Friday, November 8, 2013

Reducing the Federal Prison Population

The number of Americans in federal prison has risen more than eight-fold in the last three decades, from 25,000 in 1980 to 213,000 today. While it seems plausible that some the increase has been useful and justified as part of an effort to reduce crime, it also seems plausible that the increase has probably gone too far. Julie Samuels, Nancy La Vigne and Samuel Taxy of the Urban Institute lay out this view in "Stemming the Tide: Strategies to Reduce the Growth and Cut the Cost of the Federal Prison System."  Basically, their plan is to reduce prison time for nonviolent drug offenses.

What factors are driving the higher federal prison populations?

"The short explanation for the rapid prison population growth is that more people are sentenced to prison and for longer terms. In fiscal year (FY) 2011, more than 90 percent of convicted federal offenders were sentenced to prison, while about 10 percent got probation. By comparison, in 1986, only 50 percent received a prison sentence, over 37 percent received probation, and most of the remainder received a fine. Though the number of inmates sentenced for immigration crimes has also risen, long drug sentences are the main driver of the population’s unsustainable growth. In 2011, drug trafficking sentences averaged 74 months, though they have been falling since 2008. Mandatory minimums have kept even nonviolent drug offenders behind bars for a long time. The average federal prison sentence in 2011 was 52 months, generally higher than prison sentences at the state level for similar crime types. This difference is magnified by the fact that, at the federal level, all offenders must serve at least 87 percent of their sentences, while, at the state level, most serve a lower percentage and nonviolent offenders often serve less than 50 percent of their time. ... Before the Sentencing Reform Act of 1984 and mandatory minimums for drug offenses, a quarter of all federal drug offenders were fined or sentenced to probation, not prison. Today 95 percent are sentenced to a term of imprisonment. The average time served before 1984 was 38.5 months, almost half of what it is now. 

Samuel, La Vigne, and Taxy emphasize that the current system has a fair amount of discretion in how it handles drug offenders. For example, there is discretion over whether drug offenders will be prosecuted in the federal system or in a state system, where offenders are less likely to end up in prison. Prosecutors might choose to bring lesser charges, or judges might find various reasons to impose lower penalties. The Obama administration is putting this approach into effect, as Samuels, La Vigne and Taxy explain:

"Until recently, some lower-level, nonviolent drug offenders were subject to mandatory minimum penalties regardless of their role in the organization. As mandatory minimum penalties were originally intended to target “serious” and “major” offenders, these terms of  imprisonment may be unnecessarily lengthy with no added benefit to public safety. Attorney  General Holder’s 2013 Department Policy Memo directs prosecutors to refrain from charging lower-level, nonviolent drug offenders with drug quantities that would trigger a mandatory minimum sentence.
But I confess that this  emphasis on discretion makes me a little queasy, because in a big country, discretion inevitably means that those who committed the same crime will end up being treated very differently by the criminal justice system, depending on accidents of geography and jurisdiction and which prosecutor and judge they face. Greater use of discretion now is probably a lawsuit for unequal treatment waiting to happen in the future. In addition, relying on the discretion of law enforcement is a way for legislators to duck responsibility.

Thus, I would favor changing the sentencing laws directly. For example, among the policies they discuss is one that would reduce mandatory minimum sentences fro some drug offenders: say, cutting certain 5-year minimums to 2, 10-year minimums to 5, and 20-year minimums to 10. They estimate that over time, this change would "would have a monumental effect on the prison system."

There are also a range of options other than sentencing an offender to prison, like halfway houses and home confinement. However, it seems that increased use of probation is the only way to save money. Samuels, La Vigne, and Taxy explain:

"The average cost of housing an inmate in a BOP facility in FY 2012 was over $29,000 annually. ... [M]uch of these average costs of housing an inmate are fixed, as they go toward maintaining and staffing facilities (which are unlikely to close as a result of a shrinking prison population). Thus, the average marginal cost of increasing or decreasing the population by one inmate is $10,363. Average annual cost per inmate housed in a Residential Reentry Center (RRC, also known as a half-way house) for the BOP is $27,003. The BOP also has custody over offenders on home confinement, for which it pays contractors a flat fee for each offender. As documented by the GAO, the reimbursement rate to contractors for each inmate in home confinement that the BOP pays is pegged to half the overall per diem rate of an RRC, or over $13,500 annually. Any policy change that transfers an inmate from a BOP facility to home confinement would, under current contracting arrangements, cost more than keeping the inmate in a BOP facility ($13,500 versus $10,363, respectively). The annual cost of supervision by probation officers, however, is about $3,347 per offender.We estimate that augmenting this traditional probation with electronic monitoring to verify home confinement would cost a total of $5,890 annually ..."
In good cost-benefit fashion, Samuels, La Vigne, and Taxy frame their report in terms of cost savings and reducing prison overcrowding. While these benefits are not to be scorned, they fall short of capturing the main policy problem here. When the government of land of the free is locking up eight times as many people as 30 years ago--and remember, this article is only about federal prisons, not state ones--imprisonment becomes a central life experience for a larger share of the population, for their extended families, and for their communities. This is much more than a budgetary issue. Changing the law so that many fewer nonviolent drug offenders are in federal prison, along the lines suggested here, would still mean that the federal prison population was, say six times as high as in 1980, instead of eight times as high.

For some earlier posts on U.S. rates of imprisonment, see "Too Much Imprisonment" (November 30, 2011) and "U.S. Imprisonment in International Context: What Alternatives?" (May 31, 2012).