Towards Mass De-Incarceration

New data shows continued contraction in the number and rate of Americans being imprisoned. Let’s break it down.

The federal system helpfully updates its inmate count every week, so I was able report on the system up to the present moment. Because the Congress has been so dysfunctional in recent years — even unable to pass criminal justice reform when every key player in the House and Senate supports it — the federal system had been lagging the state-level move toward cutting prison populations, which started around 2008. However, thanks mainly to changes in drug-related sentencing (with icing on the cake from Obama’s clemency initiative), the system is now declining in size at last. The data are summarized below. You can read my full analysis at Washington Post Wonkblog, and read a thoughtful critique of my analysis by Professor Douglas Berman here.

Reporting of state data has a significant lag, so 2016 year end data just came out. What happens in the states matters way, way, more than the federal system because that’s where 7/8 of the prisoners are held. I break down the newly released data in another Washington Post Wonkblog about the now nearly-decade long contraction in that population.

What jumps out the most is the African-American imprisonment rate is at a more than quarter-century low. For Black males, the size of the drop from the peak is equivalent to the entire Black male population age 16 and older of Washington, D.C. plus that of Dallas. There’s a long way to go but this is a big change.

p.s. In addition to comments on the issue, I welcome comments on my charts. Since my Wonkblogs generally only appear on line, I decided to start color coding the text in the chart label to help people tell which line is which. What do you think? Also, like Kevin Drum, I think dual Y-axis charts are boffo. What do you think? Last, I am not mad about the year being in the middle rather the bottom of my federal chart, if anyone knows how to do that I’d appreciate it.
p.p.s. I have updated the second chart based on suggestions. I was not able to implement the suggestion to have only some years on the bottom because when I selected any interval for labels I selected it left off the 2016 label making it seem the data didn’t go as far as it did.

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.

14 thoughts on “Towards Mass De-Incarceration”

  1. Keith, if you're using Excel (I'm using Office 2010, not 2016 which is filled with bloatware), right-click on the vertical axis to open a menu. At the bottom you'll find a place to select where you put the horizontal axis. In your case, setting at -15000 would clean up the figure. I don't know where it is in other versions of Excel. In addition, you might color the vertical axis labels the same color as the plot lines. And to avoid kinking my neck, please make the captions etc. horizontal!

    One further suggestion: in the lower graph include a grid line every four years and label only those years.

  2. Thanks Mike, I love the idea of making the two y-axis labels the same color as the lines — will do that in future.

    And thanks for the horizontal axis dropping advice. Kevin Drum emailed me a method in which I right click on it, enter labels and just select the word "low" and it looks nice.

  3. Read Tufte. He may not be always right but he is definitely the place to start. He regards pie charts as useless. I had a go here at following his advice. See also this for a horror story, but it's far too easy to find similar train wrecks.

    Tufte has also written a devastating hatchet job on PowerPoint, or rather its use to replace rather than serve as a modest adjunct to written reports.

  4. Read Tufte. He may not be always right but he is definitely the place to start. He regards pie charts as useless. I had a go here at following his advice. See also this for a horror story, but it's far too easy to find similar train wrecks.

    Tufte has also written a devastating hatchet job on PowerPoint, or rather its use to replace rather than serve as a modest adjunct to written reports. The ultimate PP trainwreck slide is of course this.

    1. I read all his books. My favorite is his greatest graphic of the 19th century, Napolean's invasion and retreat from Russia.

  5. What's your view on the use of really invasive and detailed probation, like what has been used in Hawaii, to replace incarceration? Wrist monitors, chipping, probably we can learn even more from the Chinese stranglehold on its citizens? Would this kind of a regimen, with swift, certain, and small punishments, enable offenders to live better lives and cost the rest of us less?

    1. I have helped spread intensive monitoring for substance-involved offenders (i.e., most offenders) around the country and also abroad. They don't involve a tracking chip nor do I see why one would need one, but they do require as often as twice a day in person reporting (e.g., to take a breathalyzer).

      New technology can make this even simpler — there is a cell phone with an inbuilt breathalyzer. Probation officer pings the person being monitored and they have 10 minutes to take a breath test into their phone, with the phone camera recording that it's them.

      I suspect physical confinement will always be with us because no mobile tech can stop people from being violent, and contrary to what many people believe, violent crime is the driver of US imprisonment.

      1. I would sort of assume you would want a chip so the phone (Parole-O-Phone) would know whether you were near it, and would tell your PO that you had left it behind? I am thinking also from the point of view, how can we protect Megan Kankas and still let Jesse Timmendequas have something vaguely like a normal life and hold a job. This phenomenon of sex offenders living under freeway bridges because there is no space from which they aren't excluded can't be desirable. And, either there is a record of where you were when the armed robbery took place, or there is a record that your phone was not near you when it took place. Obviously, this needs thought, and you have to get over the threshold of deciding that society has a right to be that invasive into the lives of parolees.

        1. I am not aware of any evidence that sexual offender registries improve public safety, so we could probably get rid of those (or curtail them dramatically). What does work for prevention is DNA registries

          The increased possibility from offender monitoring from being detected should deter some crimes, as you note. Certainly things like bank robberies where there is always footage and a clock to synch to offender GPS. But it's imperfect deterrence, but as we both know. All those domestic abusers who shoot their ex-girlfriend/ex-wife and then themselves will not be deterred by GPS. Neither will most people with poor impulse control who for example get into an argument with a random stranger that escalates into violence when they are on parole.

          1. Timmendequas is fairly permanently imprinted on the public memory (at least for old guys like me) but there keep being incidents:… ;… which build the idea that guys with a sexual need of a particular kind keep trying to fill it. You think there is some other way, besides shooting them after a first offense, to protect Megan Kankas? You think registries don't work? I don't really know what to do.

          2. DNA registries need to be watched. Law enforcement would like to expand them to minor offenders, possibly everybody. The UK national DNA database already has over 3 million named entries, 5% of the entire population (Wikipedia). Presumably the benefit in crime prevention is subject to steeply diminishing returns with extension, as the loss of civil liberties rises. Perhaps there should be time limits on storage, as with phone traffic data, or procedural controls on access to DNA data?

          3. Sure.

            The tradeoffs here are genuinely difficult. The databases you study are of everyone arrested for felonies, not convicted. What of those who weren't charged? Of charged and acquitted? These groups include a fair number of individuals who actually committed the crime but got off for one reason or another, as well as the innocent arrested and even charged by mistake. Different observers will disagree on the likely proportions. These non-felons will, in the USA, disproportionately be from minorities. Should their DNA data be kept? If so, for how long?

            DNA evidence has been particularly important in securing convictions for rape. I'd like to hear from our female commenters or cobloggers on the DNA database issue.

            We can probably narrow down the issue by stipulating that DNA evidence has to be kept if used at trial, and it can always be accessed by a later investigation. But that doesn't make a directly searchable database without a lot of processing.

Comments are closed.