Bail and Public Safety in Santa Clara County

As I have been developing the reading assignments for class (and doing my own research), I have been amazed that critics of the bail bond industry have not put front and center the argument that bail bonds do not promote public safety. Instead, they seem to take for granted that the argument should only be about which form of pretrial release is best at preventing a defendant’s failure to appear (FTA) in court.

I can understand why proponents of commercial surety bail—that is, bail bonds, as opposed to other forms of pretrial release that, technically, are also “bail”— focus primarily on FTA. Bail bonds seem to do comparatively well at that. Empirical work by Thomas Cohen and Michael Block focus on FTA (and long-term fugitive status) and show that bail does well at preventing FTA compared to own-recognizance release (OR) and supervised OR. The main study that includes pretrial rearrest rates as an outcome, however, concludes that rearrest rates by release type “vary little.” (pdf, FN 58)

Yet the most comprehensive critique of these empirical studies (and others) makes no argument about FTA as the predominant metric, choosing instead to criticize studies’ methodologies and conclusions. The public undoubtedly cares about not just whether defendants appear in court, but what they do while they’re out. If we are concerned about public safety, then we must discuss what effect bail bonds have on public safety, measured most easily by pretrial arrest. In Santa Clara County, defendants released on commercial surety bail are roughly ten times more likely to be arrested than those released on OR or supervised OR.

A few words initially about the way commercial bail bonds work, as opposed to an individual paying the full amount of bail in cash. This is all based on California, but other than the states that have outlawed commercial bail (Wisconsin, Oregon, Illinois, and Kentucky) it’s a decent approximation of what happens elsewhere. Let’s say bail is set at $10,000, and our defendant—call him Johnny Cash–is going to pay the bail money himself. He has to put up the full ten grand in cash (or property equivalent) and deposit that with the court. If he fails to appear, he loses all of that money immediately. If he appears, it takes him some time to get that money back from the court.

Let’s now look at another defendant, call him—of course—James Bond. He finds a bail bond company that is willing to post a bond for his release. This doesn’t require any money from the bail bond company to be deposited with the court—they just have to write an instrument promising to pay if James doesn’t appear, and to have that amount in the bank. They don’t actually pay the court any money. Instead, they get it—from James. He has to pay the bail bond company a non-refundable ten percent deposit. Once he pays it, he’s never getting it back, whether he appears in court or not. If James doesn’t appear, the bond company gets an extension to look for him, but as long as he eventually appears, bail is exonerated and the bail bond company owes nothing to the court. To summarize: with cash bond, individuals pay the full amount to the court, lose all of it the moment they fail to appear, and have to do something to get their money back even if they’re entitled to it. With bail bonds, the bail bond company gets ten percent immediately, doesn’t have to pay anything to the court, and gets an extension if James Bond fails to appear.

You can see initially that the commonly held notion that commercial bail gives the defendant “some skin in the game” is actually not true. The defendant’s deposit, once paid, is gone. It is true that sometimes a defendant’s relatives will pledge property that costs the full amount ($10,000, in our hypothetical), and that this can be collected on—but it’s also true that sometimes bail bond companies cut installment plan deals where a defendant can pay much less than ten percent to get out. The only real skin in the game comes with cash bail—but most defendants don’t have ten thousand dollars (or whatever amount) lying around. Officials in Santa Clara have told us that very few people pay cash bail.

Bail bond companies have no incentive to keep people out of trouble while they’re out in the community. As long as James Bond shows up to court, the bail bond company won’t have to pay the full amount. It does not forfeit the bail amount if he gets arrested. (As future posts will show, of course, actually collecting the full money bail amount from a bail bond company—a process called bail forfeiture—is extremely complex, while collecting from an individual is easy—the court just keeps the money.)

From the bail bond companies’ perspective, it might, ironically, be preferable for a defendant to get arrested for a new offense. Going back to our hypo, James Bond got released on charge one because the bail bond company pledged that he would appear in court to face charge one. If, a week later, James gets arrested and brought to jail on a second charge, the chance that he is not going to appear in court on the first charge is zero. He is under lock and key in the county jail, so he will certainly make his court appearance on the first charge. This removes any risk that the bail bond company will have to pay the full $10,000 bail on the first offense.

So what do the numbers say? As I mentioned, there aren’t many studies that actually look into this. In addition to the study referenced earlier, there are some federal data on the subject (pdf, go to page 9). According to these numbers, OR has a 17 percent rearrest rate; commercial surety bail has a 16 percent rearrest rate. But this encompasses a variety of release screens and a variety of pretrial services—or perhaps no pretrial services at all.

In Santa Clara County, however, defendants are assessed with a risk assessment tool, and some are given additional conditions designed to keep them out of trouble. On a public safety basis (measured, again, by pretrial arrest rates), commercial surety bail loses bigtime. Defendants released through pretrial services on their own recognizance (or supervised OR) get a new arrest 1.7 percent of the time (or 3.6 percent of the time) respectively. The number for bail is 29.4%. Let me repeat: defendants released on commercial surety bail are roughly ten times more likely to be arrested while on release than those on OR or supervised OR. The FTA rates are comparable; the public safety figures (using new arrests) are a slam dunk against commercial bail. Lest one think this is simply cherry-picking on behalf of pretrial services, the total release figures are about 1260 for recognizance of both types and 269 for money bail.

Looking at the federal study, then, even if you just let people go, the rearrest rates are similar. If you’re judicious about who you let go—even if you let go of a lot of people—you can do much better than commercial bail to improve public safety.

It’s clear that the discussion needs to include public safety, not just FTA. I certainly think that most citizens care what the accused are doing when they’re out, not whether they show up for court (and, moreover, some folks don’t show up for court not because they’ve absconded, but because they forget, they have transportation issues, etc.). So why don’t we talk about it?

One reason might be that the United States Supreme Court cases on bail—at least bail as it is mentioned in the Eighth Amendment of the Constitution—frame it only in terms of FTA. (See, e.g., Stack v. Boyle: bail set at an amount higher than that “reasonably calculated to fulfill this purpose [appearance in court] is ‘excessive’ under the Eighth Amendment.”) I should note, however, that Stack v. Boyle predates the 1984 Federal Bail Reform Act, which, inter alia, authorized pretrial detention if no release conditions would reasonably assure “the safety of any other person and the community.” The part of the statute authorizing pretrial detention on the basis of public safety was upheld in United States v. Salerno.

In California, though, the answer is much easier: Penal Code (s) 1270 explicitly says that public safety is not merely a consideration, it “shall be the primary consideration” when setting bail (emphasis added). A state court Judge (Curtis Karnow) has written an accessible, persuasive argument about why this is practically impossible, but it at least suggests that, when California is considering how to release people, it think about more than just FTA. Considering the public safety effects of the commercial bail industry is not just good policy for California—it’s the law.



3 thoughts on “Bail and Public Safety in Santa Clara County

  1. Has any research been done on the practical implementation of California’s public safety first priority? Anecdotal evidence from the few counties I’ve appeared in would suggest that, at least in misdemeanor cases, public safety is not considered at all. Rather, the sole considerations appear to be 1) the defendants prior FTA history, and 2) The defendants probation or parole status neither of which take into consideration if the defendant is likely to commit a crime again.


    1. Thanks for the insight, Reed. The best thing we’ve read about it thusfar is Judge Curtis Karnow’s article “Setting Bail for Public Safety.” That’s the theoretical argument. We’re going to do some court-watching to see what happens in practice.


Comments are closed.