’t of Just., Waukesha Bar Owner Sentenced to 15 Years for Arson and Related Charges (July 16, 2020), https://www.justice.‌gov/usao-edwi/pr/waukesha-bar-owner-sentenced-15-years-arson-and-related-charges [https://‌].
  5. . William Ortman, Second-Best Criminal Justice, 96 Wash. U. L. Rev. 1061, 1084 (2019) (“The academy’s renewed attention to plea bargaining reform is cause for celebration . . . .”).
  6. . Missouri v. Frye, 566 U.S. 134, 144 (2012) (emphasis in original) (quoting Robert E. Scott & William J. Stuntz, Plea Bargaining as Contract, 101 Yale L.J. 1909, 1912 (1992)); Jed S. Rakoff, Why Innocent People Plead Guilty, N.Y. Rev. Books (Nov. 20, 2014) (“In actuality, our criminal justice system is almost exclusively a system of plea bargaining, negotiated behind closed doors and with no judicial oversight.”).
  7. . Nat’l Ass’n of Crim. Def. Laws., The Trial Penalty: The Sixth Amendment Right to Trial on the Verge of Extinction and How to Save It 11 (2018),‌getattachment/95b7f0f5-90df-4f9f-9115-520b3f58036a/the-trial-penalty-the-sixth-amendment-right-to-trial-on-the-verge-of-extinction-and-how-to-save-it.pdf []; see Ronald F. Wright, Trial Distortion and the End of Innocence in Federal Criminal Justice, 154 U. Pa. L. Rev. 79, 86 (2005) (defining “trial penalty” as “the differential between the sentence after plea and sentence after trial”); Nancy J. King & Rosevelt L. Noble, Felony Jury Sentencing in Practice: A Three-State Study, 57 Vand. L. Rev. 885, 896 (2004) (“The difference between the sentence imposed after a jury trial and the sentence imposed after a guilty plea is commonly known as the ‘plea discount,’ or, less kindly, the ‘trial penalty’ or ‘trial tariff.’”).
  8. . Ortman, supra note 5, at 1086 (arguing that “overlapping offenses and draconian sentencing laws (including but not limited to mandatory minimums) . . . are the driving forces behind large trial penalties, which are in turn the driving forces behind the deep pathologies of status quo plea bargaining”); Russell D. Covey, Fixed Justice: Reforming Plea Bargaining with Plea-Based Ceilings, 82 Tul. L. Rev. 1237, 1243 (2008) (“Most commentators that have advocated reform of plea bargaining have begun by criticizing the dramatic gap between plea and trial sentences.”).
  9. . William Ortman, Probable Cause Revisited, 68 Stan. L. Rev. 511, 555–56 (2016) (“Although the harshest posttrial sentencing laws are directly implicated only in the tiny percentage of cases that go to trial, their presence is felt in the myriad cases bargained in their shadow.”).
  10. . About, Prosecutors Alliance of Cal., [https://‌].
  11. . Rachel Swan & Megan Cassidy, S.F. Parolee Accused of Killing Pedestrians Faced Life Sentence in Earlier Case, but Got Five Years, S.F. Chron., https://www.sfchronicle.‌com/crime/article/Fatal-hit-and-run-DA-Boudin-charges-suspect-15845917.php [‌4D7K-PP7Z] (Jan. 8, 2021, 4:28 PM).
  12. . These events received attention because, after his release, McAlister struck and killed two pedestrians while driving drunk in 2021. Id.
  13. . Press Release, U.S. Dep’t of Just., United States Attorney McSwain Delivers Remarks on the Ongoing Public Safety Crisis in Philadelphia (Sept. 14, 2020), [https://‌].
  14. . John N. Mitchell, U.S. Attorney Files New Charges in Krasner Case, Calls Him ‘Too Lenient’, Phila. Trib. (Feb. 28, 2019), []; see 18 Pa. Cons. Stat. § 1102(c) (2012) (imposing up to forty years for attempted murder with serious bodily injury).
  15. . Emma Kaufman, The Prisoner Trade, 133 Harv. L. Rev. 1815, 1875 (2020); see Thea Johnson, Public Perceptions of Plea Bargaining, 46 Am. J. Crim. L. 133, 135 (2019) (“[M]ost academic literature and an increasing amount of popular media paints a dark picture of a practice pockmarked with fatal flaws . . . .”); Molly J. Walker Wilson, Defense Attorney Bias and the Rush to the Plea, 65 U. Kan. L. Rev. 271, 271 (2016) (“The pervasiveness of plea bargains in our criminal justice system is of concern to many, and scholars in the legal academy have often been critical of the practice.”).
  16. . Daniel Epps, Adversarial Asymmetry in the Criminal Process, 91 N.Y.U. L. Rev. 762, 766 (2016); see also H. Mitchell Caldwell, Coercive Plea Bargaining: The Unrecognized Scourge of the Justice System, 61 Cath. U. L. Rev. 63, 86–89 (2011) (summarizing a wide variety of divergent reform proposals); Stephen J. Schulhofer, Plea Bargaining as Disaster, 101 Yale L.J. 1979, 2009 (1992) (“Plea bargaining is a disaster. It can be, and should be, abolished.”).
  17. . See Susan R. Klein, Aleza S. Remis & Donna Lee Elm, Waiving the Criminal Justice System: An Empirical and Constitutional Analysis, 52 Am. Crim. L. Rev. 73, 77 (2015) (“There is significant scholarly disagreement on the advantages of plea bargaining . . . .”); Oren Gazal-Ayal, Partial Ban on Plea Bargains, 27 Cardozo L. Rev. 2295, 2297 (2006) (“Very few issues in the American criminal justice system generate such fierce controversy as plea bargaining . . . .”).
  18. . George Fisher, Plea Bargaining’s Triumph, 109 Yale L.J. 857, 865–66 (2000); see also Jennifer L. Mnookin, Uncertain Bargains: The Rise of Plea Bargaining in America, 57 Stan. L. Rev. 1721, 1722 (2005) (reviewing George Fisher, Plea Bargaining’s Triumph: A History of Plea Bargaining in America (2003)) (“At the most basic level, Fisher’s argument is that when parties have both the incentive and power to bargain, they will, almost inevitably, bargain.”); Andrea Kupfer Schneider & Cynthia Alkon, Bargaining in the Dark: The Need for Transparency and Data in Plea Bargaining, 22 New Crim. L. Rev. 434, 435 (2019) (“Whereas early critics called for the abolition of or limits on plea bargaining, most now recognize that plea bargaining is an unavoidable part of the criminal legal system.”); Ortman, supra note 9, at 554 (“[F]or better or worse, abolition is politically a nonstarter.”); Russell D. Covey, Plea Bargaining and Price Theory, 84 Geo. Wash. L. Rev. 920, 962–63 (2016) (concluding that banning plea bargains “is not . . . a realistic option” because “prohibition of plea bargaining is as difficult to enforce as any other prohibition in the face of strong demand, and only channels demand into black markets”). This appears to be the consensus view. One exception was Albert Alschuler, who previously argued to the contrary but has since converted. Compare Albert W. Alschuler, Implementing the Criminal Defendant’s Right to Trial: Alternatives to the Plea Bargaining System, 50 U. Chi. L. Rev. 931, 936 (1983) (introducing reforms that might allow for termination of plea bargaining), with Albert W. Alschuler, Lafler and Frye: Two Small Band-Aids for a Festering Wound, 51 Duq. L. Rev. 673, 706 (2013) [hereinafter Alschuler, Lafler and Frye] (“The time for a crusade to prohibit plea bargaining has passed.”).
  19. . “A few jurisdictions have attempted abolition, at least in part, but those experiments have produced no lasting successes and certainly have not inspired widespread emulation. Most observers of the criminal justice system have come to believe (rightly or wrongly) that abolition is simply not feasible.” Covey, supra note 8, at 1240 (footnote omitted); Michael M. O’Hear, Plea Bargaining and Procedural Justice, 42 Ga. L. Rev. 407, 409 (2008) (“[D]espite the strenuous objections of prominent academic commentators, plea bargaining seems to be growing only more entrenched over time.” (footnote omitted)); Jenia Iontcheva Turner, Judicial Participation in Plea Negotiations: A Comparative View, 54 Am. J. Comp. L. 199, 217 (2006) (describing the rise of plea bargaining in an inquisitorial system—Germany—where it “developed organically, without direction from any written rule change or centrally established policy”).
