A New Lease on Life

I am what time, circumstance, history, have made of me, certainly, but I am also much more than that. So are we all.

— James Baldwin

Table of Contents

Findings and Recommendations
What is Recidivism?
Review of the Criminological Evidence on Recidivism
American Exceptionalism
Coming Home with Low Risk and High Needs
Conclusion

I. Findings and Recommendations

A dramatic consequence of America’s investment in mass incarceration is life imprisonment. Today there are more people serving life sentences alone than the entire prison population in 1970, the dawn of the mass incarceration era. Though life sentences have always been allowable in the U.S., it is only in recent decades that these sentences have become normalized to such an extent that entire prisons are now filled or nearly filled with people serving life terms.

Despite a cultural tendency for Americans to view the U.S. crime and criminal legal system as “exceptional,” other countries have experienced ebbs and flows in crime rates but have not resorted to the levels of imprisonment, nor the lengths of prison sentences, that are commonplace in the U.S. To the contrary, restoration of human dignity and the development of resilience are at the core of an evolved criminal legal system; systems elsewhere that emphasize the responsibility of government support to returning citizens serves as a model for the U.S.

In this report we set out to accomplish two tasks. First, we examine reoffending rates among people released from prison after a violent crime conviction and review research on the topic, covering both domestic and international findings. Second, we provide personal testimony from people who have left prison after a violent crime conviction. Inviting impacted persons to share their transition experiences serves policymakers and practitioners in strengthening necessary support for successful and satisfying reentry from prison. This report focuses on the outcomes of a narrow segment of the prison population: people convicted of violent crimes, including those sentenced to life and virtual life sentences, who have been released to the community through parole or executive clemency. People with violent crime convictions comprise half the overall state prison population in the U.S. They are depicted as the most dangerous if released, but ample evidence refutes this.

Findings:

• We can safely release people from prison who have been convicted of violent crime much sooner than we typically do. Most people who commit homicide are unlikely to do so again and overall rates of violent offending of any type among people released from a life sentence are rare.
• Definitional limitations of the term “recidivism” obstruct a thorough understanding of the true incidence of violent offending among those released from prison, contributing to inaccurate estimates of reoffending.
• People exiting prison from long term confinement need stronger support around them. Many people exhibit a low crime risk but have high psychological, financial, and vocational demands that have been greatly exacerbated by their lengthy incarceration.
• People exiting prison after serving extreme sentences are eager to earn their release and demonstrate their capacity to contribute in positive ways to society. Prison staff and peers view lifers as a stabilizing force in the prison environment, often mentoring younger prisoners and serving as positive role models.

We make five recommendations that, if adopted, will advance our criminal legal system toward one that is fair, efficient, and humane.

1. Standardize definitions of recidivism. Authors of government reports and academic studies should take great care to standardize the definition of criminal recidivism so that practitioners, policymakers, the media, and other consumers of recidivism research do not carelessly interpret findings on reoffending statistics without digging into either the meaning or the accuracy of the statements.
2. Insist on responsible and accurate media coverage.
   Media consumers and producers alike must insist on accurate portrayals of crime despite the temptation to skew media coverage so that rare violent crime events appear as commonplace. Heavily skewed media coverage of rare violent crime events creates a misleading view of the frequency of violent crime. Add to this the overly simplistic assumption, allowed by inarticulate reporting, that people released from prison have caused upticks in violence.
3. Allow some level of risk. Reset the acceptable recidivism rate to allow for reasonable public safety risk. The public’s risk expectation is currently set at zero, meaning that no amount of recidivism is politically acceptable in a system that “works” even though such expectations are not attainable in any sphere of human endeavor or experience. But this expectation is largely based on highly tragic and sensationalized events that are falsely equated as the result of releasing people from prison. We have to balance our aspirations for a crime-free society with reasonable approaches to public safety and human rights considerations for both those who have caused harm and those who have been victimized by it.
4. Reform and accelerate prison release mechanisms. Decisionmakers considering whether to grant prison release rely too heavily on the crime of conviction as the predominant factor under consideration. This approach is neither fair nor accurate. It is unfair because it re-punishes the individual for a crime for which they have already been sanctioned. Risk of criminal conduct, even violent criminal conduct, closely tracks aging such that as people age into adulthood there is a sharp decline in proclivity to engage in additional acts of violence.
5. Substantially improve housing support.
   Inability to secure housing after release from prison was mentioned frequently by people we interviewed for this report. Failure of the correctional system to ensure stable housing upon exit from decades-long prison sentences imposes unnecessary challenges. Though some released persons will be able to rely on nonprofit charity organizations, shelters, or family, the most vulnerable people will fall through the cracks. We have both a public safety and a humanitarian obligation to avoid this result.

