A commentary on CCJ’s recommendations for sentencing reform

In Spring 2022, the Council on Criminal Justice (CCJ) launched a Task Force on Long Sentences with the aim of assessing our nation’s use of long prison terms (i.e., 10+ years) and the impact on public safety and justice. Most recently, the Task Force released a report detailing their 14 recommendations about how to reduce mass incarceration without negatively impacting public safety.

The recommendations are questionable, though, as they seem overly optimistic about the state of the research. For example, they propose various alternatives to incarceration that are supposedly effective, though they make the research sound much more conclusive than it actually is. Perhaps they ought to be reminded of the old adage: “If it seems too good to be true, it probably is.” In this post, I will highlight some of the more controversial “recommendations” and provide some points for consideration. In a future post, I will critically assess each recommendation in more detail and provide additional points for consideration.

Most importantly, I felt that many of CCJ’s statements regarding the research were too declarative and misinterpret the current state of the research. For example, anytime I see the word “proven”, I am somewhat skeptical (technically, we cannot say anything with 100% certainty; we can only say whether it is more or less likely). That being said, it concerns me that the CCJ used such strong language in their report. In regards to what the evidence actually shows, these claims are not always upheld. Thus, it concerns me that the CCJ widely circulates policy recommendations based on this level of evidence. The American criminal justice system has a notorious history of implementing policies rapidly without considering the possible detrimental effects to public safety.

Out of the 14 recommendations, there are three that I found particularly problematic that I will discuss here (i.e., recommendations 1, 8, and 11).

Recommendation 1

The first recommendation is for states and the federal government to reduce the number of people serving long sentences, and use the cost savings for programs that have been proven effective at reducing violent crime. According to CCJ, this is because “the final months–and even years–of long prison terms produce diminishing public safety benefits” because people are incarcerated “beyond the point at which they are likely to continue committing crimes.” They also argue that there are myriad alternatives (e.g., community-based violence reduction programs, law enforcement interventions) that are “effective” at reducing violence.

Regarding length of incarceration, as the CCJ even admits to in their prior research, there are many violent repeat offenders who benefit from incarceration. Many of the studies on incarceration length have mixed results, and there are several studies showing stronger deterrent effects from lengthier periods of incarceration, particularly for violent and repeat offenders. For more information on the research regarding length of incarceraton and recidivism, check out our recently-published research review (free version available here). In addition, official data collated by the Sentencing Commission suggests that lengthier prison terms are benefical for decreasing recidivism among federal offenders.

Further, the claim that community-based interventions are a viable and effective way of reducing violence is unfounded. For some reason, community-based alternatives to incarceration are often touted as effective in reducing recidivism, despite the fact that the research on these program remain limited. For a better understanding of some of that research, see my previous paper about violence interruption programs, as well as my review of California’s similarly-themed Prop 47 programs.

CCJ’s implementation plan for community-based alternatives is also fairly lacking. Unfortunately, they give no detail regarding which strategies should be funded. This is disappointing, especially considering that the research on these topics is very mixed, and we know that not all people will respond to these types of programs. If the CCJ are going to be making these types of recommendations, there should be more detailed guidance on which strategies should be implemented and for whom.

Recommendation 8

The eighth recommendation argues for increased use of individualized risk assessments when applying sentencing enhancements based on criminal history. To support this recommendation, CCJ claims that people’s criminal histories are not always reflective of the individual’s current or more recent behavior. While they agreed that criminal history is an important consideraton for sentencing, they argued that they can cause some people to be incarcerated for longer periods of time than are justified by crime prevention concerns. They also argued that people naturally “age out” of crime, which makes earlier criminal histories less relevant to sentencing decisions.

When it comes to the relationship between age and crime, they make one particularly bold statement: “research is definitive that criminal offending is a young person’s game.” For starters, anytime I see the word “definitive,” I am skeptical. Secondly, this statement is simply not true. Regarding age-related desistance from crime, it is important to note that current research on predictors of desistance tends to be correlational and relatively weak. First, there is little consensus on how to define or operationalize desistance, which makes it difficult to study. . Second, there is not a lot of strong evidence between many of the suggested links between supposed predictors of desistance (e.g., employment, sobriety) and criminal behavior itself.

Third, there are always going to be chronic offenders who commit crime at a high rate even as they age, though, they can be harder to predict. This is because there’s not a lot of work directly studying how imprisonment shapes the desistance process for these high-rate chronic offenders. But this is an important oversight, because chronic offenders are the most relevant when it comes to desistance from crime — both for society (because they commit a large number of crimes) and for themselves (because their criminal activity often dovetails with other antisocial behaviors that impede their well-being).

