Billy Bentley, 37, a florist from Philadelphia's Logan section, has heard about the radical experiment of electing civil rights lawyer Larry Krasner district attorney, and about Krasner's pledge to end cash bail for certain low-level offenders.
It didn't help him personally, though. When he was locked up in December for gun possession, his bail was set at $300,000 — way beyond his reach. He sat in jail for months without even a preliminary hearing before the Philadelphia Community Bail Fund stepped up to post his bond this month.
While Krasner's office estimated that his bail-reform initiative could affect 4,000 defendants a year, advocates say it was actually a moderate step, noting that just about 150 people are in jail now who fit the criteria. But Krasner said it was just phase one of a larger vision to ensure that people like Bentley aren't jailed simply because they can't afford to pay. That's about 20 percent of Philadelphia's jail population.
"The goal is a system like the one in the District of Columbia, where they never use cash as an aspect of bail. That does not mean everyone gets to go home," Krasner said. Most would, though: In Washington, more than 90 percent of defendants are released pretrial.
Krasner made his remarks at a recent community meeting in West Philadelphia, speaking alongside leaders from the courts, city administration, and City Council all committed to bail reform.
Clearly, there's political will. But is Philadelphia finally prepared to end cash bail?
Leaders say the infrastructure is not yet in place — but much of it could be within a year.
That includes risk-assessment software, a controversial move but one that's been adopted in virtually all jurisdictions that have moved away from money bail. It could also include new day-reporting centers; the city has ordered a feasibility study to begin this spring. And, down the road, it will likely involve a new type of pretrial detention hearing for those defendants prosecutors believe cannot be safely released.
"The announcements that have been so public in Philly are great examples of raising a flag," said Cherise Fanno Burdeen, chief executive of the Pretrial Justice Institute. "The devil is in the details."
The shift away from money bail, already underway in such places as New Jersey and Kentucky, reflects a growing recognition of the harm the practice has caused. Even short jail stays are often destabilizing — causing a defendant to lose a job or a home — and so, counterintuitively, can actually increase recidivism.
Just a few days in jail can increase re-arrest rates by 40 percent for low-level offenders, one study found.
Another found pretrial detention increases the likelihood of a guilty plea by 16 percentage points and nearly doubles the average minimum sentence imposed.
By contrast, chief public defender Keir Bradford-Grey said, letting defendants out pretrial and, where appropriate, referring them to programs can significantly improve outcomes.
"We're hoping, if someone is making progress, we can go to a reasonable DA and say: 'This guy is doing well. Can we status this for a while and figure out: Do we need him in the criminal justice system?' "
Right now, that's not how it usually works.
Take Bentley. A magistrate set his bail without hearing Bentley's side — that he supports his kids, that he's starting a flower business, that the gun police found wasn't even his.
"On my paperwork, it states that the bail for my charges should be between $3,000 and $12,000," Bentley said.
One analysis found that 62 percent of defendants in Philadelphia were assigned money bail – and that outcomes varied widely depending on which bail magistrate was on duty.
At the community meeting, a public defender, Naji Fenwick, said he's observed those variations firsthand, in the basement courtroom where magistrates set bail for defendants who appear by videoconference.
"The magistrates don't understand the rules. … I've been told by magistrates from the bench that the Constitution doesn't apply in arraignment court," he said. "One of the rules is that we are required to be able to speak fully and confidentially with our clients prior to these preliminary [arraignment] hearings, and they do not allow us to do that."
Martin O'Rourke, a spokesman for the court, said that arraignment magistrates, in addition to being trained and certified, receive a week of continuing education each year, and that the defender is invited to those training sessions to address areas of concern.
Recent changes — part of a coordinated effort, backed by the MacArthur Foundation, that has reduced the jail population by 26 percent in two years — don't solve the problems with that system. But they do attempt to moderate it.
One was the creation of early bail review hearings, an initiative designed to release people charged with nonviolent crimes who are unable to post moderate bail amounts (under $50,000) after five days.
Since July 2016, 82 percent of people who received early bail review were released. "It's saved over 113,000 bed days up on State Road," said Michael Bouchard, who heads pretrial services in Philadelphia. He said there are discussions about expanding the program to include more defendants.
