In July 2015, President Obama paid a press-saturated visit to a federal penitentiary in Oklahoma. The cell blocks that Obama toured had been evacuated in anticipation of his arrival, but after talking to six carefully prescreened inmates, he drew some conclusions about the path to prison. “These are young people who made mistakes that aren’t that different than the mistakes I made and the mistakes that a lot of you guys made,” the president told the waiting reporters.

The New York Times seconded this observation in its front-page coverage of Obama’s prison excursion. There is but a “fine line between president and prisoner,” the paper noted. Anyone who “smoked marijuana and tried cocaine,” as the president had as a young man, could end up in the El Reno Federal Correctional Institution, according to the Times.

This conceit was preposterous. It takes a lot more than marijuana or cocaine use to end up in federal prison. But the truth didn’t matter. Obama’s prison tour came in the midst of the biggest delegitimation of law enforcement in recent memory. Activists, politicians, and the media have spent the last year broadcasting a daily message that the criminal-justice system is biased against blacks and insanely draconian. The immediate trigger for that movement, known as Black Lives Matter, has been a series of highly publicized deaths of black males at the hands of the police. But the movement also builds on a long-standing discourse from the academic Left about “mass incarceration,” policing, and race.

Now that discourse is going mainstream. As the press never tires of pointing out, some high-profile figures on the right are joining the chorus on the left for deincarceration and decriminalization. Newt Gingrich is pairing with left-wing activist Van Jones, and the Koch brothers have teamed up with the ACLU, for example, to call for lowered prison counts and less law enforcement. Republican leaders on Capitol Hill support reducing or eliminating mandatory sentences for federal drug-trafficking crimes, in the name of racial equity.

Ed Kashi/VII/Corbis
Ed Kashi/VII/Corbis

At the state and city levels, hardly a single criminal-justice practice exists that is not under fire for oppressing blacks. Traffic monitoring, antitheft statutes, drug patrols, public-order policing, trespass arrests, pedestrian stops, bail, warrant enforcement, fines for absconding from court, parole revocations, probation oversight, sentences for repeat felony offenders—all have been criticized as part of a de facto system for locking away black men and destroying black communities.

There may be good reasons for radically reducing the prison census and the enforcement of criminal laws. But so far, the arguments advanced in favor of that agenda have been as deceptive as the claim that prisons are filled with casual drug users. It is worth examining the gap between the reality of law enforcement and the current campaign against it, since policy based on fiction is unlikely to yield positive results.

Two days before his Oklahoma penitentiary visit, Obama addressed the NAACP national conference in Philadelphia and raised the same themes. The “real reason our prison population is so high,” he said to applause, is that we have “locked up more and more nonviolent drug offenders than ever before, for longer than ever before.” This assertion is the most ubiquitous fallacy of the deincarceration movement, given widespread currency by Michelle Alexander’s 2010 book, The New Jim Crow. That a president would repeat the myth is a demonstration of the extent to which ideology now rules the White House.

Pace Obama, the state prison population (which accounts for 87 percent of the nation’s prisoners) is dominated by violent criminals and serial thieves. In 2013, drug offenders made up less than 16 percent of the state prison population, whereas violent felons were 54 percent of the rolls and property offenders, 19 percent. (See graph below.) Reducing drug admissions to 15 large state penitentiaries by half would lower those states’ prison count by only 7 percent, according to the Urban Institute.

Chart by Alberto Mena
Chart by Alberto Mena

True, drug traffickers make up a larger (though declining) portion of the federal prison population: half in 2014. But federal prisons hold only 13 percent of the nation’s prison population. Moreover, it is hardly the case that “but for the grace of God,” as Obama put it, he could have been incarcerated in Oklahoma’s El Reno for getting stoned as a student. Less than 1 percent of sentenced drug offenders in federal court in 2014 were convicted for simple drug possession, according to the U.S. Sentencing Commission, and most of those convictions were plea-bargained down from trafficking charges. Contrary to the deincarceration movement, blacks do not dominate federal drug prosecutions. Hispanics were 48 percent of drug offenders sentenced in federal court in 2013, blacks were 27 percent, and whites 22 percent.

Even on the state level, drug-possession convicts are relatively rare. In 2013, only 3.6 percent of state prisoners were serving time for drug possession, often the result of a plea bargain, compared with 12 percent of prisoners convicted for trafficking. Virtually all the possession offenders had long prior arrest and conviction records. The meth users that Tustin, California, police officer Mark Turner encountered in his undercover narcotics days were sentenced to drug classes. “Then they would skip out of the classes and always re-offend,” he says.

Nor is it true that rising drug prosecutions drove the increase in the prison population from the late 1970s to today. Even during the most rapid period of prison growth—from 1980 to 1990—violent prisoners accounted for 36 percent of the rise in the state prison population, compared with 33 percent from drug offenders. (See “Is the Criminal-Justice System Racist?,” Spring 2008.) From 1990 to 2000, violent offenders accounted for 53 percent of the census increase and all of the increase from 1999 to 2004.

Obama and other incarceration critics have targeted mandatory minimum sentences for federal drug crimes. The current penalty structure is hardly sacrosanct, but mandatory sentences are an important prosecutorial tool for inducing cooperation from defendants. The federal minimums are also not lightly levied. A ten-year sentence for heroin trafficking, for example, requires possession of a kilogram of heroin, enough for 10,000 individual doses, with a typical street value of at least $70,000. Traffickers without a serious criminal history can avoid application of a mandatory sentence by cooperating with investigators. It is their choice not to do so.

