Criminal justice reform is having a moment. The protests after the death of George Floyd, in police custody in Minneapolis, focused mostly on aggressive policing but also highlighted how policing tactics contribute to mass incarceration. The First Step Act, signed into law in December 2018, revised mandatory minimum sentences, provided avenues for early release for some drug offenders, and required the federal Bureau of Prisons to undertake reforms designed to decrease recidivism. In states—which incarcerate the vast majority of America’s 2.3 million prisoners—sentencing and prison reform efforts have moved forward, albeit in fits and starts. In general, there is a growing consensus that America incarcerates too many people for too long, which has resulted in growing efforts to reform sentencing and reduce prison terms.

Many legislators looking to reduce the ranks of the incarcerated turn their attention to community supervision, where offenders serve all or part of their sentence in the community, subject to monitoring and other conditions. Though the number of supervisees has fallen in recent years, this population still dwarfs the number of persons in prisons and jails. Yet community supervision has attracted relatively little attention from criminal justice reformers and policy makers, compared to sentencing and policing reforms.

As former federal prosecutors, we believe that the systems of community supervision in the United States need reform just as much as other aspects of the criminal justice system. In fact, considering the massive number of people involved in these systems, the need for reform may be even greater. The goal of community supervision is to reduce recidivism and reintegrate those who have been convicted back into society, helping them to break cycles of addiction, access employment, and develop pro-social habits and mindsets. Yet community supervision as currently practiced in the United States fails quite spectacularly at these goals. Originally conceived as a flexible and low-cost alternative to incarceration, community supervision has become an extension of the carceral state, with far too many low-risk offenders subjected to overly harsh conditions of supervision for far longer than is necessary. As a result, too many supervisees end up back in jails and prisons for violations of supervision that present little or no risk to public safety. Meanwhile, supervision officers have less time and capacity to deal with higher-risk supervisees, who need more intensive monitoring and attention. Recidivism among supervisees is high, and the costs of these failures continues to climb, both directly (in the costs of re-incarceration) and indirectly (in terms of wasted human potential). The results of this broken system—re-incarceration and increased crime rates—end up back on the plates of prosecutors, who nonetheless have shown little appetite for reforming supervision. We think they should be more open to smart, evidence-based reforms that can reduce recidivism and make our supervision system work more efficiently and effectively.

I. The Systems

“Community supervision” is a generic term for probation and parole. But these are different systems, and it is critical to note their differences.

Probation is a form of community supervision that offenders often serve in lieu of or in addition to a prison or jail sentence, often for low-level offenses. At the court’s discretion, and on the basis of a showing that incarceration is not warranted, probationers are permitted to serve out some or all of their sentences in the community. They are supervised and monitored by probation officers, who ensure their compliance with conditions of supervision.

Most states impose “standard” conditions on probationers that apply across the board—usually including attending school regularly or maintaining employment, avoiding criminal activity or association with felons, testing for drugs and alcohol, checking in periodically with the probation office, and remaining in the state or district. Other “special” conditions, which are imposed by the court in particular cases, generally relate to the offense and are meant to preserve public safety. Common special conditions include drug, alcohol, or psychological treatment; no-contact orders with victims; restrictions on internet or electronic media use; prohibitions on certain forms of work or employment; home arrest; or electronic monitoring. If a person violates a standard or special condition, a court may revoke that person’s probation and impose a community sanction, jail time, or even a term of incarceration.

Parole is granted after a defendant has already served part of their sentence and the delegated body—usually a parole board—agrees that rehabilitative and public safety factors support early release into the community. Like those on probation, individuals on parole will likely meet regularly with supervision officers and be subject to standard and special conditions. Parole usually lasts for the remainder of an offender’s unserved sentence. In recent decades, the federal prison system and some states—notably Florida—have abolished parole in favor of early release after a fixed percentage of a sentence (usually 85 percent) has been served with no infractions.

