The best hope for federal prison reform: a bill that could disproportionately help white prisoners

Is criminal-justice reform the one big thing that could pass Congress before 2016? Maybe — but any bill would have to go through the Senate Judiciary Committee and its new chairman, Sen. Chuck Grassley (R-IA), who's a known skeptic (at best) of reducing prison sentences. The bipartisan reform bill that's generally agreed to have a chance of passing this year is the smallest and most moderate of the bills that have been on the table — and one that will help disproportionately white and white-collar inmates.

That bill's called the CORRECTIONS Act, and it was introduced Tuesday by Sens. John Cornyn (R-TX) and Sheldon Whitehouse (D-RI).

(Tom Williams/CQ-Roll Call)

The bill would allow some prisoners to participate in "recidivism reduction" education programs. In exchange for completing the programs, or for "productive activities" like prison jobs, they'd get a certain amount of time off their sentences. How much time an inmate got off would depend on how much of a risk he was at to recidivate: lower-risk inmates would get more time off their sentences, higher-risk inmates less — or none at all.

It's a nice idea. And it's one that might actually become law. No one disputes that, thanks to Grassley's ascension, the Senate is less amenable to prison reform than it was last year, but there's some hope that he'd embrace this particular bill — after all, he voted for an earlier version of it in the last Congress. And given the recent bipartisan momentum around sentencing reform — and Cornyn's influence — there's definitely a path forward for the CORRECTIONS Act in both the broader Senate and the House. National Journal's called it "a rare bill that could actually get through Congress and be signed by the president."

But the bill's impact would be limited: only some prisoners would be able to reduce their sentences under it. And the prisoners it would help aren't necessarily the prisoners who need it most — or the ones people think of when they think about reducing mass incarceration.

Who the CORRECTIONS Act excludes

(Sarah L. Voisin/Washington Post)

The federal prison population as a whole is disproportionately people of color: 37.5 percent of all federal prisoners are black, according to the Bureau of Prisons, and 34.4 percent are Hispanic. But only some of them will even be eligible to get risk assessments for potential sentence reduction under the CORRECTIONS Act. And some of the exclusions raise serious questions about the racial makeup of the prisoners who will be helped.

"Career offenders." The bill excludes any inmate with a "criminal history" that places them in the highest category under the federal sentencing guidelines. The problem is that someone gets placed in that category automatically if they're labeled a "career offender," which just means three convictions at either the state or federal level for drug or violent crimes. Most "career offenders," according to the US Sentencing Commission, are African Americans — simply because it's easier to arrest and prosecute them for "offenses that take place in open-air drug markets, which are most often found in impoverished minority neighborhoods… [This] suggests that African-Americans have a higher risk of conviction for a drug trafficking crime than do similar White drug traffickers." In 2000, 69 percent of newly-sentenced "career offenders" were black. (Interestingly, only 17 percent were Hispanic.)

Anyone with more than one federal conviction. This exclusion doesn't mean much for most inmates. It's rare that someone gets convicted of a federal crime on two separate occasions. The exception to this is Native Americans — since crimes that take place on reservations, even crimes that would normally get prosecuted in state courts, get prosecuted in federal court instead. So this would basically exclude any Native American inmate with a prior criminal record from getting his sentence reduced.

Anyone convicted of participating in a "continuing criminal enterprise." This is another label that's typically applied to drug offenders — anyone who's an "organizer, supervisor or manager" of a group of five or more people dealing drugs can be hit with a conviction for a "career criminal enterprise." The statute isn't used that often — only 239 people were convicted under it from 2006 to 2013, according to data from the US Sentencing Commission. But 77 percent of the time, it was used against black or Hispanic defendants.

Anyone serving life without parole. According to a 2013 report from the Sentencing Project, 62.3 percent of federal prisoners serving life sentences are African American — and 16.3 percent are Hispanic.

(Andrew Burton/Getty)

Anyone convicted of a "major fraud." This exclusion is likely to target white inmates — there isn't data on federal prisoners, but over 50 percent of all state prisoners serving time for fraud in 2013 were white. But there are studies showing that black and Hispanic defendants get longer sentences for white-collar crime than white defendants do. And regardless, this won't actually affect that many people — there were only 8,800 people of any race in federal prison for fraud in 2013, and it's not clear how many of those counted as "major fraud."

It's true that Hispanic inmates are actually underrepresented in the excluded groups. But that doesn't mean that Hispanic inmates are going to benefit from sentence reduction. Many Hispanics in US prisons are noncitizens — and many of those noncitizens are either unauthorized immigrants (most of whom are serving time for immigration offenses) who'll be deported once they complete their prison terms, or formerly legal immigrants who lost their legal status because of their criminal conviction. In both of these cases, getting a sentence reduction would just mean getting deported more quickly.

