Sunday, February 22, 2009

Debunking myths about incarceration, and promoting some

Doc Berman points us to an article in Slate by Prof. John Pfaff of Fordham University law school identifying what he says are "Five Myths About Prison Growth Dispelled." The piece makes for an interesting starting point for discussion, but also promotes some mythology of its own.

The first "myth" Pfaff describes is entirely on point, disputing the contention that "Long sentences drive prison growth" Says Pfaff, "Our data on time served is imperfect at best, but it appears that the time served by the median prisoner is about two years, sometimes much less." That more or less jibes with Texas' experience (see pp. 36-37 of TDCJ's annual statistical report for data on time served). Pfaff writes:
So what is actually driving prison population growth? Admissions. Far more offenders who in the past would have received nonprison sentences are being locked up for short stints, driving up the overall population. Stop admitting as many people, and the prison population would shrink rapidly. Cutting back on long sentences is far less likely to have the same meaningful effect.
This is an excellent observation - the enormous volume of prisoners we see today comes from applying incarceration to more petty offenses than in the past, not from punishing people for longer periods. In fact, large numbers of admissions for penny ante crime tends to push more serious offenders out of the system.

The second "myth" Pfaff describes, though, doesn't seem so mythological here in Texas: That "Low-level drug offenders drive prison population growth." Pfaff claims that "most of the drug offenders are in prison for distribution, not possession," but that's not true here at all.

Among Texas' state jail felons convicted of drug offenses, 87.3% are in for possession of less than a gram of a controlled substance according to TDCJ, while 58.6% of more serious drug cases were possession-only offenses, not for "distribution." Perhaps it's true that some or even many of those charged with "possession" were really full-blown drug dealers, but that assumes facts not in evidence; it's not the offense for which most of them were convicted.

Pfaff's third "myth" is that "Technical parole and probation violations drive prison population growth," but once again, for Texas that's just not accurate (I can't speak for elsewhere). According to the Council of State Governments, "Between 1997 and 2006, the number of probation revocations to prison [in Texas] increased 18 percent, despite a three percent decline in the total number of persons under community supervision."

Indeed, Texas' experience on this score controverts both Pfaff's second and third "myths," since drug offenders make up the largest portion of those who enter prison because their probation is revoked. Meanwhile, reduced probation revocations are credited with Texas' successful efforts to stave off a projected need to build new prisons in recent years.

If reducing revocation rates empirically reduced Texas' rate of prison population growth, it stands to reason higher revocation rates would increase it. It's possible that Pfaff's assertion would be more accurate if he'd limited the observation to parolees, but Texas' experience with probation precisely contradicts his assertion.

Pfaff's attempts to dispel the fourth "myth" he describes are the most ... well ... mythological. He contends that the United States has not "newly diverged from the rest of the world on punishment," declaring that "if we look back historically at the lockup rate for mental hospitals as well as prisons, we have only just now returned to the combined rates for both kinds of incarceration in the 1950s."

Certainly it's true that a large number of Texas prisoners are mentally ill, but Pfaff's historical comparison ignores the history of who populated mental institutions in the first half of the 20th century compared with who's in prison today. Yes, many more people were locked up in mental institutions in 1950 than now, but US mental institutions back then were mostly filled with white women, while in today's prisons it's black men who are demographically overrepresented.

The dissolution of Jim Crow had a lot more to do with high incarceration rates than reduced reliance on mental hospitals, but apparently that's not the history they're teaching at Fordham.

Pfaff's fifth "myth" is a red herring, an argument he proposes and debunks that nobody really makes out in the world: "The incarceration boom has had no effect on crime levels." Except nobody claims that.

Having set up a straw man, Pfaff debunks this "myth" by declaring that "The best numbers available, controlling for a host of challenging statistical problems, suggest that the growth in prison populations contributed to up to 30 percent of the crime drop during the 1990s." Of course, "up to 30%" means most estimates are actually lower.

Even so, few serious observers I'm aware of claim incarceration has "no effect" on crime. Instead, the more common argument is that most crime reduction witnessed in the '90s and since the turn of the century was not due to high rates of imprisonment - a fact with which Prof. Pfaff entirely agrees.

He argues that prisons are "not the most efficient tool we have" to reduce crime. "A dollar spent on police, for example, is 20 percent more effective than a dollar spent on prisons," he says. (And that's if you accept the highest estimates for prisons' crimefighting effectiveness.) So Pfaff is claiming to debunk a "myth" promoted by critics of high incarceration rates, while actually adopting those critics' position and claiming they said something they didn't.

In addition, Pfaff ignores research showing that, after a certain point, incarceration can actually become a counterproductive factor causing increased crime. According to a 2007 meta-study by the Vera Institute:
Raymond Liedka, Anne Piehl, and Bert Useem have confirmed, moreover, that increases in prison populations in states with already large prison populations have less impact on crime than increases in states with smaller prison populations. States experience “accelerating declining marginal returns, that is, a percent reduction in crime that gets ever smaller with ever larger prison populations,” they argue. Thus, increases in incarceration rates are associated with lower crime rates at low levels of imprisonment, but the size of that association shrinks as incarceration rates get bigger. Eventually, they say, there is an “inflection point” where increases in incarceration rates are associated with higher crime rates. This inflection point occurs when a state’s incarceration rate reaches some point between 325 and 492 inmates per 100,000 people. In other words, states with incarceration rates above this range can expect to experience higher crime rates with future increases in incarceration rates.
In other words, in a state that incarcerates relatively few people, there are higher safety gains from relative increases in incarceration. But in a state like Texas where more than 1 in 100 are already in prison 4.6% of adults are under control of the justice system, increased incarceration produces much less public safety bang for the buck and may even increase crime by imposing untenable collateral consequences on more and more low-level offenders.

