Getting Answers: What's Missing From California's Inmate Recidivism Report
SACRAMENTO — The state's prison system is run by the California Department of Corrections and Rehabilitation (CDCR), but it turns out that only about half of the state's inmates are actually participating in rehabilitation and other programs in prison. That's according to a report that CDCR recently released following repeated records requests from CBS Sacramento.
The report also finds that those who do participate in in-prison rehabilitation are only 1.6% less likely to be convicted of new crimes after release. Critics say the findings call into question the effectiveness of the programs, which cost taxpayers hundreds of millions of dollars annually (in addition to the cost of the new crimes). However, the report fails to answer basic questions and is based on data that is more than five years old.
Follow Our Continuing Coverage: Investigating Rehabilitation & Recidivism Data
The Questions
For more than a year, CBS Sacramento has been working to answer the question: "Are prison reform laws in California leading to more rehabilitation and fewer felons reoffending after release?" Unfortunately, we still don't know because, as we've learned, the state isn't analyzing, or won't release, crucial data.
After years of delay, the Department of Corrections finally released its 2017-18 recidivism report, which leaves more questions than answers. In addition to being dated, the report lacks critical information to effectively analyze the impact of the CDCR's rehabilitation programs. While the agency has now commissioned a third-party nonprofit to evaluate its rehabilitation programming, we've learned that the CDCR is providing those contracted researchers with outdated data as well.
CBS Sacramento has conducted dozens of interviews with former inmates, crime victims, prosecutors, public defenders, lawmakers, prison reform advocates, victim advocates, criminal data researchers, and CDCR insiders. Many, if not all of them, would also like the data and the answers we've been working to get.
We're now bringing our unanswered questions to lawmakers and asking: Who can force the CDCR to release the crucial public safety information that is needed to evaluate California's landmark prison reform laws?
Please click here to add your questions to the list.
Please click here to read the growing list of viewer questions at the bottom of the page.
What we know
- The CDCR's most recent data reveals that roughly half of the inmates who were released during the 2017-18 fiscal year participated in rehabilitation and other in-prison programming before release.
- Those who did earn programming credits were 1.6% less likely to be convicted of a new crime within three years of release.
- The data is based on inmates who were released between June 2017-18, which were the first cohort to receive enhanced early release credits under Prop 57.
- Recidivism has largely held steady since 2012, hovering between 44.6%-46.1%.
- The CDCR's recidivism reports analyze three-year recidivism rates, so there is naturally a three-year delay in recidivism data.
- The CDCR does have access to detailed current programming data. They have yet to share it with CBS Sacramento or the CDCR's own contracted researchers.
What we don't know
- Who participated in the 2017-18 cohort programs and who did not (i.e. inmate demographics, sentence types, sentence lengths, prison yards, etc.)?
- Which programs did each inmate participate in to earn credits? (Did inmates participate in programs without earning credits? Why? Is there a rubric to illustrate the number of credits, and how inmates earn them, for each program or group of programs?)
- Why did so many inmates fail to participate (personal choice, lack of programs, long waiting lists, shortened prison terms, etc.)?
- How many inmates are currently participating in programs that offer Milestone Completion Credit (MCC), Rehabilitative Achievement Credit (RAC), or Educational Merit Credits (EMC) (i.e. inmate demographics, sentence types, sentence lengths, prison yards, etc.)?
- Who has access to these programs, and who does not? (How many spots are available at each prison? How are inmates selected for each program? Who is being excluded and why?)
- How do inmates specifically earn credits for each individual course offered? (Do they get credits for simply signing up? Must they complete the course? Must they demonstrate a skill or pass a test?)
- What is the average completion rate for the programs? (Does completing a program impact recidivism more so than just temporarily participating? Does that differ between programs?)
- What percentage of early release credits are inmates earning through programming vs. what percentage of early release credits are automatic Good Conduct Credits (GCC)? (Nearly all inmates automatically receive GCC, which shortens violent sentences by 33.3% and nonviolent sentences by 50-66.6% the day they arrive in prison.)