II. The Failures of Community Supervision

The Supreme Court has said the purpose of community supervision is to “help individuals reintegrate into society as constructive individuals as soon as they are able.”[1] But our current system, which supervises too many low-risk people too strictly and for too long, is at cross-purposes with that goal. Community supervision has grown dramatically in recent decades as part of a deliberate state strategy to reduce prison crowding and save money. But there is little evidence that supervision as currently practiced actually works to improve criminal justice outcomes by reducing recidivism and promoting reintegration into society of those who have been convicted of crimes. And increasingly, the failures of the supervision system are putting financial pressure on states as supervisees cycle back into prison for “technical” violations that have little impact on public safety.

Each year, millions of Americans are put on community supervision as probationers or parolees. Currently, about 4.5 million adults—2 percent of the adult population—are under supervision.[2] But the failure rate of this system is astounding. About a third of supervisees end up back in prison at some point.[3] Nationwide, about 45 percent of prison admissions are the result of supervision failures; in 20 states, more than half are.[4] These supervision failures are costing taxpayers more than $9 billion annually.[5]

What’s more, well more than half of these prison admissions from supervision are for “technical” violations of supervision conditions—conduct that does not constitute a new criminal offense—such as failing to report in, adhere to a curfew, or remain in the jurisdiction.[6] The cost of sending supervisees back to prison for missing an appointment, staying out past a curfew, failing a drug test, or committing another technical violation adds up to more than $6.5 billion annually.[7] That’s about 5 percent of the amount that all states spend in total on law enforcement—all salaries, benefits, equipment, and capital outlays combined.[8]

A leading meta-analysis found that, as currently practiced, supervision systems have essentially no impact on reducing recidivism.[9] And we have no data at all on whether supervision helps people reintegrate into society, such as by helping them to access substance abuse treatment or find meaningful long-term employment. As currently practiced, supervision is largely a box-checking exercise where one box—did the supervisee go back to prison?—is left unchecked one third of the time or more.

Why is community supervision as currently practiced so often a failure? The surprising answer is that supervision is at once both over- and under-inclusive, catching too many low-risk people in its net for too long under too-harsh conditions, while failing to engage in the kind of intensive supervision that higher-risk supervisees need to become successful. In almost all jurisdictions, the current supervision model features long supervision sentences, long lists of standard conditions, and frequent testing and reporting for almost all supervisees. Low-risk supervisees are over-monitored and subject to overly harsh conditions that impede their reintegration back into society. At the same time, the massive number of cases means that parole and probation officers are spread too thin to be effective in supervising higher-risk individuals. Reforming the system to divert low-level offenders and move people out from under supervision more quickly would free time for officials to focus on high-risk supervisees and repeat violators, which would improve public safety.

The inefficiencies in our current supervision systems come in three flavors. First, the current model of supervision often treats supervisees as though they are all alike or divides them into overly broad categories. Many jurisdictions have imposed a blanket “tough on crime” approach, in the mistaken belief that stringent supervision conditions and intensive supervision for compliance will yield better outcomes. In fact, rather than preventing criminal behavior, research strongly suggests that overly harsh supervision can actually prompt it by limiting a person’s ability to find housing, obtain employment, and rebuild community connections.[10] The office that administers federal probation programs notes in its program manual that “excessive correctional intervention for low-risk defendants may increase the probability of recidivism by disrupting prosocial activities and exposing defendants to anti-social associates.”[11] Meanwhile, reducing the number of probation officer contacts for low-risk offenders has been shown to have no effect at all on recidivism, re-arrest rates, or public safety.[12]

By contrast, we know that repeat offenders and supervisees with lower educational levels, lower levels of familial support, and fewer social ties to the community will fail supervision at much higher rates than people with jobs, education, and deep ties to their community. Intensive supervision of these high-risk persons can reduce recidivism by up to a third.[13] Calibrating supervision to address a supervisee’s “criminogenic needs,” or the social and lifestyle factors that need to be addressed to reduce the chance of recidivism (such as housing, employment, anti-social attitudes, or addiction or mental health treatment), is a critical task, one that often goes overlooked in the rush to impose blanket supervision conditions.

