The Surprising Racial Disparity in De-Incarceration

So there I was working on a piece for Zach Goldfarb, my editor at Wonkblog, on who is getting out of prison in this era of de-incarceration. I ran the race data using five Bureau of Justice Statistics reports and found a huge racial disparity that surprised almost everyone I talked to about it. This graph is a taster. Full account at the Washington Post.


If you think you can explain these trends, the comments are open for your ideas.

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.

21 thoughts on “The Surprising Racial Disparity in De-Incarceration”

  1. On first reading, I missed the word "on" in the first sentence and thought "Hey, it's great that Wonkblog is doing their bit to employ ex-cons."

  2. Okay, I'll take a crack at explaining this. Might miss by a mile, but …

    De-incarceration has now been going on for quite a few years, as your chart above shows. But, wow, look at that racial disparity: the number of white women behind bars has drifted upward — indisputably so, albeit not a lot (what? maybe 15%) — while imprisonment rates for black women have dropped by close to half. What could possibly account for it? How about this: for decades, black women were being imprisoned at far, far higher rates for far, far less serious crimes than white women ever were. (In other places, we've all read or heard comments about wild behavior by white teens resulting in a night in jail and a stern lecture from a cop and the parents, while the exact same offense by a black teen can well lead to years in the justice system). So as de-incarceration has proceeded, it's effect has been powerfully felt by black women who were previously being jailed for trivial, non-violent, and/or victimless offenses, while white women are hardly affected by the new trend since they weren't being jailed much for that stuff to begin with.

    (It would be interesting if incarceration rates for white men versus black men, and white versus black teens, shows a similar trend).

    What white women DID get incarcerated for in the past were more serious crimes. And they still are getting jailed for such crimes, probably at about the same rate. These are crimes serious enough that they are largely not affected by the de-incarceration trend.

    But why hasn't the rate for white women — even accepting my point above — just stayed flat instead of going up 20% over these years? No obvious answer comes to mind, but a two factors might play a role: (1) Years of complaints about disparities in prosecutions, convictions and sentences between blacks versus whites has actually had a modest impact nationally, so today white women are getting charged, convicted, and sent to prison somewhat more than in the past; (2) With the great overall reduction in crime that's occurred in recent years, police resources have been freed up, resulting in a somewhat higher solve rate of more serious crimes, meaning more white women are getting nailed for acts that in the past they might have gotten away with; this affects African-American women too, but that counter trend is swamped and hidden by their overall decline in incarceration rates. I doubt these two things explains it all, but might explain a small percentage of the increase among white women.

      1. Thanks for the clarification. I was just eye-balling the chart sans numbers, and so there the change looked much more modest for white females. 56% is a heap more than I thought, and kind of undermines a lot of what I was speculating about.

        1. You seem to be modifying your views in light of data, which disqualifies you from being a crime policy analyst… : )

  3. Thesis 1: Decline in the crack cocaine epidemic (which impacted primarily Black people) and rise in the prescription painkiller abuse epidemic (which impacts primarily White people).

    Thesis 2: There is a long term trend toward social and familial chaos which provides upward pressure on the crime rates of both Black and White people, but those trends hit Black communities earlier and thus are largely "played out" whereas they have only more recently hit White communities (thus explaining the rise in the White incarceration rate). But there is a simultaneous long term trend toward lead abatement, which provides downward pressure on the crime rates of both Black and White people, but a much bigger impact on Black communities because, being disproportionally urban, those communities were significantly more exposed to fuel emissions and industrial emissions containing lead (thus explaining the net downward Black incarceration rate).

  4. "Pfaff said that “law enforcement attitudes getting tougher in rural areas and softer in urban areas may be contributing to this change"

    That jives with my own experience working in a mid-sized city in which it's surprisingly difficult to go to prison (generally your first violent firearm felony gets you probation after you plea to a misdemeanor, your second gets your probation violated and might send you to prison, and your third will definitely send you to prison), surrounded by rural counties where people go to prison for stealing a car. In our city we have people released on their own recognizance who have 4 violent firearm felonies (all separate incidents) working their way through the court system, but in the neighboring rural county you have to make bail if you get picked up for shoplifting. It's like two separate countries.

      1. Speaking as somebody who grew up in one of those rural areas… They're Republicans.

        Which is to say, when somebody steals a car, or shoplifts, they don't go all root causes and understanding. They figure you knew in advance you weren't supposed to steal a car, and now it's time to convince you that it's not just morally wrong, it's a Really Bad Idea. Or, you didn't know in advance it was wrong to do that, in which case you're just a dangerous animal that needs to be caged. Either way, the legal system exists to protect everybody from you, and leaving you walking around won't do that.

