Attorney General Jeff Sessions has said over and over again that he sees the uptick in violent crime in a few major cities in the US as the start of a "dangerous trend." Let's take a look at the numbers. (Daron Taylor/The Washington Post)

Attorney General Jeff Sessions wants to reverse the policies of the Obama administration, by prosecuting more cases involving guns and drugs, and seeking more mandatory minimum sentences. John Pfaff is a professor of law at Fordham University, and the author of “Locked In: The True Causes of Mass Incarceration — and How to Achieve Real Reform,” a new book on America’s prison system. I interviewed him via email about the causes of incarceration, and the likely consequences of Sessions’s policy shift.

Many people have argued that the “war on drugs” has led to the great increase in the prison population. You argue that this theory doesn’t explain most of the increase. Why not?

At its simplest, it’s just a matter of numbers. Over half of all people in state prison are there for violent crimes, and over half the growth in state prisons since 1980 is due to locking up people for violent offenses. As of 2015, only about 16 percent of those in prison are there for drugs crimes. Of course, it’s true that drug prohibition can cause non-drug crimes, ranging from theft to fund a (more-expensive) habit to murder over a drug deal gone bad, so not everyone in prison as a result of the “war on drugs” is there for a drug crime. But studies suggest that ending the war on drugs would have complicated, off-setting impacts. For example, there would be fewer deaths over drug deals but more murders committed by people while abusing (some but not all) drugs.

There would be no more people arrested for selling drugs — and almost everyone in prison for drugs is there for dealing, not possession — but many of those who currently sell would still struggle to find gainful employment and would thus likely turn to other forms of crime to make ends meet.

One problem that both scholars and reformers face is that they think of the justice system as just that — a system with a coherent logic, design and goals. You’ve argued that it’s something much more messy — a kind of Kafkaesque ecology, in which unintended outcomes happen all the time. How does that ecology work, and how has it led to more people in prison?

The fairly incoherent way we divide responsibility across cities (which run police departments), counties (which elect prosecutors and judges and pay for jails), and states (which fund prisons and whose governors control the parole process) leads to all sorts of moral hazard risks by haphazardly separating cost and benefits. My “favorite” example is that county-elected prosecutors face no limits on how many people they can send to state-funded prison. Prosecutors get all the tough-on-crime credibility from sending people to prison, but their counties bear none of the financial cost. In fact, it’s “cheaper” for county prosecutors to charge someone with a more-serious felony (which sends the defendant to state prison) than with a lesser misdemeanor (which lands the defendant in county-funded jail or probation).

Electing prosecutors at the county level also creates a dangerous split in costs and benefits within the county, which helps explain racial disparities in punishment. In more-urban counties, the whiter, more suburban areas have a lot of political power, and they likely play an outsized role in electing the prosecutor, who in turn tends to enforce the law in poorer, more minority urban areas. Those suburbanites feel the benefits of reduced crime but few if any of the costs, which are borne by a population that they are divorced from socially, culturally, economically and geographically, in no small part because of our history of red-lining and other forms of racial exclusion. We should accordingly expect prosecutors to pay too little attention to the costs of aggressive enforcement.


Attorney General Jeff Sessions discussing his decision to instruct prosecutors to pursue stricter punishments for drug trade crimes. (European Pressphoto Agency)

One key group of actors in this ecology are prosecutors — you argue that their incentives are one of the key factors driving the increased prison population. Why is this so?

At least since crime and arrests started to drop in the early 1990s, the main engine driving prison growth has been an increased willingness on the part of prosecutors to charge more and more arrestees with felony charges. We lack almost any data on prosecutors, so it’s hard to say with any certainty why this change happened.

I have a lot of plausible theories, but right now the one that seems like the most important is a boring-but-critical story of employment. Between the early 1970s and 1990, as crime rose steadily, the number of prosecutors rose from 17,000 to 20,000; between 1990 and 2008, as crime dropped, we expanded the number of prosecutors by three times as much, to 30,000. There’s no evidence I’ve seen that individual prosecutors are more aggressive today than in the 1990s or even 1970s. We just have a lot more of them who need to justify their positions. We arrest over 10 million people every year: There are plenty of cases for them to take if they need to.

Jeff Sessions has just announced a “tough on crime” order, intended to push prosecutors to seek longer sentences. How consequential is this order going to be, and for whom?

When it comes to federal policy, it’s important to realize that the federal system is fairly small, holding only 12 percent of U.S. prisoners, and federal policies cannot apply to the states. So while Sessions’s new rule may cause an increase in the size of the federal prison (where, unlike the states, about half the inmates are in for drug crimes), its direct impact on the states will be nil, and thus its direct impact on the overall U.S. incarceration will be slight.

More concerning is any sort of “bully pulpit” effect: Will Sessions’s “tough on crime” and Trump’s “carnage in America” rhetoric shape how local county prosecutors use their vast discretion? There’s no rigorous data on this, but my sense from the snippets of data we have is that any such effect will be slight. Prosecutors, as far as I can tell, focus very much on local conditions and local politics, and people’s attitudes towards crime appear to be fairly local. There are a lot of way prison reform can fail or falter in the years ahead, but I don’t think the tough talk coming out of D.C. right now will matter much.