Even though America has by far the highest incarceration rate in the world, Barry Latzer says it is “clear that we aren’t incarcerating too many people.” Why? In a nutshell, because we lock up fewer people than actually commit crimes, and “we don’t overpunish” by locking them up for too long. Ergo, he suggests, mass incarceration is a myth and not something we should be worried about. I disagree.

The truth is, no one knows exactly how many people should be locked up at any given time. But it is certainly less than the total number of “offenders,” and it is therefore entirely possible to have an over-incarceration problem even if, as Latzer correctly notes, “many offenders go unpunished.”

Thus, I think Latzer’s view is mistaken for three main reasons. First, not all criminal offenses merit incarceration, which means you will have an over-incarceration problem if you lock up every “offender.” Second, our Constitution constrains the government’s ability to imprison people by recognizing certain rights and establishing procedures that the prosecution must comply with before a defendant can be convicted and punished. If those constraints are ignored—as they routinely are in our system—then some prisoners will not have been lawfully adjudged guilty before being sentenced, and you will again have an over-incarceration problem. Finally, if you lock people up for longer than they truly deserve, as we routinely do, then you will have an over-incarceration problem.

Starting with the first point, it is axiomatic that in a vastly overcriminalized society like ours, perfect enforcement would be an absolute disaster. Indeed, there is an entire sub-genre of legal scholarship devoted to cataloging the dizzying array of trivial offenses for which people could, in theory, be incarcerated, and many scholars have argued quite plausibly that ordinary citizens regularly commit crimes without even knowing it.[1] Simply put, if we locked up every single person who committed a criminal offense, the social and economic disruption to the country would be cataclysmic, and it is doubtful whether there would even be enough non-offenders left to operate the machinery of criminal justice, let alone any other industry.

So let’s stipulate that we do not want to see all offenders locked up, but rather some subset of offenders. Well, which subset, exactly? We could say “serious offenders,” many of whom, as Latzer notes, are never identified, let alone punished. But who exactly counts as a “serious offender”? It’s not an easy question. For example, despite the move towards legalization in the states, the federal government still outlaws marijuana and imposes severe penalties just for growing it, ranging from five years to life in prison. It seems unlikely that most people would consider growing a few marijuana plants to be a “serious offense,” but the federal government sure thinks so. Likewise, consider the millions of teens who engage in sexting and consider it a bit of relatively harmless fun. Do they realize that when they use an electronic device to create images featuring the “lascivious exhibition of the genitals or pubic area” of a person under 18, they are committing the federal crime of producing child pornography for which the mandatory minimum sentence is a whopping 15 years? Finally, consider the massive and well-documented noncompliance with gun laws involving so-called assault weapons, high-capacity magazines, and concealed carry. Should every one of those violations be considered a “serious offense” meriting incarceration, even when committed, as they often are, by otherwise law-abiding men and women who are pillars of their communities and exercising what they take to be their constitutional right to keep and bear arms?

We could try focusing instead on punishing violent crimes, but even that doesn’t resolve the conundrum of which offenses merit incarceration. Among other things, “violent crime” is a famously elastic concept, and not all crimes denominated as “violent” truly are. According to a piece published by the Marshall Project last spring titled When Violent Offenders Commit Nonviolent Crime, trafficking in stolen identity and selling drugs within 1,000 feet of a school are both considered violent crimes in North Carolina; in Kentucky a second conviction for possession of anhydrous ammonia in an unapproved container with the intent to manufacture methamphetamine is categorized as a violent crime; and in New York it is a violent felony to possess a loaded gun illegally—something plenty of otherwise law-abiding New Yorkers do for self-defense given that it is nearly impossible for non-celebrities to obtain concealed-carry permits in New York City.

Thus, while Latzer is surely correct when he notes that “thousands” of demonstrably serious crimes, including rapes, robberies, and even murders, go unsolved every year, it is by no means clear that if we managed to convict and incarcerate all of those offenders—while releasing from prison everyone who doesn’t truly deserve to be there—“[o]ur incarceration rates would be even higher.” So how do we decide who deserves to be locked up and who doesn’t?

America Has Learned to Mass-Produce Convictions

Fortunately, the Founders provided us with a time-tested mechanism for making that very determination: the jury trial. Indeed, so central to the Founders’ vision of just government was the jury trial that it is the only right mentioned both in the body of the Constitution and the Bill of Rights, which devotes more words to that subject than any other. The centrality of the citizen jury to the Founders’ vision is unmistakable. And yet jury trials are now nearly extinct on American soil, which means that, contrary to the Founders’ meticulous design, ordinary citizens play virtually no role in the administration of criminal justice. How can that be?

Quite simply, prosecutors and judges have effectively hacked the criminal justice system by replacing constitutionally prescribed jury trials with extra-constitutional—and often extraordinarily coercive—plea bargaining. Some 95 percent of convictions today are obtained through guilty pleas, and in the federal system it’s more than 97 percent.

