The average stay in Philadelphia jails drags on 95 days, four times the national average. Per capita, the city has more than twice as many inmates as the national average. And though fewer violent crimes were committed in 2015 than in 1985, the jail population has doubled.
Last month, the city's criminal-justice practitioners - courts, prosecutors, defenders, and others - vowed to change that.
Through reforms funded by a $3.5 million grant from the MacArthur Foundation, they aim to reduce the jail population, about 7,500, by 34 percent in three years. With 60 percent of jail inmates awaiting trial, it would require a radical rethinking of pretrial justice.
They did not mention that just five years ago, the courts had announced a very different overhaul of pretrial justice. That time, after a 2009 Inquirer report that found 47,000 fugitives were on the street, it was more or less a crackdown: higher bails, accountability for bail-jumpers, and a new court just to bring in fugitives. The city's jail population had fallen from 9,787 to 7,575 over two years:
"That extra capacity should be used . . . to penalize effectively those defendants who fail to appear," the courts argued in a report on what was called the Reform Initiative.
As the city embarks on this latest set of reforms, it seems worth asking how we got here.
"In the world of politics, that would be called a flip-flop," said Cherise Fanno Burdeen, executive director of the Pretrial Justice Institute, "but I think it's more enlightened than that. It's people, in the face of new information, having the courage to say, 'Holy crap, I was wrong' . . . and as a team making an important choice about the purpose of the local prison in Philly."
It reflects, she said, a fundamentally new understanding of how holding people before trial can have lasting impacts.
Recent research has shown that people who remained in jail until their trial were at least three times as likely to be sentenced to prison or jail and received sentences more than twice as long as those who had been released. It found the longer a low- or moderate-risk defendant was detained, the more likely he or she was to re-offend or miss court dates. And it found money bail wasn't keeping communities safe: Half of the highest-risk defendants were able to make bail.
Longer jail times, Burdeen explained, often means lost jobs, benefits, or housing.
Even by talking about bringing in fugitives, the courts were missing the point, she said.
"The vast majority of people who fail to appear in court are not . . . trying to evade justice. For the most part, these are people who the courts don't provide robust reminder systems, much like you or I get for haircuts or doctor's appointments. The courts didn't provide practices that doctors' offices and salons learned a long time ago can nearly eradicate failure to appear."
Philadelphia now intends to add this mild-mannered approach to its arsenal.
This new era has drawn praise but also skepticism.
Angus Love of the Pennsylvania Institutional Law Project, who has worked on lawsuits concerning jail overcrowding for years, is wary.
"The track record is not good," he said. "How can you expect the people that are the problem to create the solution?"
The city has been defending itself against jail-overcrowding lawsuits for decades. Civil rights lawyers first sued the city in 1971, when the jails housed 2,700 inmates. So, over the years, the city built more jails, including the 650-bed Philadelphia Industrial Correctional Center in 1986 (now housing 981), and the 2,016-bed Curran-Fromhold Correction Center in 1994 (home to 2,733 today).
The most recent lawsuit was filed in 2008, when the inmate population reached 9,300.
Pew Research Initiative found the growth had been driven in part by a rise in pretrial inmates, higher average bail amounts, and slow court processing that meant 15 percent of inmates were in jail for 120 days or more.
After the lawsuit, the prison population fell - but then the First Judicial District declared the system broken and began its reforms.
Among them: In 2012, Bench Warrant Court was created to haul in those who missed court dates and, often, raise their bail amounts. The courts also handed over bail debt to collection agencies, raised standard bail amounts, and controversially paved the way for the return of private bail firms.
But defense lawyers complained that they'd been shut out of the restructuring. And soon, the prison population swelled again, so much so that a judge reinstated the 2008 lawsuit.
Last year, City Councilman Bobby Henon proposed legislation to buy land for another prison to replace the crowded and, according to officials, dangerously outdated House of Correction. But, unlike what happened in the '80s and '90s, the plan was crushed by opposition.
Finally, said the Defender Association's Tom Innes, there's a new mind-set.
"We were spending so much money on jailing people, we never had the opportunity to go back and look at the process from the beginning of the pipeline," he said. "What we're looking to do now is to prevent people from going into that jail pipeline."
For the first time, officials have started analyzing the data, using a one-day snapshot of the jail population to understand who's there, for how long, and what's keeping them.
They've started to identify where bottlenecks exist and how to address them.
Mark Houldin, policy director at the Defender Association, said: "The previous efforts were tinkering with a system that didn't work. What we're doing now is a fundamental change."
The largest segment of the MacArthur funds, $2.6 million, is dedicated to "alternatives to cash bail." Much of that will go to updating an electronic monitoring system that now relies on landlines - specifically, the copper lines that are rapidly being replaced by fiber-optic networks. The Centers for Disease Control and Prevention estimates that 59 percent of poor households do not have landlines.
The grant allocates $100,000 for a pretrial risk-assessment tool by University of Pennsylvania statistician Richard Berk that would use historical data to predict whether a defendant would flee or re-offend. The courts announced that Berk would create this five years ago, and in fact, he submitted a prototype.
Municipal Court Judge Marsha Neifield said the "drastic change in pretrial practice" could not be pursued then because prosecutors and defenders weren't at the table. Now, they are.
Love hopes this is just the start. "New York City," he said, "has done what we need to do." It nearly halved its jail population from its 1990 peak to about 10,000. Given that Philadelphia's population is less than one-quarter of New York's, he said, "they could bring it down to 4,000 or 5,000."
Even if that happens, those who hope for an era when Philadelphia can go from building jails to shutting them down will likely have to wait.
Julie Wertheimer, chief of staff for criminal justice in the Managing Director's Office, said: "Mathematically, the 34 percent goal, which is an ambitious but achievable goal, would not allow us to shut down the House of Correction. The numbers aren't there yet."