Andrew Hundley

In the summer of 1997, 15-year-old Andrew Hundley killed another teen in Mowata, Louisiana. His case went to trial and he received a life-without-parole (LWOP) sentence. In 2016, at age 35, his sentence was commuted to time served and he was released. Hundley was the first of more than 200 Louisianans serving LWOP for crimes committed while young who have been released since a series of landmark Supreme Court rulings invalidated the LWOP sentences of some 2,000 people.¹

From the time of his release Hundley has devoted his freedom to helping remaining qualified lifers earn a meaningful opportunity for release. Within a year of his own freedom he founded and now runs the Louisiana Parole Project, a 501(c)(3) organization that serves as a critical bridge between prison and life on the outside. The organization defines itself as a human services provider and advocacy organization, working to reduce recidivism through second chances for released lifers and others who have served 20 years or more. As in other states, those exiting prison from a former life sentence exhibit very low rates of reoffending.²

II. What is Recidivism?

Understandably, policymakers, practitioners, and researchers seek results-oriented crime policies. Unfortunately, the key measure of “successful” imprisonment — recidivism — is frequently poorly constructed. Definitional issues plague an accurate understanding of what is meant by recidivism and measurement errors abound in the research. Florida State University criminologist Gerald Gaes and colleagues from the Bureau of Justice Statistics and Abt Associates, which collect and analyze the nation’s largest corrections datasets, write extensively on these and other pitfalls of “relying blithely on recidivism data without investigating the underlying criteria.”³

Sometimes recidivism refers to arrest, other times it is reconviction, and for others it is a return to prison either for a parole violation or a new crime conviction. All of these measures of recidivism tell a different story. For instance, one might be arrested but never convicted, as is often the case, so if recidivism is considered only at the point of arrest it is a gross overestimate of criminal conduct.

Another problematic feature of many recidivism studies is the inclusion of technical violations. Sometimes technical violations of parole are included in studies of overall recidivism rates but not always.⁴ These have the effect of distorting outcomes as well, since technical violations are noncriminal activities that violate the conditions of parole such as leaving a certain radius without first obtaining a “travel pass” or failing to register a new email address, but these activities are not unlawful in and of themselves.⁵ They are also largely irrelevant as a measure of public safety. Decomposition of prison return data often reveals that a high proportion of returns to prison originate from technical violations.⁶

Reincarceration for such rule infractions is problematic and discouraged by international bodies. In its 1994 guidance report on life imprisonment, the United Nations cautioned against returns to prison which were not entirely necessary for public safety: “No assessment procedure can guarantee that a released prisoner will not relapse into crime…[the process of returning someone to prison] requires the most stringent application of the principles of fairness. Those to whom it applies have already served the period of imprisonment deemed sufficient as punishment, and have been assessed as posing no further risk to society. There should therefore be powerful, yet challengeable reasons for re-detention.’”⁷

Another inconsistency in recidivism research is the degree to which crime types are specified.⁸ Some studies make no delineation regarding crime type, others make a binary distinction of violent/nonviolent, and others provide detailed specifications of crime type.

Consider research findings by John Moore and Jacob Eikenberry which analyzed outcomes of 18,947 released individuals from the Iowa Department of Corrections over a three-year period. Crime type was critically important, with the highest proportion of those who returned to prison with a new crime having been convicted of a drug crime, much more so than those initially imprisoned for a violent conviction.⁹

Recidivism figures, especially those utilized by media, do not routinely distinguish violent from nonviolent reoffending but there are critical reasons to do so. Even within the category of crimes classified as violent there are important distinctions to consider. Acts of violence like homicide committed spontaneously, out of passion, are different from those with the premeditated intention to cause harm to another. Both types are decidedly different from conduct that constitutes a means to an end, such as a robbery committed to obtain money to serve an underlying drug addiction but that results in an unplanned homicide (e.g., felony murder).¹⁰ The criminal legal system response–particularly during incarceration–should not be a one-size-fits-all approach wherein all crimes are responded to identically and interpreted as presenting the same level of risk of offending.

Research that disentangles types of homicide shows important differences in recidivism. Pieter Baay, Marieke Liem, and Paul Nieubeerta’s research distinguishes between four underlying conditions in murders in their recidivism study: intentional homicide, felony murder, family violence, and arguments with those outside the family (e.g., barroom brawl). They conclude that specificity matters: people who had been convicted of an unplanned murder committed spontaneously during a felony were less likely to recidivate with a new violent offense and those whose homicide was related to family violence were also less likely to recidivate than those who committed intentional homicide.¹¹

Sophisticated risk categories based on homicide type have been established by New Jersey researchers using administrative crime data with typologies grouped into one of various categories. Findings showed that both rates of recidivism and types of recidivism varied with differences in original homicide, again illustrating the need for specification of crime type and recidivism type.¹² The best studies disentangle all these possibilities because of their unique contribution to the incarceration-crime relationships.

Another inconsistency in defining recidivism comes about as a result of the misuse of prosecutorial discretion. Specifically, in instances where prosecutors may decide against charging an individual for a new crime because of the cost and time saved from doing so; instead, more prison time is added to the revocation itself, so admissions might be coded as revocation as the cause when in fact a new crime was committed but is not specified in the prison records.³

The time frame between prison release and return to the criminal legal system also matters. Though most studies observe conduct for up to three years¹⁴ sometimes five years and even as far as nine years have been used.¹⁵ In well-articulated research, all of these differences are clearly articulated. Standardization should be the goal.

The Role of the Age-Crime Curve in Understanding Recidivism Trends

Some of the reluctance to release people with violent convictions originates from a misunderstanding, promulgated by media sensationalism of select crimes, that all persons released from prison run the same risk of committing a new crime. Related to this is the assumption that crime rises are caused by people who have been released from prison. Neither is empirically supported.