I have explained elsewhere the fundamentally coercive nature of American-style plea-bargaining, and I will not repeat those arguments here—the numbers speak for themselves.[2] Instead, I will focus on two particularly relevant consequences of the fact that, as the Supreme Court itself has recognized, “criminal justice today is for the most part a system of pleas, not a system of trials.”[3]

First, with the near-complete displacement of trials by plea bargaining, it is no longer ordinary citizens who decide what conduct merits punishment, but prosecutors. Prosecutors decide which cases to pursue and which to decline; they decide—often with great creativity—what charges to bring and what charges not to bring; and they largely dictate the punishment a defendant will receive after pleading guilty. Those are emphatically not decisions the Founders meant to place in the hands of a single government actor, let alone one who cannot remotely claim to be neutral. Accordingly, rather than reserving the criminal sanction for those occasions where the community as a whole, speaking through the supermajoritarian institution of a unanimous jury, agreed it was appropriate, we now have a system in which a finding of guilt is largely determined by how much pressure prosecutors can bring to bear on a defendant. And given that the courts make no effort whatsoever to ensure that plea bargaining does not become coercive, the typical answer these days is “Whatever it takes.”[4]

The Expensive but Constitutional Design of Jury Trials

Besides displacing the constitutionally designated adjudicator, plea bargaining also enables the government to convict—and incarcerate—far more people than it could using the constitutionally prescribed mechanism of a criminal jury trial. Jury trials are expensive, inconvenient, and chancy. Guilty pleas, by contrast, are cheap, efficient, and certain. As a matter of simple economics, when the cost of a particular good—whether it be automobiles or convictions—comes down, consumption will rise. And as Professor Darryl Brown argues persuasively in his essay, “The Perverse Effects of Efficiency in Criminal Process,” there is every reason to believe that plea bargaining has had precisely that effect on our system. Brown observes that “[c]heaper adjudication can lead to even more cases entering the system”—and, by logical extension, more convictions than we would be willing to underwrite if we had to pay the full constitutional freight. It seems entirely reasonable to characterize that dynamic as a form of over-incarceration.

Finally, Latzer argues that “we don’t overpunish.” That is of course a value judgment, and to say, as he does, that “actual time behind bars is in line with historical practice in the United States” tells us nothing about whether those practices are fair. Latzer argues that our substantially more punitive approach to incarceration compared with Canada, Australia, and European countries reflects a kind of “American Exceptionalism” driven by “[t]hree key influences on American incarceration rates[:] recidivism, guns, and murder.” But he omits another example of American exceptionalism, the so-called “trial penalty.” The trial penalty refers to the differential between the time a defendant will serve if he agrees to plead guilty and the time he will serve if he exercises his right to trial and loses. As documented by the National Association of Criminal Defense Lawyers in their recent study on the trial penalty, those differentials can be staggering, with those convicted after a trial serving sentences two to three times longer than those who pleaded guilty.

Relatedly, in Prisoners of Politics, Professor Rachel Barkow has shown that a significant role of mandatory minimum prison sentences is precisely to strengthen prosecutors’ hands in plea negotiations. Between the trial penalty and evidence that at least some sentencing policy is informed by goals other than ensuring a just punishment, we can say with confidence that some fairly significant number of prisoners are serving longer sentences than they truly deserve, which is, again, a form of over-incarceration.

If the question posed is whether, in some absolute sense, America imprisons too many people, the answer is that we simply cannot say without a coherent theory of who truly deserves to be locked up. Mr. Latzer doesn’t offer one, and no surprise—it’s an extraordinarily complicated issue. But if the question is whether our system has become utterly cavalier in its use of the criminal sanction; whether it has abandoned constitutionally mandated procedures for deciding who gets convicted and who does not; and whether it routinely hands out unjust punishments in order to facilitate the machinery of coerced adjudication—then the answer is yes, absolutely.

[1] See, e.g., Mike Chase, How to Become a Federal Criminal: An Illustrated Guide for the Aspiring Offender (2019), or see Chase on Twitter; Harvey Silverglate, Three Felonies a Day (2011); Go Directly to Jail: The Criminalization of Almost Everything (Gene Healy ed., 2004); Misha Tseytlin and Alex Kozinksi, You’re (Probably) a Federal Criminal (Timothy Lynch ed., 2009); Glenn Reynolds, “Ham Sandwich Nation: Due Process When Everything Is a Crime,” 113 Columbia L. Rev. 102 (2013).

[2] See also Clark Neily, “Jury Empowerment as an Antidote to Coercive Plea Bargaining,” 31 Federal Sentencing Reporter 284, (April/June 2019).

[3] Lafler v. Cooper, 566 U.S. 156 (2012).

[4] See, e.g., Bordenkircher v. Hayes, 434 U.S. 357 (1978) (rejecting the argument that a prosecutor’s threat to increase defendant’s exposure from ten years to life in prison if defendant refused to plead guilty violated due process).