        What I find facinating: Why didn't you ask why they're so lenient in urban areas? Why the assumption that just blowing off crimes is the normal thing to do?

        1. The city I work in is part of a larger county. The majority of the county is suburban and Republican. The majority of the judges are Republican, the DA is Republican, the county executive is Republican, etc.

        2. What a hilariously naive and childish essentialism from someone who claims to have studied anthropology.

          A few relevant points:

          1. These areas are hardly monolithically Republican. I grew up in rural northeast Ohio — in a village that was once home to the only area chapter of the KKK — and it was filled with people who had lived through the Depression and the New Deal and practically worshipped FDR and, to a lesser extent, JFK. The same is true of rural counties around the eastern and midwestern parts of the country. But if you'd like to defend a claim that the rural south, full of majority-black counties, is Republican, go nuts! (NB: In 2008, this same rural county voted for Barack Obama by less than 1%. In 2012, they voted for Romney by less than 1%.)

          2. To the extent that anything you say about their ideas on crime and punishment comport with reality (NB: not much), they don't arise from Republicanism, they arise from religiousity.

          3. In rural areas, the most common thing determining how severely one is punished for crimes, in contrast to the stream of nonsense you post above, is Who You Know. This I know from deep personal experience with both friends and relatives.

      2. Hello Keith,
        I can think of many reasons. #1, rural areas can afford to be tougher. With more crime in the city where I work, if we sent people to prison for things as minor as a joyride, it'd practically be a population transfer. And we couldn't afford to keep all those people in jail waiting for their trial.
        #2, norms of behavior. With a lower crime rate, things like an assault shooting really stand out. But in my city of about 200,000, we have between 200 and 300 shootings a year. Judges see people arrested for felony assaults all the time and I imagine they become inured to it. But go in front of a judge who hasn't seen a (non-domestic) felony assault in a year, and that judge will be OUTRAGED!
        #3, I don't think judges in our city identify with victims very much. They never live in the same neighborhood, our victims are frequently drug dealers or high-school dropouts, occasionally don't speak english, etc. I think that part is human nature.
        #4, our prosecutors are overloaded. They offer plea-downs to probation partially because they are simply unable to take everything to trial. And just like judges, prosecutors become accustomed to horrible behavior. "robbery/stabbing, OK, victim didn't suffer any permanent physical damage, I guess I can live with probation." I've also assisted DAs in rural areas, and the amount of time they can devote to each case is amazing to me. These are part-time DAs to begin with because they have so little work to do.
        #5, because of the high amount of violence in my city, very few violent crimes get real media attention beyond the initial blurb parroting the police press release. So there's nobody watching to see what happens with the plea, trial or sentencing.

        I'm sure there are other reasons as well.

          1. Interesting question, I think:

            Is the lower crime rate *enabling* the strict enforcement? Or are they mutually creating each other?

            What I mean is, you can imagine two alternate states, both stable: Low crime/high enforcement, and high crime/low enforcement. The rural areas have low crime rates, so they can afford strict enforcement, but arguably, strict enforcement maintains the low crime rates, too.

            While the urban areas have high crime rates, so they can't afford strict enforcement, but the lack of enforcement feeds back into the crime rate being high.

            Now, the real question is, is it possible to 'flip' urban areas into the low crime/high enforcement state? It's obviously better, the people aren't being preyed on.

          2. There is a literature on this under the term "enforcement swamping" which is part of how crime causes more crime.

          3. And guess who authored the first result, when I googled the term? Our host.

            Still leaves the question open: Could the urban areas be flipped? What would it take?

  5. Just to clarify: The statistics are, I believe, for the *stock* of prisoners, not for the *rate* of incarceration. So you should look at the disparity in sentence lengths as well.

    1. Hi Mike

      The figures are based on the imprisonment rate (pegged to the population of people of that race in the country) on the last day of each year.


  6. Your guess is as good as (or better than) mine.

    It's nice to see a post about some good news on this front. I find it endlessly frustrating how little press this type of good news gets. If you reported the opposite it would be featured all over the place yet the impact is just as consequential.

    Most people don't know about the large decreases in teen pregnancy or the decrease in the divorce rate. Here in California the black/white disparity in the child welfare system narrowed from 54:8 in 2000 to 24:5 in 2014 (both are in-care rates per 1,000 kids). And nobody talks about this. There's plenty of hand wringing about the remaining disparity (and appropriately so) but we should celebrate and try to learn from positive trends!!!

    1. Yes, and I have noticed this is particularly true of good news about Black people. There was a story out a few years ago that African-American women were the most educated group, and the reporter quoted an activist saying how sad it was because this meant they had student loan debt, something that has never been said in a million stories of whites being more educated than Blacks.

Comments are closed.