It is a criminological fact that violent conduct occurs in somewhat predictable ways over the life course, with proclivity toward criminal behavior among at-risk individuals rising from late adolescence to the mid-20s and dropping precipitously after. Robert Sampson and John Laub’s seminal research on crime over an individual’s life course identified six distinct age-crime patterns, ranging from a peak crime age of 16 among those identified as “low-rate chronic” to “a peak age of 40 among those identified as “high-rate chronic.”¹⁶ But even among so-called “chronic offenders’’ it appears that the vast majority will stop committing crime by their 40s and their later offenses are typically low-level “nuisance crimes.”¹⁶

Research on minors who have frequent involvement with the system starting from an early age suggests that even these individuals eventually desist from crime. Though it may take longer, desistance from crime is the typical outcome. Some crime will happen. Rather than assigning blame to the individuals themselves when new offenses are committed by youth who were previously incarcerated, we should acknowledge the harms done by incarceration at an early age. Their experiences while incarcerated may explain why young people released from prison sometimes take longer to desist.¹⁸

Despite these known trajectories of crime, prisons are increasingly filled with elderly persons who pose little threat to public safety. A 2013 analysis by the Bureau of Justice Statistics found that two of every three persons serving a state prison sentence for a violent crime was at least 55 years old.¹⁹ Our existing research on life sentences finds that nearly one third of those serving life sentences are elderly.²⁰

Some people released from prison will recidivate, and sometimes their crime will include violence. When people released from prison commit crime—especially violent crime—there are good reasons to question what went wrong and who is responsible. For the most part these questions are not delved into deeply enough and the system of correction itself is rarely held accountable for its contribution.

In many other Western democracies, programming is the central component of imprisonment, the underlying philosophy of this approach being that it is the function of the institution to reform the individual. In the U.S., by contrast, few policymakers question the logic of simply increasing lengths of incarceration rather than investing in programming and training to prepare incarcerated people to return safely to the community. Most American officials falsely conclude that recidivism is the result of not enough punishment and so more is applied. In contrast, the science on the efficacy of applying additional punishment as an effective deterrence is straightforward: more punishment does not lead to less crime.

The Impact of America’s Legacy of Racism on Criminal Punishment

Racism has consistently implicated policy decisions about crime and punishment throughout U.S. history. Indeed, overblown portrayals of violent crime and racist assumptions about people who commit violent crime has been and continues to be an easily manipulated political factor in the build-up of mass incarceration and the extreme punishment paradigm that supports it.

One need look no further than the story of William (or “Willie,” as he was renamed by the media) Horton to see how this has played out.

The infamous national story started with a Massachusetts-based prison furlough²¹ program that collapsed shortly after Horton escaped, fled to Maryland, and committed a series of violent crimes in 1987. Horton became the focus of the ongoing presidential campaign, which led to the political downfall of presidential hopeful Michael Dukakis. Dukakis, then the governor of Massachusetts, had publicly expressed support for the program, which had been a standard corrections practice in more than half the states and the federal government at the time.

Though difficult to fathom in today’s distorted punishment system, furlough, or work-release, programs allowed persons serving life sentences for first degree murder to leave the prison grounds on a regular basis and work in the community.²² This facilitated hands-on training that often led to employment opportunities after release.²³ The practice of rejoining the community for small segments of time provided individuals a chance to transition to their eventual freedom. It also allowed the preservation of family and peer relationships that are often critical to success after a conviction, including lower rates of recidivism.²⁴ After Horton’s crime, departments of corrections largely shuttered their furlough programs around the country and they are rarely used to this day.

Horton’s crimes were indeed tragic but they were an anomaly in an otherwise successful program which maintained a voluntary return-from-furlough-rate in the range of 99% year over year.²⁵ But his 1987 crimes occurred at a time when crime policy was just becoming more deeply enmeshed with political jockeying for who could be the toughest on crime by doubling down on punishment. The successful defeat of Dukakis solidified even further both excessive punitiveness as a political agenda and the use of crime policy as a racist dog whistle in American politics. The political reaction to Horton’s crime became a precursor to election campaigns in the coming decades, helping shape the policies of the 1990s and early 2000s that greatly accelerated mass incarceration and further entrenched political racism as a driving force in crime policy.

The exploitation of his crimes focused considerably on racial dynamics. Horton is Black and his victims were white. The focus on race in the endorsement of lifelong confinement cannot be overstated. Today one of every five Black men in prison is serving a life sentence.²⁶ Bush’s campaign used this incident to seal the public’s association between Blackness and criminality thus ensuring that harsher sentences would be favored and exit-options for the incarcerated would be closed.²⁷

Left out of heated accusations claiming that African Americans’ possessed a unique proclivity toward violence was the radically different outcome for nearly everyone else who participated in the furlough program at the time. In fact, Horton was one of about 600,000 persons released that year nationwide, the vast majority of whom returned voluntarily and without incident.²⁸

Research establishes the racist tendencies driving reactions to crime and crime policy, especially crimes of violence. Harvard University race scholar Khalil Gibran Muhammad observes that violence committed by Black people evokes a different public reaction than violence committed by white people. ²⁹ Black people are portrayed as dangerous and violent in comparison to white people, whose acts of violence are more easily interpreted as aberrant and situational. This perception allowed the proliferation of the concept of a youthful superpredator ³⁰ to emerge with little controversy at the time, though it has now been thoroughly debunked. Author David Sklansky, whose legal scholarship specializes on the definition of violence and its intersectionality with race, makes a similar connection.³¹ He notes that crime committed by whites is often attributed to situations associated with the commission of crime whereas crimes committed by Blacks are attributed to a fundamental nature within them as dangerous and violent.

As with all stages of the criminal legal system, race-based assumptions about African Americans who commit crime subjects Blacks to greater scrutiny and ultimately more punishment than whites.³²

Michael Mendoza

At 15 years old, Michael Mendoza sat in the backseat of a car while the front-seat passenger shot and killed someone in a gang-related murder. Mendoza was prosecuted in criminal court as if he was an adult, convicted of second degree murder, and sentenced to life in prison with the possibility of parole.

