John F. Pfaff’s Locked In: The True Causes of Mass Incarceration and How to Achieve Real Reform is probably the best book on so-called mass incarceration to date. Its great strength is that it is empirically grounded. Pfaff, a professor of law at Fordham, doesn’t cherry-pick data to support some a priori theory, he grapples with the hard realities that the data present. As he well understands, this makes his argument for reducing imprisonment a very tough sell.

After all, if violent crime and other serious offenses are the primary reasons for incarceration then why should we reduce imprisonment? And if we still wish to disincarcerate, despite the compelling reasons for all the lockups, then how might this be achieved? Unfortunately, Pfaff’s answers to these questions are the least persuasive portions of his book.

The author’s big point is that the usual explanations for the rise in imprisonment—the “standard story” as he calls it—are not only wrong, they are destructive of efforts to disincarcerate. The standard story has three components, each said to contribute mightily to mass incarceration: 1) the War on Drugs; 2) long sentences; and 3) private prisons. Each of the three, this book demonstrates, was a secondary contributer at best.

Pfaff quickly disposes of the threat from the “prison industrial complex.” He points out that privately managed prisons house only around 7 percent of all U.S. inmates and that their management policies are no worse than those of the public sector. Focusing on such a minor player simply diverts attention from the real problem: publicly run prisons.

The drug war is a more serious contender. After all, and especially in the late 1980s, when crack cocaine terrorized the nation, drug convictions put thousands behind bars. In Michelle Alexander’s much-ballyhooed 2012 book New Jim Crow we read: “The impact of the drug war has been astounding. In less than thirty years, the U.S penal population exploded from around 300,000 to more than 2 million, with drug convictions accounting for the majority of the increase.”[1]

Well, not quite. As Pfaff demonstrates, at the peak of the great crack-down in 1990, inmates serving time in state prisons for drug crimes accounted for 22 percent of the state-prison population, whereas violent inmates comprised 47 percent—more than twice as many. When prison populations increased dramatically, from 1980 to 1990, drugs accounted for a third of the increase. But even in that lock-’em-up decade, crimes of violence put a greater number, 36 percent, behind bars.[2]

And in the decades before and after the more punitive 1980s, the case for emphasizing drug imprisonments is even weaker. A survey of state prisoners in 1974 found that only 10 percent were in for drug offenses, 52 percent for violent crimes.[3] From 1990 to 2009, Pfaff found, drugs contributed only 14 percent to state confinement; violence a whopping 60 percent.[4]

Pfaff thinks that the false claims about drug imprisonment actually hurt the case for disincarceration. Why? Because reducing drug imprisonment leaves the main culprit, violent crimes, unscathed. If tomorrow we freed all drug-law violators (and most of those in prison are traffickers, not simple users, so this is not going to happen), then the number of inmates in our state prisons would drop only 16 percent. Eighty-four percent of them would remain. That’s over 1.1 million prisoners.[5] A pyrrhic victory at best. And as Pfaff notes, the structural basis for another period of increased imprisonment in response to the next crime wave remains.

When imprisonment rates rise, it must be due to one of two reasons or some combination of the two: either more people are being incarcerated, or they are being confined for longer periods. Pfaff denies that the latter factor is to blame for mass incarceration. Indeed, he says that time served in the United States is “surprisingly short, and there’s no real evidence that it grew much as prison populations soared.” He’s right. Armed robbers, to take a particularly scary crime of violence, were released in a median 2.3 years (in 2010). Moreover, actual prison time (very few prisoners serve their maximum sentences) has decreased in the last few years.[6]

Pfaff proves that time served has not increased in recent decades by graphing annual admissions to prison along with annual releases over a 36 year period, 1978 to 2014. We can readily see that sentences must not have lengthened much. If they had, prisoners would be confined longer, releases would be slowed, and the plot lines would diverge as the admissions line would rise faster than the releases line.[7]

So then, Pfaff concludes, since imprisonment rates rose substantially, but sentences didn’t get much longer, more people must have been sentenced to prison. This brings him to the villain of the incarceration buildup as he sees it: prosecutors. Who decides whether or not a defendant goes to prison? The general public probably would point to the judges and, perhaps, the sentencing laws that guide them. But Pfaff asserts that the role of prosecutors is more significant. They, after all, select the charges and negotiate the plea that determines both guilt and the boundaries of the sentence.

Pfaff’s quantitative proof for his hypothesis that prosecutors were responsible for increasing the number of people sent to prison is, as he recognizes, feeble. He does show that prison admissions per prosecutor rose from nine in 1974 to 25 in 1990. But this isn’t a decisive argument, as other factors could have been in play. For instance, sentencing statutes with mandatory prison provisions could have become more significant by 1990, especially given the growing number of defendants with criminal records. And he admits that the enormous increase in assistant prosecutors alone (17,000 in 1970 to 30,000 in 2007) could explain why more people were sent to prison—even without any change in prosecutorial attitudes.

