By David M. Greenwald
Woodland, CA – Every month the Yolo County DA’s Office has put out a press release that makes the claim, something along the lines of: “Over 628 New Crimes Committed by Individuals Released on $0 Bail.” After they put that release out on February 8, 2021, the Vanguard filed a records request from the county to get their underlying data to make those claims.
It took a few weeks, but we received that data in late February and it took us about six weeks to reconcile two different spreadsheets. What we found after careful scrutiny is that the base claims by the DA’s office are largely correct. We were never able to get their exact numbers, but what we did get we believe is close enough to the reported number as to render their numbers plausible.
The DA reports: “Since April 13, 2020, 464 individuals have been arrested and released on $0 bail a total of 583 times, with some benefitting on multiple occasions.” They found about 40 percent of those released on zero bail would reoffend at least once. That’s about what we found.
In short, their numbers themselves are accurate. Where I have problems with their analysis is what they tacitly infer from that data. It is interesting that this week the DA’s office in Yolo County has made a big deal out of their transparency portal—but to us, what it appears is they really need a good data analyst because, while their raw numbers are fine, they fail to do very basic things to be able to properly analyze that data.
Claim: “The individuals who have reoffended since their release on $0 bail in Yolo County have committed over 628 new crimes in Yolo County alone. Those new crimes include 256 felonies and 372 misdemeanors…”
There is a bit of a definitional issue here. The DA appears to be calling each new charge a new crime. You can make that case somewhat, but it does overstate the impact of this policy, because in a lot of cases each discrete act can have a stack of multiple charges. We found that roughly 50 discrete acts (an individual on a single day) had at least four charges, 100 had had three charges and 200 had at least two charges.
Overall 509 people were released under zero bail in the year that we looked at the policy. About 197 of those people committed some sort of crime. The recidivism rate here was 37 percent.
There were more serious crimes. The DA’s press release notes: “[C]rimes such as Attempted Murder (2 new victims), Assault with a Deadly Weapon or Assault by Means of Force Likely to Product Great Bodily Injury (18 new victims), Robbery (10 new victims), Burglary (39 new victims), felony domestic violence (12 new victims), possession of an assault weapon…”
We calculate about 202 of these were “serious” felonies—the charges listed above.
But 60 percent of them were not. Some of these weren’t even necessarily new crimes, they were technical probation violations. The biggest single category is possession of a controlled substance—so they were arrested on drug charges, released, and then they got picked up again on drug charges. The second biggest single category was violation of probation or parole.
The DA of course emphasized in his press releases serious crimes, but as the numbers show those are the exception not the rule.
None of this is particularly surprising or even alarming. We can quibble over the emphasis or over the use of the term “new crime” but we kind of expect this DA to play up these sorts of issues.
The problem in the analysis is the next wave of issues.
Just to point out this one quickly: “Of those released on $0 bail, 40.5% have reoffended at least one time, the highest percentage thus far during the $0 bail period.”
That’s true by definition, almost. As time goes on, more of your old cases will have people who reoffend, so you would kind of expect that number to rise over time.
The first major problem is that they never put that recidivism into any sort of historical perspective. So they can point out that the rate is 40.5 percent, or we can find 37 percent, but without some sort of a baseline this means very little.
PPIC (Public Policy Institute of California) in 2019 for example found that the recidivism rate—measured by rearrest within two years—was around 66 percent. But given that most of our folks are six months to one year, it’s hard to gauge. Nevertheless, we would expect nothing particularly unusual with around 40 percent of those arrested to be rearrested.
In other words, what happened over the first year of zero bail is about what we would expect to happen.
By not having a baseline, they portray this as though the occurrence were unusual or alarming rather than predictable and expected.
Second, and probably a bigger problem is there is the implicit assumption that we are simply generating new crimes that would not have existed before, because the person would be in custody.
But there are at least two very serious data analytical flaws with that analysis.
First, they never attempt to assess what percentage of these people would be released anyway. That means anyone with a misdemeanor charge is automatically released and anyone with a non-serious felony is now getting released.
Guess what—many of those people are not going to be held in custody even during non-COVID times. I have no idea what percentage, but the fact that the DA doesn’t baseline that in his analysis is another serious flaw.
We didn’t do a deep dive here, but we noticed some people were arrested, released, and then rearrested—but it was seven months later. We suspect, under normal times, those people would probably have had their case adjudicated and they would have been out by that point anyway.
Others we noticed were picked up, released and almost immediately reoffended. Some of those were drugs and some were failures to check in with probation.
But, again, the analysis by the DA’s office never attempts to reconcile that, so we have no way of knowing how many people were actually not going to be in a position to commit a new crime and only had the opportunity to do so because of the zero bail law.
Finally, the DA ignores something else that is also very important. The purpose of zero bail was to prevent the transmission of COVID in jail to incarcerated populations as well as staff. And for almost a year, while places like CDCR and Santa Rita Jail have been ravaged by COVID, Yolo County has had very few cases and no inmate deaths due to COVID.
But the DA’s office has never once pointed this out in their press release.
There is a second bottom line here—is the crime rate going up due to these policies?
We ironically pulled this data off the transparency portal. But again, the finding is that the overall trend is fairly flat. This is the number of felonies referred to the DA’s office by outside agencies, which has the benefit of being independent of charging policies by the DA.
And you can see that over the COVID period it fluctuates greatly from month to month, but the trend is flat. There is therefore no evidence that these releases are contributing to a rising crime trend in Yolo County.
On April 2, the DA reported the number was now 831 “new crimes” and a 42 percent recidivism rate.
At the end of their release, they write: “The continued use of the Emergency Bail Schedule and the creation of annual bail schedules are determined by the Yolo County Superior Court. Anyone with concerns about the ongoing use of the Emergency Bail Schedule or the Proposed bail schedule should contact the Yolo County Superior Court.”
We have pointed out previously that the DA had to consent to this policy and has never formally requested that Yolo County cease zero bail. Our analysis suggests that zero bail is functioning about as you would expect, and does not appear to be contributing to a rising crime rate.
—David M. Greenwald reporting
Thank you to Morgan Poindexter, Vanguard Board Member and Hongyi Wen and Feuerstein, Vanguard Interns for helping to compile and analyze the data.
To sign up for our new newsletter – Everyday Injustice – https://tinyurl.com/yyultcf9
Support our work – to become a sustaining at $5 – $10- $25 per month hit the link: