From President Obama on down, politicians are saying that America needs to stop locking up so many people. And from President Obama on down, they're saying that the way to do that is to reconsider the way we treat nonviolent drug offenders — to stop giving them long sentences, and to offer them alternatives, like drug treatment, so they can deal with the underlying problems that drove them to crime.
It's a view that has long had public support (even at the height of the war on drugs). The public believes that "nonviolent drug offenders" are just a different kind of person than real criminals, and politicians are playing into that to encourage criminal justice reforms that target nonviolent offenders and rely heavily on a treatment-heavy approach to drug prisoners.
If you believe all that, it's easy to believe that the prison problem is easy to fix. But it isn't — because there aren't that many nonviolent drug offenders in the grand scheme of the criminal justice system, and because there isn't as firm a distinction between "nonviolent drug offenders" and everyone else as the public likes to believe.
Only 20 percent of prisoners in the United States are serving time for drug offenses. The number of prisoners that politicians are willing to consider "low-level, nonviolent" drug offenders is far smaller. And those offenders often aren't addicted to drugs, so drug treatment programs won't help them.
If the problem is that too many people are in prison, focusing on nonviolent drug crimes isn't going to get very far toward addressing it. To really address America's over-incarceration problem, politicians and the public need to stop thinking that "nonviolent drug offenders" are different from, and safer than, people who have committed other kinds of crimes.
Drug offenders are only 20 percent of prisoners — and they're not all nonviolent
The majority of prisoners in the US are in state prisons. The majority of state prisoners are in prison for violent crimes. Drug offenders simply aren't the majority of American prisoners. They're about 20 percent.
Drug offenders are a plurality of federal prisoners, which is one reason politicians and the press focus on them so much: If Congress wanted to reduce the federal prison population, it would get the biggest bang for its buck by reducing drug sentences. But politicians at the state level often focus on drug offenders, as well, even though they make up only about 16 percent of people in state prisons. That's because they know the public is more sympathetic toward low-level drug offenders than to other types of criminals.
In 2013 there were about 308,000 people serving state or federal sentences for drug crimes, according to the Bureau of Justice Statistics — about 20 percent of the total US prison population. But how many of those count as "low-level," "nonviolent" drug offenders, in the eyes of policymakers or the public? How many of them are the prisoners who America has agreed aren't a threat?
Unfortunately, federal statistics don't provide a precise answer to this question. But the statistics we do have can give us a rough estimate. The government's prison statistics count people based on the most serious crime for which they're serving a sentence. So by definition, none of the 308,000 drug prisoners in the US are currently in prison for committing a violent crime. For sentencing reform advocates, that's the same as saying they're nonviolent. But many politicians disagree. Prosecutors emphasize that many drug offenders are in prison for drug trafficking (which they consider "inherently violent"), and some are high-level operatives. Other government officials, including those working on sentencing reform in the Senate, think that "nonviolent" should exclude inmates whose crimes involved a weapon (which often just means they had weapons on them when arrested).
How many people does that leave? Here's one clue: When the US Sentencing Commission considered allowing drug prisoners to appeal for shorter sentences last year, the Obama administration urged them to limit it to certain nonviolent offenders without significant criminal histories — and the exclusions they included would have cut the eligible number of prisoners from 51,000 to 20,000. While that doesn't encompass all federal drug prisoners, it seems that as many as half of federal drug prisoners wouldn't count as "low-level, nonviolent." At the state level, a 2002 study by the Sentencing Project found that 58 percent of state drug prisoners had "no history of violence or high-level criminal activity."
Let's assume, conservatively, that only 40 percent of drug prisoners would be considered "high-level" or "violent," or have long criminal histories. That still means that America's "nonviolent drug offenders" account for a paltry 12 percent of America's prisoners.