Moreover, the case is even more complicated than Pfaff recognizes. He didn’t examine the number of felony charges resulting in probation, a get-out-of-jail-free sentence unquestionably under prosecutorial control. In 1986, for instance, 31 percent of all adults convicted of felonies got probation.[8] So prosecutors kept nearly one in three felons out of prison at a time when rates of imprisonment were rising sharply—which hardly suggests greater punitiveness.

Whether prosecutors caused the massive rise in imprisonments is debatable. One could as easily focus on different points in the causal chain leading to incarceration. In a time of rising crime—and the late 1960s to the early 1990s was unquestionably such a time—one would have expected the criminal justice system to imprison more people. More crime, more arrests, more prosecutions, more incarcerations. In fact, Pfaff observes that rising crime in the 1970s and 1980s explains approximately half of the prison-population increase in those decades.[9] Arguably crime, and the fear it generated, led to the increase in the number of key actors in the criminal justice system as well as their more punitive attitudes. So one can just as readily argue that crime is the true explanation for mass incarceration.

But since Pfaff thinks that prosecutors are to blame, his solution is to rein them in by such structural changes as plea-bargaining guidelines, local elections to give inner cities more control over who prosecutes (though he also urges, contradictorily, that prosecutors be appointed), and sentencing commissions as a bulwark against public pressure to escalate punishments. As he seems to appreciate, however, none of these proposals is likely to work even if they were politically feasible. The reason is straightforward: The overwhelming majority of prisoners are guilty of serious and usually violent crime, and few prosecutors are going to decline to lock up such defendants no matter how the prosecutor is selected or constrained.

This brings us to a deeper problem with the author’s arguments. He says that prison growth is a “national shame.”[10] But if the vast majority of prison inmates are guilty of violent or serious nonviolent crimes, and if sentences are not unacceptably lengthy, then why should we disincarcerate? Currently, at least 88 percent of state prisoners are in for extremely serious offenses, including violent crimes such as murder and rape (53 percent); property crimes like burglary and grand larceny (19 percent); weapons possession (4 percent); and yes, drug trafficking (12 percent).[11]

Is it wrong to imprison such people? Pfaff’s answer is that the costs of incarceration—more single-parent families, assaults on inmates, prisoner health problems, and an array of post-release issues, including unemployment, drugs, and the need for more social services—outweigh the benefits. But some of the problems he lists are not attributable to prisons. The type of person who goes to prison for serious crime is not likely to be highly employable, drug-free, or a responsible parent.

Even if we grant that prisons exacerbate inmate problems, they provide advantages that should be part of any cost-benefit analysis. Pfaff gives short shrift to the benefits side of the equation. As he concedes, the best analyses indicate that, starting in the early 1990s, increased imprisonment reduced crime. Yet he downplays the costs of a post-decarceration crime wave—a risk he cannot deny. These costs—the injuries and deaths due to assault; the fear engendered by unrelenting violence; the risk of ruining young lives through the temptations of gangs and drugs, to say nothing of dysfunctional schools; the depressed property values; and the discouragement to businesses, in turn creating more unemployment and greater demand on limited community services—are most likely to be borne by low-income and minority communities.

Ironically, by presenting the hard realities of imprisonment, Locked In makes the case against it more difficult to mount, not less. This is its great virtue, though. Discussion of this contentious issue will, I hope, be taken to a higher level by Pfaff’s work.

[1] Michelle Alexander, The New Jim Crow: Mass Incarceration in the Age of Colorblindness (New York: The New Press, 2012), p. 6.

[2] Pfaff, Locked In, pp. 32-3.

[3] U.S. Department of Justice, Law Enforcement Assistance Administration, Profile of State Prison Inmates: Sociodemographic Findings from the 1974 Survey of Inmates of State Correctional Facilities (1979), p. 2.

[4] Locked In, p. 33.

[5] The latest figures available are for prisoner counts as of December 31, 2014. U.S. Department of Justice, Bureau of Justice Statistics, Prisoners in 2015 (2016), app table 5.

[6] Locked In, pp. 52 and 56.

[7] Ibid., pp. 57-8.

[8] U.S. Department of Justice, Bureau of Justice Statistics, Felony Sentences in State Courts, 1986, p. 4.

[9] Locked In, p. 4.

[10] Ibid., p. viii.

[11] U.S. Department of Justice, Bureau of Justice Statistics, Prisoners in 2015 (2016), table 9.