Crime rates are increasing across the United States. If history is any indication, many policymakers and Americans will look to our prisons and jails in the next few years to stop the crime rise.
But how do we know whether incarceration works as intended?
The existing standard for measuring whether someone is successful after release from prison is recidivism — a simple binary “yes or no” if a person commits another crime after leaving prison. Recidivism is used as a key metric for evaluating the effectiveness of prisons and reentry programs and as a marker of public safety.
The historical emphasis on recidivism among policy analysts, practitioners and scholars stems, at least in part, from a desire to establish a shared indicator of “success.” If you were rearrested, reconvicted or reincarcerated after leaving prison, you failed. If not, you succeeded. But binary measures of post-release success ignore decades of research on how and why individuals actually cease to commit crimes.
The National Academies of Sciences, Engineering, and Medicine recently convened an expert committee, on which we served, to study alternative measures of success for the more than 600,000 persons who reenter society each year after leaving prison. The committee consisted of and consulted with research scholars from a variety of disciplines, including practitioners who work with individuals reentering society from prison and with victims of crime, and persons with lived experience of incarceration.
Researchers who study crime know that abandoning a criminal lifestyle is a gradual process that often takes years, and can depend on every aspect of a person’s life — from their health to their housing and family. This process of ending criminal activity is referred to as “desistance.” Like recovery from addiction, illness or disease, desistance can involve relapses.
But binary recidivism measures treat any return to crime as a failure, even if the return occurred in the midst of or even indicated progress toward eventual cessation — such as committing fewer or less serious crimes. For example, an individual who went to prison for a violent crime and is later re-arrested for low-level drug possession could be seen as taking an uneven path toward cessation, rather than pure and simple failure.
Recidivism can also be over-inclusive. For example, an individual who is incarcerated for a technical parole violation, like missing a report date or leaving the jurisdiction without permission, may be classified as a “recidivist” for a non-criminal offense.
At the same time, recidivism is inadequate as a measure of success. Common measures of recidivism ignore other progress in community reintegration or general well-being, including employment, education, housing, health and family relationships. Individuals who avoid crime may be barely surviving, much less thriving, in other important aspects of life.
In a recently published report, our committee outlined four recommendations. Our first recommendation calls for supplementing recidivism measures with more accurate, precise and nuanced measures of desistance from crime. We also recommend better measurement of progress after release from prison.
Third, the committee recommends a thorough review of existing measures of structural barriers to progress, such as licensing and housing restrictions, and the development of new measures as needed. This review should consider how such barriers affect the lives of historically marginalized populations.
We do not propose that our recommendations about measuring recidivism and successful reentry be taken as the last word on these demanding topics. On the contrary, our fourth recommendation is that federal agencies and private foundations use our findings as points of departure for evaluating the kinds and quality of data underlying current recidivism measures. These same agencies and foundations should work towards the development of uniform national standards for measuring desistance from crime and successful reentry.
We recognize that this is a tall order. Recidivism is deeply embedded in correctional philosophy and practice. Even if our recommendations are taken up by private and government stakeholders, it may be years before they issue their own findings and recommendations.
What is to be done in the meantime? We urge everyone who cares about what happens to the people who pass through the nation’s prisons to ask themselves the same questions we did as we were writing our report: Do current measures of recidivism tell us what we need to know about life after prison? If not, how can we do better?
Richard Rosenfeld, professor emeritus of Criminology and Criminal Justice at the University of Missouri-St. Louis, chaired the National Academies committee that wrote “The Limits of Recidivism: Measuring Success After Prison.” Elsa Chen, professor of Political Science at Santa Clara University and Ronald F. Day, vice president of Programs and Research at The Fortune Society, were members of the committee.