Drug treatment won't help drug prisoners who aren't addicted
Identifying who should benefit from more lenient criminal treatment is only half the battle. Figuring out how to deal with those prisoners instead is the other half. And when it comes to drug offenders, the primary answer offered by politicians and the public alike is "drug treatment." Drug courts, which provide an alternative to prison for some drug offenders who agree to rehab and intensive probation, have been around since the late 1980s. It's taken them a very long time to catch on, or to be expanded beyond the most minor cases, but they've hit the mainstream as part of the current wave of state criminal-justice reforms. Some states, like New Jersey (under a plan championed by Gov. Chris Christie), make treatment mandatory for people convicted of drug crimes.
But drug treatment only makes sense as an alternative for prisoners who are addicted to drugs. And we don't know how many nonviolent drug offenders are also drug addicts. There are plenty of drug offenders out there — including small-time dealers, who are often the people who get incarcerated — who are in the drug industry to make money, whether it's their only path to making a living or they're simply hoping to get rich. Those people aren't motivated to commit crimes by their addictions. They don't benefit from treatment-focused interventions: The National Association of Drug Court Professionals encourages drug courts for people who were "dealing drugs to support an addiction," but not those dealing "solely for the purposes of financial gain." And we just don't know how many people fall into each of those camps.
So while legislators are finally coming around on the "low-hanging fruit," we don't actually know how many people fall into that category — as opposed to the politically more problematic categories of drug addicts who've convicted violent crimes, or drug offenders who were simply involved in the business for the money.
At a glance, it looks like most prisoners are tied up in drug use: A majority of state and federal prisoners tested positive for using drugs or alcohol in the 30 days before they were arrested. But not everyone who uses drugs or alcohol once a month is an addict. The most recent statistics on drug dependence (using the psychological definition from the Diagnostic and Statistical Manual) come from a study of prisoners in 2006 from anti-drug group CASAColumbia. It found that while about half of all federal prisoners have been convicted for drugs, only 37 percent of all federal prisoners were substance-dependent (on illegal drugs or alcohol). In state prisons, meanwhile, only 16 percent of prisoners were serving time for drugs — but 48 percent were substance-dependent. That raises big questions about just how much overlap there is between "nonviolent drug offenders" and the drug addicts who could benefit from treatment.
In fact, there's some evidence that if prisoners know that drug addicts get lighter penalties or access to more prison services, they have an incentive to say they're addicted to drugs — even if, by medical standards, they're not. The CASAColumbia study found that 163,196 federal, state and local inmates with "substance use disorders" (including both dependence and abuse) had gotten professional services for their addictions in the last year. Meanwhile, 23,498 inmates who didn't have a substance disorder had gotten addiction services. Since not everyone who qualifies for "substance abuse" may actually be addicted, that probably understates the amount of non-addicted inmates getting treatment. And the more that individual prisoners are incentivized to seek drug treatment as a path to an easier sentence, the more possible that is.
What if the people who really need drug treatment are violent offenders?
The premise of favoring treatment over prison is that drug addicts will commit crimes to support their addictions, so curing the addiction is the best way to prevent future crime. But the crimes that addicts commit aren't just drug crimes. CASAColumbia's study found that 77 percent of prisoners who'd committed property crimes, and 65 percent of prisoners who'd committed violent crimes, were either under the influence of alcohol and drugs when they committed the crime, were committing the crime to support their addiction, or had a history of drug and alcohol abuse.
But the public's interest in lighter penalties and alternatives to incarceration for nonviolent offenders doesn't extend to violent offenders at all. Studies have found that Americans take a punitive attitude toward drug trafficking when it's connected to violence, and that the same people who want alternatives to incarceration for nonviolent offenders want longer prison sentences for violent ones. And criminal justice reformers have often reduced sentences for nonviolent offenders, or provided them with alternatives to prison, while keeping the status quo for violent crimes.
As both Vox's German Lopez and I have pointed out before, it's going to be nearly impossible to reduce state prison populations substantially without cutting prison sentences for violent offenders. The hypothetical, drug-addicted "nonviolent drug offender" is the archetype politicians and the public have used to pursue criminal justice reform. But there simply aren't enough of them to save America from mass incarceration. And they may not be the ones who most need to be guided to treatment.