  20. . For example, a rigorous study of Philadelphia murder cases found that defendants who were represented by public defenders (as opposed to appointed attorneys) received shorter prison sentences, primarily because they were more likely to take plea deals. James M. Anderson & Paul Heaton, How Much Difference Does the Lawyer Make? The Effect of Defense Counsel on Murder Case Outcomes, 122 Yale L.J. 154, 159, 187 (2012).
  21. . Michael L. Rubinstein, Stevens H. Clarke & Teresa J. White, U.S. Dep’t of Just., Alaska Bans Plea Bargaining vii–viii (1980) (finding that sentences became more severe for offenders without criminal records and those charged with less severe offenses and remained the same for serious offenses). A similar dynamic appeared in El Paso, Texas, where defendants continued to plead guilty, although at a lower rate than before the ban, and “[d]efense attorneys were commonly of the opinion that the judges sentenced much more harshly.” Robert A. Weninger, The Abolition of Plea Bargaining: A Case Study of El Paso County, Texas, 35 UCLA L. Rev. 265, 293, 303 (1987) (reporting that “after the ban a large proportion of defendants—more than a third—still admitted their criminal liability,” but only reporting anecdotal accounts of sentencing due to the complexity of the necessary statistical analysis). Although the Alaska ban continued in effect to varying degrees, it was modified to allow broad categories of plea bargains in 1980, and as time went on, its effects became more difficult to distinguish from the effects of other changes, like changes to Alaska’s sentencing laws. See Teresa White Carns & John Kruse, A Re-Evaluation of Alaska’s Plea Bargaining Ban, 8 Alaska L. Rev. 27, 29, 32 (1991) (“In the face of a new criminal code, new provisions for presumptive sentencing and a drastically altered state economy, it was unclear whether the ban persisted and evolved or decayed.” (footnotes omitted)); see also Bryan C. McCannon, Alaska’s Plea Bargaining Ban (Jan. 7, 2021) (unpublished manuscript at 3), https://‌ [] (expressing a lack of clarity around the effect of the ban on the prevalence of plea bargaining).
  22. . See Rubinstein et al., supra note 21, at 12 (noting the expressed purpose of the ban was to “bring about longer sentences”); see also Will Bain, Plea Bargaining, Legislative Limits, and the Separation of Powers, Colo. Law., Mar. 2003, at 63, 67 (stating that Colorado’s legislative limits on plea bargaining in domestic violence and drunk driving cases were intended “to ensure the vigorous prosecution” of such crimes and counteract “the perception that district attorneys and judges were treating drunk drivers and domestic violence defendants too leniently”).
  23. . See Joint Comm. on N.Y. Drug L. Eval., Ass’n of the Bar of N.Y.C., The Nation’s Toughest Drug Law: Evaluating the New York Experience 4 (1978) (describing the statute’s mandatory prison terms and plea bargain limits).
  24. . Cal. Penal Code § 1192.7(a)(2) (West 2022) (“Plea bargaining in [certain cases] is prohibited, unless there is insufficient evidence to prove the people’s case, or testimony of a material witness cannot be obtained, or a reduction or dismissal would not result in a substantial change in sentence.”); see also Marc Mauer, Why Are Tough on Crime Policies So Popular?, 11 Stan. L. & Pol’y Rev. 9, 10 (1999) (describing New York’s Rockefeller Drug Laws, which “called for harsh sentences for drug crimes and limits on plea bargaining,” as “the most severe in the nation at the time”); cf. Scott W. Howe, The Value of Plea Bargaining, 58 Okla. L. Rev. 599, 599 (2005) (“The public tends to believe that bargaining treats defendants too leniently.”). A short ban in El Paso, Texas, arose when judges resisted negotiated plea deals in burglary cases and the district attorney responded by barring plea bargains altogether. Weninger, supra note 21, at 275–76.
  25. . See Peter A. Joy & Rodney J. Uphoff, Sentencing Reform: Fixing Root Problems, 87 UMKC L. Rev. 97, 108 (2018) (“In California, the parties struck plea bargains early in the process at a stage where the ban did not apply, and in New York, prosecutors and defense lawyers used pre-indictment plea agreements instead.”).
  26. . Cf. People v. Cardoza, 207 Cal. Rptr. 388, 391 n.4 (1984) (noting that the purpose of the provision was “to prevent the prosecutorial abuse of the plea bargaining procedure, i.e., plea bargains which are too lenient” (emphasis in original)).
  27. . Alschuler, Lafler and Frye, supra note 18, at 707 (concluding that those aggrieved by plea bargaining should “resist over-criminalization (especially the expansion of federal criminal law)” and “oppose severe punishments”).
  28. . Scott & Stuntz, supra note 6, at 1965 (proposing the elimination of “mandatory sentences that attach to overbroad criminal statutes”).
  29. . Máximo Langer, Rethinking Plea Bargaining: The Practice and Reform of Prosecutorial Adjudication in American Criminal Procedure, 33 Am. J. Crim. L. 223, 242 (2006) (arguing that plea bargaining becomes coercive and pleas involuntary whenever a “prosecutor’s . . . plea proposal . . . includes [(1)] binding over for trial a defendant that no reasonable jury could find guilty beyond a reasonable doubt, [(2)] a sentence that would not be fair for the particular case, or [(3)] overcharging”).
  30. . See, e.g., Covey, supra note 18, at 968 (“Obvious steps in this direction would include abolishing mandatory minimum sentences, habitual offender and three strikes laws, and the like, and reducing maximum sentences across the board.”); Rakoff, supra note 6 (“If there were the political will to do so, we could eliminate mandatory minimums, eliminate sentencing guidelines, and dramatically reduce the severity of our sentencing regimes in general.”).
  31. . See infra subpart II(D).
  32. . See Alice Woolley, Hard Questions and Innocent Clients: The Normative Framework of The Three Hardest Questions, and the Plea Bargaining Problem, 44 Hofstra L. Rev. 1179, 1181 (2016) (discussing circumstances where “accepting the plea is in the [innocent] client’s best interests”); Scott & Stuntz, supra note 6, at 1961 (“The situation is a sad one, but preventing the offer only makes it sadder.”).
  33. . See infra subpart II(B).
  34. . See generally Jeffrey Bellin, Mass Incarceration Nation: How the United States Became Addicted to Prisons and Jails and How It Can Recover (forthcoming Cambridge Univ. Press 2023) (discussing how the confluence of these problems contributes to mass incarceration).
  35. . For a distinct take, see Albert W. Alschuler, Plea Bargaining and Its History, 79 Colum. L. Rev. 1, 3 (1979). See id. (“Plea bargaining consists of the exchange of official [concessions] for a defendant’s act of self-conviction.”).
  36. . See, e.g., Scott Daugherty, He Cooperated with Prosecutors. Now He’s Going to Prison for a Crime a Jury Said Never Happened., Virginian-Pilot (June 19, 2019, 11:00 AM), https://www.‌ [https://perma.‌cc/Q7XX-NU4U] (recounting a case where, after eight defendants were acquitted, the one codefendant who pled guilty was the only person punished for the crime).
  37. . Frank H. Easterbrook, Plea Bargaining as Compromise, 101 Yale L.J. 1969, 1976–77 (1992).
  38. . Cf. Langer, supra note 29, at 232 (pointing out that “[t]here are many situations in which we receive proposals that we cannot rationally refuse, yet we still consider them noncoercive,” and providing as an example a doctor’s proposal of an expensive treatment for a life-threatening ailment).