The critics of “mass incarceration” love to compare American incarceration rates unfavorably with European ones. Crime is inevitably left out of the analysis. Jeremy Travis and Nicholas Turner, head of the John Jay College of Criminal Justice and the Vera Institute, respectively, penned a classic treatment of this theme in the New York Times in August 2015. Germany’s incarceration rate is one-tenth that of the U.S., they fumed. “To be sure,” they acknowledged, “there are significant differences between the two countries.” And might those “significant differences” have anything to do with crime, perhaps with the fact that the U.S. rate of gun homicide is about 17 times higher than that of Germany? Of course not. No, for Travis and Turner, the key difference is that “America’s criminal justice system was constructed in slavery’s long shadow and is sustained today by the persistent forces of racism.” The same people who denounce American gun violence and call for gun control in a domestic context go silent about gun violence when using Europe as a club to cudgel the American prison system. The U.S. homicide rate is seven times higher than the combined rate of 21 Western developed nations plus Japan, according to a 2011 study by researchers of the Harvard School of Public Health and the UCLA School of Public Health. This disparity is largely fueled by the American firearm homicide rate: 19.5 times higher than in the comparison high-income countries, according to 2003 data. Among 15- to 24-year-olds, Americans kill with guns at nearly 43 times the rate of their counterparts in those same industrialized nations. Since the American prison system is driven by violent crime, it is not surprising that America’s incarceration rate is higher than Europe’s.

Contrary to the advocates’ claim that the U.S. criminal-justice system is mindlessly draconian, most crime goes unpunished, certainly by a prison term. For every 31 people convicted of a violent felony, another 69 people arrested for violence are released back to the streets, according to a 2007 Bureau of Justice Statistics analysis of state courts. That low arrest-to-conviction rate reflects, among other reasons, prosecutors’ decisions not to go forward with a case for lack of cooperative witnesses or technical errors in police paperwork. The JFA Institute estimated in 2007 that in only 3 percent of violent victimizations and property crimes does the offender end up in prison.

Far from being prison-happy, the criminal-justice system tries to divert as many people as possible from long-term confinement. “Most cases are triaged with deferred judgments, deferred sentences, probation, workender jail sentences, [and] weekender jail sentences,” writes Iowa State University sociologist Matt DeLisi in a forthcoming issue of the Journal of Criminal Justice. Offenders given community alternatives “are afforded multiple opportunities to violate these sanctions only to receive additional conditions, additional months on their sentence, or often, no additional punishments at all,” DeLisi adds. In 2009, 27 percent of convicted felons in the 75 largest counties received a community sentence of probation or treatment, and 37 percent were sentenced to jail, where sentences top out at one year but are usually completed in a few weeks or months. Only 36 percent of convicted felons in 2009 got a prison term. Among convicted violent felons in 2009, 17 percent received community supervision and 27 percent were sentenced to jail, leaving 57 percent on their way to prison. (The numbers have been rounded by the Bureau of Justice Statistics.) A 17-year-old gang member in Tustin, who has just been arrested for stealing a bike and leading the police on a chase through residential backyards, tells of a friend who stole a car and took off on the freeway with the police in hot pursuit. His friend had a gun at the time. Though this car thief already had a serious felony on his record, he was given a ten-month jail sentence and was out in five months, hardly an overly harsh sentence for the public danger he caused. The bike thief himself has a long record of burglaries, assault, and absconding but has never gone to jail.

The vast majority of felony defendants whom a district attorney decides to prosecute rather than divert out of the system have an extensive criminal history, yet were still in the community committing crime. Half of the defendants charged with a felony in 2009 in the 75 largest counties had five or more prior arrests, and 36 percent had ten or more. About three in five had at least one prior conviction, and 30 percent had multiple felony convictions, with 11 percent of felony defendants having five or more previous felony convictions. Yet the majority of those offenders will still not get a prison term. Among those who wind up sentenced to prison, the prior records are even longer. The average number of prior convictions for inmates released from state prison in 2005 was five; the average number of prior arrests was more than ten.

The Los Angeles County Probation Department has supervised a “frequent flier” whose extensive arrest record includes multiple charges of assault with a deadly weapon, grand theft auto, taking a vehicle without the owner’s consent, threatening a crime with intent to terrorize, robbery, escape from custody, failure to appear, driving without a license, possession of a controlled substance, possession of drug paraphernalia, false imprisonment, exhibiting a deadly weapon, and murder. He has twice been sentenced to prison for those crimes, but he is out on the streets as often as not. In August 2015, he was in jail after getting arrested for another violent crime.

Steve, a 49-year-old convict in Santa Ana, is a typical career criminal who is unconfined and still offending. He has a long rap sheet for burglary and firearms charges. His last prison stint ended in 2013, with a three-year term of supervision; recently, a case implicating him in 12 burglaries in nearby Irvine was thrown out by the district attorney because of technical flaws in the police report. In August 2015, he sat with his brother, son, and a friend on the cement back porch of his classic California bungalow, surrounded by a Mercedes 300 SL, a pickup truck, and a jumble of household detritus, as probation officers searched the bungalow’s dark interior for contraband and other occupants. The probation team found a semiautomatic handgun in a backpack and a 30-year-old female probationer hiding in a bathroom. She was absconding from her probation officer and high on meth. Steve claimed that he had found the backpack on the front porch a few days earlier and put it in the front hall closet but that he had no idea that it was still there and that it contained a gun. “If it was my pack, I wouldn’t keep it in the hall closet,” he told the officers. The syringes for his diabetes medicine that were also in the backpack seemed to belie his claim that the pack was not his.