Other exclusions — such as the exclusion of federal sex offenders — would disproportionately target white inmates. So it's not clear how the pool of prisoners eligible for the CORRECTION Act's reforms would look, on the whole. Fifty percent of the people who were convicted in 2013 for things that would make them ineligible for the CORRECTIONS Act are white, which is much greater than the proportion of whites in federal prison. But that doesn't look at all the prisoners sentenced before 2013, during the boom years of mass incarceration.

Paul Hofer, of the Sentencing Resource Counsel Project for the Federal Public and Community Defenders, has conducted preliminary estimates looking at more of the data. He says, "It seems likely there will be an adverse impact on minorities from the exclusions."

Risk assessment could exacerbate racial gaps

But being eligible for reductions is only the beginning. How much time prisoners can get off their sentences depends on their risk of recidivism. Inmates who are deemed to be "low risk" can get 10 days off their sentence for every 30 days they spend in an education program or "productive activity"; inmates who are deemed to be "medium risk" get 5; and the highest-risk inmates can't get any time off their sentences at all (unless they can change their risk category).

But how does the government determine how likely someone is to recidivate? The bill tells the federal government to come up with a risk assessment tool. These tests are used in several states and in federal court to figure out how best to manage an inmate's case — or to determine whether someone should be put on probation instead of prison to begin with. But most states shy away from using them to determine the length of an inmate's sentence.

Prisoners conduct agricultural labor, which could make them eligible for reduced sentences under the new bill. (Ann Hermes/Christian Science Monitor)

And there's a reason for that. Some of the factors used to determine recidivism risk are "dynamic" — they're factors that an individual can change over time. But others are "static" factors: they say more about the environment where an inmate lives, or where he grew up, than about his own behavior.

One of the major risk-assessment tools treats drug use, low education level, and frequent changes in residence as factors that put someone at higher risk to recidivate. Even factors that look fair on the face of it, like the age an inmate was when he was arrested for the first time, can just mean that the inmate lived in a neighborhood where teenagers (or younger) were under police suspicion.

The CORRECTIONS Act tells the federal government to "emphasize" dynamic factors in its risk assessments, according to Sen. Whitehouse's office, "so that all inmates can lower their risk of recidivism." But it's not clear what that would look like — most existing assessments use a mix of the two. And even if static factors aren't disqualifying, including them at all could put inmates who aren't affluent or educated at a disadvantage.

This is being hailed as a big deal because the goalposts have shifted

The CORRECTIONS Act is what's known as a "back-end" prison reform: it reduces the sentences of people who are already in prison. But it doesn't address the front end: how many people are getting sent to prison to begin with, and for how long.

Criminal justice reformers agree that both ends are necessary, but that when it comes to reducing mass incarceration, it's the front-end reforms that are really important. And Sen. Whitehouse's office acknowledges that without front-end reforms, the CORRECTIONS Act is hampered by "the broader problem in our criminal justice system of minority communities having much higher rates of arrest and conviction."

Sen. Whitehouse's office "hopes our bill can jumpstart a debate on" front-end reforms as well as back-end ones. But last Congress, the Senate Judiciary Committee looked at the two together: in addition to an earlier version of the CORRECTIONS Act, the committee passed the Smarter Sentencing Act, a bill to reduce mandatory minimum sentences for various crimes. In fact, Smarter Sentencing, the front-end reform, was the higher-profile bill.

Now, the situation appears to have reversed: back-end reform is leading the way. Smarter Sentencing is also being reintroduced in the Senate, with another bipartisan group of senators. But Judiciary Chairman Grassley is skeptical, and the Smarter Sentencing Act hasn't gotten the "could pass" momentum that the CORRECTIONS Act has.

The fact that the CORRECTIONS Act has gone from a modest supplement to sentencing reform to the star of the show means that the goalposts have quietly shifted on criminal justice reform from last year to this year. Policymakers may be saying that criminal justice reform is the one big thing that can get done in the 114th Congress, but they've reduced their expectations about what can actually pass. Before Chairman Grassley's Senate Judiciary Committee even passes any bills, he's already changed the debate over what Congress can do to reduce mass incarceration — and who it ought to help.

Explanatory journalism is a public good

At Vox, we believe that everyone deserves access to information that helps them understand and shape the world they live in. That's why we keep our work free. Support our mission and help keep Vox free for all by making a financial contribution to Vox today.

One-Time Monthly Annual

$95/year

$120/year

$250/year

$350/year

Other

$

Yes, I'll give $120/year

Yes, I'll give $120/year

We accept credit card, Apple Pay, and Google Pay. You can also contribute via