Pfaff's article is a useful starting point for discussing soaring incarceration rates and how to reduce them - particularly his observation that admissions, not sentence length, drive expanding incarceration rates. But I'm afraid he's promoted nearly as many "myths" in this piece as he's dispelled - at least as far as the Texas is concerned.

9 comments:

  1. One thing I would be interested in as a historical matter, how likely were highly racist police departments to simply ignore black on black crime and only attempt enforcement if crime leaked out to strike the white community? If such an attitude was previlent then trying to make comparisons between modern and past practices would be made all the more difficult by missing data that there is likely no way to generate even rough approximations for. Under-reporting seems like it would be a serious issue here, is a family likely to report a crime if they know no action will be taken?

    Another possibility is that people were fearful enough of racist violence that they were far more careful than today.

    ReplyDelete
  2. Thanks for demythologizing some of the myths asserted by Prof. Pfaff about prison growth. Loosey-goosey scholarship has no place in an area as high-stakes as prison policy.

    ReplyDelete
  3. Pfaff used data from 11 states (not including Texas). If you look at historical arrest rate data (1960-2005) for those states there are significant differences between states. I doubt that all his conclusions are valid for states not included in his data set.

    It appears that his conclusions the first two are true in Iowa prisons but are irrelevant for jails. The third conclusion is not true for Iowa jails or prisons.

    It is true that if a mentally ill person requires 24 x 7 supervision it will be hard to find a residential bed other than a jail or prison. But the number of MH hospital inpatients in 1955 is several times the number of mentally ill prisoners now in Iowa prisons. With respect to the fourth conclusion he may be somewhere in the ballpark.

    The frequency of serious crime types depends on local circumstances so one should expect significant state to state variations in the dependence of prison admissions on crime rates. As a consequence I don't think his last conclusion is valid for any state (including the states in his data set).

    ReplyDelete
  4. We can conduct all kinds of scholarly research and produce myriad data to support or debunk just about anything we please. It is next to impossible to test the validity of such studies in a truly scientific manner.
    I believe crime, causes and solutions will be debated ad infinitim.
    My simple mind prefers the K.I.S.S. method. I believe there are universal truths that can be applied to human behavior.. We start learning about consequences at an early age. That is our way of learning self preservation and that will never change. Beyond the primordial, we develop a moral compass that dicates how we live our lives. Who we become has alot to do with nature and a little to do with nurture. Some of us have a stronger drive and character than others and some of us need the law of the land to prvent us from harmng others. No one can explain the resiliance factor, it just is. We waste a great deal of time looking for causes and explanations. Some of that is beneficial but in the end all we can do is offer hope and a chance for change and at the same time keep public safety in mind.

    ReplyDelete
  5. On the drug thing. . . The report doesn't differentiate, but it seems certain that most of the possession cases were "possession-with-intent-to-deliver" situations. TDCJ would just lump them together as possessions (since that's what they are) but it is a higher grade of felony than your garden variety personal-use kind of drug possessor. I'm speaking anecdotally, but small time possessors almost never go to prison -- well, I mean first time possessors.

    ReplyDelete
  6. That's just not accurate, 1:11, at least not universally. In Harris County, e.g., they charge paraphernalia as a SJF because of trace amount of drugs on a crack pipe. Thirty percent of Harris County's felony docket is filled with less than a gram drug cases.

    ReplyDelete
  7. I know I'm speaking from Dallas Co., but it may as well be Harris Co; it's the same ol', same ol'.
    This stuff just fries my pop tarts everytime I read about all this over a small amount of some 'possession'.
    It reminds me every single time of the dinner date I had with a friend, along with his friend, Billy, the public defender at Frank Crowley Courthouse. This had to have been late 2003 to early 2004 - I could check my calendar to find a more specific time point. Anyhow, poor Billy was feeling a little down because his nephew had been delivered to him by DPD for having a bag of weed on him and trying to board a flight at DFW.
    I mean, all this cost this kid was a missed flight and put Billy in a bad mood. Of course I imagine at that point, for sure, Billy owed the DA's office...
    I am just so tired of the extreme injustice in this area. All this talk is over people who don't have the money or the proper connections. There should be some kind of preface to these articles, you know ?

    ReplyDelete
  8. "they charge paraphernalia as a SJF because of trace amount of drugs on a crack pipe."

    Well,that's because it's a felony. We weren't talking about overcharging. The issue is whether those offenders routinely are sent to prison (particularly, as is often reported in progressive circles, if they're first-time offenders). Felony or not, they aren't sentenced to prison (or state jail) in droves.

    ReplyDelete
  9. They're the only county in the state that does it, 11:44, and Harris accounts for a disproportionate share of SJF commitments to TDCJ.

    Also, where does this come from?

    "(particularly, as is often reported in progressive circles, if they're first-time offenders)"

    Who, or what, exactly, are you talking about?

    I worked on passing the legislation in 2003 that gave mandatory probation for state jail felony drug possession on the first offense, so you've never heard that misinformation here. I'm happy to defend things I've written or said, but if you just make up stereotypes and strawmen, I don't feel particularly obliged to debate them.

    But a dismissive tone still doesn't change the fact that 40% of those in state jails are less than a gram drug offenders, or that drug offenders dominate probation revocations.

    BTW, I should point out for those not familiar with the lingo that a "state jail felony" is the equivalent of a 4th degree felony. There was some confusion by an emailer that I was talking about county jails while Pfaff was talking about prisons, but these are people with felony convictions serving in secure TDCJ facilities.

    ReplyDelete