  39. . Cf. Adam M. Samaha, Randomization in Adjudication, 51 Wm. & Mary L. Rev. 1, 4 (2009) (chronicling historical resistance to randomization of judicial merits decisions); Stephanos Bibas, Plea Bargaining Outside the Shadow of Trial, 117 Harv. L. Rev. 2463, 2528–29 (2004) (recognizing “uncertainty” about likely outcomes as one of a number of “pervasive influence[s]”). For a proposal that would push criminal justice in the other direction, see generally Kiel Brennan-Marquez, Darryl K. Brown & Stephen E. Henderson, The Trial Lottery, 56 Wake Forest L. Rev. 1 (2021) (proposing that a small percentage of plea deals be randomly selected for a compulsory trial).
  40. . See infra subpart II(A).
  41. . See infra subpart II(B).
  42. . See infra subpart II(D).
  43. . See Bellin, supra note 34, at 184–85, 190–93 (proposing reduced sentences and increasing the rate of early release, and urging prosecutors to dismiss cases with insufficient evidence to avoid convicting the innocent).
  44. . Cf. Carlos Berdejó, Criminalizing Race: Racial Disparities in Plea-Bargaining, 59 B.C. L. Rev. 1187, 1190–92 (2018) (finding racial discrimination in plea bargaining in Wisconsin).
  45. . See Laurie L. Levenson, Peeking Behind the Plea Bargaining Process: Missouri v. Frye & Lafler v. Cooper, 46 Loy. L.A. L. Rev. 457, 467 (2013) (noting that “by 1879, 70 percent of all pleas were guilty”).
  46. . See Corinna Barrett Lain, Accuracy Where It Matters: Brady v. Maryland in the Plea Bargaining Context, 80 Wash. U. L.Q. 1, 1 (2002) (“Since at least the 1960s, the plea bargaining rate for American criminal cases has been around 90%.”).
  47. . See William J. Stuntz, The Pathological Politics of Criminal Law, 100 Mich. L. Rev. 505, 552 (2001) (describing benefits to legislators of guilty pleas).
  48. . See, e.g., Weninger, supra note 21, at 303 (reporting that after a ban in El Paso, Texas, “a backlog of cases quickly accumulated and . . . substantial delays were encountered in the administration of justice”).
  49. . See Miriam Hechler Baer, Cooperation’s Cost, 88 Wash. U. L. Rev. 903, 930 (2011) (“Whatever its drawbacks, cooperation’s mix of transparent and unobserved policing improves the government’s ability to deter and incapacitate offenders.”); Jonathan Abel, Cops and Pleas: Police Officers’ Influence on Plea Bargaining, 126 Yale L.J. 1730, 1787 (2017) (arguing that “police officers in many jurisdictions do play a significant role in plea bargaining”).
  50. . Brennan-Marquez et al., supra note 39, at 14 (discussing this and other benefits to defendants of plea bargaining); Ortman, supra note 9, at 555 (“For defendants, plea bargaining means avoiding harsh posttrial sentences.”).
  51. . Brennan-Marquez et al., supra note 39, at 9–10 (“Plea bargains mean less work for lawyers and judges, they provide certainty of outcome to both parties, and they help judges prevent backlogged court dockets.”).
  52. . Fisher, supra note 18, at 1075.
  53. . John H. Langbein, Torture and Plea Bargaining, 46 U. Chi. L. Rev. 3, 12–13 (1978) (footnote omitted).
  54. . See, e.g., Ortman, supra note 5, at 1078 (“As John Langbein memorably argued, the coercive logic of plea bargaining is different in degree, but not in kind, from the medieval continental practice of securing confessions by torture.”); Gregory M. Gilchrist, Trial Bargaining, 101 Iowa L. Rev. 609, 631 (2016) (“Fundamentally, plea bargaining is coercive. Langbein famously likened the coercive impact of plea bargaining to that of medieval European torture . . . . The analogy will seem overwrought to some—it is not.”); Allegra M. McLeod, Exporting U.S. Criminal Justice, 29 Yale L. & Pol’y Rev. 83, 118 (2010) (“Langbein makes a compelling case that the widespread reliance on plea bargaining . . . is morally wrong because it is coercive.”).
  55. . Scott & Stuntz, supra note 6, at 1909–10. A study of defendants’ perceptions of plea bargaining similarly suggests greater acceptance of the practice by those affected by it. Jeanette Hussemann & Jonah Siegel, Pleading Guilty: Indigent Defendant Perceptions of the Plea Process, 13 Tenn. J.L. & Pol’y 459, 490 (2019) (“Defendants charged with both felony and lesser charges articulated positive perceptions of the plea process (62 percent) and outcome (62 percent and 81 percent, respectively).”).
  56. . Easterbrook, supra note 37, at 1975.
  57. . Schulhofer, supra note 16, at 2009.
  58. . See Russell M. Gold, Carissa Byrne Hessick & F. Andrew Hessick, Civilizing Criminal Settlements, 97 B.U. L. Rev. 1607, 1628 (2017) (“[P]rosecutors often need not accept a plea bargain for anything less than the charges or the punishment that they want.”); Covey, supra note 18, at 921 (“[T]he deck is stacked to ensure that prosecutors usually get what they want (convictions accompanied by substantial punishment) without giving up any real bargaining concessions.”). But see Epps, supra note 16, at 795 (“[E]ven when prosecutors prefer a higher punishment, they are often willing to settle for less, because they value reducing process costs—both because they want to preserve resources for other cases, and because they have self-interested reasons for reducing their workload.”).
  59. . Gold et al., supra note 58, at 1617 (“Prosecutors have massive leverage to force defendants to plead guilty rather than proceed to trial.”); Jed S. Rakoff, Why Prosecutors Rule the Criminal Justice System—And What Can Be Done About It, 111 Nw. U. L. Rev. 1429, 1430 (2017) (“The plea bargain is the ultimate source of this ever-increasing prosecutorial power.”).
  60. . Rachel Elise Barkow, Prisoners of Politics: Breaking the Cycle of Mass Incarceration 38, 43, 54 (2019).
  61. . See generally Brandon L. Garrett, Convicting the Innocent: Where Criminal Prosecutions Go Wrong (2011) (examining the more than 250 cases where DNA testing exonerated actually innocent people).
  62. . It is also a critical consideration that the “accused often has a large array of facts and adverse evidence against him,” which make acquittal unlikely. Am. Bar Ass’n, Standards Relating to The Administration of Criminal Justice 108 (1974).
  63. . Jeannie Suk, Criminal Law Comes Home, 116 Yale L.J. 2, 56 (2006) (discussing this dynamic in misdemeanor domestic violence cases).
  64. . Cf. Josh Bowers, Punishing the Innocent, 156 U. Pa. L. Rev. 1117, 1149 (2008) (“Time-served pleas or the equivalent are so prevalent because bargains that require jail time are more likely to prompt defendants to balk.”).
  65. . Bibas, supra note 39, at 2467 (“[M]any plea bargains diverge from the shadows of trials.”).
  66. . Id.
  67. . Id.
  68. . Cf. Allison D. Redlich, Miko M. Wilford & Shawn Bushway, Understanding Guilty Pleas Through the Lens of Social Science, 23 Psych. Pub. Pol’y & L. 458, 462 (2017) (“Many of the issues raised by Bibas can be dealt with in the shadow model.”).
  69. . Bibas, supra note 39, at 2468; cf. Molly J. Walker Wilson, Defense Attorney Bias and the Rush to the Plea, 65 U. Kan. L. Rev. 271, 276 (2016) (“In sum, defense lawyers are prone to biased decision-making and are incentivized to make concessions to get their clients’ cases settled efficiently.”). Criminal law scholars often draw a contrast to civil settlements, which they perceive to, in fact, mirror the governing law, but research suggests that is not the case. See, e.g., Nora Freeman Engstrom, Run-of-the-Mill Justice, 22 Geo. J. Legal Ethics 1485, 1489 (2009) (analyzing “settlement mills [that] operate in a manner that bears little resemblance to—and thus implicitly challenges—conventional notions of bargaining”).
  70. . Cf. Model Rules of Pro. Conduct r. 1.4 cmt. 2 (Am. Bar Ass’n 2020) (“[A] lawyer who receives . . . a proffered plea bargain in a criminal case must promptly inform the client of its substance . . . .”).
