Recidivism — the rate at which people who have been released from custody commit further crimes — appears across these episodes chiefly as a premise of policy rather than a measured quantity. Guests dispute how large the rate actually is, what drives it, and whether the policies justified by fear of it produce any safety gain.
Recidivism and long sentences
In a discussion of prison costs, Veronique de Rugy argues that long sentences defeat rehabilitation: when people are locked up for so long, the idea that they can come out, be rehabilitated and make a living is nonsensical. She states that four out of ten people who come out of jail end up back in jail in the short term, attributing the return to the impossibility of life on the outside without a job or the ability to pay medical bills Crime Down: Prison Costs Up (2011). The same episode notes an aging prison population — people in their 50s, 60s and 70s — and the medical and other expenses of incarcerating senior citizens, with statistics on the threat posed by that group left unfinished in the excerpt.
Rachel Barkow approaches the question from sentencing policy. She describes the Federal Sentencing Commission, created by Congress in the 1980s, as unable to determine the right sentence length: economists and a retributive justice expert were brought in, and in the end the commission averaged past sentences by judges. Legislators setting ranges in statutes, she says, pluck numbers out of thin air and favor round numbers — five-year, ten-year and twenty-year mandatory minimums — with no evidence behind the lengths Rachel Barkow on Fixing the Fundamentals of Criminal Justice (2019).
Barkow offers a natural experiment as evidence that sentences were set higher than public safety required. In 2007 the Commission adjusted crack sentences — which she calls notoriously long and disproportionately affecting black men — and made the change retroactive, letting people in prison go before a judge for a reduction. Comparing those released early with otherwise matched people who had already served their full terms, and following them for five years after release, the earlier-released group had a lower recidivism rate. She generalizes: states that have lowered sentences do not see crime go up, often see it go down, and save money. She also notes that people who got out earlier freed prison bed space and money for programming, and that even on pure retribution the system is out of whack, since viewing child pornography online draws longer sentences than abusing a child, and selling drugs longer than killing or raping.
Bob Zadek’s framing in that episode is that the prison system does not do a very effective job at rehabilitation, thereby inducing recidivism and further increasing cost, and that sentencing is a lottery when liberty is too dear for one. He proposes that taking five years at random off every sentence would yield large economic savings and a freer society without added risk.
The registry and the recidivism myth
Two episodes examine recidivism as the stated rationale for public sex-offender registries. Sarah Stillman, asked by Zadek about the fear that a young person who commits one sexual offense will become a predator, says there are mistaken impressions about the nature and location of child sexual assault. Recidivism rates among juveniles, she says, are lower than people tend to imagine and lower than for a number of other crimes Sarah Stillman on Minors on the Sex-Offender List (2016). She also challenges the “stranger danger” premise of registries, saying most sexual harm occurs in the household and comes from someone already known to the victim, and that registries are not the effective approach even for young people with serious behavior problems, for whom good treatments exist.
Stillman’s episode also illustrates how registry exposure can compound: a young person in Texas, in a relationship both parties saw as consensual but that the law did not, was placed on ten years of probation and, after technical violations in the ninth and a half year, was locked up for a decade. She notes that some states charge sexting as possession or distribution of child pornography, and that in Louisiana a juvenile can have the word “sex offender” placed on their license.
Lenore Skenazy states the recidivism figure directly: the rate of doing another crime is about 5%, the lowest of any criminals except murderers, while people believe anyone on the registry is compelled to offend again A Sunday Morning Chat with The World’s Worst Mom (2018). She cites a New Jersey study, in the state where Megan Kanka was murdered, finding no difference in crime before and after the registry. Her argument is comparative: there are no registries for murderers or drug offenders, because those who have served their sentence and finished probation are treated as rehabilitated, while anyone who committed a sex crime is treated as incapable of rehabilitation. Zadek’s framing in the same episode is that a non-existent fear produced a police tool that became public and then ever more restrictive, depriving almost a million people of rights others take for granted, with no proof that the damage to the individual solves the problem.
Pre-trial detention as a recidivism driver
Barkow extends the causal claim to the front end of the system. Cash bail, she says, detains people in America because they are poor rather than because of safety or flight risk, and those who can pay are released even if dangerous. She describes jails as holding facilities that intermix people accused of everything from jumping a turnstile to violent robbery, with no classification system and almost no programming. Detention even for a few days, she says, can cost a person their job, housing and custody of their children. When researchers compare two people accused of similar crimes with similar records, detaining one pre-trial and releasing the other, the detained person is more likely to commit crimes after eventual release — which she attributes to the upending of that person’s life. She calls pre-trial detention a public policy failure of the widest scale Rachel Barkow on Fixing the Fundamentals of Criminal Justice (2019).
Across episodes: what changed
The topic recurs in four episodes across eight years, and the treatment shifts in emphasis rather than in conclusion. The 2011 episode treats recidivism as a consequence of long incarceration and the impossibility of reentry, with de Rugy supplying the four-in-ten figure Crime Down: Prison Costs Up (2011). By 2016 the question is whether the fear of recidivism justifies lifetime registration, and Stillman answers with juvenile rates lower than commonly imagined Sarah Stillman on Minors on the Sex-Offender List (2016). In 2018 Skenazy puts a number on the registry population’s reoffending — about 5% — and adds the New Jersey before-and-after finding A Sunday Morning Chat with The World’s Worst Mom (2018). In 2019 Barkow supplies the strongest causal evidence in the set, the 2007 retroactive crack-sentence reduction and the lower recidivism of those released early, and carries the argument backward to pre-trial detention Rachel Barkow on Fixing the Fundamentals of Criminal Justice (2019). What changes is the direction of the causal arrow: the earlier episodes argue that recidivism is lower than assumed, while Barkow argues that punishment practices themselves raise it.
What the sources do not cover
The excerpts give no definition of recidivism as a measured statistic, no time window for the four-in-ten or 5% figures, and no source or methodology for either. They do not name the New Jersey study, the Texas case, or the statutes and amendments involved in registry or bail law. The 2011 excerpt breaks off mid-sentence on statistics about offenders in their 50s and 60s, and the 2018 excerpt ends with Zadek’s label and no further statement.