The biggest myth about the criminal-justice system is not that it mindlessly metes out overlong sentences but that the disproportionate number of blacks in prison reflects bias by police, prosecutors, and judges. “The bottom line is that in too many places, black boys and black men, Latino boys and Latino men experience being treated differently under the law,” President Obama told the NAACP conference in July, echoing a line he has made with increasing frequency over the last year. Incarceration “disproportionately impacts communities of color,” Obama said. “African Americans and Latinos make up 30 percent of our population; they make up 60 percent of our inmates.”

Naturally, Obama said nothing about crime rates. It is not marijuana-smoking that lands a skewed number of black men in prison but their elevated rates of violent and property crime. A 2011 study of California and New York arrest data led by Pennsylvania State University criminologist Darrell Steffensmeier found that blacks commit homicide at 11 times the rate of whites and robbery at 12 times the rate of whites. Such disparities are repeated in city-level data. In New York City, blacks commit over 75 percent of all shootings, according to the victims of and witnesses to those shootings, though they are only 23 percent of the city’s population. They commit 70 percent of all robberies. Whites, by contrast, commit under 2 percent of all shootings and 4 percent of all robberies, though they are 34 percent of the city’s population. In the 75 largest county jurisdictions in 2009, blacks were 62 percent of robbery defendants, 61 percent of weapons offenders, 57 percent of murder defendants, and 50 percent of forgery cases, even though nationwide, blacks are 12 percent of the population. They dominated the drug-trafficking cases more than possession cases. Blacks made up 53 percent of all state trafficking defendants in 2009, whites made up 22 percent, and Hispanics 23 percent, whereas in possession prosecutions, blacks were 39 percent of defendants, whites 34 percent, and Hispanics 26 percent.

Repeated efforts by criminologists to find a racial smoking gun in the criminal-justice system have come up short. If the prison population were not a reminder of a reality that the political and academic establishment would rather cover up—the black crime rate—it is unlikely that the deincarceration movement would have generated the same momentum. After all, the nearly fourfold rise in the prison population since the early 1980s played a major role in the record-breaking crime drop since the early 1990s. That prison buildup represented a backlash against the anti-confinement ideology of the 1960s and 1970s that had lowered the incarceration rate, as crime was exploding in cities across America. Many of the same alternatives to penal custody that are now being proposed had been put into place in the late 1960s and early 1970s to keep criminals out of prison. But these alternatives lost support as crime spun out of control. Legislators started lengthening sentences, especially for repeat felony offenders, and pressing for a greater confinement rate. During the 1980s, crime rates fluctuated as the prison population steadily grew; it was only in the early 1990s that crime began a steady downward trajectory, ultimately to be cut in half by the mid-2000s. Anti-incarceration advocates point to the divergent paths of crime and imprisonment in the 1980s to argue against the role of prison in the 1990s crime drop; University of California at Berkeley law professor Franklin Zimring, however, has argued that it was not until the 1990s that the prison buildup reached its most effective incapacitative strength and kicked in as a sustained antidote to lawlessness.

Statistical war continues to be waged over incarceration’s role in the last two decades’ crime decline, with all activists and many academics still denying that incarceration contributed to the crime drop. Given the nonstop pressure from the Black Lives Matter movement, we may be embarking on another real-world experiment testing the relationship between incapacitation and crime. If the country is really serious about lowering the prison count, however, it is going to have to put aside the fictions about the prison population. The legendary pot-smoker clogging up the rolls is long gone, if he were ever there. Cutting the prison population will require slashing the sentences of violent criminals and property offenders (many of whom have violent histories) and keeping more of them in the community after their convictions. The problem is not the “Michelle Alexander story that we have too many harmless people in prison,” says New York University public-policy professor Mark Kleiman. “Most of the problem is that we have too many murderers in prison.”

Compared with the rhetoric around “mass incarceration,” current sentences do not seem outrageously high. In 2009, the median sentence length for all felony convictions was 30 months. For violent felonies, the median was 48 months, and for nonviolent felonies it was 24 months. In 2011, according to the Bureau of Justice Statistics, 43 percent of new admissions to state prisons were sentenced to two to four years; 57 percent of all prisoners had sentences of four years or less. About 42 percent of incoming prisoners had sentences of five years or more. Whether you find those numbers shocking depends on your view of retribution and incapacitation. To be sure, some very long sentences are meted out. California, for example, has one of the strictest sentencing-enhancement laws in the nation for the use of guns during felonies. Rob someone with a knife, and you may get two years in prison. Threaten your victim with a gun, however, and you may, depending on your criminal history and plea bargaining, face an additional ten years. In Iowa, class B felonies like armed robbery have a 25-year prison sentence, of which at least 70 percent must be served.

Still, it will take a lot of sentence cutting and diversion to the community to make a difference in the prison population. Cutting the time served by violent felons in New Jersey state prisons by 15 percent, for example, would lower the prison population there by only 7 percent by 2021. Cutting violent felons’ time served by half would still only bring down the population by 25 percent, according to the New York Times’s Erik Eckholm, using an Urban Institute estimation tool. Such measures will hardly end the era of “mass incarceration.” To get back to our historical level of incarceration, we would need to reduce the prisoner headcount by 80 percent.

Some deincarceration advocates argue that increased social programs for criminals can significantly reduce the risks of letting offenders out early or not confining them in the first place. We have entered the era of “evidence-based practices,” or EBP, they say. Evidence-based practices are social-services and therapeutic programs delivered to the “at-risk” population that have allegedly been scientifically shown to reduce offending. The EBP movement represents an “embrace of scientific data and expertise” and a “rejection of penal populism and of ill-informed common sense,” writes Stanford University law professor Joan Petersilia. Of course, it was the “expert”-run corrections regime of the 1960s and 1970s that ushered in “penal populism” and “ill-informed common sense,” in response to the ensuing crime wave.