He first went before the parole board in 2010, where he provided evidence of his personal growth in prison and his readiness to return home. He was denied. In 2014, he received his second chance as a result of California’s Senate Bill 260, which created a separate, age-appropriate parole review process for youth sentenced to life imprisonment. Successfully appearing before this board allowed for his release in 2014.

A condition of Mendoza’s parole is a lifetime of supervision by California’s Division of Adult Parole, an agency within the California Department of Corrections and Rehabilitation. Such supervision often imposes requirements that make a successful transition to life in the community difficult. For instance, Mendoza was initially required by his parole conditions to stay within a 50-mile radius of his residence, which left him unable to visit family. Yet maintaining family bonds serve as a strong protective factor against committing crime. The myriad limitations set by parole restrictions motivated Mendoza to pursue a career advocating for formerly incarcerated people.

“Being engaged with these policies just by simply sharing my own personal experiences of what it was like to grow up in incarceration as a Mexican-American kid gave me so much confidence and experience that I needed to really succeed in this world,” said Mendoza.

Today he is the National Policy Director at the Anti-Recidivism Coalition (ARC) in California, where he is responsible for expanding the organization’s policy priorities. Mendoza hopes his experiences will serve as a way to help other formerly incarcerated people “continue to change the narrative” and give them the opportunity to “show that we are not ex-cons, we are not felons, we’re not inmates, we’re people that have a way to give back.” The importance of the lived experience of imprisonment in earning trust and support of newly released lifers is critical.

Mendoza’s work is just one of the ways he is providing others with the same opportunities he’s been given. He recently adopted a dog and highlights how the experience has impacted him by being able to serve as an advocate in a new way.

“[S]he’s teaching me a lot of patience, humility, love and for me, for someone like myself and the traumas that I’ve experienced, it’s been really helpful. She’s amazing. She’s smart, she’s well-behaved. And I think it’s because she did time, too. She did about a year in an animal shelter before I found her.”

“We are not ex-cons, we are not felons, we’re not inmates, we’re people that have a way to give back.”

III. Review of the Criminological Evidence on Recidivism

National Evidence Shows Low Rates of Violent-Crime Recidivism

The Bureau of Justice Statistics is responsible for the collection and analysis of state corrections data and the agency’s reports are heavily relied on by scholars, advocates, and policymakers for understanding national crime policy and corrections trends.

Researchers at the BJS tracked the arrests of 404,648 people exiting prison in 2005 across 30 states; within three years, 42% were rearrested and within five years just over half had been rearrested. Twenty percent of all individuals released from prison were arrested for a new violent offense within three years. The majority of these were for assault, 1% included a homicide, and 2% included a sexual assault/rape. Among those who had initially been convicted of a homicide, 2% committed a subsequent homicide. As depicted in Table 1, these individuals were less likely to commit any other violent offense than released persons who were initially convicted of a nonhomicide. ³³

Similarly, a 2002 Bureau of Justice study of 272,111 prison releases across 15 states found that persons exhibited low public safety risk following release after a homicide conviction.³⁴ Among those released after serving time for murder, 1% were arrested for another murder and 17% were arrested for another type of violent offense. One percent of people released from prison after serving time for a violent crime were subsequently arrested for a murder and 28% were arrested for another violent offense. These rates fall far below new arrests among those convicted of other crime types.³⁴ Persons released after a homicide conviction were rearrested at a considerably lower rate (41%) than released prisoners generally (68%).

Despite these relatively low rates of recidivism, this is not the portrayal of murder or other violent crime that media consumers receive. Instead, the most sensational murders are characterized as commonplace.

Reproduced from Durose, M., R., Cooper, A. D., & Snyder, H. N. (2014). Recidivism of prisoners released in 30 states in 2005: Patterns from 2005 to 2010. Bureau of Justice Statistics.
Notes: BJS defines murder as inclusive of murder, voluntary manslaughter, vehicular manslaughter, negligent manslaughter, nonnegligent manslaughter, unspecified manslaughter, and unspecified homicide. BJS measures recidivism as arrest rather than return to prison on a new conviction. Because arrest frequently does not lead to conviction and imprisonment, this is likely to be a substantial overestimate of criminal offending. This study captured data from 30 states.

John Pace

Left image: John Pace early on in his life sentence.
Right image: John Pace today

John Pace committed attempted robbery in 1985 and his victim died from related injuries ten days later. At 17 years old, Pace was convicted of second-degree murder, which requires a mandatory life without parole sentence in Pennsylvania. A Supreme Court decision in 2016 allowed for Pace to be resentenced and granted parole a year later.

“Seeing the pain of your loved ones, particularly in my case, my mother,” Pace said. “Seeing these kinds of things, I think those were the kinds of things that really resonated with me and said I want to do something different.”

Maintaining an emotional and physical connection to family makes a difference. A Canadian study of 86 people convicted of homicide who subsequently recidivated identified the loss of community and family support as a result of their incarceration as the primary explanation for reoffending.³⁶

Pace was originally denied programming because of his life sentence; some administrations see it as a waste of money to provide programming to those who will never be released. Eventually he participated in the Inside-Out Prison Exchange Program which brings together incarcerated and traditional university students. He eventually earned a bachelor’s degree from Villanova University.