- Are the increased GCC under Prop 57 leading to shorter sentences and, as a result, less opportunity to participate in in-prison rehabilitation? (How long is the current waiting list for each program? What is the average length of the wait? What percentage of inmates on waiting lists are released from prison based on GCC before a spot opens up? For people released early on the basis of GCC, how early [months/days] were they released?)
** NOTE: We will be updating this list with additional questions from viewers, researchers, and other interested parties.
Please click here to add your questions to the list.
Please click here to read the growing list of viewer questions at the bottom of the page.
The Backstory
Tens of thousands of California felons have been re-arrested in recent years, after serving a fraction of their sentence under a series of prison reform laws that were intended to reduce California's prison population (AB 109 Realignment, Prop 36, Prop 47, Prop 57).
The most recent law, voter-approved Proposition 57, was passed in November 2016. It was, in part, supposed to reduce the prison population and incentivize rehabilitation by offering early release credits in exchange for program participation. Supporters said Prop 57 would ultimately reduce recidivism (the number of felons who commit new crimes after release) by encouraging inmates to take rehabilitation and education courses in exchange for early release credits.
Inmates were already eligible to earn credits for good behavior and rehabilitative and educational achievements. Prop 57 increased the number of credits that they could earn through the Good Conduct Credit and Milestone Completion Credit programs; it also added two new types of credits: Rehabilitative Achievement Credit and Educational Merit Credits.
The 2017-18 release cohort was the first full cohort of inmates who were eligible for the enhanced credits, which were implemented in May 2017.
ALSO WATCH: "Secret" Prop. 57 prison credits: Are most felons really "earning" early release?
Is Prop 57 Working?
After a half-dozen years and hundreds of millions of tax dollars annually, the state still can't say if Prop 57, or more specifically, the CDCR's implementation of Prop 57, is working the way that voters intended.
The most recent data is based on inmates who were released between July 1, 2017, and June 30, 2018.
The data show recidivism largely held steady between 2012 and 2018 and it reveals that only half of the released inmates had participated in any in-prison programming. Those who did were only 1.6% less likely to be convicted of a new crime within three years of release.
Critics note that the data is dated and incomplete, so it's not clear if the CDCR's programs contributed to the slightly lower recidivism rate among that group. CDCR's public data does not include any demographics to indicate which courses and which inmates have the best outcomes or why 50% of inmates did not participate.
And a lot has changed since 2018.
ALSO WATCH: CDCR's California inmate recidivism report leaves basic questions unanswered.
Turning to lawmakers for help
"When it comes to accountability, any department in the State of California should be able to give you data immediately, said Assembly Public Safety Chair, Reggie Jones-Sawyer (D-Los Angeles). But he acknowledges, that's not the way California's state agencies work.
Jones-Sawyer has chaired public safety committees for the better part of a decade and helped to shape CDCR's budget. He says lawmakers are also struggling to get the answers they need from CDCR to identify how prison reform laws are working and what needs to be improved.
"I need that data. The rest of the legislature needs that data," Jones-Sawyer said.
In a recent interview with CBS Sacramento, Jones-Sawyer said he would be submitting a State Audit Request in an effort to get that data. He agreed to include our unanswered questions in that request, along with relevant questions from our viewers.
The state auditor is an independent government watchdog that lawmakers can commission to get answers about state agencies and programs when the agency won't provide answers on its own.
Jones-Sawyer dodged questions about why lawmakers don't just ask the governor to demand answers from CDCR. The governor oversees the agency and appoints the CDCR Secretary. He is on his third CDCR Secretary in three years after the previous two Secretaries abruptly retired.
The agency has denied our repeated requests for an interview with the current Secretary, and the Secretary before him.
Lawmakers and journalists aren't the only ones struggling to get answers and comprehensive data from the CDCR.