The problem with the EBP movement is that there is not much E for the P. As Petersilia herself acknowledges, few programs have been shown to work. And if a program produces an effect in its initial iteration, that result may not be replicable, especially at a larger scale. None of the six programs evaluated by the Justice Department for prisoner reentry was rated as effective. Two had no positive results, while the efficacy of the others had not been established. The federal government funded a large “collaborative” reentry program for serious and violent offenders. Though “collaborative” is almost as favored a term as “evidence-based,” the program had no impact on employment or the rearrest and re-incarceration rates of the ex-cons.

Even programs concentrating on work may not have lasting effects. Fifty-five percent of ex-offenders placed in government-subsidized jobs in Chicago, Detroit, Milwaukee, and St. Paul had been rearrested two years after the program ended, compared with 52 percent of ex-offenders in a control group who were not placed in jobs, the MDRC found in an evaluation. Twenty-nine percent of the subsidized jobs recipients had been reconvicted two years out, compared with 27 percent of the control group.

Moreover, it is hard to find an offender who has not already been given programs galore, whether “evidence-based” or not. “These guys have been through so many programs,” says an Orange County probation officer. The officer is checking up on a heroin dealer and user in Santa Ana. “I’ve offered this guy programs, but he’s declined. I’ve forced him into residential programs. We tell them to get counseling, they don’t show up. I offer people resources, but they don’t follow through because they’re addicts.” The dealer is not home, but his sister complains that nearby Saddle View Park is a favorite hangout for druggies and an easy place for her brother to get high.

The female meth user hiding in Steve the burglar’s house during the Santa Ana probation check had previously been given a government-subsidized job with the department store Marshalls as a “women’s associate” in the handbags section. She has also received residential treatment for drugs and alcohol use and been placed in a maternity home. When the job subsidy ended, the store cut back on her hours, and the probationer, who falsely gave her name as “Yvette” during the probation check, stopped showing up. She was fired. Six months later, she returned to a practice that she had begun at age 15: stealing cars, this time from an auto dealership when she noticed a bunch of keys left unattended.

Other deincarceration advocates are frankly skeptical about programs as a means of reducing the prison population. “To lower the prison population we need to change the penal code,” says James Austin, president of the JFA Institute. “Don’t talk to me about programs. We need to bring sentences back to a rational level.” The advocates even admit that letting prisoners out after a shorter time in prison will lead to more crime, though such acknowledgments rarely make it into the public discourse. But under a cost-benefit analysis, a crime increase may be an acceptable result, if the incarceration savings are put to better uses, they argue—though here, deincarceration advocates seemingly reimport a belief in programs. “If we let everyone out six months earlier, some guy will throw a little old lady off the roof,” says Michael Jacobson, executive director of the Institute for State and Local Governance at the City University of New York. “The substantive argument to be made is that reinvesting the enormous savings from reduced prison populations into programs that we know effectively reduce crime will make us all safer in the end.” Fordham law professor John Pfaff says: “If we are experiencing more $30 thefts because we aren’t spending $6,000 or $7,000 per year to lock someone up, that could be an efficient reallocation of costs,” especially if the savings are put toward greater treatment options.

In defense of this bracingly honest argument for shorter sentences, one has to recognize that all sentences are arbitrary to begin with. Though there is political risk in reducing sentences once they have been established at a certain length, if the sentence had always been set at the reduced level, no one would notice or complain. Even deincarceration advocates ignore the inherent arbitrariness of sentences. In the American Society of Criminology newsletter, Jeremy Travis and Bruce Western recently called for sentences to be “proportionate” to the crime, echoing a 2014 National Academy of Sciences panel that they chaired. This is a meaningless principle, since no objective, “proportional” relationship between a crime and its punishment exists.

But though we have no ideal, Platonic length for sentences, we have arguably arrived at our current sentences through trial and error. During the halcyon days of “expert”-driven corrections in the 1960s and 1970s, crime was raging. Sentences got longer until, in conjunction with a policing revolution that began in New York City, they finally put a lid on crime, ushering in the biggest national crime drop in recorded history.

Further, the costs of prison are comparatively modest, contrary to deincarceration advocates on both the right and the left. The states spent $48.5 billion on corrections in 2010, the last year for which a full breakdown of corrections expenditures is available. Never acknowledged is the fact that more than one-fifth of that amount goes to noninstitutional oversight, such as probation and parole, as well as to training. The amount spent on operating prisons and jails was about $37 billion in 2010. The 2010 budget for the federal Bureau of Prisons was $6.1 billion, bringing total federal and state expenditures on institutional confinement that year to $43 billion. (Groups such as the Koch brothers–supported Coalition for Public Safety regularly claim $80 billion in annual prison spending.) That $43 billion is a small fraction of the $1.9 trillion that the states alone spent in 2010, an outlay dominated by education and welfare payments. In 2011, the states contributed $283 billion to federal means-tested welfare programs like Medicaid and Temporary Assistance to Needy Families cash aid. Los Angeles has proposed a $5.8 billion budget to host the 2024 Summer Olympics, an amount lowballed by several billion. Americans spend $7.4 billion on Halloween, according to the National Retail Federation. By comparison, $43 billion nationally to incapacitate serious offenders seems a bargain.

