The Relatively Small, Atypical Federal Prison System

The federal prison system is a small and atypical part of the U.S. prison system

I praised Attorney General Holder’s proposed changes to mandatory minimum sentencing procedures in federal drug cases. That said, the way some media outlets touted them in the subsequent news cycle grossly overstated their impact by not understanding the nature of the federal prison system. If you want to comprehend the realities of correctional policy in the U.S., not just regarding Holder’s proposals but more broadly, it is essential to appreciate that the federal prison system is a relatively small and atypical part of the U.S. prison system.

It is I suppose natural to assume that what is “federal” is large and what is “state” is small, and in some public policy areas (e.g., health care) this heuristic holds. But in prison policy, it’s the other way round. At any given time between 85-90% of the U.S. prison population are in state facilities. To give a comparison point, over the years the State of California’s prison system has often held nearly as many people as does the entire federal prison system.

So when the BBC World Service reports that the Attorney General’s proposals will have a huge impact because “half of U.S. prisoners are serving time for drug offenses”, they are being hyperbolic. Drug offenders constitute a small and declining proportion of the U.S. prison population. Half of federal prison inmates are serving time for a drug conviction, but that’s only about 6% of the US prison population as a whole. The federal percentage is high because of the unusual mandate of the federal system, which picks up the farrago of cases that are not charged by states. As a result of violent criminals rarely being charged at the federal level, the percentage of all other types of crimes in the federal system goes up.

Granted, it is easy to get this nuance wrong both because it is counter-intuitive that the federal system is such a small part of the national prison system, and, because some people quote federal prison statistics in intentionally misleading ways. For example, some people with axes to grind invoke federal prison statistics to imply that half of all U.S. prisoners are serving time for drug offenses or that a sixth of all U.S. prisoners are serving time in private facilities, neither of which is true.

But even taking these possible reasons for error into account, I must say the following to the BBC World Service: When your producer calls me before a broadcast and makes this mistake about the federal prison system, and I take the time to patiently explain the difference between the federal and state systems, and then your program host repeats the false claim on the air anyway, a little bit of my wonky heart dies.

Author: Keith Humphreys

Keith Humphreys is the Esther Ting Memorial Professor of Psychiatry at Stanford University and an Honorary Professor of Psychiatry at Kings College London. His research, teaching and writing have focused on addictive disorders, self-help organizations (e.g., breast cancer support groups, Alcoholics Anonymous), evaluation research methods, and public policy related to health care, mental illness, veterans, drugs, crime and correctional systems. Professor Humphreys' over 300 scholarly articles, monographs and books have been cited over thirteen thousand times by scientific colleagues. He is a regular contributor to Washington Post and has also written for the New York Times, Wall Street Journal, Washington Monthly, San Francisco Chronicle, The Guardian (UK), The Telegraph (UK), Times Higher Education (UK), Crossbow (UK) and other media outlets.

27 thoughts on “The Relatively Small, Atypical Federal Prison System”

  1. But isn’t this also ignoring another important distinction, the one between prisons and jails?
    This just means that those serving post-conviction sentences of more than a year for drug offenses are declining. But how many are in jail before trial, found guilty, and released with time served (and a felony conviction), or sent back to jail for another six to nine months (with a felony conviction).
    I don’t actually know, just curious if that is something being included in the statistical analysis of different systems. I’d think that small time drug offenders typically go to jail, not prison.

  2. You could analyze city and county jails + state prisons + federal prisons all in a lump, and there are probably some policy questions for which that makes some sense. But being in jail is a light years’ difference from being in a state prison in terms of who you are locked up with, how long, impact on your life, distance from family, cost to the system, deterrent value etc. And we also know that there are policy interventions that can trade one off from the other…HOPE probation and 24/7 sobriety which Mark and I have written about a lot increase the likelihood of spending a night in jail and lessen the risk of spending a year in prison.

    And one language quibble. When you say “This just means”…you have to think of it from the point of view of the prisoner when you say “just”…I have the same reaction when people say that the prison population “only” dropped by 50,000 people…think of those people and this isn’t a “just” or “only” kind of problem, though I may be over sensitive about it because I have been inside so many prisons.