ALSO WATCH: Public Safety Chair to Include CBS Sacramento's Questions in CDCR Audit Request
Outdated information
"We don't know if (CDCR's programs) are working," said Heather Harris, a principal investigator for the Public Policy Institute of California (PPIC). The CDCR has hired the PPIC to evaluate how its programs are working.
Following a scathing state audit in 2018, lawmakers passed a bill requiring the CDCR to track the effectiveness of its rehabilitation programs. Gov. Gavin Newsom vetoed the bill in 2019, stating:
"This bill would require the California Department of Corrections and Rehabilitation to contract with a researcher to conduct a recidivism analysis of the effectiveness of rehabilitation programs and to submit a report to the Legislature.
The goal of this bill can be accomplished administratively."
Now, years later, the CDCR has contracted with an outside researcher anyway.
The PPIC is in the early stages of a multi-year study and they have been given unprecedented access to CDCR data. Harris is the principal investigator on the project and has been working for years to get all the data she needs from the CDCR.
"There are just too many unanswered questions to really say whether the incentive structure is effective and also whether the programs themselves are effective," Harris said.
Unfortunately, Harris said, the data they're getting is also dated and stops at inmates who were released in 2019. By the time the study is complete, the information could be nearly a decade old.
Since 2019, tens of thousands of felons have been released early with credits and then rearrested for violent crimes.
Harris says she has requested more current data from 2022-23 but, so far, CDCR has not agreed to provide it. While the PPIC wouldn't be able to analyze three-year recidivism rates from the 2022-23 programming data yet, there is a lot they could learn about the evolution and effectiveness of CDCR programming.
Based on CDCR's public data, it's not clear which policies are effective, which courses and which inmates have the best outcomes, or why 50% of inmates are not participating. Those are all questions Harris hopes to answer.
Long Waiting Lists
Nearly all of the former inmates we spoke with complained about long waiting lists to get into rehabilitation programs.
"It's not that they don't want to (participate in the programs)," explained Jose, one recently released inmate we interviewed. "There's waiting lists. It might take a year."
We've requested waiting list data from the CDCR and are still waiting on their response. The CDCR website indicates that inmates do get some early release credit just for signing up for programs and sitting on the waitlist.
However, inmates tell us that sentences are often cut so short with standard good conduct credit, that they are released before they ever get off the waiting list, let alone complete the programs.
"The truth is, about two-thirds of people only serve prison sentences that are less than two years," Harris said. "About a quarter of people are in for less than one year."
She clarified that those are not the inmates' full sentences. Most inmates serve a fraction of their full sentence.
With or without programming credits, nearly all nonviolent offenders get their sentence cut in half with automatic Good Conduct Credits (GCC) the day they walk in the gates. Most violent offenders get their sentence cut by a third.
Groups like the prison guards say the automatic Good Conduct Credits incentivize good behavior.
"What the credits are intended to do, is incentivize program participation. And we don't know whether they're doing that because we don't know why people are enrolling in programs, why they aren't, whether or not they were more likely to enroll in programs before there were credits," Harris said.
"It would be to everyone's benefit to understand how credits can be earned and then whether they actually are being earned."
Fighting for answers
For more than a year, CBS Sacramento has been investigating violent crimes that were allegedly committed by felons who were previously released early from prison with credits.
That includes:
- The man who killed Kate Tibbits and her dogs
- The man accused of dismembering an elderly Sacramento woman in her home
- The man accused of killing Selma Police Officer Gonzalo Carrasco Jr.
- And the Sacramento mass shooting suspect who was denied parole due to his risk of violence and criminal activity just months before he was released early anyway with good conduct credits. He would have been in jail the night of the mass shooting if he had served even half his previous sentence for domestic violence.
Based on the data the CDCR is willing to release, we may never know if those felons received any effective rehabilitation at all.
ALSO READ: Where's California's recidivism data?
The CDCR continues to refuse our requests for access to the CDCR secretary and the people running the in-prison reform programs. We recently asked the Governor's office to step in and help facilitate an interview. His office has largely ignored that request as well.