The costs of uncontrolled crime dwarf $43 billion—or $80 billion. Efforts to estimate those costs inevitably fall short. Immeasurable is the psychological toll of feeling unsafe in your own neighborhood. It is conventional in anti-incarceration circles to dismiss property crime as “nonserious” and an acceptable consequence of lowered law enforcement. But a street experiencing home or car break-ins is under siege, its residents restricted in their freedoms and well-being. Add violence, and the inhibition on lawful civic and commercial activity intensifies. The loss of business-generated wealth and tax revenue in crime-plagued inner-city areas across the country has spurred usually useless government spending to try to jump-start those crime-strangled economies. That spending eclipses prison outlays. The federal Housing and Urban Development agency alone spent $88 billion in 2014 on Community Planning and Development grants to troubled communities.

Per-prisoner costs are also exaggerated. A widely quoted figure is $2,600 a month per prisoner. But that is an average that includes fixed capital costs and wages. The marginal cost of each new prisoner is closer to $500 a month, according to John Pfaff, at least until a threshold is crossed that either allows the shutting down of a wing or facility or requires the addition of a new one.

A regime of work and high-quality vocational training—such as this program teaching California inmates underwater welding—would be expensive but might reduce recidivism. (MONICA ALMEIDA/THE NEW YORK TIMES/REDUX)
A regime of work and high-quality vocational training—such as this program teaching California inmates underwater welding—would be expensive but might reduce recidivism. (MONICA ALMEIDA/THE NEW YORK TIMES/REDUX)

The current case against incarceration may have been built on multiple fictions, but prison unquestionably is, on average, a squalid, spirit-killing enterprise that can turn borderline offenders into more hardened criminals. (Research is divided on whether incarceration in the aggregate increases recidivism: some studies find increased lawbreaking among ex-prisoners, some studies find no effect, and some find a decrease in recidivism. The impact on future employment and earnings is also contested, with some studies finding no negative effects and others even finding a short-term bounce in employment upon release.) If there were alternatives to arresting and confining criminals that provided the same anticrime benefits, they should be implemented.

California provides a test case for how not to go about deincarceration and decriminalization. In November 2014, voters passed Proposition 47, a ballot measure to reclassify retroactively many drug and property felonies as misdemeanors. All thefts under $950, including of someone’s car or of an illegal gun, or yanking a handbag or laptop from someone’s hands, would now be a misdemeanor, which can be punished, at most, only by time in jail, not prison. In fact, misdemeanor convictions only infrequently yield jail time. Misdemeanor offenders are not put under probation or parole supervision in the community, which means that they are not subject to search by probation officers; they cannot be ordered into drug treatment. DNA cannot be collected from misdemeanor suspects, diminishing law enforcement’s ability to solve past and future crimes. Many officers have stopped making arrests for a range of drug and property offenses, since the “juice is not worth the squeeze,” as a Santa Ana gang detective put it: the time spent processing a case exceeds the consequences to the offender. Prosecutors previously could file a shoplifting incident as a felony commercial burglary if the facts warranted it and the thief had a serious criminal history. They have lost that tool when the goods stolen are worth less than $950. “Now many so-called misdemeanor offenders are hard-core criminals,” says Jennifer Contini, an assistant district attorney in Orange County.

Prop. 47 was sold to voters as a way to remove from offenders the stigma of a felony record and to lower the prison and jail populations, with their attendant racial disparities. Someone arrested for a misdemeanor, if he has identification and no outstanding warrants, is cited in the field and asked to come back to court on another day, rather than being taken into a police station or jail for booking. As a result, the state’s jail population dropped after Prop. 47 passed, though it is starting to rise again, thanks to crime increases. The measure also promised to reroute the money saved on incarceration into truancy, treatment, and mental-health programs, starting in 2016.

Crime increased immediately after Prop. 47 passed. “We had 10 years of crime reductions,” Los Angeles county sheriff Jim McDonnell told the Associated Press in August, “and all of a sudden, right after November when 47 kicked in that changed and fairly dramatically, very quickly. It would be naive to say that 47 didn’t play a major role in that. . . . People are no longer incarcerated, they’re not in treatment, they’re out reoffending on the street.” In the city of Los Angeles, violent crime rose nearly 20 percent through August 22, 2015, compared with the same period in 2014; property crime was up 11 percent. Shooting victims were up 27 percent. Arrests were down 9 percent. In Santa Ana, felony crime was up 33 percent in May 2015, compared with May 2014. Violent crime was up 28 percent, property crime up 43 percent, and robbery up 89 percent. In nearby Costa Mesa, violent crime increased 47 percent, and theft was up 44 percent, through late July, compared with the same period in 2014. In San Francisco, violent crime was up 13 percent, and property crime up 22 percent, through June 2015 over the previous year. Granted, cities across the U.S. have experienced a sharp crime increase during the last year, as officers back off of proactive policing in response to the anti-cop calumnies of the Black Lives Matter movement. But the addition of Prop. 47 in California appears to be adding to law enforcement’s challenges.

The criminal world is well versed in the new regime. “Sure, I know about Prop. 47,” says Mitchell, a 62-year-old vagrant hanging out in Santa Ana’s perennial Civic Center homeless encampment. Mitchell, who sports sunglasses, cargo shorts, and a ponytail, has spent 22 years in prison for 24 felony convictions, including for burglary and meth trafficking. “I’ve seen 47 in action,” he says. “If someone is busted, the police cite and release them right there. People [i.e., criminals] are getting a little sloppier. If it’s a felony and I’m sitting there with the cops, I’m going to be a little nervous. Now it’s just a ticket.” Mitchell winces: “I think that’s a little lax.” Theft should be serious, he says. Even when theft was a felony, the system used discretion in prosecuting: “If you’re caught at Kmart, you’re not going to do time for your first offense. If it’s your second, maybe you’ll do 30 days in jail.” Now there’s more dope flowing and the drug trade is picking up, Mitchell says. “There’s more people on the streets. It’s fast living and a fast life.”