So far, 89 percent of those released have showed up for court. But additional outcomes data were unavailable, he said.
As a sample, the Inquirer and Daily News reviewed docket entries for 25 people who received bail adjustments in early 2017. Of those, all but two showed up for court. For eight, charges were withdrawn or dismissed. But 12 were involved in new cases before the first one was even resolved, ranging from retail theft to drug possession to driving while intoxicated. One man, Joshua Cardona, received early bail review hearings on three different drug arrests in three months; on his third visit to the courtroom, the judge at last declined to reduce his bail.
Such cases offer a window into the types of challenges Philadelphia is likely to face if it does eliminate money bail.
According to the Philadelphia Reentry Coalition, 36.9 percent of people released from Philadelphia jails will be arrested again within a year. The city has also historically struggled to get defendants to appear for their court dates — and previous solutions, such as the creation of bench-warrant court, have typically fueled the same spikes in jail population the city is now trying to avoid.
But, it was clear at early bail review hearings that many defendants were struggling with homelessness, addiction, mental-health issues, and chaotic lives.
Criminologists believe that most people who skip court dates do so not willfully but because they are disorganized, or couldn't get off work or get child care. Solutions may come down to logistics: text reminders and more efficient case processing to allow for fewer court dates.
Julie Wertheimer, chief of staff for the deputy mayor for criminal justice, said an extensive text, phone and email reminder system has pushed Philadelphia's court attendance rate above 90 percent.
Accountability also plays a role. In the District of Columbia, a combination of support and sanctions is used to sustain a similar attendance rate there even as more defendants are released, said Michael Williams, deputy associate director of pretrial services there.
"We try to mitigate that noncompliance as best we can. And if we can't, we ask the court to intervene," he said.
Still, D.C has far more resources for pretrial services than Philadelphia does; it has 65 pretrial officers to Philadelphia's 22.
Philadelphia officials point to this deficit as a barrier.
But, Burdeen said, "It's almost better if you're on a budget. You're less likely to misappropriate electronic monitoring or to put people on high levels of supervision. It's better to be in a situation where you have to be very cost-effective because that will compel you to comply with the law, which requires the least restrictive conditions."
Still, the city is making modest investments in pretrial services.
The MacArthur grant funded the purchase of new electronic monitoring equipment, replacing decades-old technology that required a landline. It also will provide funds to hire a social worker who can undertake a needs assessment for some defendants. And, it's supporting a pilot bail-advocate program to allow a staffer from the public defender's office to confer with some defendants pretrial.
The next big investment, though, is the most controversial: a complex risk-assessment tool that Bouchard expects to be up and running by September.
Burdeen said such programs are critical because they relieve judges of the "risk hot potato."
"What we've found is the best way to change judicial culture is to demonstrate to them, using a tool, just how un-risky people are," she said.
But activists have been critical, noting that research has found that judges don't reliably follow the recommendations, that the tools sometimes appeared to lead to more people being jailed, and that existing tools yield racially biased results.
Hannah Sassaman, who has advocated around the issue as policy director of the Media Mobilizing Project, points to an analysis of the risk-assessment tool used for probation decisions in Pennsylvania: It found zip code was a highly weighed factor, which is a problem because it can be a proxy for race in segregated Philadelphia.
"These tools are making predictions that take the racial bias in our society, clean it through a computer, and spit it out on a piece of paper," she said.
That's the worst-case scenario. The best is that the tools could actually be used to monitor and correct for racial bias in the system.
Wertheimer said the city has already pledged not to use zip code in its assessment tools. And, Bouchard added, the intention is to continuously evaluate the tool and make sure it's fair.
Activists like Sassaman are now lobbying for transparency and citizen oversight to make sure that happens.
In any case, all sides agree that risk-assessment tools should be used only to guide levels of supervision – not to decide whether to detain a defendant. Once the tool is in place, it's likely more people would be released without bail. Bouchard suggested early bail review hearings could then be replaced by pretrial-detention hearings, within four days of a defendant's arrest, where a judge would determine whether it is safe to release.
As Sassaman put it, "We need to set our community standards ourselves about what risk we're willing to tolerate."