The CDCR has responded, in writing, to many of our requests for information and public records. However, in most cases, they tell us that the data we are requesting is not currently available or they provide piecemeal answers and carefully crafted statements that only partially answer our questions. Those statements, drafted by unnamed agency officials, often result in convoluted multi-part email chains, with back-and-forth questions, that never fully provide meaningful answers for our viewers.
In at least two instances, the CDCR has provided false information. In both cases, they later acknowledged the inaccurate information and apologized.
We will now be forwarding our questions to lawmakers in hopes that they are included in an upcoming state audit of CDCR's programs. Notably, the new Chief of Communications for the California State Auditor was previously the Governor-appointed CDCR Press Secretary.
We will continue updating our list of questions for CDCR and we encourage former inmates, families, crime victims, prosecutors, public defenders, lawmakers, prison reform advocates, victim advocates, criminal data researchers, and anyone else with questions for the CDCR to add your questions to the list.
Viewer Questions Continued:
- How many sex offender treatment programs are there specifically for sex offender treatment (not a related course)? (Is there one in every prison? Is there capacity for every incarcerated sex offender?)
- How many paroled sex offenders actually went through the sex offender treatment program before being released?
- The CDCR published two reports on recidivism at the same time, one for inmates released in FY 2016-2017 and one for inmates released in FY 2017-2018. Why was the earlier-period report not published the year before?
- If a crime is committed within the 3-year follow-up period but the conviction is later, that still counts as recidivism. The follow-up period for FY 2017-18 releases ended June 30, 2021, and they published the report less than 2 years later. When (what cutoff date) do they actually stop looking for new convictions?
- According to the report, inmates earning "enhanced credit" were less likely to re-offend by a small margin (1.6 percentage point difference) -- but how does Secretary Macomber know whether this was due to the effectiveness of the programs? We don't know much about who was participating in these programs, and there were likely various differences between the groups even before the programs started. For example, those who chose to participate may have been more motivated to change than those who did not. How much of the recidivism decrease could be due to factors such as these?
- Does CDCR have rubrics that illustrate the number of credits and how inmates earn them for each program or group of programs?
- What statutory or regulatory scheme is being used by CDCR to award conduct credits to violent and serious offenders?
- What specific criteria are used in assessing whether to award conduct credits to an inmate?
- Does an inmate have to proactively do anything to earn conduct credits, or are credits simply awarded based on time being served?
- What steps, if any, does CDCR take to notify victims or next of kin when an offender is set to be released from custody? (Specifically, were those steps followed prior to the release of Smiley Martin, based on his Sacramento conviction in docket 17FE008296? Can the DA get a copy of that notification?)
- How many inmates had their sentences reduced and were released due to CDCR's emergency measures? How many of those inmates have been subsequently arrested for a new offense? (NOTE: On Friday, April 20, 2021, "emergency" regulations were passed allowing CDCR to award additional conduct credits to inmates.)
- In the past 3 years, what percentage of released inmates had no current address upon release? How many were released on a transient status?
- Considering the recent increase in unhoused individuals in California, does the state have an accounting of the number of former inmates who end up unhoused (homeless) within three years of release?
- What is the efficacy of CDCR programming vs. community-based organization (CBO) programming? The research we do have on CBO programs shows that they are quite effective at ensuring that people who are released are better equipped to return to their communities, but CBOs face information access obstacles that hinder them from evaluating their own programs' efficacy in a more comprehensive manner.
- Will California really shift to a Norway model that outsources in-prison programming to CBOs? The Governor recently announced plans to launch a new "California Model" of incarceration, created in the image of the Norway model. Does that mean California will fundamentally shift toward offering programming run exclusively by CBOs to lower recidivism rates?
UPDATE:
In response to this reporting CDCR provided the following statement: "CDCR is committed to providing all Californians with accurate recidivism data that is complete and has been thoroughly vetted and analyzed. Our 2017-18 report was published in a timely manner, following the national standard of reporting arrests, convictions and returns to prison on a three-year cycle."