  71. . See Bibas, supra note 39, at 2493 (“[P]retrial detention places a high premium on quick plea bargains in small cases, even if the defendant would probably win acquittal at an eventual trial.”). It is important to emphasize that this situation is a travesty. But the villain here is pretrial detention, not plea bargaining.
  72. . See Adam J. Kolber, The Bumpiness of Criminal Law, 67 Ala. L. Rev. 855, 874 (2016) (“The law’s bumpy treatment of uncertainty at trial stands in stark contrast to its treatment when plea bargaining.”); see also Eisha Jain, Prosecuting Collateral Consequences, 104 Geo. L.J. 1197, 1239 (2016) (“[P]rosecutors as well as defendants and defense attorneys need to be aware of collateral consequences during the plea bargaining process.”).
  73. . See Easterbrook, supra note 37, at 1976 (criticizing academic critics as committing, “the Nirvana Fallacy, comparing an imperfect reality to a perfection achievable only in imaginary systems”); see also Gerard E. Lynch, Screening Versus Plea Bargaining: Exactly What Are We Trading Off?, 55 Stan. L. Rev. 1399, 1407 (2003) (critiquing academic commentators as being “too quick to accept the conventional view that ‘plea bargaining’ is a bad thing, without focusing carefully enough on what aspects of the plea bargaining system are undesirable, and why they desire to eliminate it”).
  74. . Norman L. Reimer & Martín Antonio Sabelli, The Tyranny of the Trial Penalty: The Consensus that Coercive Plea Practices Must End, 31 Fed. Sent’g Rep. 215, 215 (2019).
  75. . Nat’l Ass’n of Crim. Def. Laws., supra note 7, at 11; Andrew St. Laurent, Anatomy of a Plea, The Champion, June 2019, at 42, 42 (“The ‘trial penalty’ refers to the substantial difference between the sentence proposed in a plea offer prior to trial versus the sentence a defendant receives after trial.”); United States v. Holloway, 68 F. Supp. 3d 310, 313 (E.D.N.Y. 2014) (“The difference between the sentencing outcome if a defendant accepts the government’s offer of a plea bargain and the outcome if he insists on his right to trial by jury is sometimes referred to as the ‘trial penalty.’”); Clark Neily, A Distant Mirror: American-Style Plea Bargaining Through the Eyes of a Foreign Tribunal, 27 Geo. Mason L. Rev. 719, 730–31 (2020) (“This so-called ‘trial penalty’ is defined as ‘the substantial difference between the sentence offered prior to trial versus the sentence a defendant receives after a trial.’” (quoting Nat’l Ass’n of Crim. Def. Laws., supra note 7, at 11)).
  76. . Mary Price, Weaponizing Justice: Mandatory Minimums, The Trial Penalty, and the Purposes of Punishment, 31 Fed. Sent’g Rep. 309, 309 (2019).
  77. . Tina M. Zottoli, Tarika Daftary-Kapur, Georgia M. Winters & Conor Hogan, Plea Discounts, Time Pressures, and False-Guilty Pleas in Youth and Adults Who Pleaded Guilty to Felonies in New York City, 22 Psych., Pub. Pol’y, & L.250 (2016) (“[T]he discrepancy between the penalty threatened if the defendant loses at trial and the penalty offered in a plea deal . . . is often referred to as the ‘trial penalty’”); see also JaneAnne Murray, Ameliorating the Federal Trial Penalty Through a Systematic Judicial “Second Look” Procedure, 31 Fed. Sent’g Rep. 279, 279 (2019) (defining “the trial penalty” as “the threat of a substantially more severe conviction or longer sentence after trial than the one available upon a guilty plea”).
  78. . Jamie Fellner, An Offer You Can’t Refuse: How U.S. Federal Prosecutors Force Drug Defendants to Plead Guilty, 26 Fed. Sent’g Rep. 276, 276 (2014) (“In essence, it is the price prosecutors make defendants pay for exercising their right to trial.”).
  79. . Joy & Uphoff, supra note 25, at 101 (“But many judges do signal, directly or indirectly, that a much longer prison sentence will be forthcoming should the defendant go to trial and lose. This is commonly referred to as a ‘trial penalty’ for exercising the right to go to trial.”).
  80. . See Josh Bowers, Plea Bargaining’s Baselines, 57 Wm. & Mary L. Rev. 1083, 1088–89 (2016) (explaining the need for a baseline in distinguishing penalties from discounts); Darryl K. Brown, The Case for a Trial Fee: What Money Can Buy in Criminal Process, 107 Calif. L. Rev. 1415, 1416 (2019) (“In plea negotiations, the government can straightforwardly impose a cost on defendants’ decisions to go to trial: a ‘trial penalty’ in the form of greater liability and punishment if convicted.”). Nat’l Ass’n of Crim. Def. Laws., supra note 7, at 6 (“This ‘trial penalty’ results from the discrepancy between the sentence the prosecutor is willing to offer in exchange for a guilty plea and the sentence that would be imposed after a trial.”).
  81. . See Christopher Slobogin, Plea Bargaining and the Substantive and Procedural Goals of Criminal Justice: From Retribution and Adversarialism to Preventive Justice and Hybrid-Inquisitorialism, 57 Wm. & Mary L. Rev. 1505, 1510 (2016) (“[F]or bargaining to work, there must be a significant divergence between the sentence that results from a plea and the sentence that results from trial.”); Nancy J. King, David A. Soulé, Sara Steen & Robert R. Weidner, When Process Affects Punishment: Differences in Sentences After Guilty Plea, Bench Trial, and Jury Trial in Five Guidelines States, 105 Colum. L. Rev. 959, 962 (2005) (“We found that a significant plea discount—the difference between the average sentence given after a guilty verdict and the average sentence given after a g[u]ilty plea for the same offense—is evident for most offenses in all five states . . . .”).
  82. . King et al., supra note 81, at 962–63; see also Brian D. Johnson, Trials and Tribulations: The Trial Tax and the Process of Punishment, 48 Crime & Just. 313, 313–14 (2019) (“The trial penalty is consistently found across jurisdictions, offense types, and over time, and it is among the most robust findings in the empirical sentencing literature.”).
  83. . “[S]entencing judges tend to validate and encourage bargains through a ‘plea discount’ (or a trial penalty): They impose lighter sentences on those who waive their right to trial.” Ronald Wright & Marc Miller, The Screening/Bargaining Tradeoff, 55 Stan. L. Rev. 29, 40 (2002) (suggesting that limiting plea bargaining would result in increased case screening); see also Nancy J. King & Ronald F. Wright, The Invisible Revolution in Plea Bargaining: Managerial Judging and Judicial Participation in Negotiations, 95 Texas L. Rev. 325, 326–27 (2016) (concluding based on detailed interviews with judges in ten states that “the judge’s participation in negotiations has matured into a standard managerial tool”); Albert W. Alschuler, The Trial Judge’s Role in Plea Bargaining, Part I, 76 Colum. L. Rev. 1059, 1151–52 (1976) (“The experience of jurisdictions like Houston suggests that, even when judicial plea bargaining is formally disapproved, it is difficult to prevent off-the-record agreements between trial judges and especially favored defense attorneys.”).
  84. . Weninger, supra note 21, at 295.
  85. . See Levenson, supra note 45, at 467 (noting that “by 1879, 70 percent of all pleas were guilty”).
  86. . Vanderheyden v. Young, 11 Johns. 150, 151, 160 (N.Y. Sup. Ct. 1814) (rejecting post-sentence challenge by militia member who failed to heed the militia’s call to arms during the War of 1812 on the ground that the private had “waived every objection by pleading guilty, and throwing himself on the mercy of the court”).
  87. . Cf. Donald A. Dripps, Overcriminalization, Discretion, Waiver: A Survey of Possible Exit Strategies, 109 Penn St. L. Rev. 1155, 1171–72 (2005) (observing that “imposing a higher penalty on those who plead guilty is not wrong on principle”).
  88. . United States v. Goodwin, 457 U.S. 368, 372 (1982) (quoting Bordenkircher v. Hayes, 434 U.S. 357, 363).