Second, supervision terms are far too long across the board, which strains resources and unnecessarily sets supervisees up for failure. Probation and parole terms are often five years or longer. In Georgia (which has the highest supervision rate in the country, at 1 in every 18 adults), nearly three-quarters of all felony probationers have sentences that are longer than five years, and 37 percent have sentences that exceed 10 years.[14] Yet the vast majority of supervision failures occur during the supervisee’s first or second year, and there is no evidence that extending supervision terms much beyond that period reduces recidivism.[15] In fact, after two years, re-arrest rates plummet.[16] Continued supervision after this period not only has less potential to depress criminality, it deprives people of their full liberty unnecessarily while straining corrections resources—all for no benefit.

Third, the explosive growth of supervision in the United States means that probation and parole officers simply cannot do their jobs effectively. Caseloads regularly reach 100, 200, or even more persons per officer.[17] The American Probation and Parole Association recommends that caseloads per officer not exceed about 20 high-risk, 50 moderate-risk, or 200 low-risk supervisees.[18] But in Georgia, parole and probation officers who manage low-risk offenders have an average caseload of 290 people, while officers monitoring a mix of standard and high-risk cases typically supervise an average of 130 people, nearly double the recommended amount.[19] In Louisiana, officers supervise about 123 cases at a time;[20] in Maryland, that number exceeds 200.[21] What’s more, statewide averages can mask vast disparities among local departments. For example, in the populous Delaware County in suburban Philadelphia, caseloads have reached 318 per officer.[22]

The problem of overextended caseloads is more than just a numbers game. Research strongly suggests that “intensive” compliance monitoring that is not accompanied by assistance to help supervisees access education, employment, and treatment is counterproductive, actually driving up re-arrest rates.[23] Reducing the use of supervision for misdemeanors, shortening supervision terms, and discharging those who have demonstrated that they can remain compliant allows officers to focus on problem cases and direct supervisees to services that can help them put their lives in order, which yields dividends in terms of both public safety and improved outcomes.

III. Reforming Supervision

Four broad reforms to supervision systems will reduce recidivism, help to reintegrate offenders, and increase public safety.

A. Adopt Supervision Techniques That Work

Effective community supervision uses techniques that have been shown to work to nudge supervisees toward complying with conditions and avoiding recidivism. Indeed, a federally funded study found that reliance on evidence-based approaches such as risk assessment, specialized case management for different kinds of offenders, and cognitive and behavioral therapy, as well as reduced caseloads “led to significant reductions in the risk of recidivism for medium and high-risk probationers” in two localities.[24]

One area where evidence should guide supervision is in using positive incentives for compliance, such as gradually relaxing supervision conditions or earning credit towards early termination of supervision. Studies show that providing positive incentives—rather than solely threatening punishment—increases the likelihood that community supervision will be successful.[25] A study of intensively supervised persons in Wyoming, almost all of whom committed some kind of technical violation of their probation (but few new offenses), found that individuals were far more likely to successfully complete probation if their supervision featured regular rewards for good behavior—in most cases rewards as simple as verbal praise, though also permission to attend special events or level down the intensity of supervision—in addition to punishments for non-compliant behavior.[26] Incentives are especially important to low-risk supervisees, who identify the opportunity for early termination of their supervision as a major incentive for compliance.[27]

The federal probation system has used another evidence-based technique to shift probation officers away from serving solely as compliance officers and toward applying a case management approach. The case management approach focuses on addressing “criminogenic” factors—such as housing instability, lack of education, negative mental processes, and poor interpersonal relations—that can prompt recidivism. To identify these factors, federal probation officers use a tool called the Post Conviction Risk Assessment to tailor supervision techniques to the specifics of a case and the responsiveness of the person under supervision. Over the past decade, the federal courts have provided probation offices with grant money to use this tool, driving the re-arrest rate among federal probationers down by more than 20 percent.[28]

Indeed, comprehensive risk and needs assessments are foundational to any smart supervision approach. Courts and supervision officers can use these assessments to craft effective packages of supervision conditions tailored to supervisees’ risk levels and needs. For example, in Iowa, supervisees are classified by risk levels that are reassessed every six months.[29] These classifications drive the level of contact supervisees have with probation officers, ranging from no contact to contact twice a month.[30] This approach allows supervision officers to protect public safety without imposing overly-onerous conditions on supervisees.