Regarding the use of risk assessment instruments, it is still unclear whether risk assessment tools are sufficient in predicting dangerousness and recidivism likelihood. Further, it is hard to know how accurate these instruments are for individual subgroups, such as violent offenders or chronic offenders. In the CCJ’s proposed implementation plan, they state that determinations of risk should refer to assessments from “actuarial instruments that are transparent, open to public scrutiny, and have been validated for predictive accuracy,” and “where such a transparent and validated assessment instrument is not available for use, authorities should develop one.” They unfortunately give no detail on the types of risk assessments that they would recommend or how to even how to know if a risk assessment is validated. The statement about “authorities should develop one” also seems very ill-informed. Developing a risk assessment tool is challenging and time-consuming, and it seems very misleading to frame this as if it is a simple or easy task.

Recommendation 11

Their eleventh recommendation is for officials to consider expanding sentence credit opportunities for people serving long sentences. Supposedly, this to incentivize good behavior and encourage participation evidence-based rehabilitative programming (e.g., educational, vocational, mental health and substance use disorder treatment) to promote rehabilitation. This sounds great in theory, except for the fact that not much is known about the effectiveness of rehabilitative prison programs, despite the fact that the CCJ claims such programming is “proven effective.” The CCJ even admits that there is a need for more research on sentence credit programs, but they are still apparently willing to create a policy recommendation based on a possible indication that such programs could reduce recidivism. This  seems too big of a jump for me.

While I agree that there are likely some rehabilitative programs that have positive impacts for some people, the CCJ seems to oversell the state of the research. In particular, it can be very difficult to know what types of correctional programming are effective in reducing recidivism, and for whom. When it comes to knowing “what works” in terms of correctional programming, there are many types of correctional programs with a variety of components, which makes it even more difficult to study their effectiveness. It’s also important to note that the type of people participating (and completing) these types of programs are likely different from those who do not, so it is hard to know whether successes are related to the programs themselves or if there is something about those individuals (e.g. greater motivation to change) that makes them more successful. For example, individuals may participate in a program for external reasons (e.g., for good time credits), but those who actually complete it are usually intrinsically motivated (e.g., they enjoy it) and realize that they find value in the programming.

Per the research, some interventions may improve certain post-release outcomes, such as obtaining employment (albeit short-term) or reducing substance use, though they tend to have little impact on recidivism in the aggregate. Certain factors might increase their efficacy, but unfortunately, these are not detailed in the CCJ report. For example, there is a growing body of evidence that job-related programs could be helpful in reducing recidivism, but this seems contingent on the individual obtaining high-quality employment (not just any employment) after release. Similarly, educational opportunities may reduce recidivism for certain individuals, such as people with existing education deficits, but this effect is not noted much in the general offending population.

To summarize, correctional programs look vastly different from one another, and there is a lack of specificity reagrding the different  components of each program. This makes it even more difficult to know which components are helpful in reducing offending, and whether these effects apply to all offenders or just a subset of offenders. Taking this into consideration, it is premature for CCJ to simply point to “evidence-based correctional programs” without giving detail on the types of programs to implement and for whom.


Unfortunately, it is a common theme among advocacy groups (such as CCJ) to oversell certain ideas and claim that they are “evidence-based” without acknowledging the limitations of that research. In this post, I reviewed CCJ’s recent article about reducing prison sentences and pointed out a few examples of where this seems apparent. Overall, I felt that many of their statements were too declarative in nature and incongruent with the actual state of the research. For example, strong words like “proven” and “definitive” came up multiple times, which immediately draws my skepticisim regardless of the topic at hand.

Additionally, the claims are oversimplified and missing so much necessary context. There is a lot of room for nuance regarding the tradeoffs between public safety benefits and potential harmful effects of incarceration. We also know that offenders are heterogenous and will respond to incarceration in different ways. Yet, CCJ’s report is crafted in a way that leans toward a certain narrative — that longer sentences are bad — while omitting a lot of context. Sadly, many will not notice the missing context and will take these findings at face value. Despite being misleading, many people will not question the findings as they appear “convincing enough.” Unfortunately, this is a concerning trend with different advocacy groups. In fact, we saw a similar theme in a recent report by the Vera Institute, which also focuses on sentencing reform efforts as a way to reduce reliance on incarceration. For more information on that report, see my previous post.

The American criminal justice system has a notorious history of implementing policies rapidly without considering the full breadth of the research, and they often cite reports like this one instead of looking toward published, high-quality studies. Considering that many people do not have the time to dive into the past research on this topic, it is very likely that non-researchers will take these findings at face value. CCJ knows this, and uses it to their advantage.

Stay tuned for part two of this post, where I will perform a deeper assessment of the report in its entirety and comment on each of the 14 recommendations.