On Friday morning in the bowels of the Frank Crowley Courts building, Dallas County's leading criminal justice policymakers reminisced about the bad ol' days at the Lew Sterrett jail. County Commissioner John Wiley Price recalled the time five years ago when the county, faced with a choice between drastically trimming its jail population or shipping surplus inmates to a Louisiana jail, released nearly 1,500 inmates in a single day with little more than a reminder that they show up for court. "There was a period of time when we probably had the largest incarcerated population in the world here in Dallas County," added Ryan Brown, Dallas County's chief budget officer. In those days, meetings of the county's jail population committee were tense, "kind of like those cage matches where whoever comes out alive wins," according to county criminal justice director Ron Stretcher.
The atmosphere at Friday's meeting was considerably more relaxed. When jail population coordinator Etho Pugh announced that the average jail census in October was 5,509 — down 450 people from the same period a year earlier and significantly lower than at any time in recent memory — there was a murmur of approval. Partly the decreased jail population can be chalked up to a reduction in crime, which has meant that about 26 percent fewer inmates per month are being booked into jail in 2015 as were arriving in 2008. It's also a result of meetings like this, where all of the independent pieces that constitute the local criminal justice system — prosecutors, judges, defense attorneys, the sheriff's office, county commissioners — come together to pore over data and negotiate ways to shrink the number of people Dallas County is locking up. Fewer people in jail means fewer people losing their jobs, being evicted from their homes or having their families disrupted, which means fewer repeat arrests. It also means Dallas doesn't have to pay the $69 per day it costs to house and feed each of them.
"We have made a lot of strides in the last few years to kind of stabilize and manage our jail population," Stretcher explains. "And that's kind of led us to look at some of our process issues." The past few years the county has primarily been focused on ironing out bureaucratic inefficiencies, things like getting cases to grand juries faster and making sure inmates bound for state prison are shipped there promptly. But there's a limit to how low the jail population can get just by making sure the assembly line is running at optimum speed, and Dallas County has about reached the floor. "We've kind of gotten all of the efficiencies that better processing can get for our jail population," Stretcher says. "Now, we're really having to get into policy drivers."
Despite the recent progress, Dallas County still keeps more people in its jail per capita (25 inmates for every 10,000 residents) than the national average (23.1). Dallas County's incarceration rate is also higher than Texas' other urban counties, as the following chart, prepared by the Council of State Governments' Justice Center, makes clear:Particularly troublesome is the rate of pretrial detention. These are people who haven't yet been convicted of a crime but must nevertheless cool their heels in jail until their case goes to trial or their families can scrounge together enough bail. Right now, more than two-thirds of the people in Lew Sterrett are awaiting trial. That's two-and-a-half times the rate two decades ago.
And the raw number of pretrial inmates has doubled over the same period, even as the total jail population has decreased.
Put bluntly, this is a stupid way to run a criminal justice system. Even setting aside the whole innocent-until-proven-guilty thing, locking up all of these people is expensive. It's also counterproductive. Advocates of pretrial reform point to a study conducted by the Houston-based Laura and John Arnold Foundation, which found that the longer offenders are kept in pretrial detention, the more likely they were to re-offend. The connection was strongest for lower-level misdemeanor offenders.
Dallas County officials aren't quite sure why the pretrial population has skyrocketed, which has happened even as the overall jail population has dropped. They are hoping an analysis of county incarceration data, currently being carried out in partnership with the Meadows Mental Health Policy Institute under a Caruth Foundation grant, will provide answers. The reason Dallas County does so much worse on pretrial detention than Texas' other big counties is easier to identify.
"The short answer is we do not provide any supervision to those released on pretrial bonds," Stretcher, the county's criminal justice director, wrote in an email. (Pretrial release, which involves an interview or other form of risk assessment, is an alternative way out of jail for defendants who can't afford a cash bond.) "We have been cautious about creating a big pretrial operation that supervises a lot of people, at a significant cost to the county, and still have an overcrowded jail."
Leah Gamble, Stretcher's deputy, gave the Observer a tour of Dallas County's pretrial operation on Friday. It did not take long. The office, housed on the bottom floor of the Crowley courthouse, consists of four officers and one clerk. Every morning, the officers print off a list of all inmates still in jail after being booked in the previous day. The officers go through the list and cross off anyone deemed ineligible for pretrial release under a 1999 Dallas County Commissioners Court order that lists 25 disqualifying offenses, some of which make a certain amount of sense (e.g., capital murder) while others ("prohibited sexual conduct," aka. incest) seem to have little rationale beyond emphasizing the county's absolute opprobrium. The list of inmates is thus whittled down from more than a hundred to perhaps a dozen (Friday's count of 20 was unusually high, Gamble said). The pretrial officers head to the jail to interview the remaining inmates, mainly to determine if they have a permanent address. If their permanent address can be confirmed, then they are eligible for pretrial release.
Other places are more progressive. San Antonio's Bexar County has a significantly smaller jail population but has about five times the pretrial releases as Dallas County, which released 2,270 defendants on pretrial bond over the past year. Bexar County also supervises people released pretrial, which research shows increases the chance an individual will show up for court and decreases the chance they will commit a crime while awaiting trial. Bexar County's recidivism rate is 21 percent, compared with 27 percent in Dallas County.
Such vigilance is expensive on the front-end, but Rebecca Bernhardt, executive director of the Texas Fair Defense Project, says Dallas County is being "penny wise, pound foolish ... Pretrial supervision is massively cheaper on a per person basis."
The more fundamental problem, Bernhardt says, is that Dallas has been slow to shift away from the traditional system of forcing defendants to post cash bonds in order to get out of jail. "The problem with the current system is bail amounts are generally set to offense level as opposed to risk," Bernhardt says. Judges in Dallas County generally must adhere to bond schedules — one for felonies and another for misdemeanors — when determining the amount of a defendant's bail. Gamble says the judges can sometimes set bail higher, but not lower, than the amount on the schedule. The bond schedule is based primarily on the crime an individual is charged with, but the charge alone is a poor predictor of whether a defendant will skip out on a court date or commit other crimes. Cash bonds are much better at sorting out those with means, who can buy their way out of jail even if they're all but guaranteed to reoffend, from those without, who can languish in jail indefinitely. A robust system of pretrial release can ameliorate that effect, but Dallas County doesn't have one.
On the plus side, Dallas County policymakers are re-examining they way they handle pretrial release and considering doing a better job of incorporating risk-assessment into decisions about whom to release. Gamble is particularly intrigued by a widely adopted risk-assessment tool developed by (once again) the Laura and John Arnold Foundation. The tool uses nine pieces of data covering criminal history and community ties to determine a defendant's likelihood of skipping court or committing a crime while awaiting trial. The tool has been shown to perform as well as or better than in-person risk-assessment interviews, which are labor-intensive and expensive. Kentucky has adopted the model statewide, releasing low- and moderate-risk defendants without forcing them to put up cash and basing bond amounts on an inmate's recidivism risk. As a result, pretrial detention there is exceptionally low — 43 percent of jail inmates there are awaiting trial — even while recidivism remains well below the national average. If Dallas County wants to continue to lower its jail population, it should consider doing the same.