  89. . See, e.g., United States v. Mendez–Zamora, 230 F. App’x 792, 794 (10th Cir. 2007) (“Mendez–Zamora portrays downward departures for pleading guilty and cooperating with the government as punishment for taking his chances with a jury trial, but in fact, these departures were rewards offered for accepting culpability.”); Steven P. Grossman, Making the Evil Less Necessary and the Necessary Less Evil: Towards a More Honest and Robust System of Plea Bargaining, 18 Nev. L.J. 769, 770 (2018) (“American courts at every level have engaged in all sorts of verbal and conceptual gymnastics to avoid acknowledging this reality.”); Bowers, supra note 80, at 1092 (“A threat depends on the existence of a penalty; in turn, a penalty is the product of a threat. The Court has made no serious attempt to resolve the circularity.”).
  90. . E.g., United States v. Whitton, No. 2:19-cr-00224, 2020 WL 1991585, at *2 (E.D. Wis. Apr. 27, 2020) (rejecting challenge to trial penalty, in part, because “a person who refuses to take responsibility for a crime stands in a different position than a person who pleads guilty and expresses remorse”).
  91. . State v. Walton, 91. P. 490, 493 (Or. 1907).
  92. . U.S. Sent’g Guidelines Manual § 3E1.1(a) (U.S. Sent’g Comm’n 2018) (providing a two-level decrease for acceptance of responsibility); id. (providing an additional decrease for “timely notifying authorities of his intention to enter a plea of guilty”); id. at cmt. 3 (“Entry of a plea of guilty prior to the commencement of trial combined with truthfully admitting the conduct comprising the offense of conviction, and truthfully admitting or not falsely denying any additional relevant conduct . . . will constitute significant evidence of acceptance of responsibility . . . .”).
  93. . United States v. Crawford, 906 F.2d 1531, 1534 (11th Cir. 1990).
  94. . Fed. R. Crim. P. 11 advisory committee’s note to 1974 amendment. This narrative is, of course, complicated by the bargained for aspect of most modern guilty pleas. Steven P. Grossman, An Honest Approach to Plea Bargaining, 29 Am. J. Trial Advoc. 101, 123–24 (2005) (“[A]lmost all criminal defendants are acting purely in their own perceived self-interest in deciding whether to accept a plea bargain . . . . [I]t is quite clear that almost all defendants who choose to plead guilty do so for reasons having nothing to do with any such step on [the] road [to rehabilitation].”); cf. Brandon L. Garrett, Why Plea Bargains Are Not Confessions, 57 Wm. & Mary L. Rev. 1415, 1443 (2016) (“A guilty plea is final because it is legal, not because it is a confession, or even necessarily a particularly accurate or complete admission.”).
  95. . See Langer, supra note 29, at 242 (evaluating prosecutors’ responsibility not to offer coercive plea deals).
  96. . Ben Grunwald, Distinguishing Plea Discounts and Trial Penalties, 37 Ga. St. U. L. Rev. 261, 281 (2021); see also Langer, supra note 29, at 242 (proposing three conditions to “set[] the baseline to apply in the context of plea dispositions”: “If the prosecutor’s . . . plea proposal . . . includes [(1)] binding over for trial a defendant that no reasonable jury could find guilty beyond a reasonable doubt, [(2)] a sentence that would not be fair for the particular case, or [(3)] overcharging”); Richard L. Lippke, Plea Bargaining in the Shadow of the Constitution, 51 Duq. L. Rev. 709, 719 (2013) (“Trial penalties . . . are added sanctions assigned to offenders not for their crimes but to discourage or punish them for the exercise of their constitutional rights.”). For a similar argument, see Brown, supra note 80, at 1449 (noting that when two options are so disparate, “no theory of criminal responsibility would say either sentence is as appropriate as the other”).
  97. . Grunwald, supra note 96, at 267, 281 (recognizing that this is the “tricky part”).
  98. . Grossman, supra note 89, at 781 (arguing that “in the vast majority of cases” it is “absurd to attempt to claim that there is a regular sentence”).
  99. . Bowers, supra note 80, at 1089 (emphasizing this as the courts’ current approach).
  100. . Gold et al., supra note 58, at 1617.
  101. . Covey, supra note 8, at 1254.
  102. . See supra note 18 and accompanying text.
  103. . Gold et al., supra note 58, at 1617–18; see also Carissa Byrne Hessick, Punishment Without Trial: Why Plea Bargaining Is a Bad Deal 44 (2021) (asserting that “lawmakers and prosecutors argue that these punishments are necessary to make defendants cooperate with law enforcement and to plead guilty”).
  104. . Rachel E. Barkow, Administering Crime, 52 UCLA L. Rev. 715, 728 (2005).
  105. . See Muscarello v. United States, 524 U.S. 125, 132 (1998) (explaining severe mandatory gun enhancement as intended by Congress “to combat the ‘dangerous combination’ of ‘drugs and guns’” by persuading “‘the man who is tempted to commit a Federal felony to leave his gun at home’” (first quoting Smith v. United States, 508 U.S. 223, 240 (1993); and then quoting 114 Cong. Rec. 22,231 (1968))). Studies suggest that such incentives are poor deterrents in a system where arrest for many offenses is unlikely, but this message has never penetrated the legislative (or popular) psyche. See generally Daniel S. Nagin, Deterrence in the Twenty-First Century, 42 Crime & Just. 199 (2013) (summarizing the empirical literature on deterrence and concluding the “evidence in support of the deterrent effect of various measures of the certainty of punishment is far more convincing and consistent than for the severity of punishment”).
  106. . Cal. Penal Code § 667(g) (West 2019) (“Prior serious or violent felony convictions shall not be used in plea bargaining . . . .”); see also Wash. Rev. Code § 9.94A.421(6) (2010) (explaining that “in no instance may the prosecutor agree not to allege prior convictions” as part of a plea agreement). California’s Three Strikes law was enacted both by legislation and a voter initiative. See People v. Superior Ct. (Romero), 917 P.2d 628, 630 (Cal. 1996) (explaining the legislative history behind the Three Strikes law).
  107. . Gold et al., supra note 58, at 1618 n.32 (quoting 161 Cong Rec. S955-02, S963 (daily ed. Feb. 12, 2015) (statement of Sen. Grassley)); Barkow, supra note 104, at 728 n.25 (noting legislative resistance to reduced sentences “in part because it would reduce incentives for defendants to cooperate with prosecutors”); Rachel E. Barkow, Institutional Design and the Policing of Prosecutors: Lessons from Administrative Law, 61 Stan. L. Rev. 869, 880 (2009) (“Representatives from the Department of Justice and the various United States Attorneys’ Offices often argue before Congress that legislation with inflated or mandatory punishments should be passed or retained because those laws give prosecutors the leverage they need to exact pleas and to obtain cooperation from defendants.”).
  108. . See, e.g., Michael Pope, Did Virginia Prosecutors Get What They Wanted from Jonnie Williams?, Am. Univ. Radio (Aug. 4, 2014),‌denies_he_had_intimate_contact_with_ex_governors_wife/ [] (describing immunity grant received by Jonnie Williams in return for his cooperation in the prosecution of former Virginia Governor Bob McDonnell).
  109. . See supra note 96 and accompanying text.
  110. . For example, in the early 2000s, the United States Attorney’s Office for the District of Columbia where I worked tried to enforce a policy that no plea deal would be offered to defendants in any gun case unless the defendants agreed to reveal how they obtained the gun they possessed illegally.
  111. . Deprived of the efficiencies of plea bargaining, judges might become more aggressive at resolving cases through other mechanisms.
  112. . Cf. Tracey L. Meares, Rewards for Good Behavior: Influencing Prosecutorial Discretion and Conduct with Financial Incentives, 64 Fordham L. Rev. 851, 869 (1995) (“Overcharging is also due in part to an abhorrence of losing that is central to prosecutorial culture.”).
  113. . Allison Orr Larsen, Bargaining Inside the Black Box, 99 Geo. L.J. 1567, 1569 (2011).