Los Angeles’s Skid Row is the most anarchic and squalid homeless colony in the nation, compared with which the tormented figures of a Boschian hellscape might as well be in a fête galante. (See “The Reclamation of Skid Row,” Autumn 2007.) Through August 22, 2015, violent crime in the area was up more than 57 percent over the previous year, shots fired were up 350 percent, and property crime up over 25 percent. In July, a man was nearly decapitated with a machete. “I see the effects of 47 every day. People are emboldened,” says Wendell Blassingame, the self-described mayor of San Julian Park (known as “marijuana park”), in the heart of Skid Row. Blassingame is seated at a cardboard table with flyers for social programs, as mentally ill addicts stumble past headed for the park’s picnic tables. Prop. 47 has made it harder to keep order, he says, because police can’t ask the gang members who prey on the local population if they are on parole or probation. It has led to the “WDNC phenomenon: ‘We do not care,’ ” says Blassingame. “People say: ‘What can they do to me?’ Everyone knows they’re not going to prison. Even if they commit a violent crime, the DA may let them plea out. And they’re back on the streets.”

The proponents of Prop. 47 say: not to worry. By 2016, the promised savings from prison and jail diversion will have materialized and been redirected to treatment programs. This reassurance overlooks the fate of another California prison-diversion program, Proposition 36, which has fallen out of official memory. That ballot initiative, passed in 2000, gave nonviolent drug offenders the option of free treatment in lieu of incarceration. One-quarter of defendants who chose treatment never showed up; less than a third who did start treatment completed it. Arrests increased, even among those who completed treatment, according to Angela Hawken, a public-policy professor at Pepperdine University. Prop. 36 has quietly been shelved, but 47 seems to be treading the same path by removing the threat of confinement as a means of getting people to change their behavior. The number of offenders enrolled in California’s drug courts has dropped sharply since Prop. 47, since they no longer face the threat of prison time for most drug and property crimes.

Deincarceration advocates still applaud Prop. 47 anyway. The fact that prosecutors have lost discretion to charge a felony for most theft and drug offenses is a good thing, says John Pfaff, since prosecutors needed reining in. Their excessive zeal to prosecute was a significant cause of “mass incarceration,” Pfaff and others argue.

California’s experience with Prop. 47 to date suggests that a wholesale downgrading of offenses is a reckless solution to “mass incarceration.” There might be another way to keep people out of prison while also constraining crime, however: tight supervision in the community, accompanied by infallible but modest sanctions for slipping up. A movement known as Swift and Certain (SAC) argues that what changes criminal behavior is not the severity of a punishment—its length—but its certainty and the swiftness with which it is imposed after the offense. Since most criminals have short time horizons, telling them that after six arrests they may face a prison sentence of five years is not as much of a deterrent as telling them that as soon as they offend, they will go to jail, if only for a day or two, according to SAC proponents.

The crown jewel of the SAC movement is the HOPE (Hawaii’s Opportunity Probation with Enforcement) program, developed by Hawaii superior court judge Steve Alm. Alm noticed that probation officers would regularly come into his court seeking to revoke probation for their clients in punishment for repeated meth use, which violated the conditions of their probation. But the probation officer would show up to Alm’s chambers only after the offender had accumulated his sixth or so dirty urine test—at which point, the exasperated officer would announce, in essence: “That’s it, no more Mr. Nice Guy. I’m sending you to prison on your original felony sentence” (which could be five or ten years for such offenses as sexual assault or burglary). This pattern was the opposite of how best to modify behavior, Alm concluded. It sent the message that the offender could expect to get away with drug violations almost indefinitely, until some arbitrary and unpredictable moment when the system would come down hard by reimposing the original long prison term. By contrast, we train teenagers by meting out punishment exactly as promised, after, say, a weekend curfew violation. The longer that punishment is deferred, the less relationship it seems to have to the underlying behavior and the less deterrent and retributive effect it possesses.

Alm devised HOPE as a fundamentally different probation regime. Probationers would be randomly tested for drug use six times a month—a more frequent testing regime than usual. At their very first dirty urine, they would immediately be sent to jail for a few days. Other probation violations, such as missing an appointment with a probation officer or skipping out on mandated treatment, would also immediately be sanctioned with a short jail stay. Subsequent violations would bring lengthening jail commitments, ultimately culminating in a probation revocation to prison. Alm called every probationer entering the HOPE program into his court and explained the system, so that the probationer would know exactly how to avoid sanctions and what to expect if he violated the rules.

The results were startling. Half of the probationers in Alm’s experimental program never tested dirty for meth again. Another quarter of the experimental population stopped using meth after one trip to jail. Those who continued to use after repeated short stays were ordered into treatment. Arrests for new crimes also dropped in the HOPE population. One-fifth of probationers in the HOPE program were rearrested after a one-year follow-up, compared with nearly half of the probationers in a control group given traditional probation without swift and certain sanctions.