Pace’s participation in the Inside-Out program afforded him the opportunity to secure a job with the program upon his release, eliminating the barrier to employment that many returning citizens face. Pace says he’s blessed to have the opportunities that he has had upon his release, but still faces challenges. He is under lifetime parole supervision with strict guidelines.

“You’d like to think that you’re free, but you’re really not and I think you’re reminded of that,” Pace said about parole.

Today, Pace works as a reentry coordinator for other people coming out of prison. “I like to speak to young people, particularly young people who come from marginalized communities, that probably don’t think there’s a way out of this,” said Pace. “Being able to provide my perspective to them, I think I provide them hope that there are ways that you don’t have to go through the same experience I went through in order to get it.”

State-Level Recidivism Data Supports Low Levels of Reoffending for Violent Crime

When Louisiana abolished parole in 1971, it foreclosed any possibility for release for persons convicted of first or second degree murder, before which time Louisiana often granted release after approximately ten years. Louisiana has one of the largest populations of life and virtual life-sentenced prisoners; one in five people in Louisiana prisons has a life sentence.³⁷

A number of recent legal challenges have led the way for a new era of reform to take hold.³⁸ The earned release of hundreds of people originally sentenced to life with no chance for parole has allowed researchers to observe outcomes for these released prisoners.

Louisiana State University researchers tracked arrest, conviction and reimprisonment of 205 released people who had been convicted of murder or armed robbery. Both three and five-year reimprisonment rates were examined revealing a 5% and 8% reimprisonment rate, respectively.

At its peak, Louisiana had the world’s highest per-capita rate of people sentenced to life without parole for crimes committed while under 18. Recent legislative reforms now allow this group parole consideration after serving 25 years. According to news reports from the end of April 2021, since the Louisiana legislature extended parole eligibility to this subset, the board has granted parole to 68 people and not a single one has been rearrested.³⁹

Similar results are evident in Michigan, another state that relies heavily on parole ineligible life sentences as a public safety tool but people paroled in this state between 2007 and 2010 with convictions for second-degree murder, manslaughter, or a sex offense were about two-thirds less likely to be reimprisoned for a new crime within three years as the total paroled population, according to a 2014 study by researchers at the Citizens Alliance on Prisons and Public Spending. Over 99% of these individuals had not been re-imprisoned for a similar offense within the three-year study period.

Slightly higher rates of violent recidivism are evident from a study in New Jersey of 320 people who were sentenced for a homicide conviction, imprisoned, and released between 1990 and 2000. Reoffending data during a five-year follow up period revealed that 48% of the sample did not recidivate and another 27% violated their parole. Of the remaining, 6% committed a property offense, 7% committed a violent offense, 10% committed a drug offense and 3% committed a weapons-related offense. None committed another homicide.⁴⁰

New York has a population of persons serving life sentences that is 69% greater than its entire prison population of 1970 at the start of the mass incarceration era. Nearly 8,300 people are serving parole-eligible life sentences in New York, representing one in 5 people in prison. Characteristics of many states, the majority of people serving life sentences in New York, 93% have been convicted of a violent offense, including 71% for a homicide.

John Carner, former spokesperson for the New York State Division of Criminal Justice Services, observes: “Individuals who are released on parole after serving sentences for murder consistently have the lowest recidivism rate of any offenders.”⁴¹

Figure 1. New York Trends in Lowering Life Imprisonment Amid Declines in Violent Crime

Source: Federal Bureau of Investigation (n.d.) Crime Data Explorer; Nellis, A. (2021). No end in sight: America’s enduring reliance on life imprisonment. The Sentencing Project.
New York has been safely reducing its reliance on life imprisonment since 2004 and maintained a declining violent crime rate over the same period of time. Between 2004 and 2020, the state has declined its life-sentenced population by 5,000 people.

Analysis of New York prison release data by University of Michigan Law School’s J.J. Prescott, Benjamin Pyle, and Sonja Starr found that reimprisonment rates among people previously convicted of murder or nonnegligent manslaughter in New York were less than half that of the general population released from prison during the three years following their release.⁴² Moreover, homicide convictions among those who were aged 55 and older, and released during the study period between 1991 and 2014, were very rarely imprisoned (0.2%) for the same offense.

Repeat offending among persons released from prison after a murder conviction is rare in New York. “Of 368 convicted murderers granted parole in New York between 1999 and 2003, six [people], or 1.6% percent were returned to prison within three years for a new felony conviction–none of them a violent offense.”⁴³ A separate study of persons released between 1985 and 2012 fewer than 2% were returned to custody.⁴⁴

Jacob Brevard

Jacob Brevard was 19 years old when he arrived in prison to serve a parole-eligible life sentence for a first degree murder conviction.

After 25 years behind bars, Brevard was granted release in 2014. He attributes his personal transformation to his mother’s death that occurred while he was incarcerated. In that moment, he says that he made a promise to himself that he was going to show up differently in life. He has kept this promise and now uses his experiences as the Associate Director of Inside Programs with the Anti-Recidivism Coalition (ARC), where he runs character development and rehabilitative groups inside California prisons. Through his work Brevard is able to impact the lives of incarcerated people

“You know, you go in and the guy’s real pessimistic…and after taking programs and taking groups, you see this same person and he’s like, ‘I have to change the way I think. I have to change my mentality,’” said Brevard. “When people get it and come home, that’s the most fulfilling thing that can happen in my life and it’s wonderful.”

Brevard’s success counters the common narrative that those who caused harm in the past will always be on the brink of causing harm again so they must stay in prison. He encourages his clients in showing their capacity for change and advocates for offering meaningful opportunities for second chances.