  114. . See United States v. Miller, 471 U.S. 130, 145 (1985) (rejecting challenge to conviction after charges dismissed prior to trial on the ground that “[t]he variance complained of added nothing new to the grand jury’s indictment and constituted no broadening”); id. at 136 (noting that when “the crime and the elements of the offense that sustain the conviction are fully and clearly set out in the indictment, the right to a grand jury is not normally violated by the fact that the indictment alleges more crimes or other means of committing the same crime”).
  115. . See Oren Bar-Gill & Omri Ben-Shahar, The Prisoners’ (Plea Bargain) Dilemma, 1 J. Legal Analysis 737, 758 (2009) (“The prosecutor’s imperfect information leads her to exercise more restraint and therefore shifts the equilibrium plea bargains downward.”); Josh Bowers, Grassroots Plea Bargaining, 91 Marq. L. Rev. 85, 88 (2007) (“As police turn up enforcement pressure, prosecutors may feel the need to pull back on the punishment throttle to ensure that these communities accept—or at least tolerate—hard-nosed police tactics.”); William J. Stuntz, Plea Bargaining and Criminal Law’s Disappearing Shadow, 117 Harv. L. Rev. 2548, 2554–56 (2004) (contending that “there is no good reason to assume that their preference is always for the harshest sentence they can possibly get” and that “very often” prosecutors “want to impose lower sentences than the law allows, sometimes lower than the law mandates” due to the law’s severity and docket pressures); cf. Paul T. Crane, Charging on the Margin, 57 Wm. & Mary L. Rev. 775, 783 (2016) (“Prosecutors will sometimes exercise their charging prerogative by filing a lesser charge and, in so doing, gain the strategic advantage that comes from significantly reducing a defendant’s procedural entitlements.”).
  116. . Bowers, supra note 64, at 1149. See also id. at 1141 (arguing that prosecutors “care little, if at all, about maximizing plea prices and ultimate sentence length”).
  117. . The jurisdiction where we would most expect to find an effect would be one with lots of severe mandatory minimum sentences that did not track judicial sentencing practices. See Jeffrey Bellin, Reassessing Prosecutorial Power Through the Lens of Mass Incarceration, 116 Mich. L. Rev. 835, 852 (2018) (laying out the requirements for scenarios where prosecutors would exercise the most control over plea bargaining).
  118. . See Weninger, supra note 21, at 299 (“The tests offer only weak support for the interview data and interpretations that prosecutors may have charged more conservatively after the ban.”).
  119. . See Carns & Kruse, supra note 21, at 43 tbl.1 (reporting that the rate of convictions per case referred for prosecution dropped slightly after that state’s ban, from 50% to 47% and then gradually rebounded to 49%).
  120. . Id. (reporting that an increase in charges screened out, from 11% to 31%, was accompanied by a decline in charges dismissed, from 40% to 22%).
  121. . See id. at 60 tbl. 4 (showing that the mean sentence for all offenses in Anchorage, Fairbanks, and Juneau, Alaska, increased from thirteen months to thirty-one months between 1975 and 1986); see supra note 21.
  122. . Rubinstein et al., supra note 21, at viii.
  123. . See supra notes 24–25 and accompanying text.
  124. . See supra notes 24–25 and accompanying text; Jeff Brown, Politics and Plea Bargaining: Victims’ Rights in California, 45 Hastings L.J. 697, 700 (1994) (book review) (discussing evasion of California restrictions).
  125. . Brian D. Johnson, Plea-Trial Differences in Federal Punishment: Research and Policy Implications, 31 Fed. Sent’g Rep. 256, 256 (2019) (quoting Nat’l Ass’n of Crim. Def. Laws., supra note 7, at 6).
  126. . See Jeffrey Bellin, The Power of Prosecutors, 94 N.Y.U. L. Rev. 171, 180–81 (2019) (discussing some of these checks).
  127. . See, e.g., Shima Baradaran Baughman, Subconstitutional Checks, 92 Notre Dame L. Rev. 1071, 1073 (2017) (arguing that “these constitutional checks are not functioning . . . in the criminal justice system”); Glenn Harlan Reynolds, Ham Sandwich Nation: Due Process When Everything Is a Crime, 113 Colum. L. Rev. Sidebar 102, 106 (2013) (noting the “longstanding aphorism that a good prosecutor can persuade a grand jury to indict a ham sandwich” and offering origin story).
  128. . See, e.g., Fed. R. Crim. P. 11(c)(3) (requiring that judges approve plea deals that include a contemplated sentence or be given discretion to decide the sentence under the parties’ deal).
  129. . See John B. Meixner, Jr., Modern Sentencing Mitigation, 116 Nw. L. Rev. 1395, 1437 & n.221 (2022) (reviewing sample of plea bargains in federal court and finding only a small percentage included a set sentence).
  130. . See Grossman, supra note 89, at 779 (“In almost all non-capital cases for which there is not a precise mandatory sentence, it is the judge who ultimately decides the defendant’s sentence.”).
  131. . Emily Bazelon, Charged: The New Movement to Transform American Prosecution and End Mass Incarceration 239, 248–49 (2019). For a detailed analysis of Kevin’s case, see Jeffrey Bellin, Defending Progressive Prosecution: A Review of Charged by Emily Bazelon, 39 Yale L. & Pol’y Rev. 218, 228–35 (2020) (book review).
  132. . Kevin took the deal, completed the program, and avoided both a prison term and a criminal record. See Bazelon, supra note 131, at 248–49 (noting dismissal of charges after Kevin fulfilled the conditions of the plea deal).
  133. . Richard A. Oppel, Jr., Court Approves Plea Deal with No Jail Time in Baylor Rape Case, N.Y. Times (Dec. 11, 2018), [].
  134. . Broad gaps actually create two potential problems: excessive plea discounts and unwarranted trial penalties. Most academic commentary targets the latter.
  135. . See supra note 130 and accompanying text.
  136. . Schulhofer, supra note 16, at 2007.
  137. . See Reimer & Sabelli, supra note 74, at 215 (summarizing other articles in a special issue of the Federal Sentencing Reporter and noting that “the ‘innocence problem’ is perhaps the most tragic and compelling reason why the trial penalty must be purged from the U.S. justice system”).
  138. . Rakoff, supra note 6.
  139. . Id.
  140. . Cf. Bowers, supra note 64, at 1119 (“On balance, plea bargaining is a categorical good for many innocent defendants, particularly in low-stakes cases.”); Covey, supra note 18, at 922 (“Abolition of plea bargaining would not necessarily improve the lot of innocent defendants.”).
  141. . ABA Standards for Criminal Justice Prosecution Function and Defense Function standard 3-3.9 cmt. at 77 (Am. Bar Ass’n, 1993); see also Memorandum from Att’y Gen. John Ashcroft to All Fed. Prosecutors Regarding Pol’y on Charging of Crim. Defendants (Sept. 22, 2003), [https://‌] (echoing earlier guidance and instructing prosecutors that “charges should not be filed simply to exert leverage to induce a plea”).
  142. . 434 U.S. 357, 358 (1978).
  143. . Id. at 358 n.1.
  144. . Id. at 359.
  145. . See, e.g., William Stuntz, The Collapse of American Criminal Justice 258 (2011) (explaining that “Bagby’s threat to use Kentucky’s three-strikes law was probably unfair even in Bagby’s eyes”); Bowers, supra note 80, at 1128 (“In Bordenkircher, the prosecutor purposefully disregarded his moral obligation to do equitable justice. He thereby acted impermissibly.” (footnote omitted)).
  146. . Hayes was indicted in January 1973. The inflation adjustment was calculated using the most recent data available (August 2022) from the U.S. Bureau of Labor Statistics. CPI Inflation Calculator, U.S. Bureau of Lab. Stats., [https://‌].
  147. . See, e.g., Jeffrey Bellin, Theories of Prosecution, 108 Calif. L. Rev. 1203, 1225 (2020) (considering the ethics of a “a prosecutor who is confident of the defendant’s guilt but seeks to spare a child-victim the trauma of trial,” deciding to “‘pile on’ charges to coerce a plea”).
  148. . Hayes v. Cowan, 547 F.2d 42, 44 (6th Cir. 1976) (citation omitted), rev’d, Bordenkircher v. Hayes, 434 U.S. 357 (1978).