Some states and localities have made great strides in using risk and needs assessment effectively. But even where risk assessment is practiced, supervision officers often fail to bridge the gap between the assessment and case management. One large study that tape-recorded and coded interactions between officers and their charges found that officers spent little time addressing the results of risk and needs assessments and instead focused on enforcing supervision conditions. In many communities, the authors found, “[a]ssessments are completed according to policy but much of the information from the assessment fails to make it into the Intervention Plan and even less is dealt with in the sessions” with supervisees.[31] In many instances, another author found, states and localities that claim to use risk assessments “fail to use them to adjust supervision commensurate with risk,” seeing supervisees at the same rate every month and “generally concentrat[ing] on monitoring compliance with conditions of supervision, rather than on targeted, proactive efforts to reduce risk.”[32] It also turns out that half of states fail to validate their risk assessment models by comparing the predictions to actual rates of recidivism and supervision failure.[33] Risk assessment that isn’t calibrated to real-world outcomes isn’t evidence-based.

One benefit of employing evidence-based approaches to supervision is that the savings generated by reducing re-incarceration can be reinvested into hiring officers to reduce caseloads and expanding approaches that prove to be effective, creating a “virtuous cycle” where the best approaches can scale up across jurisdictions. California tried this in 2010 by enacting SB 678. Under that law, counties that reduced probation revocations received 40 to 45 percent of the money saved, which they could reinvest in expanding evidence-based supervision programs. Probation revocations declined by 23 percent within the first year, saving the state $179 million, during a period of time in which state crime rates continued to fall across the board.[34] Likewise, after North Carolina modernized its supervision practices and reduced caseloads, it was able to hire 175 new supervision officers with the money it saved.[35] Instead of serving as a conveyor belt to prison, effective and efficient supervision can free up money to support rehabilitation and crime prevention.

B. Limit Revocations for Technical Violations

As noted above, one out of every two times a person under supervision gets sent back to prison, it’s for a technical violation—a failure to comply with supervision conditions that is not itself a new crime. Common technical violations include missing appointments with supervision officers, missing curfew, leaving the state or district, or failing a drug or alcohol test. Many states leave the decision whether to revoke supervision and order a person’s arrest for a technical violation to the discretion of supervision officers. In New York, a parole officer can order an arrest on the spot without affording the supervisee the chance to respond to charges.[36] Revocations of this kind cost states more than $6.5 billion per year. And there is little evidence that they reduce recidivism. Randomized studies of intensively supervised parolees, for example, find that jailing supervisees does not increase the likelihood that they will successfully complete supervision or reduce the time until their next violation.[37]

Some scholars and commentators argue that revocations and imprisonment for violating supervision rules are far less common than is supposed and are an important tool: failure to follow the rules speaks to a supervisee’s attitude and willingness to conform his behavior to societal expectations.[38] And there is some evidence that revocations, even for technical violations, often occur when a supervisee has violated more than one supervision condition.[39]  But the prevalence of what might be called “multi-factor” revocations surely has something to do with the multiplication of standard supervision conditions, which has made it increasingly difficult for even well-intentioned supervisees to comply. Some standard conditions, such as prohibitions on drinking alcohol, are imposed regardless of whether the condition is indicated by the supervisee’s crime or history. Others, such as prohibitions on “associating with felons,” are worded vaguely enough to bring innocuous or even pro-social behavior within their ambit, such as a son living with a parent who has a criminal record.

A better approach to technical violations would be to develop a graduated system of sanctions that increase in severity, leading up to revocation. Maryland, for example, has an extensive sanctions matrix that delineates a detailed list of violations as minor, intermediate, or technical, with an escalating set of sanctions for each depending on an offender’s risk level.[40] Sanctions range from “[p]ositive reinforcement for compliance with other conditions” to more drastic punishments, such as electronic monitoring, home arrest, and restrictions on travel. In California, which has a similar system, supervision officers even have the option to seek “flash incarceration” for willful, repeat violators, jailing them for up to 10 days to focus their attention on compliance.[41]

Limiting supervision revocations to serious or repeated violations that endanger public safety lays the foundation for constructive community supervision reform. It keeps offenders in the program, simplifies the rules supervisees need to understand to comply with supervision, and preserves state resources for evidence-based, rehabilitative programming. Restricting revocations to public safety-related violations provides the most direct link between the act committed and the punishment doled out.

