In August, Governor Jerry Brown signed legislation that abolishes cash bail in the state of California. The legislation contains several important features. First, it eliminates bail schedules, which tie bail amounts to the charged offense, and, as currently utilized in California, are likely unconstitutional. Second, it gets rid of the bail bonds industry, known for its corrupt and predatory practices. Third, it replaces a money based pretrial release system with a risk based system. However, the bill that was passed, SB 10, is not without controversy. Not surprisingly, it is opposed by the bail bond industry. More importantly, as its critics (including the ACLU and other advocates) have pointed out, the legislation is problematic in how it relies on risk assessment tools, and in the broad discretion it gives to judges to preventively detain defendants. As I will discuss below, these concerns are valid, and thus while many hope that the legislation is an important step toward a fairer system of pretrial justice in California, its success will depend on counties and local judges.

The decision as to whether to release an individual while they are awaiting trial, and if so, under what conditions, is one of the earliest and most important stages of the criminal process. At a defendant’s arraignment, or first appearance in court, judges typically make a series of decisions. First they must decide whether a defendant will be detained, because they pose either a risk to public safety or of failing to appear in court; this is known as preventive detention. Preventive detention in California is rare; just 5% of defendants were denied release between 2000 and 2009. More often, defendants are given a release option and judges must decide whether to a) release a defendant on their own recognizance, meaning that they simply promise to show up for their court hearings, or b) release them conditionally, typically by setting money bail, which they post as a guarantee that they will make their court appearances. If they are released on bail, a judge must then determine what amount of bail to set.

In making these decisions, current California law requires judges to consider a wide variety of factors, including characteristics of both the defendant and the offense. In addition, each county in California uses bail schedules, which consist of a list of offenses and a presumptive bail amount for each offense. Although bail schedules are supposed to be just one factor considered by judges, empirical research conducted in Southern California found that they were the main factor determining pretrial research decisions. Instead of the individualized determination that defendants are entitled to, must judges simply applied the bail schedule. This means that the charged offense was the most important factor in setting bail (despite the fact that this has no impact on either public safety or a defendant’s likelihood to appear for hearings), and a defendant’s ability to pay the bail amount set determines whether or not they are released.

Currently in California, most defendants are granted release, but most remain behind bars while they await trial, generally because they cannot afford to pay the money bail set in their case. Between 2000 and 2009, just 41% of defendants obtained pretrial release. The impact of this high rate of pretrial detention is huge. Not only are defendants deprived of their liberty (and subject to the pains of confinement in often dangerous jails) before being convicted of any offense, but they are more likely to plead guilty, more likely to be convicted, more likely to be sentenced to jail or prison, and more likely to receive a longer sentence than similarly situated defendants who are released pending trial. All of these consequences fall more heavily on the poor and racial and ethnic minorities.

The inequities of pretrial justice have led to reforms in numerous jurisdictions around the country over the last few years, including in Harris County, Texas, where in 2017, a federal judge found the county’s misdemeanor bail system unconstitutional, and in New Jersey, where a broad coalition of advocates and policymakers worked to pass state-wide bail reform legislation, which went into effect in 2017. Given that California has one of the highest rates of pretrial detention in the country, there is no doubt that reform was necessary. In its 2017 report to the Chief Justice, the Pretrial Detention Reform Workgroup, a group of state judges, concluded: “California’s current pretrial release and detention system unnecessarily compromises victim and public safety because it bases a person’s liberty on financial resources rather than the likelihood of future criminal behavior and exacerbates socioeconomic disparities and racial bias.” Legislation was first introduced in 2017, and initially was supported by a wide coalition of reformers. However, most of those groups pulled their support for the new law as a result of amendments that they argue give too much discretion to judges to detain defendants, as well its heavy reliance on actuarial risk assessment tools. Both of these criticisms have merit.

First, there is no doubt that there are potential problems with the use of pretrial risk assessment tools. These tools need to predict a) a defendant’s risk of flight from a jurisdiction, b) a defendant’s risk of failing to appear at a court hearing, and c) their risk of committing a new crime while awaiting trial.  Then, even if they are validated on the relevant population to predict all of these risks, they need to be both properly administered, and properly interpreted and used by pretrial decision-makers. Finally, and most importantly, it is important that risk assessment tools are not biased, in particular against non-White defendants. While the use of pretrial risk assessment tools is growing, there is little empirical research on their validity. For example, the most widely used tool is the Laura and John Arnold Foundation’s Public Safety Assessment. However, no peer-reviewed validation studies for the tool are publicly available, and although the studies that have been released show good predictive validity, one recent study (albeit not peer-reviewed) concluded that failure to appear scores were more accurate predictors of case outcomes for White defendants than Black defendants. Despite these concerns though, while many advocates oppose the use of risk assessments for making pretrial release decisions, because of their potential to increase existing racial disparities, there is no evidence that replacing often arbitrary judicial decision-making with actuarial risk assessment tools increase disparities.

Further, evidence from jurisdictions that rely on risk assessment tools shows that their use typically does reduce the number of people detained pretrial. For example, six months after it replaced cash bail with risk assessment, New Jersey had one-third fewer people in jail awaiting trial. While risk assessment is not the only alternative to money bail, it is the most prevalent one, and thus far, no jurisdiction has been willing to adopt widespread release of defendants on their own recognizance as some advocates have suggested. SB 10 probably gives too much discretion to counties, which will be able to choose their own tools, but the Judicial Council will maintain a list of approved tools, and a panel of experts will assess each county’s tool and guidelines.

More worrying is the presumption against release (prior to court hearing) for a large number of defendants. While SB 10 requires many people assessed as low risk to be released on their own recognizance prior to arraignment, it prevents the pretrial services agency from releasing individuals charged with certain offenses (both felonies and misdemeanors) or who meet certain criteria, even if those individuals are assessed to be low-risk.  The new law also allows courts to establish rules limiting the pre-arraignment release of medium-risk defendants. While most of these defendants may be released by a judge at arraignment, SB 10 gives judges a considerable amount of discretion to detain defendants, and indeed includes a rebuttable presumption that defendants who are charged with a violent felony (regardless of assessed risk) and who are assessed to be high risk and meet certain conditions, including having a recent conviction for a serious felony. Thus, even though the new law purports to move from a system of money bail to a risk-based system, in certain circumstances, it directs both pretrial services agencies and judges to ignore assessed risk. Further, judges tend to be risk averse, and thus concerns that judges will exercise their discretion to lock up more people prior to trial are well-founded.

Whether SB 10 will go into effect in October 2019 as intended is unclear. In addition to criticism from bail reform advocates, the law is being challenged by bail agents, a coalition of which is seeking to put the law on the November 2020 ballot. Further, while SB 10 may moot some pending constitutional challenges, both federal and state cases challenging existing law remain active. If it does go into effect, what impact it will have – whether it will make a difference to the thousands of people currently locked up awaiting trial simply because they are poor, or whether it will make an already bad system worse – will depend on the actions of counties and judges.