  149. . Bordenkircher, 434 U.S. at 365.
  150. . For example, while there are over 3,000 counties in the country, Paul Mackun, Joshua Comenetz & Lindsay Spell, More than Half of U.S. Counties Were Smaller in 2020 than in 2010, U.S. Census Bureau (Aug. 21, 2021), [‌82H2-M6L6], the largest 75 counties generate almost half of the nation’s serious crime. George Ebo Browne & Matthew R. Durose, State Court Processing Statistics, Bureau of Just. Stats. (May 26, 2009), [https://‌].
  151. . See 1 Wayne R. LaFave, Jerold H. Israel, Nancy J. King & Orin S. Kerr, Crim. Proc. 168 (4th ed. 2015) (describing the vertical system as “standard for small offices” and the horizontal system, described above, as more common in larger offices).
  152. . Bordenkircher, 434 U.S. at 364 (“[B]y tolerating and encouraging the negotiation of pleas, this Court has necessarily accepted as constitutionally legitimate the simple reality that the prosecutor’s interest at the bargaining table is to persuade the defendant to forgo his right to plead not guilty.”).
  153. . Alschuler, Lafler and Frye, supra note 18, at 705.
  154. . See Levenson, supra note 45, at 460 (“It is a system that is tolerated because, without it, our criminal justice system would be so overwhelmed that it would collapse.”).
  155. . See generally Bellin, supra note 34 (discussing mass incarceration and offering a path to reform).
  156. . See Benjamin Levin, The Consensus Myth in Criminal Justice Reform, 117 Mich. L. Rev. 259, 268 (2018) (noting that the appearance of consensus with respect to overincarceration “gloss[es] over real differences” in views about what conduct should be criminalized and why).
  157. . See Stuntz, supra note 47, at 520 (“Substituting an easy-to-prove crime for one that is harder to establish obviously makes criminal litigation cheaper for the government.”).
  158. . See generally Michelle Alexander, Go to Trial: Crash the Justice System, N.Y. Times (Mar. 10, 2012), [] (presenting a proposal by Susan Burton); Andrew Manuel Crespo, No Justice, No Pleas, 90 Fordham L. Rev. 1999 (2022) (elaborating on Susan Burton’s proposal).
  159. . Jeffrey Bellin, Modern Justice and the Bill of Rights, Daily Press (Aug. 2, 2013, 12:00 AM), [].
  160. . Id. (“[I]t is difficult to believe that James Madison and his colleagues would have gone through all the trouble of placing the trial rights at the heart of the Bill of Rights if they knew that the invocation of those rights would be the (rare) exception rather than the rule.”).
  161. . See David S. Abrams, Putting the Trial Penalty on Trial, 51 Duq. L. Rev. 777, 780 (2013) (“The mistake that is frequently made is considering the conditional expected sentence in the context of a defendant with a plea offer, a setting where the unconditional expected sentence is the appropriate one.”); Uzi Segal & Alex Stein, Ambiguity Aversion and the Criminal Process, 81 Notre Dame L. Rev. 1495, 1495 (2006) (“In a paradigmatic jury trial, the probability of the defendant’s conviction is profoundly ambiguous.”).
  162. . See Bellin, supra note 117, at 851 (noting that mandatory minimum sentences, while problematic, are not as widespread as some commentary suggests, particularly in state systems responsible for most criminal cases).
  163. . See King & Wright, supra note 83, at 375 (emphasizing the importance to the parties of uncertainty about a post-plea sentence and efforts to minimize that uncertainty). There is also the possibility that by rejecting one plea offer, the defendant will receive a subsequent more (or less) favorable offer.
  164. . Gold et al., supra note 58, at 1645 (“Adopting more liberal discovery rules could also facilitate and improve plea bargaining.”); Erica Hashimoto, Toward Ethical Plea Bargaining, 30 Cardozo L. Rev. 949, 950 (2008) (“[P]roviding defendants with access to information about the case—in particular exculpatory or impeachment information—before the guilty plea is critical to ensuring an accurate and equitable plea process.”); Bibas, supra note 39, at 2469 (recognizing that “better discovery can reduce the influence of uncertainty on bargaining”).
  165. . Jenia I. Turner & Allison D. Redlich, Two Models of Pre-Plea Discovery in Criminal Cases: An Empirical Comparison, 73 Wash. & Lee L. Rev. 285, 289 (2016) (“While discovery rules continue to vary significantly from state to state, a recent trend has been in the direction of earlier and broader discovery.”).
  166. . See Daniel S. McConkie, Structuring Pre-Plea Criminal Discovery, 107 J. Crim. L. & Criminology 1, 15 (2017) (discussing a “common objection to granting broad defense discovery of the prosecution’s case: if factually guilty defendants know what they did, don’t they already know the government’s case against them?”); Hashimoto, supra note 164, at 951–52 (responding to this objection by noting that innocent defendants and some guilty defendants will be unaware of the core facts and that defendants need the information to “obtain pleas that accurately reflect the strength of the government’s case against them”); Kevin C. McMunigal, Disclosure and Accuracy in the Guilty Plea Process, 40 Hastings L.J. 957, 978–79 (1989) (responding similarly to this argument).
  167. . Cf. Jenia I. Turner, Transparency in Plea Bargaining, 96 Notre Dame L. Rev. 973, 976–77 (2021) (arguing for the “recording of plea offers, charging decisions, sentencing outcomes, and other key facts about a criminal case in digital databases that are searchable and available to prosecutors, defense attorneys, and judges” that “would help promote fairness and equal treatment of defendants by educating lawyers and judges about plea precedents and facilitating a more informed analysis of plea offers”); Michael P. Donnelly, Truth or Consequences: Making the Case for Transparency and Reform in the Plea Negotiation Process, 17 Ohio St. J. Crim. L. 423, 436 (2020) (asking, “What resources would it take to create a centralized database to track every criminal sentence issued in our state with information available to advocates and judges to allow for meaningful comparison and analyses promoting both consistency and proportionality throughout the state of Ohio?”); Schneider & Alkon, supra note 18, at 456–57 (describing data that are and could be collected regarding plea bargaining and outcomes).
  168. . Cf. King et al., supra note 81, at 992 (concluding in a study of various jurisdictions that discounts for pleading guilty are “far from uniform” and describing this variation as “perhaps our most notable finding”). If prosecutors in the jurisdiction routinely offer multiple plea offers, data on those plea offers and their timing should be made available as well.
  169. . See Bellin, supra note 147, at 1231 (discussing normative approaches to plea bargaining).
  170. . Lafler v. Cooper, 566 U.S. 156, 168 (2012) (“It is, of course, true that defendants have ‘no right to be offered a plea . . . nor a federal right that the judge accept it.’” (quoting Missouri v. Frye, 566 U.S. 134, 148 (2012))).
  171. . See supra note 131 and accompanying text.
  172. . See David N. Dorfman & Chris K. Iijima, Fictions, Fault, and Forgiveness: Jury Nullification in a New Context, 28 U. Mich. J.L. Reform 861, 887 (1995) (reporting on acquittal rates as high as 56% in gun possession cases).
  173. . See Shawn D. Bushway, Allison D. Redlich & Robert J. Norris, An Explicit Test of Plea Bargaining in the “Shadow of the Trial”, 52 Criminology 723, 741 (2014) (reporting striking similarity between defense counsel and prosecutor estimates of likely trial results and proper plea deals).
  174. . Rubinstein et al., supra note 21, at 8.
  175. . See John Gramlich, Only 2% of Federal Criminal Defendants Go to Trial, and Most Who Do Are Found Guilty, Pew Rsch. Ctr. (June 11, 2019),‌2019/06/11/only-2-of-federal-criminal-defendants-go-to-trial-and-most-who-do-are-found-guilty/ [] (reporting 83% federal conviction rate at trial); Jeffrey Bellin, The Silence Penalty, 103 Iowa L. Rev. 395, 411 (2018) (reporting on statistical analysis that showed that “in California the rate of conviction at trial in felony cases is reportedly higher than 80%; in Florida it is around 73%”).