C. Shorten Supervision Terms and Expand Good-Time Credit for Compliance

Finally, supervision terms should be no longer than necessary to rehabilitate and reintegrate people into society. Any longer simply wastes time, money, and human potential.

At the beginning of supervision, courts should impose the shortest appropriate sentence—generally three years or less. Many judges and prosecutors impose long supervision sentences, believing that the length of the sentence will give the criminal justice system leverage over supervisees, ensuring their continued good behavior. But as noted above, there is no evidence that long supervision sentences have a significant effect on recidivism: According to the U.S. Bureau of Justice Statistics, more than half of recidivist arrests occur within a year of release, and nearly 72 percent within two years. Arrest rates for released persons plummet to less than three percent in years four and five after release.[42] Data from Minnesota,[43] Wisconsin,[44] Iowa,[45] and the federal supervision system[46] replicate this pattern: significant supervision failure rates in years one and two, with steep drops in the years after those. What the data suggest is that people who are going to fail, fail quickly; continuing to subject people who have shown they can comply and are at little risk of recidivism is costly and intrusive, and may actually be counterproductive, especially for lower-level offenders. Supervision periods should have a relatively short maximum term limit, with the opportunity to terminate short of that cap when people under supervision have achieved the specific goals mapped out in their individualized case plans. This latter milestone ought to be marked by a special ceremony to highlight the event’s significance. In short, reduced terms of supervision can focus resources where they make a difference—the first year of the supervision term and in cases of repeated, willful violators of supervisory conditions.

At the end of supervision, states should set up systems to allow for earned time for early discharge. Not surprisingly, no incentive is as compelling to supervisees as the opportunity to earn time toward completing supervision by complying with conditions. A 2007 study of parolees found that the chance to get off supervision early was “one of the strongest motivators” for compliance.[47]

Some states are trying earned-time programs, with notable success. For example, in 2012, Missouri established an “earned compliance credits” policy that allows supervisees to shorten their time on probation or parole by 30 days for every full calendar month that they comply with the supervision conditions, with the possibility of losing the earned time they have accrued towards a shortened sentence if they violate supervision conditions or are arrested. As a result, 36,000 probationers and parolees reduced their supervision terms by an average of 14 months, driving down the state’s supervision rate by 18 percent and reducing caseloads without increases in recidivism or general crime rates.[48] The Missouri earned-credit law was limited to low-level offenses and only available to those who had already spent two years on supervision, but the program could be expanded to a broader range of low-risk offenders and implemented earlier—since, as the research shows, people who have successfully completed two years of probation are very unlikely to re-offend.[49] In 2019, a group of New York state senators introduced a bill, S1343, that would create a similar earned-time credit system for probationers in that state.[50]

IV. The Prosecutor’s Case for Reform

As former federal prosecutors, we can say that it’s no secret that prosecutors have been largely hostile to a host of criminal justice reforms.[51] But calls for reform to supervision have mostly gone unremarked by this influential group. Based on our experience, most prosecutors likely see supervision and the attendant questions of effectiveness, recidivism, and reintegration as the work of supervision officers and courts.

This seems short-sighted, if for reasons of efficiency alone. Though day-to-day supervision is handled by probation and parole officers, prosecutors are often called upon to participate in sanction and revocation hearings when violations occur. In 2017 alone, there were more than 137,000 re-incarcerations for supervision violations (not counting federal probation), almost all of which required some form of court hearing.[52] This figure suggests that prosecutors are spending substantial amounts of time on issues related to supervision failure. If this figure can be reduced without compromising public safety, the result would be better outcomes for supervisees and more time for prosecutors to focus on their primary job.

But the case for prosecutorial support for supervision reform runs deeper than simply avoiding work. Reintegrating defendants back into society and putting them on the road to sobriety, stable housing, and meaningful employment lowers the crime rate and increases public safety. These imperatives lie at the heart of the prosecutorial calling. Efficient, effective supervision systems can help millions achieve these goals. Everyone has a stake in improving the outcomes of probation and parole, and prosecutors most of all.

Image credit: BrunoWeltmann