    1. How does the quality of life differ among county jails, state prisons and federal prisons? I can see someone preferring prison to jail. Especially if their crime would put them in a lower security unit.

      1. How does the quality of life differ among county jails, state prisons and federal prisons?

        You could write a book on this, probably a few people have. I could write a book based on my experiences in these three incredibly different settings, but I don’t have the time.

          1. Netflix users have rated it ahead of just about all of Netflix’s hundreds of other streaming video titles.

      2. General rule of thumb: Federal minimum security prison are the nearest things to open prisons in the American penal system; relatively save and pleasant. A lot of the amenities from the Seventy’s like tennis courts are gone and they are more like prisons now but, from what I’m told, still the best place to do time.

        State prisons are a mixed bag but generally rougher, more dangerous and more prison-like. Some places like Hawaii and Angola are genuine hellholes. Jails are a mixed bag. People tell me that they are generally rougher and more unpleasant than state prison. The only jail I was really familiar with and spent an real time in, Orleans Parish Prison, was a really terrible place.

        Also, in many states, it isn’t the quality of the jail or prison that’s important to the convicts but sometimes it’s the way you get credit for time served. In my time, prisoners in Louisiana didn’t start to accrue certain kinds of time (work time, for example) or got less of certain other kinds of credits (my memory is that convicts got 1:1 in OPP but 1:3 in a DOC facility)so they were anxious to get to prison for the improvements in living arrangements and the increased credits available.

    2. In the comments section of the link you provided about the declining portion of drug prisoners you stated, “there would be some people (50,000 maybe?) in prison on drug charges for long stretches who had never had a violent conviction, e.g., someone who ran a huge cocaine ring but never actually shot/beat/kidnapped anyone in the course of the business. That doesn’t bother me.”

      I was curious why the 50,000 you mention that are in prison that didn’t commit violent offenses doesn’t bother you, yet here you mention 50,000 is a big deal when you think of each prisoner as a person. Maybe I am misinterpreting your earlier comment, or is that you feel that non-violently running a cocaine ring is something that deserves prison.

      Also, as a reply to Agorabum’s initial statement, the stat about the portion of drug offenders in prison declining does not mean the population of drug prisoners is declining. That’s just a percentage of the whole. Both are growing is my understanding.

        1. That depends on the start date you wish to choose. The date range you specified in the post you referenced, was 20 years. Prison population:
          1991: 792,535
          2011: 1,504,150
          Increase: 711,615

          Contrast that with the recent 3-year decrease of about 17,000 prisoners. That increase is a better indication of the trend, consistent with the period you initially specified.

          That second number is hopeful of course, but it’s important to keep it all in perspective.

  3. “The federal percentage is high because of the unusual mandate of the federal system, which picks up the farrago of cases that are not charged by states.” That does not answers the question why drug trafficking and dealing should be part of that federal grab-bag. Treason has to be because it’s a crime against the State, not a state. Would an interstate car theft or paedophile ring be prosecuted for federal crimes?

    One part of the problem may be that the FBI is strictly a law-enforcement agency, which can only investigate federal crimes, which therefore have to be invented. The mandate of other countries’ security services like MI5 or Shin Bet is countering internal and external threats, not convicting criminals. MI5 reputedly tries to avoid prosecutions, which bring an unwelcome spotlight. Shin Bet’s motto translates as “The unseen shield.”

    1. The federal system has historically been very selective in the cases it accepted for prosecution, preferring to focus on high-end white collar crimes, organized crime or criminal activity of special relevance to some federal interest such as bank robbery. Consequently, the federal prison population has tended to be small and nonviolent. The gateway agency is actually the DOJ and the US attorneys because they, and not the FBI, DEA, etc decide who gets prosecuted in federal court. A significant number of arrests made by federal law enforcement agencies is sent to the state court systems for trial and incarceration.

      In recent years, there has been an significant influx—in relative but not absolute terms—of street level criminals who are prosecuted as part of some specific initiative, particularly the ATF and DEA effort to attack certain smaller, inner city drug gangs and the epidemic of drug related murders of the Eighty’s. There’s also a number of carjackers from that era and then a number of bank robbers, check forgers and so forth that are prosecuted in the federal system for a variety of reasons.

