Texas 10-12K prison beds short under Senate, House budgets, eyewitness ID clears committee

Mike Ward at the Austin Statesman had good coverage of Tuesday's Senate Criminal Justice Committee hearing, where I spent my day (on behalf of the Innocence Project of Texas) waiting to testify on legislation to require law enforcement agencies to create policies on eyewitness identification procedures and authorize a statewide model policy implementing best practices. See Mike's coverage:

• Exonerees support eyewitness ID rules
• Report: Prison cuts could cause bed shortage in two years

Faulty eyewitness IDs responsible for hundreds of false convictions
Sen. Rodney Ellis' eyewitness ID bill passed out of committee unanimously, with an an inconsequential cleanup amendment from Sen. Joan Huffman, just as Chairman Pete Gallego's companion bill passed out of House Criminal Jurisprudence Committee last week. Since that legislation two years ago died because of time as opposed to any vocal opposition (and no one testified against it in either committee this year), that bill appears to have a good chance of passage.

Exonerees who testified had spent between 13 and 30 years falsely imprisoned, and as always their testimony was powerful and moving. It's quite an honor and a humbling experience to get to work with those guys. I find it unimaginable what's been taken from them, not to mention the courage it takes to have endured such a trauma and then keep coming down to the Legislature to tell them "Change the system so this doesn't happen to somebody else." (Watch the video here; testimony on SB 121 (Ellis) begins at the 58:05 mark.)

Except it is happening somewhere in Texas, arguably, every day Texas courts are open for business. DNA testing has given us a narrow window into the causes of innocent people being convicted - mainly false eyewitness identification, mendacious snitches, false confessions, faulty forensics, and ineffective assistance by defense counsel - but those problems arise every day in many cases that don't have DNA available to clear a falsely accused of defendant. Eyewitness identification errors were involved in the vast majority of DNA exonerations (75% nationally, 80% in Texas), so this bill arguably is the most important piece of preventive policy legislation on the subject the Lege will consider this year.

That was the subject of my own testimony to the committee: The exonerated men who testified, I reminded senators, represent just a small sample out of hundreds or even thousands of other innocent people locked up who weren't so lucky. DNA evidence exists in fewer than 10% of violent crimes, and not at all in crimes like robberies where use of eyewitness identification involving strangers can be key to a conviction. So off the bat, DNA exonerations represent less than a tenth of the total number of innocent people locked up. Then consider that Dallas is the only major jurisdiction that a) kept old DNA samples and b) went back to vet old cases for possible innocence claims. If other counties had done so, there'd be a lot more exonerated men (and maybe even a few women) clamoring to testify in favor of changing eyewitness ID procedures.

Relating estimates of actual innocence from various sources (rounded up recently in this post), which range from .75% to 3.3%, one gets a range of 1,200 to 5,000 estimated innocent people currently locked up in TDCJ. Having examined every published estimate I'm aware of and their methodologies, my own personal belief is that the number of people currently incarcerated in TDCJ who are innocent of the crime they were convicted of is likely 2,000 to 3,000 - maybe higher but probably not much lower.

When you think about it, that range makes sense. Convictions are secured when there is evidence "beyond a reasonable doubt," but not "beyond any doubt." So we assume the system convicts in cases with some doubt, meaning that going in we know sometimes the system gets it wrong. Let's say for the sake of argument that "beyond a reasonable doubt" means a fact finder is 98% sure they're convicting a guilty person. Whenever that standard is applied, 98 times out of 100 the right person goes to prison and twice they convict the wrong guy. But there are a lot of people in Texas prisons, 154,000 or so, which means applying that proportion to the whole system would put the number of incarcerated innocents just over 3,000. If you think "beyond a reasonable doubt translates into 99%, figure perhaps 1,500 or so innocent people are in prison. Either way, Texas' 42 DNA exonerations so far represent just the tip of an iceberg.

Fabelo predicts prison bed shortages without policy changes
At the beginning of the hearing, the committee heard testimony from Dr. Tony Fabelo who described prison population trends and projected the number of increased prisoners faced by the state if diversion programs are cut as anticipated in the House and Senate budgets. By 2013, he said, under the House budget as filed the state would be 12,857 beds short; under SB 1 on the Senate side, TDCJ would be 9,634 beds short. In other words, under either budget Texas will need to build or lease several new prisons over the next two years. (Watch the testimony here, Fabelo is the first speaker; here's a link to his power point presentation.)

Boiling it down, Fabelo said Texas has only two real options: "Reduce the size of cuts" or "Change policies to reduce demand for prison space." That's it, pick your poison. Building or leasing new prisons isn't an option if you want to cut $786 million from TDCJ's requested biennial budget, as Gov. Perry and the House have proposed. The only remaining path is to "Change policies to reduce demand for prison space."

He offered three policy suggestions at the end of his presentation to reduce prison numbers. Option 1 involves implementation of LBB Performance Report recommendations, including a "supervised reentry program for those presently released after completing their [full] sentences." In other words, when someone has completed, say, 90% of their time or have one year to go, they'd be paroled so that their initial reentry period would occur under supervision. That would reduce the prison population by 1,800 over the next two years and up to 9,000 inmates over the next five, says Fabelo.

The second option suggested: "Reintroduce SB 1909 from 2007 and passed by the Senate that requires mandatory probation and treatment for low level drug possession." The five year fiscal benefit to the state would total $500 million. Sen. Rodney Ellis is carrying the bill again this session.

A third option might free up 6,000 additional beds, said Fabelo: Shock probation on technical revocations, capping the time in prison for property and drug offenders revoked on technical violations to no longer than 12 months. That idea is not dissimilar from Grits recent suggestion that judges be required to more frequently utilize Intermediate Sanctions Facilities for technical violators. However you decide to skin that cat, the state can find large cost reductions by changing how they deal with that category of offenders.

Fabelo also suggested taking measures to edge up parole rates toward levels anticipated by the Board of Pardons and Paroles' official guidelines, noting that a 1% increase in the parole rate would reduce the inmate population by 1,500, while a 1% decline would increase the prison population by a like amount. Presently the overall parole approval rate is just above 30%; increasing it to 35%, which is the rate suggested by the board's official guidelines, would free up thousands of beds. There are seven risk categories to which potential parolees are assigned, and it's the lowest risk parolees who are being released at below-guideline rates, i.e., the lowest-risk offenders are the ones the parole board is holding onto longer than the guidelines suggest, a situation that's been happening as long as I've paid attention to the Board of Pardons and Paroles. The BPP could singlehandedly solve this conundrum of their own accord, but they appear to have little interest in doing so unless somebody (read: the Legislature) makes them.

These aren't the only policy changes that would do the trick; indeed, they're really rather modest proposals compared to more aggressive approaches advocated on this blog, and IMO the Lege should go farther. But at least we've finally reached a point - for the first time, to my knowledge, publicly - where legislators have acknowledged that policy changes affecting sentence length must accompany cuts at TDCJ. That's been obvious for more than a year, but at least they're beginning the conversation in earnest now.