“If you want people to change and you want public safety to be paramount, we have to change the way we deal with people who are incarcerated or previously incarcerated and give them opportunities to be successful,” he says.

“I feel like I’m an ambassador for all the people that are still doing time and that my behavior is a reflection of them,” said Brevard. “If I come out here and do something stupid, some of those guys will have to pay the price for it.”

“I feel like I’m an ambassador for all the people that are still doing time and that my behavior is a reflection of them,” said Brevard. “If I come out here and do something stupid, some of those guys will have to pay the price for it.”

International Evidence Supports Low Rates of Recidivism for Violent Crime

Examining recidivism trends internationally is challenged by the fact that the U.S. incarcerates its citizens far longer than any other comparable nation. Even when examination is limited to life sentences, the range of years defined as a life sentence outside the U.S. is typically 10-15 at the most.⁴⁵

The U.S. diverges sharply from other democracies in its perspective on the purpose of imprisonment. While other nations reject outright the imposition of long term and life sentences on grounds of human rights violations, the U.S. continues to rely on them in the erroneous belief that excessively punitive sentences keep Americans safe.

Some American policymakers, academics, and corrections leaders have begun to look beyond the United States–primarily to Western Europe–for guidance on how to rightly shrink the prison system without jeopardizing public safety.

Observation of more efficient and effective correctional systems allows a view of how the U.S. system might operate if prison sentences were substantially shortened. A range of international studies shows that life imprisonment is of little utility given the extremely low rates of reoffending among people convicted for violent crimes such as murder.⁴⁶

A western Australian study examined crime outcomes of 1,088 individuals originally convicted and imprisoned for homicide. Arrest data showed that 22% of the individuals were arrested for another violent crime and among these, three individuals were subsequently charged with a new homicide.⁴⁷

Like the U.S. the Netherlands dramatically increased its incarceration rate between the early 1970s to the mid-1990s.Similarities between the Netherlands’ approach to punishment and that of the U.S. allows for comparative study, though the use of life sentences in the U.S. still far outpaces that of the Netherlands.

Researchers Pieter Baay, Marieke Liem, and Paul Nieuwbeerta examined new convictions for 621 Dutch individuals originally imprisoned for a homicide between 1996 and 2004 and released before 2008. Overall, persons released from periods of imprisonment ranging from one year to eight years for a homicide were significantly less likely to reoffend with a violent offense than a nonviolent one. After three years, 38% of those originally convicted of homicide were reconvicted for a nonviolent crime compared with 14% for a violent crime.⁴⁸

Criminologists Ben Crewe, Susan Hulley and Serena Wright documented the expansion of time-served among lifers in England and Wales in their ethnographic account, noting that the minimum time-served on a life sentence was 13 years in 2003 but has almost doubled by 2013. As in the U.S., punitive policy shifts rather than large-scale changes in crime, account for these extended imprisonment times. But an assessment of outcomes from two distinct periods in England and Wales of 2000-2001 and 2010-2011 researchers found that of the more than 6,000 murder convictions, fewer than 0.5% were committed by persons previously convicted of such an offense.⁴⁹

A second study released in 2013 of crime outcomes among those released from a life sentence in England and Wales reported that the overwhelming majority of prisoners reintegrated to the community without incident: “‘[O]nly 2.2% of those sentenced to a mandatory life sentence and 4.8% of those serving other life sentences reoffended in any way, compared to 46.9% of the overall prison population.”

The Scandinavian countries are widely regarded as being on the opposite end of the punishment spectrum as the U.S. In Sweden, for instance life terms have a maximum imprisonment of 18 years. Here, government clemency is used regularly and releases among lifers typically occur after 14-16 years.⁵⁰ Though relying on a small sample size of 26 persons released from a life sentence, researchers identified only four instances of violent offending after release.⁵¹

IV. American Exceptionalism

A common critique when comparing the US imprisonment rate with other countries is that America suffers from a higher violent crime rate and this creates a higher need for imprisonment.⁵² It is true that the U.S. has a violent crime rate that is roughly five times that of other similarly situated countries. It is also true that when it comes to nonviolent crime, America’s rates are mostly on par with rates in other countries. That is, when countries of similar size are compared (e.g., New York and England or Los Angeles and Sydney, Australia), nonviolent crimes occur with the same frequency.

One plausible factor contributing to the elevated homicide rates in the U.S. is the readily availability of firearms. The possession of a firearm during the commission of a crime allows for it to become lethal much more easily.⁵³

Government responses to crime elsewhere are also vastly different from the U.S. approach.⁵⁴ Though incarceration is still utilized imprisonment is much briefer. Prison facilities also aspire to mirror life on the outside as much as possible to ensure that incarcerated individuals are prepared to succeed when they re-enter the community.

Zerious Meadows

Left image: Zerious pictured with his sister as a young teenager.
Right image: Zerious pictured with his granddaughter in 2020.

Zerious (which means “warrior”) Meadows was released from his life-without-parole sentence in 2016 after serving 47 years in Michigan, a sentence he began at age 16. Today, at age 67, he is cared for by his devoted family, various members of whom visited him regularly over his almost half-century in prison for a crime he claims he never committed.

Meadows considers himself fortunate to have had regular visits from family, and credits their devotion to him with his staying on track. When asked what it meant to him to have frequent visits from family, he shared, “it was a lot because it kept you out of trouble. I didn’t want to worry my mother.”