HOPE revealed a previously unrecognized fact: many drug users can stop on their own, without treatment, if the right incentives are in place. Placing all drug offenders in treatment is a waste of resources; a sanctioning regime like HOPE acts as a sorting mechanism to distinguish the drug users who can control themselves from those who can’t—about 9 percent in the original HOPE sample. HOPE is crucially different in that respect from drug courts, which place every enrolled offender in mandated treatment without seeing if he can stop on his own. Drug court should be something you fail into, says New York University’s Mark Kleiman.

HOPE also validated the principle that lengthy punishment is not necessary to change behavior, at least regarding substance abuse; short sanctions can work so long as their application is certain and immediate. The question is how far the SAC principle can go in transforming the criminal-justice system. As of July 2015, 28 states had a SAC program operating within them, with interest in the concept growing daily; the largest jurisdiction so far is the entirety of Washington State, where the statewide probation department has retooled itself for immediate, no-discretion sanctions for probation violations. An evaluation of the Washington State program will be out shortly. South Dakota created a SAC program for DUI offenders that requires twice-daily alcohol testing, while otherwise allowing convicted offenders to drive so long as they blow clean. Half of the participants never skip or fail a test. Jurisdictions are experimenting with how minimal sanctions can be and still change behavior; some are assigning offenders to community service instead of sending them to jail. Others are using carrots in addition to sticks: in Washington State, for example, if a probationer complies with all the conditions of his probation for 18 months, he can free himself from further oversight. Preliminary results show that those released probationers are not rearrested.

Could the Swift and Certain principle provide the key to unlocking prisons, by so closely regulating offenders’ behavior in the community that they can remain there without needing long-term confinement in prison or jail? Perhaps, but the implementation challenges are great. Swift and certain sanctioning sounds intuitively obvious, but it takes an enormous amount of institutional buy-in and coordination. Everyone in a local criminal-justice system, including police and probation officers, prosecutors, defense attorneys, and judges, must be committed to making sure that offenders are immediately punished; if the sanctioning is not consistent, the credibility and legitimacy of the threat are undermined.

Some jurisdictions have been unable to ensure uniformity of response. Sometimes this lack of uniformity represents lack of manpower and management capacity; at other times, it reflects disagreement with the program. Many probation officers take satisfaction in the exercise of discretion regarding punishment; they see an individualized response to each probationer’s situation as a mark of justice. “I give people chances. I am fair,” says an Orange County probation officer proudly. SAC removes that discretion to give an offender a second, third, or fifth chance; every offender who violates the conditions of his freedom must face immediate and preset consequences. (The tension between uniformity and discretion pervades the criminal-justice system. Do we want police officers to arrest everyone for drinking in public, or should they make an ad hoc judgment about whether simply to pour out the liquor and warn the drinker, at the risk of unequal treatment? Judicial discretion in sentencing, once the norm, was curtailed during the 1980s and 1990s because of the perception that judges were being too lenient with criminals. Now the pendulum is swinging back.)

Taking SAC to scale in large urban jurisdictions would require a revolution in management. New York City has more than 1 million open arrest warrants for failure to appear in court or pay a fine for a low-level offense. No one is going after those absconders. In Los Angeles, 2,000 felon absconders are still at large who never checked in with authorities after a 2011 California law changed their confinement status, according to Los Angeles assistant chief Michel Moore. Under such conditions, it is almost unthinkable that someone who skips out of a Breathalyzer or drug test would be immediately picked up and brought to court. SAC advocates suggest starting small in urban areas. Probation departments would need to be enlarged. But once the deterrent effect of immediate sanctioning kicks in, the caseload requiring sanctions would drop precipitously, SAC advocates maintain.

More foundational questions arise as well about Swift and Certain’s potential to lower crime and the prison population. The trigger for SAC sanctions at present is substance abuse, as well as violations of other easily monitored probation conditions, such as showing up for appointments. There is not a technology now available for immediately detecting property and violent crimes, though GPS monitoring holds out some promise. Yes, a huge proportion of criminals abuse drugs and alcohol and are thus candidates for SAC monitoring and sanctioning: a 2009 study by the Office of National Drug Control Policy found that 87 percent of arrestees test positive for drug use. The theory is that by reducing a criminal’s substance abuse and by rigorously enforcing key probation conditions, he will have less opportunity and inclination to commit crimes and can be kept safely in the community. The theory seems plausible, but more data are needed on changes in re-offending rates among SAC enrollees.

A final question is whether short but certain punishments are always as effective as long but uncertain ones. Offenders facing their first institutional confinement will likely be traumatized by a weekend in jail. But for seasoned offenders, short-term sentences are less of a deterrent. California has a flabby version of SAC called “flash incarceration,” which allows probation officers, at their discretion, summarily to send a client to jail for ten days. “A career criminal can do ten days standing on his head,” says Steve Martin, a frequent court consultant on prison and jail management. Another Steve, the burglar on probation in Santa Ana, had recently done a ten-day flash for associating with his felon girlfriend, in violation of his probation terms. He seemed to regard it as the equivalent of a trip to a spa: “I just wasted the time, it was a chance to catch my breath and get some exercise with push-ups,” he said. “When you’re locked up, at least you’re working out inside your cell.”

The response to Prop. 47 would also seem to suggest that length of sentence matters, since criminals are scoffing at the lowered sanctions associated with misdemeanor offenses. A SAC proponent would respond that those shortened misdemeanor sanctions lack the swiftness and certainty of application essential to behavior modification. A persistent offender may brush off a short jail sentence to be imposed at some indefinite day in the future, but if you tell him that he’s going to jail now and losing his Saturday night out with his homies, you’ve got his attention, claims Kleiman.