  176. . Cf. Shawn D. Bushway & Allison D. Redlich, Is Plea Bargaining in the “Shadow of the Trial” a Mirage?, 28 J. Quantitative Criminology 437, 446–47 (2012) (concluding from empirical analysis that a shadow-of-trial theory of plea bargaining appears to hold in the “aggregate”); Mona Lynch, Realigning Research: A Proposed (Partial) Agenda for Sociolegal Scholars, 25 Fed. Sent’g. Rep. 254, 255 (2013) (“[Studies] revealed how routine criminal matters were resolved in reference to ‘going rates’ for different offenses . . . .”).
  177. . See supra note 55 and accompanying text.
  178. . Any calculation of the trial penalty ex ante should incorporate the potential for acquittal. Since this illustration considers variations on a single case, however, that potential is constant throughout.
  179. . If the defendant is not likely guilty, the case should be dismissed. Bellin, supra note 147, at 1223 (“[A] prosecutor should only charge a case when the prosecutor expects that the evidence introduced at trial will prove the defendant’s guilt beyond a reasonable doubt.”).
  180. . For discussion of Kevin’s case, see supra notes 131–132, 171–172 and accompanying text. (Kevin is a pseudonym.)
  181. . See Erik Luna & Marianne Wade, Prosecutors as Judges, 67 Wash. & Lee L. Rev. 1413, 1421 (2010) (describing the case as revealing “the rough play of American adversarialism and the extent of prosecutorial power in plea bargaining”); John F. Stinneford, Dividing Crime, Multiplying Punishments, 48 U.C. Davis L. Rev. 1955, 1963 n.38 (2015) (“The most important effect of these sentences may be the ‘terrific leverage in plea bargaining’ they give prosecutors, as exemplified by the Angelos case.”).
  182. . 18 U.S.C. § 924(c).
  183. . United States v. Angelos, 345 F. Supp. 2d 1227, 1231 (D. Utah 2004) (“[I]f he pled guilty to the drug distribution count and the § 924(c) count, the government would agree to drop all other charges, not supersede the indictment with additional counts, and recommend a prison sentence of 15 years.”).
  184. . Id. at 1232.
  185. . Brief of Appellant at 4, United States v. Angelos, 433 F.3d 738 (10th Cir. 2006) (No. 04-4282), 2005 WL 2367629 at *4 (“It is without question that Mr. Angelos was involved with marijuana, and it is undisputed that he sold eight ounces of marijuana to Lazalde on three occasions.”).
  186. . Id. at 4–5 (asserting that despite police “observ[ing]” and “describ[ing]” Angelos’s drug transactions in their reports, “[n]o police officer observed the presence of a firearm at any of these transactions, and police reported that the confidential informant was carefully observed and debriefed after each such transaction. The contemporaneous police reports make no mention of the presence of a firearm.” (citation omitted)).
  187. . United States v. Angelos, 433 F.3d 738, 753 (10th Cir. 2006) (stating that the government’s evidence “clearly established that Angelos was a known gang member who had long used and sold illicit drugs[, and] . . . at the time of his arrest, Angelos was a mid-to-high drug dealer who purchased and in turn sold large quantities of marijuana”).
  188. . Brief for United States at 11, United States v. Angelos, 417 F. App’x 786 (10th Cir. 2011) (No. 09-4224), 2011 WL 286997, at *11 (“Angelos . . . met with law enforcement to discuss the possibility of Angelos’s cooperation, but the meeting was not successful. The agents did not believe Angelos.” (citation omitted)).
  189. . Brief for Respondent at 2 n.2, Bordenkircher v. Hayes, 434 U.S. 357 (1978) (No. 76-1334) (“In Kentucky the sentencing judge is not bound by the prosecutor’s sentence recommendation reached as a result of a plea negotiation. Even though the judge rejects the prosecutor’s recommendation on sentence, the defendant is not entitled to withdraw his guilty plea.”).
  190. . See id. at 2 & nn.1–2 (noting that “Hayes may well have been skeptical of the advantages of securing the prosecutor’s recommendation on sentence” because the sentencing judge would not have to follow that recommendation).
  191. . See Dan Canon, Pleading Out: How Plea Bargaining Creates a Permanent Criminal Class 3 (2022) (reporting that “Hayes . . . made parole on his first try, nine years after his conviction”); Stephen E. Henderson, The Jury Veto, 40 Yale L. & Pol’y Rev. 488, 490 n.5 (2022) (“Hayes was paroled on first eligibility, which in those days was triggered by roughly a decade of imprisonment.”). This appears to be earlier than the parties anticipated. See Bordenkircher v. Hayes, 434 U.S. 357, 370 n.1 (1978) (Powell, J., dissenting) (“It is suggested that respondent will be eligible for parole consideration after serving 15 years.”).
  192. . See Bellin, supra note 126, at 185 (noting distinction in generating evidence and proving “civilian-initiated cases” versus those generated by police).
  193. . Bordenkircher, 434 U.S. at 369 (Powell, J., dissenting) (“I observe, at this point, that five years in prison for the offense charged hardly could be characterized as a generous offer.”).
  194. . Severity is the key factor in the perceived injustice of these cases. Cf. Laurène Soubise & Alice Woolley, Prosecutors and Justice: Insights from Comparative Analysis, 42 Fordham Int’l L.J. 587, 609 (2018) (observing that “[p]lea bargaining is a robust part of Canadian criminal justice, but an accused who goes to trial cannot risk the sorts of consequences suffered by American accused like Weldon Angelos” because of the absence of severe sentencing laws).
  195. . United States v. Tigano, 880 F.3d 602 (2d Cir. 2018).
  196. . See id. at 605 (describing the factual background of Tigano’s arrest); Brief for Defendant-Appellant at 38, United States v. Tigano, 880 F.3d 602 (2d Cir. 2018) (No. 15-3073), 2017 WL 880549, at *38 (recounting plea offer and potential twenty-year mandatory minimum sentence); Brief of Appellee at 2, United States v. Tigano, 880 F.3d 602 (2d Cir. 2018) (No. 15-3073), 2017 WL 2418187, at *2 (explaining that “the district court sentenced Tigano principally to terms of imprisonment of 240 months . . . all to be served concurrently”).
  197. . For discussion of Kevin’s case, see supra notes 131–132, 171–172 and accompanying text.
  198. . See, e.g., Smith v. United States, 549 A.2d 1119, 1121 (D.C. 1988) (explaining in case where the defendant sold one packet of drugs to undercover officer and then threw away remaining drugs during later arrest, “the government properly focused on the single packet sold as the criminal act of distribution and the thirteen packets thrown away as the criminal act of possession with intent to distribute”).
  199. . See Plummer v. United States, 870 A.2d 539, 540–41 (D.C. 2005) (rejecting challenge to thirty-year sentence with all but five years suspended for crack cocaine distribution).
  200. . An excerpt of a D.C. police officer’s testimony describing a “buy-bust operation” can be found in the dissenting opinion in Morton v. United States, 734 A.2d 178, 183 (D.C. 1999) (Kern, J., dissenting).
  201. . The author was a prosecutor in the District of Columbia. Defense attorneys would often also seek an allocution cap, an agreement that the prosecutor not ask for more than a specified amount of prison time (or take no position on sentencing).
  202. . In fact, the guidelines that judges consult in the District of Columbia include an example from a “buy-bust” operation where a defendant sells both heroin and cocaine to a police officer. The guidelines say, “The sentences imposed for distribution of heroin and distribution of cocaine should run concurrently because they are non-violent crimes that arose from the same event.” Voluntary Sent’g Guidelines Manual ch. 6.3, ex. (D.C. Sent’g Comm’n 2020), [].
  203. . See supra text accompanying note 174 for an example of an interview from Alaska.
  204. . To assess the scenarios ex ante, the trial penalty becomes the product of two variables: post-trial sentence and likelihood of acquittal. That’s why the trial penalty for Kevin is fairly low on the chart (since there was a strong chance of acquittal) and why the trial penalty for Tigano (whose conviction was not realistically in doubt) is higher than that for Angelos and Hayes even though Tigano faced a shorter sentence.