Today, Meadows does not work because of his old age and instead collects Social Security Insurance (SSI). He shared his discomfort with going from the control of the corrections system to the care of his family, expressing desire to get a chance to be on his own but knowing he may have missed the chance. He does not leave the house much and struggles with paralyzing depression; he describes some days as being like a, “a blanket over me.”

Meadows wonders how he ever lived through multiple decades in prison. Meadows’s story serves as a reminder that low recidivism rates among released persons after longtime imprisonment does not imply that life is easy on release. To the contrary, people exiting years of living in prison face substantial psychological, social, economic, employment, and housing challenges in their newfound freedom in the community. Prison is an artificial environment with few attributes that pass over to life on the outside. Most decisions are made for the residents and autonomy is discouraged. The conditions in many prisons are deplorable: unsanitary as well as physically frightening. After Meadows’s release, memories of prison riots, other men being murdered and raped, and female corrections officers being physically and otherwise abused by male officers have resurfaced repeatedly for him.

“I see how veterans come back from war, they be having flashbacks. That’s what happened to me,” Meadows said. “I guess when I was in, my body put me in survival mode.”

The psychological toll of prison is intense and long-lasting. Released lifers share that the reintegration to life outside prison raises many unforeseen psychological challenges. Ralph Brazel was released from federal prison in 2013 after serving more than two decades for a nonviolent drug offense. He recalled the following, “In prison I sometimes dreamt I was free but woke up to the nightmare of my incarceration. For a long time after my release, I dreamt I was back in prison. Fortunately I woke up to realize I had been freed.” The mental toll on people who are released is often tremendous.

The transition for long termers is disorienting; there is enormous pressure but little support. People enter an obstacle course of rules and expectations that are difficult to meet and have high stakes if they fail. Though some prison administrations provide instructions on basic daily living skills, like how to use a debit card and a cell phone, how to write a resume and complete a job application, or how to obtain official birth records, others do not.⁵⁵

“I see how veterans come back from war, they be having flashbacks. That’s what happened to me,” Meadows said. “I guess when I was in, my body put me in survival mode.”

V. Coming Home with Low Risk and High Needs

The myopic focus on commission of new crimes as the sole measurement of success ignores attention to the overall well-being of the individual leaving prison. As a result, government programs rarely provide the support needed to make a successful transition to the community and rarely address the real challenges individuals face upon reentry.

Most people who commit homicide are unlikely to do so again and overall rates of violent offending of any type is also rare. While it is important to decipher what prompts individuals to commit new crimes of violence after release, understanding what motivates them to lead law-abiding lives and contribute positively to society is equally important.

Catherine Appleton, longtime scholar on life sentences worldwide, notes, “lifers who fail on license (i.e., parole/release) attract a high level of publicity and attention, whereas day-to-day routine of good practice goes largely unnoticed.”⁵⁶ As a result, policies that would benefit the majority are too often judged on the recidivism of the few.

In addition to the fact that individuals convicted of homicide and other violent crimes rarely commit these crimes again, there is also ample evidence that these individuals are highly motivated to change negative behaviors and transform their lives. Indeed, contrary to the assumption that lifers have “nothing to lose” once they arrive in prison on a sentence that could last their natural life, dozens of studies on the lived experience of life-sentenced individuals find just the opposite. Lifers are eager to earn their release and are viewed as a stabilizing force in the prison environment.⁵⁷ This occurs in spite of, not because of, the experiences they have in prison. Prison is, after all, an artificial environment in which obedience to the institution’s rules rarely translates into challenges faced on the outside. Individuals learn to cope in prison but it is very different from the outside world; the coping skills gained in prison are not easily adaptable to society and are sometimes even counter effective.⁵⁸ Most prison programming in the U.S. is prioritized for its ability to reduce recidivism as its main objective rather than as a path to self-improvement, job training, education, cognitive behavioral improvements, and so on.⁵⁹

Critics see in-prison programming focused only on reducing risk, as well as risk assessments to estimate risk, as largely disconnected from what we know about punishment.⁶⁰ Michael Tonry, longtime scholar on sentencing, writes, “A number of states are busy at work trying to include risk predictions in their sentencing guidelines…There are several problems. First is the excessive punishment problem: given the extreme lengths of legally authorized and routinely imposed prison sentences in the United States, it is highly unlikely that sentence increases for offenders adjudged to be high risk will be consonant with proportionality constraints.” He goes on to note, as have others, the high probability of “false positives,” or the overestimation of risk which results in the excessive incarceration of individuals who would not have offended. This exacerbates the moral and human rights problems that already set the U.S. apart from other nations regarding punishment.

Another failing of American corrections is its one-size-fits-all approach to re-entry. For instance, the prison programming applied in the U.S. is rarely tailored to the conduct that landed an individual in prison. A vast literature and growing industry now exists to predict reoffending via so-called “risk assessments.” Response protocols should be tailored to the individual underlying causes of crime. Anger management and cognitive behavioral techniques would be appropriate for some and development of life skills, substance abuse treatment, trade development would be more suitable for others. Certainly, recidivism prospects would depend on whether the rehabilitation provided was relevant to the underlying causes.

In states like Arizona, people serving life sentences are pointedly excluded from participating in programming, but this is anathema in other countries. In Sweden by contrast, the legislature has explicitly required that lifers are treated similarly to all prisoners. “Every inmate, regardless of the length of the individual sentence, is obliged to take part in some form of occupation, be it in the form of work, training, or programs ‘related to crime or misuse or some other structured occupational activity’”⁶¹

Left out of meaningful policy and corrections conversations in the U.S. is how to best support people exiting prison after long-term imprisonment. Instead the reentry process is generic to all people leaving the prison doors, regardless of the number of years spent there. For people coming back after decades away, the world has transformed and relationships with family and connections on the outside have been strained or ended. Reentry needs are significant.

