People on parole or probation do not account for most new arrests in four cities in California, according to a new report. Instead, it is people who are not on supervision who are arrested most frequently.

“Most people would have stated before this report came out that probationers and parolees were the primary source of crime in their communities,” said Terri McDonald, undersecretary of operations for the California Department of Corrections and Rehabilitation. “At least in these four cities, this does not seem to be the case.”

The report showed that people on parole or probation commit only one in six violent crimes and one in five crimes overall. In comparison, more people on supervision committed drug crimes, but they were still not close to a majority of the offenders. In that category, people on probation or parole account for one in three arrests.

The report analyzed arrests between and 2008 and 2011 in Los Angeles, Sacramento, San Francisco and Redlands. It was produced by the Council of State Government’s Justice Center, a national, non-profit, non-partisan research group, at the request of the police chiefs of each city.

Though the report did not examine a representative cross-section of California, it captures what is going on in three major metropolitan areas where a significant chunk of the state’s population lives, said Robert Coombs, communications director for the Council of State Governments.

The report’s analysis covers the period just before prison realignment, or AB 109, began in October 2011. But it does highlight some of the potential that reform has to prevent people on supervision from committing new crimes, as well as ways that reform may be falling short in effective crime prevention.

“The question is one of how to best prioritize resources,” Coombs said. “Who do you provide services and programs for?” Asking where services can be cut is another important part of prioritizing scare resources, he added.

The higher number of drug-related crimes among people on supervision, for instance, indicates a need for more rehabilitative services.

Limited resources could also be used more effectively with better risk assessments, especially as the number of people on probation grows, to help decide which people need the most supervision. Risk assessments use a series of questions about criminal history, attitudes, personality and life circumstances to determine whether or not someone is likely to commit another crime.

“It’s really important that we implement a system of triage,” said Wendy Still, San Francisco’s probation chief. “The success of realignment depends on the ability of probation departments to supervise a larger number of people successfully.”

Since prison reform began, people who commit certain non-serious felonies receive jail time or split sentencing, a combination of county jail and probation, instead of state prison and parole. People who were sentenced to prison for one of these crimes before reforms began are also released to the supervision of county probation agents, rather than state parole agents.

But while the state’s parole division generally did a good job of using risk assessments to determine who was likely to commit another crime, and allocating resources accordingly, only one probation department in the study – San Francisco – used a risk assessment that was good at predicting who would reoffend.

These assessments determine how much contact people under supervision have with their probation or parole officer.

Aiming more resources at people more likely to commit crimes and less at low-risk offenders offers a better return in terms of crime reduction, Coombs noted.

Changing policing and sentencing practices to encourage more supervision for people who have been assessed as high-risk and more rehabilitation overall may prove difficult, despite the encouragement provided by reform.

It may be hard, for instance, to shake the perception that all people on supervision are high-risk, focus groups with police officers suggested. In the focus groups, conducted as part of the report’s research, police officers described situations where they arrested the same people on probation or parole multiple times. Such experiences could contribute to the widespread misconception that people on supervision are the primary source of crime, the report said. Police in the focus group also held misconceptions about what determines risk level, and had difficulty accessing information about people on probation and parole beyond a name and address.

A recent analysis of split sentencing by the Chief Probation Officers of California also suggests that judges continue to sentence people to jail much more frequently than they use other options meant to encourage the rehabilitation necessary to prevent more crime.

Split sentencing, a combination of jail and probation, is a new option under the reform law. Split sentencing is meant to offer support as well as supervision to people leaving jail as they return to the community.

Only 5,000 felony offenders affected by AB 109 have received split sentences since realignment began, according to the analysis by the Chief Probation Officers of California, which was released in December. In contrast, more than 16,000 received straight jail time.

Straight jail time means no supervision – or support and rehabilitation – from probation after release from jail, which is a critical time of transition. “This is the period when recidivism is most likely,” the probation chiefs’ report notes, adding that “the research is clear – these offenders will have a higher likelihood of committing more crime than those who have been given a split sentence.”

Data from some counties for the first year of reform indicates that many people affected by realignment who reoffend or violate the terms of their probation are committing drug-related violations.

During the first ten months of realignment in Alameda County, for instance, drug crimes accounted for 22 percent of new crimes committed by people who were under the supervision of probation because of prison reform. The next most common reason for re-arrest among people on probation instead of parole because of reform – a type of probation that is called post-release community supervision – was auto theft. It followed distantly at 9 percent.

In Fresno County, between April and August of 2012, people on post-release community supervision were sanctioned most often for drug-related violations. And during the first year of realignment in San Francisco, drug crimes were the single largest category of new offense, accounting for 48 of the 126 new crimes committed by people on post-release community supervision.

The Council of State Government’s report emphasizes the importance of substance abuse treatment options, said McDonald of the CDCR. “You cannot use a jail bed for every problem that you have,” she said.

The report’s final recommendations included improved risk assessment, more rehabilitative options and improving data sharing between police departments and departments of probation and parole.