Many criminologists and prisoner advocates resist SAC because they think that it is too punitive and because it de-emphasizes services and treatment. “Deterrence-oriented programs [should be] subsidiary to the delivery of therapy aimed at fixing the deficits (or criminogenic needs) leading to reoffending,” argued three criminologists from the University of Cincinnati in the 2014 Federal Probation Reporter. But it is precisely its simple, behaviorist approach to criminal offending that makes it so appealing. We don’t need more services—we need more immediate consequences, says Kleiman.

For all the challenges of bringing SAC principles to scale, the concept is the most promising alternative to the carceral status quo. At the end of 2013, there were twice as many criminals in the community on probation and parole than confined in prison or jail. Those 4.7 million probationers and parolees are already not being particularly well supervised. If the institutional population—2.3 million at the end of 2013—is greatly reduced, many more offenders will be in the community needing supervision. The more that probation and parole departments can embrace the idea of SAC sanctioning, the better the chances for keeping offenders out of trouble. Mark Kleiman has proposed the most radical application of SAC yet, as part of a reentry program for violent offenders. Violent felons would serve the final part of their sentence in small scatter-site apartments, where they would initially be under something close to house arrest, permitted to go out only to work or to look for work, to make necessary purchases, and to meet with their correctional supervisor. Employment attendance would be monitored. A GPS ankle bracelet and camera in their apartment would track their movements. Every day that they comply with every condition of release would gradually gain them more freedoms. Violations of those conditions would be immediately sanctioned. The costs would be offset with savings on incarceration. The idea of early release for violent felons, however, strikes even some SAC advocates as a step too far. “Let them serve their time if they’re violent offenders,” Judge Alm told me. “It’s not that easy to get into prison.” Alm predicts that well-organized victims’ rights groups would browbeat any legislator who contemplated Kleiman’s reentry program.

The other major alternative to incarceration is policing—above all, pedestrian stops and Broken Windows policing. New York’s prison population dropped 17 percent between 2000 and 2009, while the number of prisoners in the rest of the country continued to rise. The decrease in the New York prison population is all the more surprising, since the average sentence meted out to convicted felons over that period increased considerably, in violation of the deincarceration platform. The different trajectories of the New York and national prison counts reflect the onset, in 1994, of the New York Police Department’s practice of aggressively enforcing quality-of-life laws and stopping and questioning people engaged in suspicious behavior. Misdemeanor arrests in New York City doubled from 1990 to 2009, while felony arrests (and thus, felony convictions) plummeted, as documented by Michael Jacobson and James Austin, in a 2013 study for the Brennan Center for Justice. Even though convicted felons in New York were being sentenced to longer terms, there were far fewer such convicts, so the overall incarcerated population fell. And the reason for that drop in felony crime is that the NYPD was apprehending potential felons for lower-level quality-of-life offenses and getting them off the street before they had the opportunity to commit more serious crimes.

Reasonable-suspicion stops represent an even earlier intervention in potentially serious criminal behavior: questioning someone who looks to be casing a jewelry store in an area plagued by burglaries may prevent a subsequent break-in. And the possibility of getting stopped deters crime in the first place. An NYPD detective who used to work the club scene in midtown Manhattan during the Rudolph Giuliani mayoralty recalls talking to someone who had come into Manhattan from the outer boroughs to party. “We don’t carry guns into Manhattan,” the club goer said. “I’ve been stopped three times since I got off the train.” Now, according to the detective, under the Bill de Blasio mayoralty, “no one is getting stopped and everyone’s carrying.” Of course, the political opposition to policing, especially to misdemeanor enforcement and pedestrian stops, is even more pointed now than the opposition to incarceration.

No matter how effective the police are at deterring crime, there will always be criminals who should be incarcerated. It is a truism that prisons should be safe, orderly, and conducive to self-reform. But that is easier said than done, or it would have happened long ago. Ideally, all prisoners would work, since there is no better rehab program than the discipline and self-esteem that come from regular labor. The larger the prison, however, the harder it is to get the entire incarcerated population productively engaged, since the logistics of moving large numbers of prisoners from cells to a workplace without a violent incident are complex and labor-intensive. Unions fight prison labor as unfair competition. Prisoner advocates complain if prison work is not paid the minimum wage, raising its costs further. Most prisoners, however, if given the choice between earning minimum wage and earning significant time off from their sentence for a flawless work record, will unhesitatingly choose the latter option. High-quality vocational training should also be available for the off hours when prisoners are not working. Such a universal work and training regime would be expensive but may pay off in lower recidivism costs.

In the final analysis, America does not have an incarceration problem; it has a crime problem. And the only answer to that crime problem is to rebuild the family—above all, the black family. The media troll incessantly for an outlier case of a hapless bourgeois who got slammed in prison for a one-shot mistake. In fact, the core criminal-justice population is the black underclass. “Young black males between the ages of 17 and 26 drive the system,” says corrections expert Steve Martin. “Family is the solution—and the work ethic. You show me people with intact families and those folks work—their chances of ending up in prison are zero.”

The demonization of the police and the criminal-justice system must end. As the Black Lives Matter movement marches forward with no apparent diminution of strength, there are signs that the very legitimacy of law and order is breaking down in urban areas. Resistance to lawful police action is becoming routine. Officers are reluctant to engage, given the nonstop campaign against them. Homicides in 35 large U.S. cities were up nearly 20 percent by August 2015. Liberal elites have successfully kept attention focused exclusively on phantom police and criminal-justice racism while squelching even the most nascent discussion of the crime-breeding chaos of inner-city underclass culture. We are playing with fire.

Top Photo by AndrewLichtenstein/Corbis via Getty Images


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next