In Alabama, as in many other states, people leaving prison are given a small amount of “gate money,” approximately $10 and a bus ticket. Exiting individuals are provided with the following guidance:

Upon the completion of your sentence, you must be discharged from the penitentiary. In the event you do not have suitable free-world clothing available at the time of your discharge, you will be furnished clothes. You will also be evaluated for transportation needs. When you do not have transportation available, you will be provided with the least expensive kind of public transportation back to or nearest to the point of sentencing, or if paroled, to the point to which you will have to report for parole supervision.⁶²

Given such insignificant support for a life-changing event like leaving prison, it is hardly surprising that many individuals find returning to the community exceedingly difficult, if not impossible.

For example, finding a place to live after leaving prison is a common obstacle for people exiting a long prison sentence. When we spoke with Joyce Granger, a Pennsylvania lifer released after 35 years in prison, she said that if it had not been for a nonprofit organization that provided housing she would have been homeless. Her felony conviction rendered her ineligible for most housing assistance programs and she had few connections to the outside world. When we asked Andrew Hundley, Executive Director of the Louisiana Parole Project, the most common barrier faced by people leaving prison after multiple decades of incarceration, he emphasized the challenges people face in finding a place to live.

A key function of his organization’s work is to offer comfortable, decent housing to their clients. Hundley says: “The houses are not what people would imagine. We want clients to understand, this is what normal living is like. They’re nicely decorated, nicely furnished, kept clean. Because we want when people leave us, and they go into a situation that, you know, should not feel normal, we want them to know, like, hey when I was living at a transition house with the Parole Project and it was much nicer than this. This is what I want to aspire to for my living condition.”

Typical bedroom for exiting incarcerated people housed by Louisiana Parole Project.

Stability is a key component of remaining crime-free after release. Parole regulations are strict, requiring frequent check-ins and myriad stipulations. Added to this is the high turnover among parole officers which disrupts the ability to bond to someone who could otherwise be a key member of the individual’s support network. Granger shared with us that she already had four parole officers since her release in 2018. While they were helpful to her in some ways, they did not get the chance to know her. Studies of readjustment among people released from a life sentence suggest that the network of support around them is critical. Supervisory relationships are defined by trust and dignity. “This depends on a system that recognizes the importance of enabling life-sentenced prisoners, both inside and outside prison ‘to take responsibility, to strike out independently, to look beyond the prison label, and to recognize their own potential and strengths and human beings.’”⁷ In the U.S. these qualities are minimized and too often go utterly unsupported.

VI. Conclusion

A false dichotomy exists between meeting our universal need for public safety and offering second chances. Stereotypes about people who commit violent crime ignore criminological research, and overlook successful reentry stories. Instead media coverage and public discourse are too frequently dominated by the exceptional, outlier cases where crimes are committed by individuals who were formerly incarcerated for murder or other violent acts. In this way, the exception has become the rule in American crime policy by creating a system that hurts the majority while guarding against a very small minority.

In this report we have explored the experiences of people who committed violent harm in their past and been imprisoned for it. We have examined the quantitative evidence nationally and internationally that shows the minimal risk of releasing such persons after a reasonable period of time. Despite some differences in crime patterns and imprisonment trends, it is wise to look to other countries for guidance on how to shrink our prison populations while maintaining public safety. As we have shown, in most countries the presumption of release after a maximum of 15-20 years is standard. In most studies of recidivism rates of persons convicted of murder or other violence, recidivism rates are less than 10%, often as low as 1-3%.

Though efforts to shrink the size of our correctional population are gaining momentum in public discourse, too frequently reforms do not account for the need to shorten allowable prison sentences for people convicted of violent crime. Almost half of those in prison have been convicted of violent crime, and prison terms have grown so long that they exceed their anticipated public safety benefit. The national, state, and international evidence shows that we can safely release people convicted of violence far sooner than we do.

Inaccurate and sensationalized reporting impedes a complete understanding of crime risk. Media portrayals that present violent crime as commonplace and random misrepresent reality. Media producers have a responsibility to deliver accurate crime news. Consumers have a responsibility to read and watch news with greater scrutiny.

Add to this is the various definitions of recidivism that abound in various studies. As a start, definitions for recidivism should be uniform and studied more rigorously. Recidivism should also no longer be utilized as the sole measure of “success.” Factors related to social, physical, and emotional health, gaining employment, and securing housing are all factors that should be included.

Providing a “second look” to currently incarcerated people after no more than 10 years and restricting prison terms in most cases to a maximum of 20 years would effectively reduce our prison size and keep the public safe.⁶⁴

To make this a reality, states should professionalize and accelerate prison release mechanisms. Decisionmakers considering whether to grant or deny prison release rely too heavily on the crime of conviction as the harbinger of future behavior. Risk of criminal conduct, even violence, closely tracks with aging into adulthood, a statistical fact that can be trusted when adopting sentencing and reentry policies.

Most people can succeed on release but some will reoffend. Policymakers and the public must accept some level of risk. We must balance aspirations for a crime-free society with human rights considerations for both those who have caused harm and those who have been victimized by it. Investment in successful reentry will reap far greater outcomes than creating endless obstacles that set people up for failure.⁶⁵