Parents should be consistent, punishing children immediately, predictably, and fairly when they misbehave. The same logic, some believe, should guide our criminal justice system.
The problem with the way probation and parole violations are addressed — according to Bret Bucklen, director of research at the Pennsylvania Department of Corrections — is that authorities act like lousy parents. "The equivalent to business as usual is like telling your kid if they did something wrong, 'You've got a 40 percent chance of going to your room in five months.' "
That idea of "swift and certain" punishment, in the form of very brief jail stays for minor violators, is baked into a framework unveiled Monday by Pennsylvania officials to reduce the prison population by a thousand inmates by 2020, and reinvest a portion of the $108 million saved to reduce recidivism and increase public safety.
Known as justice reinvestment, this approach to corrections budgets was first deployed in Pennsylvania in 2012. The 2012 Justice Reinvestment Act, among other changes, allowed counties to voluntarily adopt the swift-and-certain model for probation — but only a handful, including Chester County, did so.
The report released Monday outlined steps lawmakers and criminal-justice leaders recommend for a second justice-reinvestment round. One part of that agenda would be to double down on the swift-and-certain vision, expanding it to the state parole system and, backers hope, to probation departments statewide.
But despite some impressive results, recent research suggests the model may not work any better than existing practices.
"This is not a proven method of probation or parole," said criminologist Francis Cullen, research professor emeritus at the University of Cincinnati. "It's gotten a lot of hype. … But any probation department that puts this into place, it's as likely not to work as it is to work. That's the scientific evidence right now."
Evidence-based reform is core to the justice reinvestment model, launched in 2010 with funding by the federal Bureau of Justice Assistance and supported in Pennsylvania by the Council of State Governments Justice Center.
The plan unveiled Monday would generate savings through an incremental change: Automatically paroling people serving short sentences after they completed their minimum sentence, as long as they behaved well. Excising the red tape of applying for parole will get them released five months earlier on average.
The report also recommends a slew of other changes, some modest and others likely to be controversial.
One is expanding the use of risk-assessment tools to determine bail or conditions of release pretrial. Such tools have been touted as a safe way to empty jails of low-risk offenders, but they have also been decried as racially biased. Only 12 of Pennsylvania's 67 counties currently use them. Philadelphia courts announced a plan to begin using pretrial-risk-assessment software last year, as part of a plan to reduce the jail population by a third.
Other components deal with strengthening county probation departments by creating a state-level governing body for probation and by increasing funding. Currently, the state contributes just $100 toward the $830 annual cost per probationer to Pennsylvania counties; comparable states contribute eight to 10 times more than that.
"Pennsylvania is spending $37 million on probation – but $200 million on incarcerating folks that don't do well on it," CSG Justice Center's Patrick Armstrong said.
So the hope is to do it better — in part by implementing swift-and-certain punishments for violators.
Cullen, a longtime critic, said the popularity was unsurprising: "Liberals like it because they support anything that keeps people out of prison. Conservatives like it because it's punitive. It's something for everyone."
His concern is that punishment-based corrections efforts have consistently failed to produce results.
A randomized, controlled trial of a swift-and-certain program in Delaware last year reported no improvement over existing probation practices. And a national demonstration in four different states, sponsored by the National Institute of Justice and the Bureau of Justice Assistance, found no effect on new arrests and, in some cases, an increased rate of probation revocation.
Still, in Chester County, which in 2016 launched such a program for high-risk drug offenders on probation, officials expressed optimism.
Those in the program must submit to random drug tests two or three times a week. If they admit using drugs, they spend two days in jail right away. If they deny use and test positive, they get five days in jail. That's compared with a previous average of nine months in jail for probation violations.
So far, the probation revocation rate has fallen from 59 percent to 40 percent, and the positive drug test rate is down from 41 percent to 7.5 percent.
"One young lady said she would be dead if she wasn't in this program," said Chris Pawlowski, a section director for adult probation in Chester County. "The random testing, in light of the heroin crisis we have, is really saving lives."
Jordan Hyatt, a professor of criminology at Drexel, said whether such models work depends on context.
"I think it's a good idea that should be engaged with cautiously," he said.
He helped pilot a swift-and-certain program for inmates in halfway houses in State Intermediate Punishment, an alternative sentencing program. Arrests fell 12.9 percent.
Philadelphia has declined to create such a program. For now, probation detainers continue to swell the jail population. About half the jail population is there on detainers, which make them ineligible for bail.
Jason Ziedenberg of the nonprofit Justice Policy Institute pointed to research showing even a brief stay in jail will increase a person's chances of being arrested again.