Truth in Sentencing and Early Release: A Follow Up

Last week, I spoke on truth in sentencing at Marquette Law School as part of Mike Gousha’s “On the Issues” series.  My PowerPoint slides and a video of the event are available here.  Alan Borsuk summarized some of the key points in this blog post.

If you watch the video, you will see that time constraints caused me to skip over a couple of slides.  I’ll fill in those gaps here and then suggest where I would like to see Wisconsin go with early release.

First, I think one of the most interesting and puzzling aspects of our polling research is that substantial numbers of Wisconsin voters say they support both truth in sentencing (“TIS”) and release from prison when an offender can demonstrate that he is no longer a threat to public safety, even though these two policies are in tension with one another.  My Marquette colleague Darren Wheelock and I have been studying this “both-and” group to better understand what underlies their thinking.  Comprising more than one-quarter of our polling sample, I think of this group as the TIS swing voters — the people who support TIS, but not so strongly as to categorically rule out support for a well-designed, well-justified early release program.  

We don’t have have any definite conclusions about the swing voters yet, but Darren’s multivariate regression analysis produced some intriguing results.  Holding all else constant, the swing voters are more likely than the rest of the sample to have these demographic characteristics: religious, single, minority, low income, and low education.  The religiousness of the swing voters may provide further support for my suggestion in the video that early release reforms should not be framed merely as cost-saving measures, but should also engage with the moral dimension of criminal punishment.

What to make of the other distinguishing demographic characteristics?  It occurs to me that all of these characteristics are also ones that would be disproportionately associated with having personal experience with the criminal justice system, either from being arrested oneself or from having a close friend or relative caught in the system.  I would hypothesize that, for our swing voters, TIS is not so abstract a policy question as it is for many other TIS fans, but has a real human face that tends to moderate support for TIS.  If this is true, it might serve as another reminder for reformers of the importance of humanizing the problems of mass incarceration, rather than casting them solely in statistical or budgetary terms.

In addition to the regression analysis of the swing voters, we’ve also tried to figure out what attitudes correlate with support for TIS more generally.  Two of the strongest correlations (both statistically significant) are (1) a belief that prison officials are too lenient, and (2) support for uniform statewide sentencing policies, even at the cost of local discretion.  The first seems unsurprising, given historical concerns that parole was operated more to serve the needs of prison officials than broader public safety interests.  It is possible, then, that TIS supporters would be more open to early release programs that were administratively independent of the Department of Corrections or that did not depend on discretionary decisions by the Department.

The second attitude (support for sentencing uniformity) recalls another common criticism of the old discretionary parole systems that were replaced by TIS in Wisconsin and many other states: it was said that the lack of objective standards produced unwarranted disparities in the treatment of similar offenders.  Again, I would suggest that TIS supporters, even if strongly opposed to old-school parole, might be more open to other sorts of early release programs in which the scope of discretion is narrower, discretion is exercised using more transparent or objective criteria, and institutional checks and balances provide more protection from arbitrary outcomes.

So, what should Wisconsin do?  Let me suggest two reforms that would facilitate moving low-risk inmates out of prison and into less expensive forms of punishment, but that would still look very different than old-school parole and likely have more credibility with the public.

First, reinstitute good time, the program that permitted inmates to earn credits toward early release through good behavior.  Many other states have joined Wisconsin in abolishing discretionary parole, but very few of these also abolished good time, which is generally regarded as a helpful tool to incentivize good behavior in prison.  In this article, I provide a more extended defense of good time.

How much time off should inmates be permitted to earn?  While most states have retained good time, there is tremendous variation in the generosity of the programs.  Discounts in the range of 25% to 40% seem most common.  European equivalents of good time typically provide a one-third “remission” for prison terms.  The U.S. federal system — no one’s idea of soft on crime — offers a 15% discount.

It seems to me that a 25% discount could, over time, make a real dent in the state’s imprisonment rate without coming close to rendering the sentences pronounced in court meaningless.  By contrast, the old parole system might give inmates a 75% discount (release at the one-quarter mark in the sentence).

If there is concern that good time might result in the premature release of inmates who, while well-behaved in prison, nonetheless seem at high risk of reoffending, then there might be a provision in the law permitting the Department of Corrections or a prosecutor to request a hearing before release.  If the state proves that the inmate will probably reoffend, then some or all of the inmate’s good-time credits could be canceled.  The decision should be based, as much as possible, on objective criteria and validated risk-assessment tools.

Second, long sentences should be regularly revisited after the ten-year mark.  Any way you cut it, ten years in prison is a long time.  Given the very  high costs — both fiscal and human — of long prison sentences, we would do well to review them on a regular basis to ensure that continued imprisonment actually serves some legitimate public purpose.

Sentence review should be conducted in the first instance by an independent, expert commission, and should make use of the best actuarial risk-assessment techniques available.  If a long-term inmate is evaluated as low-risk, then a hearing should be conducted, perhaps in front of the original sentencing judge, to determine whether there is nonetheless a good reason to continue to hold the inmate.

In the most serious types of cases, continued imprisonment might be justified beyond ten years on the ground that early release would unduly depreciate the severity of the offense.  However, such a determination should be limited to homicide and other cases of extreme violence.  Indeed, in cases of homicide, it might be appropriate to defer sentence review until the fifteen- or twenty-year mark.

An interesting question is whether release of a low-risk, long-term inmate should be denied if the inmate does not accept responsibility for his offense.  As indicated in the video, our survey research indicates that the vast majority of Wisconsin voters believe that it is “very important” for acceptance to be taken into account in early release decisions.  An acceptance component would help to give sentence review the sort of moral grounding that I think should be part of early release programs (for reasons I discuss in the video).  I’m inclined to favor including such a component, but I am troubled by the level of discretion and subjectivity that might thereby be introduced into sentence review.  Ideally, there should be some clear criteria to guide any assessment of acceptance, much as the federal sentencing guidelines set forth various criteria for assessing acceptance at the original sentencing.

In any event, ten-year sentences make up only about 10% of prison sentences in Wisconsin, but account for about 35% of prisoners, so a rigorous focus on the propriety of these sentences could deliver a real imprisonment-reduction benefit.