      I think the huge increase in drug related prisoners is the product of something I’ve previously mentioned, namely, the fact that federal law enforcement comes in contact with a lot of low-level people in organized crime or narcotics trafficking organizations and they apply pressure to those people in an attempted to get them to cooperate. If they don’t flip, the federal prosecutors really have no choice but to throw the book at them if these threats are to have any credibility. In which case, those people are stuck in federal prison for really, really long sentences with no realistic possibility of early release. And even if people cooperate, they will still serve their reduced sentences in a federal prison thereby increasing the number of prisoners serving drug sentences. But, again, with the exception of a handful of outliers, nearly everybody serving time in federal prison for a drug related crime is there because he or she was involved in trafficking.

  4. James: There are different models, MI-5/MI-6, GRU/KBG, I don’t that see that any one arrangement is the measure of the others. Hyper-centralized Britain surely can’t be the model for law enforcement in federalist systems in which law enforcement is a disagreegated activity.

    There are rational reasons for how drug trafficking cases wind up in the federal bin. If you have an organization which operates in Columbia and Jamaica, launders its money through Ghana and Switzerland, and supplies drugs to 8 U.S. states as well as several Canadan provinces, who else could go after them, even if you do not assume and even more compelling case in which the same organization has corrupted local law enforcement.

    There are also of course irrational reasons, for example someone in Congress wants to make a statement and the federal system is more within his/her control than is the home state, so they make something a federal crime.

    1. Congress seems intent on making just about everyone guilty of something with about 4,500 federal crimes on the books.

  5. And this overlooks one of the really weird thing in our Federal sentencing policies. If the judge sentences someone to a year, they will (eventually) serve 365 days. But someone sentenced to a a year and a day is eligible for good time. Assuming our inmate keeps his nose clean, he will be released substantially earlier than 365 days. Why? Because anyone sentenced to less than a year is ineligible for good time and must serve their full sentence.

    1. Dennis, what do you mean by “good time”. There is no parole in the federal system so I am not sure what you are referring to.

      1. There’s no parole, but if prisoner can keep her head down, stay out of trouble and not annoy the powers that be, there’s a formula for trimming time off her sentence.

        Ah, here it is: Good conduct time

        1. There has in the past been some controversy over how Good Conduct Time was to be calculated.The statute calls for, I think, 15%, butis unclear whether that refers to 15% of the actual sentence or 15% of the time spent. That is, if you are sentence to two years, say, can you earn 110 GCT days (15% of 730) or only 95 days (15% of 635)?

          In 2010, in Barber v. Thomas, the Supreme Court took the latter, less generous, view by a 6-3 vote with Kennedy, Stevens, and Ginsburg dissenting. One interesting point is that the BOP uses a highly complex and error-prone method of calculating GCT, when a much simpler method is available that yields the same result. One consequence is that prison officials make mistakes, and miscalculate GCT.

          (No, I haven’t been a federal prisoner, but I did once help one I knew and his lawyer work through all this and file an appeal similar to Barber’s, which went nowhere.)

          1. “…the BOP uses a highly complex and error-prone method of calculating GCT, …”

            That’s probably a feature, not a bug.

  6. I think you might be eliding (correct me if I’m wrong) the number of people who are incarcerated because their probation/parole was revoked due to failing a drug test. So they may very well be imprisoned for a non-drug offense, but the proximate cause of their imprisonment is a drug offense.

    1. There is no parole in the federal system, but it is certainly possible in states for parolees to be violated back to their original charge based on drug or alcohol use.

  7. ERIC HOLDER and Rick Perry have little in common…
    On one thing, however, the two men agree. On August 12th Mr Holder said: “Too many Americans go to too many prisons for far too long, and for no truly good law-enforcement reason.” He then unveiled reforms to reduce the number of people sent to America’s overcrowded federal prisons. In this, he was following the perfectly-coiffed Texan’s lead. Several years ago, Mr Perry enacted similar reforms in the Lone Star State, and they worked.

    An unlikely alliance of left and right: America is waking up to the cost of mass incarceration

Comments are closed.